What Is Voluntariness?

Liberal states aim to enable and support their citizens in the choices they make and the decisions they take so that citizens can lead the kind of lives they hope to live. Indeed, many civil rights in liberal states are designed to enable just that. For instance, citizens should be free to worship as they see fit or not adhere to any religion; they should be able to choose their professions as befits their inclinations and talents; and they should be able to have families or choose to be child-free. Yet choices need to have a certain quality to serve as grounds for individual autonomy. In liberal theory, voluntariness is important since only a voluntary choice is one that is an expression of the cardinal virtue of personal autonomy.Footnote ¹⁰

Many definitions of what constitutes personal autonomy abound; for my purposes, the precise definition of the parameters of autonomy can be put aside. Here, I follow a widely cited account of personal autonomy originally proposed by Joseph Raz, which suggests that individuals should at least be “part authors of their own lives.”Footnote ¹¹ This is to say that persons should have some control over the shape their lives take. To be able to exercise such authorship, specific conditions need to be in place: autonomy depends on having access to a range of options, among which is the choice of the course we want to give our lives; of being reasonably free from coercion when choosing among viable options; and of being reasonably able to implement these choices. The crucial point of this definition for the discussion of climate-induced migration is that autonomy-supporting choices should be free from coercion. This suggests that such choices ought to be made voluntarily.

How should we think about voluntariness, though? For some, voluntariness is a statement about how we make certain decisions.Footnote ¹² Agents act voluntarily if they have deliberated on their options and have decided to adopt one option over another. A “voluntary choice,” then, is the result of deliberation and the active, intentional decision to choose one option over another. Others suggest that voluntary decisions are those that are made between acceptable alternatives.Footnote ¹³ What counts as acceptable can be evaluated based on an objective standard, such as the promotion of individual well-being, basic needs, or human flourishing. Both takes on voluntariness make clear that by declaring a choice voluntary one makes reference to the motivations based on which an agent makes the choice. In contrast, to speak of a “free choice” is to say something about the circumstances within which agents make a choice. Free choices “are claims about the options an individual faces” and should be distinguished from “claims of voluntariness, which are claims about how the nature of those options affect an individual's will.”Footnote ¹⁴ This is to say that freedom and voluntariness cannot be used synonymously. However, coercion always stands in the way of voluntariness. If there is coercion, a choice cannot be taken to be a voluntary one.

We can now return to the link between voluntary choices and autonomy. If it is the case that voluntariness describes the motivations to choose one option over another, then we can accept that to designate something as voluntary is an explanatory claim: it explains how a choice was made. Autonomy then depends on having access to viable options among which to choose voluntarily. Put differently, a choice does not count as voluntary if either: (a) it was coerced; or (b) it was made from an inadequate range of options; or (c) both. I will explain below that for cases of climate-induced migration, the lack of options may be most often what renders the decision to move nonvoluntary.

So far, then, I have argued that individual agents ought to have access to the conditions of autonomy. One important autonomy-enabling condition is that individuals can make voluntary choices, which is to say that they can choose among viable options to give their lives shape. I have proposed voluntariness as a statement about an agent's motivation to choose one option over another, and I have distinguished voluntariness from freedom as a statement about the circumstances in which specific choices are made. Now, I have not stipulated yet what we should mean by “a range of viable options”; that is, what kinds of options are needed to satisfy the demands of voluntariness. For instance, do limited options in the amount of, say, two allow for voluntary choice? This is an important question when thinking about climate-induced migration: if the choice is between moving or staying and not being able to lead a reasonable life, then I believe it fair to say that the decision to move is nonvoluntary for lack of a viable option for staying. To be certain that an option is chosen voluntarily, viable alternatives—designated as such by objective criteria such as commensurability and comparability—must be available. A person also needs to have adequate information to be able to assess the possible ramifications of their decision; what we can call “the criterion of well-informedness” that voluntary decision-making demands. A voluntary choice, then, is the result of a noncoerced, intentional, and informed decision to choose one option over another among a range of viable options.Footnote ¹⁵ How to define the relevant criteria is debatable; however, individual well-being, basic needs, or human flourishing are obvious candidates for defining how to assess the viability.Footnote ¹⁶

Voluntariness in Migration?

Voluntariness has been the subject of much debate in migration studies. I follow Valeria Ottonelli and Tiziana Torresi in assuming that most migration is based on individual projects that can only be realized through migration.Footnote ¹⁷ Assuming that migration is based on a migration project does not deny that the decision to migrate is taken under conditions of constraint: there may be socioeconomic reasons why people migrate, or, in the case of climate-induced migration, there may be climate reasons. Instead, what is relevant is that the decision is a voluntary one in the sense I described above. This is the case if the choice between migrating and not migrating is one between viable alternatives, as we just saw.

To help clarify when migration is voluntary, then, we need to develop criteria to identify when options for individuals are the kind of options needed to enable voluntary decision-making. As we have seen, the options must not involve coercion. Put otherwise, if a decision is the result of coercion, it is nonvoluntary. Equally, a decision between inadequate options should not count as voluntary. In this vein, we can speculate as to what extent climate change represents coercion: I believe it is fair to say that, for some, climate change restricts the available option-set so that at some point in the future, staying in the original territory will no longer be an option.Footnote ¹⁸ The particular harm that climate-induced migrants face, such as what the Carteret Islanders will experience, is that they must make the decision to move; the viable alternative of not moving is unavailable. I assume here that coercion can be used when describing changes in the context of action; that is, the motivation to act, in particular, changes the options based on which one can act.Footnote ¹⁹ The decision to move is thus not a voluntary one.

Now, we could say, of course, that the prospect of not being able to stay should not count as a problem in our analysis of the decision to move. After all, for some atolls, such as the Carteret Islands, the prognosis that residents would have to relocate by 2015Footnote ²⁰ has proven wrong. I contend, however, that knowing that, at some point in the future, people will have to move should count as “anticipatory displacement.”Footnote ²¹ The fact that we cannot envision ourselves residing in a place in the future affects how we think of the option to stay. I will discuss below how a lack of a secure projection into the future affects individual autonomy. What is important to note here is that an option that has a time limit built in cannot count as enabling voluntary decision-making. It hampers our investment in a place, in the relationships with those there, and in its history.

What can we draw from the discussion of voluntariness and the particular discussion of voluntariness in migration for the case of climate-induced migration? I suggest that there is a scale of scenarios that determine whether or not such migration can be considered to be voluntarily undertaken. Consider, first, the case of territories where crops that have provided for the community can no longer be grown, or where livestock can no longer be kept. For instance, think of Mauritanian herders who, for centuries, have moved their herds across the Sahel to find food and water. Due to droughts, they now have to travel ever further, with the animals suffering along the way and often dying, thus depriving the herders of their livelihood.Footnote ²² In an effort to allow for pastoralists, who produce 10 percent of GDP in Mauritania, to continue their ancient practice, several NGOs have started to implement safe corridors for herds to find food and water. Rather than crossing vast areas with their herds, pastoralists now can use these safe passages to transport their animals.Footnote ²³ In this case, while the traditional lifestyle must be adapted to the challenge of climate change, individual pastoralists clearly have a viable alternative to leaving the profession and territory behind. Instead, they can adapt. If they were nevertheless to move, I believe it plausible to say that the decision to do so would be a voluntary one even though it was climate induced. In these cases, relocation can actually count as a migration project, not as displacement.

Another parallel we can think of is adaptation to changed professional circumstances. For instance, we can imagine that in some settings adaptation to changed environmental givens is not possible, as it may ultimately become for Mauritanian pastoralists, and indeed many pastoralists around the world over time. However, it might be the case that governments provide professional alternatives and support for them to learn a different trade. If people were nevertheless to move, one could still argue that their decision was a voluntary one, since it would be difficult to defend a right to exercise one profession over another. I argued earlier that voluntariness in choice demands a range of viable and adequate options. If pastoralists could not work in this profession any longer, but could find work as gardeners or game keepers, one could argue that this is a viable and adequate alternative. In other words, we can imagine adaptation measures to climate change that while prompted by climate change should nevertheless count as voluntarily chosen. In these circumstances, the decision to move for climate-induced reasons has to count as voluntary. We should then speak of climate migration, rather than climate displacement.

On the extreme end of the scale are cases in which the territory will be submerged or become otherwise uninhabitable. It seems clear that there is then no viable alternative to moving, even if the necessity of doing so is still a distant prospect. These are the cases in which the decision to move is not a voluntary one. What does this tell us about the Carteret Island example? Can we accept that the circumstances under which the islanders took their decision was a voluntary one? As I acknowledged in my introduction to this example, there were several problems with the proposal to relocate, including a lack of ties to members of the new community where relocation should occur, which is often taken to be an important element for relocation success. What I want to discuss here, though, is independent from the circumstances of relocation and focuses on the context of the decision-making process in the community. I believe it fair to say that Tulele Peisa was aiming to provide the best-possible circumstances to make the decision to relocate a voluntary one: the organization looked for the best-possible relocation territory after deliberations about the needs of the community and its members, and, once found, provided as much information as possible about the potential new territory. The NGO did this in the most transparent manner possible. However, at some point relocation will have to happen because at some point in the future, the option of staying will no longer be available; in this case, then, the decision context is coercive. Hence, I suggest that the motivation of Carteret Islanders to think about relocation is not voluntary.

Relocation: Territory, Time, and Autonomy

So far, I have discussed the condition of voluntariness in autonomy. Applied to the case of climate-induced migration, I have proposed a distinction between climate-induced displacement and climate-induced movement that is motivated by viable alternatives to the traditional use of territory or adaptation by taking up different professions or changing the way a profession is exercised. In these latter cases, climate-induced migration is not displacement. I began my account with a different case, however—that of the Carteret Islanders. In that case, as in other atoll island states facing slow submersion, I suggested that climate-induced migration is not based on viable options to choose from. Instead, it is based on a decision to take on the only available option over time, which is to leave. This is a case of displacement. I now want to discuss whether or not relocation, including carefully planned relocation, can mitigate the harm of displacement.

Addressing the harm of climate displacement was the motivation of the authors of the Cancun Adaptation Framework, which includes specific recommendations for relocation as an adaptive measure in response to climate change.Footnote ²⁴ And indeed, several governments around the world have begun to relocate threatened population groups. Proactive governments in Vietnam, Papua New Guinea, and the Maldives, and even the Netherlands and the United States, have begun planning relocation measures that climate change has made necessary.Footnote ²⁵ Yet while relocation may address some individual and collective needs such as access to secure forms of livelihoods and employment, loss of territory constitutes a specific and distinct harm. To see this, a look at recent literature discussing the moral status of territorial occupancy, and, more relevantly to my purposes here, the harm that is inflicted on those who are driven from their homeland, is helpful.

Philosophers discussing the moral role of residency in a specific territory have argued that a right to residency should be understood as “a right of non-dispossession, a right to remain, at liberty, in one's own home and community and not to be removed from the place of one's projects, aims, and relationships.”Footnote ²⁶ According to at least one account, territorial occupancy is important since it allows individuals access to important means of developing their lives according to their own plans.Footnote ²⁷ More specifically, territorial occupancy is fundamental “for an individual's personal autonomy—his ability to form and pursue a conception of the good,”Footnote ²⁸ making it impossible “to move . . . without damage to nearly all of [one's] life plans.”Footnote ²⁹ Thus, the first harm that relocation causes is the severing of ties between the individual and the autonomy-grounding link to a territory.

Furthermore, there may be collective needs, such as the need to protect cultural goods, that may require access to traditional lands and subsistence practices. This is to say that some autonomy-enabling functions, like a sense of place, importantly depend on a community being in a specific place.Footnote ³⁰ Margaret Moore argues that “if we are to have any control over our lives, we have to have control over the most fundamental elements in the background conditions of our existence, and among these is the ability to stay in our communities.”Footnote ³¹ This last concern, in particular, illustrates that climate-induced migration is often a collective phenomenon that needs a collective solution. For some climate-displaced people, the concerns about climate change are not only about lost livelihoods and loss of territory but also about cultural survival.

Now, one could argue that relocation could be considered the best way to mitigate the harm of climate displacement. After all, as the Carteret Island example suggests, the concern with providing land for the community—and taking into consideration culturally and collectively specific territorial needs, such as the need to provide land to grow subsistence foods—was one of the motivations for choosing a specific territory. So, what is amiss?

In addition to the direct harms suffered by those experiencing climate impacts, relocation inflicts yet another harm. This is clarified when we assess what leaving a territory implies; one of the wrongs that removal causes is not only the severing of territorial relationships but also the severing of important historical relationships. Historical relationships to territory are often expressed in collective accounts of identity, such as the idea that a people have been part of the territory. As a member of the Kwaaymii council in California explained:

What is lost when cultural communities are relocated is an identificatory feature of the culture that adds to its autonomy-enabling function; namely, its historical grounding in the territory. To make sense of this, let us return to the definition of autonomy I provided at the outset. I stipulated a rather basic account, suggesting that autonomy depends on three elements: (1) having access to a range of options among which to choose the course we want to give our lives; (2) being able to choose among the options voluntarily; and (3) being reasonably able to implement these choices to make us part authors of our lives.

In my discussion so far, I have explained the first two elements, examining what kind of options are required for autonomous decision-making and the condition of voluntariness in decision-making. I will now turn to discuss the third element; that is, the need for an individual to be reasonably able to implement the choices concerning the shape of their life. This suggests that personal autonomy informs and motivates personal agency. Much has been written about the connection between autonomy and agency—such as how, precisely, they hang together;Footnote ³³ suffice it to say here that for the autonomous shaping of our own lives to be plausible, a minimal sphere for the implementation of our choices about our lives must be possible. In order to understand this minimal sphere, we must understand that there is a temporal aspect to implementing our choices: this involves the possibility to project our lives into the future, to plan for things we hope to happen in our lives.Footnote ³⁴ In other words, we need to be able to imagine a future for ourselves; if we cannot imagine a future, it is not clear how we can meaningfully make decisions about the course of our lives.Footnote ³⁵

At first, the idea of imagining a future does not seem to be tied directly to a territory. After all, one could say that one can imagine a future anywhere. I argue that this is not the case, however. Rather, being able to imagine a future fundamentally depends on having a sense of place. This warrants explanation.

Cara Nine has shown that having a secure home is important for the development and exercise of our deliberation and thinking about our sense of self:

In a similar vein, Avner de Shalit suggests that being able to have a sense of place is an important capability for individual functioning.Footnote ³⁷ What determines a sense of place is being able to engage in the relationships that define a place. At least some of our planning activities are importantly tied to a web of relationships with others. To use the philosopher Margaret Gilbert's words, we “walk together” intentionally to realize individual and communal goals.Footnote ³⁸

I suggest that territory is part of a web of relationships that ties us to the community in which we can develop a sense of self that projects into the future. This is a temporal as much as an interpersonal aspect of our lives. We project our role in the community, as well as what role we hope to play and what role and place we wish for our children and their descendants. Moreover, personal agency has a backward-looking aspect to it. We are not only projecting ourselves into relationships in the future; we are also defining ourselves with reference to the past. We draw on family history, which we accept, reject, or alter to accord with our own plans and wishes. And, importantly, we also draw on the historical practices of our community. This is precisely what “home” means for many and why being homeless is a very particular harm.Footnote ³⁹ I want to suggest that at least some of these practices are territorially defined. It is not clear that individuals can develop a sense of place in just any territory.

As an example, think of many culinary traditions that define cultures, and that depend on animals, fruits, or grains to be cultivated in a given territory. Or think of artistic traditions that are deeply embedded in a territorial context: J. M. W. Turner's seascapes or the gyáa'aang (totem tradition) of the Pacific Northwest, for example. None of these would have been possible without the territorially specific landscape. Artistic expression and production, in particular, are tied to territorial cultural history. So are definitions of national cultures. For instance, many Scandinavian countries have a sea-faring history that is tied to their territories and that provides a particular kind of territorial characterization of the people who live there.

Yet the impact of history on the territory goes further. Some ethical conceptions about how to be in the world also depend on the territorial context. To see this, take the example of Native American chief Plenty Coups, the leader of Crow Nation, as depicted in Jonathan Lear's Radical Hope. Plenty Coups is quoted as saying, “When the buffalo went away the hearts of my people fell to the ground, and they could not lift them up again. After this nothing happened.”Footnote ⁴⁰ Importantly, though, Lear argues that Plenty Coups had to imagine a future away from the traditional hunting grounds, including what should count as virtuous or ethical behavior in a radically changed environment. “What counted as courage in a world dominated by tribal warfare no longer made sense . . . in the post-conquest world of the reservation (where traditional warfare is mere pointless criminality).”Footnote ⁴¹ This is to say that what use people have made of a territory—how they move on it and have over time—circumscribes many of the ethical beliefs members of communities hold.

Similarly, anthropologists have described the intimate link between territory, the resources it provides, and the social organization of communities. See, for example, the relationship between the Inupiaq in the northwestern Arctic and the whales that provided for much of their livelihood:

Now, a critic could argue that I put too much of a positive spin on temporal and historical continuity when thinking about territory.Footnote ⁴³ In the first instance, this seems to go counter to the role I attributed to voluntariness when thinking about choosing among options as a necessary feature of autonomy. Is the temporal and historical relationship to forebears in conflict with this conception of autonomy? Can we choose voluntarily if we are part of this web of relationships? In fact, one could go further and argue that, sometimes, it may not be a harm to face discontinuity with our forebears, thus making the temporal link a particular interest—historically or territorially contingent—rather than a universal one. Instead of being an important interest, it may be a relief and a new and better beginning to have discontinuity with relationships to the past. Think of the reckoning that many colonial powers have to face concerning the crimes of the past. A lot of history, as we know, is not exactly cause for pride.

In response, I maintain that it is not only positive relationships with our forebears that serve as autonomy-enabling background conditions. Consider, for instance, the change in Canadian public discourse in 2021 in the aftermath of the discovery of old mass graves in former residential school grounds, filled with the remains of First Nations children. First Nations representatives had long argued that the residential school system had the objective not only to assimilate young and very young First Nations children but also to serve as tools of cultural genocide.Footnote ⁴⁴ After these discoveries, many Canadians had to reconsider their understanding of Canadian history and the use of territory—it became clear that Canada was not postcolonial but that for many First Nations people, colonialism would be ongoing until the grave wrong that they had suffered was at last recognized and addressed. In this vein, a reckoning with a difficult and disturbing past can help to assess and develop future action.

Another question that may follow from this criticism is how much weight should be given to the interest in historical continuity in the territory. Does disruption of historical continuity interfere in important ways with our autonomy?Footnote ⁴⁵ I think it does. Consider my discussion of the role of territory in individual autonomy. One of the arguments is that territory provides for a stable background condition of autonomy. How does it do this? Not only by being there but also by providing the historical context into which individuals can insert themselves, leave their mark by changing the way the land is used, or indeed just leave; and by providing a vital point of reference for themselves from which to start autonomous decision-making.

To sum up, I argue that historical territorial traditions are part of the web of relationships that allow for personal autonomy and agency. As participants in the historical and temporal web of relationships, we have access to lives lived, and to ways of understanding and being in the world. Put otherwise, being part of historical territorial relationships allows agents to access possible options that will help them shape their own lives.

If this is accepted, then we ought to think about the normative implications of the harm of severing ties with the historical relationship. Recall that I am discussing cases of necessary relocation; that is, relocation of those hitherto territorially bound cultural groups who are climate displaced. I suggest that there is a fundamental harm in severing ties with the traditional territory. One of the normative implications of this harm is that remedial relocation ought to consider the territorial link. For instance, we can imagine that members of the cultural group should only be relocated to territories to which they already have established ties, whether, say, through trade, intermarriage, or historical alliances.Footnote ⁴⁶ This demand would not only take into consideration the web of relationships I have identified as autonomy enabling but it would also help successful relocation, as studies of the Pacific communities have suggested. In fact, one of the reasons why the Carteret Island relocation was not successful was that there were no such ties, which fueled animosity between the Carteret Islanders and the Bougainvilleans, who were supposed to accommodate the relocation of the former group.Footnote ⁴⁷

A second normative conclusion to be drawn from the harm of severing territorial relationships could be that relocation should attempt to reestablish historical relationships.Footnote ⁴⁸ In this vein, consider the many examples of members of First Nations who were displaced to territories to which they did not have spiritual or historical connections during the colonial projects of North America. One normative conclusion we can draw from my analysis of the territorial tie would be to aim to find locations of relocation that are historically relevant to those displaced, or to resettle those formerly displaced on their traditional territory.Footnote ⁴⁹

Thus, relocation can lead to several different harms distinct from the direct experiences of an altered climate. The first harm is that of severing the ties between the individual and the concrete link to a territory. The second is that of severing the tie between the community and the territory, thus making communal practices based on the territory tenuous. The third harm, as I explained earlier, is that relocation interferes with autonomy-enabling historical relationships. If the territory that has framed the history of a community is no longer available to individual members of that group, one aspect of autonomous living is compromised. Moreover, projecting our lives into the future is made more difficult since we cannot refer to the past any longer when designing our future.

Of course, one could say that the third, historical aspect of harm of relocation can be outweighed by the possible gains. For instance, to return to the Carteret Island example, we can say that some hardships arise simply because of the nature of living in a particular territory, and the droughts, floods, and food insecurity could be ameliorated in the new territory. I do not want to dispute that some gains can come from relocation that may well improve individual well-being and promote flourishing lives. Rather, my point is that weighed against the charge of involuntariness, and the harms that come from relocation, the possible gains cannot easily tip the scale. Some gains will not be able to counterbalance the harm. Although there may be material improvements to individual lives after relocation, these benefits are the minimum owed to the climate displaced and cannot make up for the loss of historical connection. This seems to indicate that other measures must be taken to further compensate these groups.

Finally, one might argue that the moral assessment changes when relocation is designed by those who will experience climate-induced migration: should relocation then count as voluntary?Footnote ⁵⁰ I maintain that while the design may be voluntary and may respond better to the wishes of those who need to be relocated, the motivation to relocate, and by extension, to design the place of relocation is not. If traditional lands must be abandoned, the fact that a new place is accepted as a place of relocation does not alleviate the problematic way the decision to move came about. In this scenario, the previously mentioned sufficiency option—which is to say that an alternative to migration must be available—does not exist.