Introduction
Rising sea-levels due to anthropogenic climate change will cause unprecedented damage to the planet. With global temperatures set to reach 2C° above pre-industrial levels, NASA predicts that sea-levels will rise several meters in the coming centuries.Footnote 1 The Intergovernmental Panel on Climate Change (IPCC) is certain that sea-level rise will continue beyond the year 2200.Footnote 2 This future is made even more certain by the current lack of political cooperation on climate change, particularly in recent years.Footnote 3
In the near future, several states are set to be submerged entirely by the rising seas, including Tuvalu, the Maldives, French Polynesia and Kiribati.Footnote 4 While political theorists and philosophers alike have begun to consider the claims of refugees, they have largely ignored the question of collective rights stemming from this loss. Little headway has thus far been made in considering what obligations, if any, remaining states will have to whole groups of refugees whose homeland has been destroyed. Moreover, these obligations have not been evaluated in light of the unfairness of this harm. Some states are more responsible for climate change than others and those states predicted to be destroyed by sea-level rise are the least responsible of all.
This paper therefore aims to build a preliminary account of reparative justice for climate refugees, answering the questions, (1) who is owed, (2) who should pay, and (3) in what form should the payment be made? I propose that climate refugees are indeed owed reparations for their loss of land, and that these reparations should be paid by those outcome responsible for this wrong,Footnote 5 namely, states who have not cut carbon emissions despite clear knowledge of their negative effects. These reparations should be paid in the most ‘appropriate’ form, potentially obliging responsible states to provide new land to communities who have been territorially displaced.
Defining ‘Refugeehood’
Notably, the existing legal definition of a ‘refugee’ places great emphasis on persecution based upon race, religion, nationality, social membership or political opinion.Footnote 6 However, from a moral perspective, it is not clear that those fulfilling this definition should have exclusive access to protection; persecution may be a sufficient but not necessary condition of refugeehood. Instead, our legal definition neglects the central normative issue of refugeehood, namely, that refugees have lost the protection of their state. Therefore, refugees are ‘persons whose basic needs are unprotected by their country of origin, who have no remaining recourse other than to seek international restitution of their needs, and who are so situated that international assistance is possible’.Footnote 7 We will continue with this broader definition of refugeehood focusing on the severance of state-citizen relation. Of course, the term ‘climate refugee’ indicates a specific cause of displacement and only applies to a subset of those who satisfy this broad definition. Climate refugees are those whose basic needs are not met by their country of origin, specifically due to climate and environmental factors.Footnote 8
Before we move on, two further clarifications should be made. First, in considering cases of total territory loss, this paper focuses on permanent climate displacement. Reparations for temporary forced displacement due to climate change must wait for another time. Second, this paper attends largely to the collective claims of those suffering total land loss. Again, addressing the needs of individuals harmed by climate change goes beyond the scope of this paper. Therefore, the term ‘climate refugees’ is often used here as a shorthand for climate refugees considered as a group.
The Harm of Climate Displacement
Building an account of reparative justice for climate refugees depends on forced displacement being harmful. Although this is often taken as given, we should clarify at least two ways in which total state submergence can be particularly harmful to groups.
First, states submerged by sea-level rise face the loss of collective self-determination.Footnote 9 We should note here an important distinction between land and territory. Territory is an inherently political concept referring to ‘the geographical domain of a political entity, typically the state’.Footnote 10 It serves as a way to delineate the boundaries of the given political entity by constructing an inside-outside distinction, allowing the state to function within an “ordered” system. The concept of territory lies at the intersection between land and the political body, and is therefore the area within which the state is usually assumed to be able to exercise legitimate force. Land, however, is simply an area of the earth's surface. Certain political meaning can be attributed to land though it is not inherently political. Instead, this meaning is implanted rather than in some way essential.
Arguably then, total territory loss constitutes a specific type of ‘state death’, usually defined as the loss of territorial control to a foreign state. This definition reflects the historical fact that most state death has resulted from invasion, occupation, assimilation or annexation by another state and, therefore, does not apply where states are submerged by rising sea-levels. Not only is the focus on exogenous actors irrelevant in the case of climate refugees, but this definition also assumes that the geography of the state remains unchanged. In more familiar cases of state death the land remains whereas the territory has changed hands.
Whether we view submergence as a form of state death depends on how we conceptualise the state. Surprisingly, there is at least some consensus among political philosophers on what constitutes statehood. The state is generally regarded as a geographically bounded territory over which a set of institutions hold jurisdiction and power. The state therefore claims a monopoly on lawmaking and enforcement for a given territory, as well as a delineated set of political subjects. Crucially, great importance is given to the role of boundaries, as states are taken to be institutions of ‘continuous public power’, for which control over territory is eminently necessary.Footnote 11 Thus, ‘[a]ny attempt to understand the state must consider its spatial and temporal dimensions’.Footnote 12
Given the centrality of territory to statehood, sea-level rise poses a particular challenge. The submergence of land has the same effect as foreign invasion, assimilation and annexation; the state loses effective control and cannot exercise jurisdiction. Submergence therefore appears relevantly similar to historic forms of state death. It seems then that the conceptual importance of state death stems not from the foreign actors, but from the loss of minimally effective sovereignty over a delineated space. To echo Hobbes, ‘The Commonwealth without Sovereign Power, is but a word without substance’.Footnote 13
On this reading, territory is necessary for self-determination. It acts as a ‘practical’ foundation through which the state exercises self-governance. Thus, a group's capacity for self-determination depends upon having territory. The loss of territory therefore heralds the loss of statehood altogether. Indeed, in the case of total land loss, there may be extra moral considerations which do not apply to historic cases of state death. In these cases, there is little or no prospect of a successor state.
Second, land in itself can have moral value depending on the practises and beliefs of its inhabitants; certain land has meaning for certain people. Climate refugees face losing their ‘political identity, political community, currency, civil society institutions, and perhaps even language’.Footnote 14 We, therefore, fail to do justice to the moral claims of particular groups when we treat the world as inherently undifferentiated. For Bell, we ‘cannot treat the disruption, degradation or destruction of particular environments or homes as nothing more than the loss of some substitutable natural resources’.Footnote 15 Therefore, discussions of justice for climate refugees must move beyond territorial claims and also highlight the importance of land. Although loss of self-determination is possibly a simpler harm to understand and therefore address, this loss of land as home must not be forgotten.
Compensation or Reparations
Those who discuss justice for climate refugees do so almost exclusively in the language of compensation, understood as the payment of money or goods to recognise certain harms, or ‘effectively transferring resources in order to put a person on the indifference curve they would have been on had the disadvantage not occurred’.Footnote 16 Goodin helpfully distinguishes between means-replacing and ends-displacing compensation.Footnote 17 Means-replacing compensation allows individuals or groups to pursue the same ends as they would have before the loss. Ends-displacing compensation conversely allows them to pursue some other end that would make them as well off as they were before the loss. In the case of climate refugees, ends-displacing compensation would demand climate refugees be given land on another state's territory. These refugees would become political members of another state, allowing them to achieve the end of self-determination through alternative means. Means-replacing compensation, in turn, may require new territory for climate refugees, rather than relying on another political body. Ödalen suggests that means-replacing compensation could also be achieved by increasing climate refugees’ autonomy in other ways. For instance, submerged states could be given resources to maintain a government in exile.Footnote 18
Nine offers a more radical discussion of compensation for climate refugees, applying the Lockean ‘enough and as good’ proviso to current territorial holdings. She argues that changes in land availability resulting from climate change demands a re-evaluation of current territorial claims. That is, given that less land will be available post sea-level rise, territorial holdings are up for potential redistribution. These submerged states may therefore ‘be candidates as sovereign over a new territory after their existing territorial lands have been lost to the rising sea’.Footnote 19
My project here is not to evaluate these accounts of compensation, but rather to question the enterprise as a whole. This is because conceptualising justice for climate refugees as a matter of compensation misses a key feature of sea-level rise. Crucially, determining what climate refugees are owed demands an understanding of how this wrong came about. Climate change did not happen by accident. Instead, it is a result of the carelessness with which (largely) so-called ‘developed’ states have polluted, through practices of burning fossil fuels, eliminating tropical rainforests, while ignoring basic emissions quotas. To put it simply, the plight of climate refugees is caused by the emissions of other states. To add to this unfairness, these soon-to-be-submerged states have historically been the lowest emitters, with Tuvalu projected to become the first zero-carbon state by 2020.Footnote 20
So why does this unjust dimension to climate change demand a shift away from the language of compensation? In short, compensation aims to remedy accidental damage or harm, whereas reparations aim to rectify injustice. We see this distinction most starkly when we note that compensation may be given to a victim by a third-party.Footnote 21 Reparations, however, can only be successfully offered by those responsible for the harm; the injury creates an ‘affinity’ between them.Footnote 22 In cases where certain agents are found to be responsible for some harm, third-party compensation may be insufficient. Therefore, rather than continuing in the language of compensation, this paper develops a preliminary account of reparations for climate refugees whose land has been destroyed by climate change.
Reparative Justice for Climate Refugees
Whilst the topic of reparative justice has only gained traction in the arena of political theory in recent decades, claims for reparations are far from new. We can see instances of reparations in the Guacaca Courts of Rwanda, the Truth and Reconciliation Commission in South Africa, and payments made to Israel by West Germany in the 1940s. There have also been persistent discussions of reparations for slavery, with Harvard Law Professor Charles Ogletree recently taking the case to court.Footnote 23 In 2015 CARICOM, a union of Caribbean states, sought reparations for slavery from Great Britain in the form of financial compensation and investment.Footnote 24
Reparations are a form of repayment, restitution, or recompense for some wrong. Reparations are therefore, at their core, historically oriented. For reparations to be owed there must have been some wrong committed. Whilst this historical focus is central to all reparative theories, there are other ongoing debates to be resolved.
First, there is a continuing struggle between forward-looking and backward-looking accounts of reparative justice. Backward-looking accounts do just that, they look backwards and only backwards. Thus, for a group to have a reparative claim, we need only consider whether a moral violation was committed. However, forward-looking conceptions of reparative justice argue that the claim must also be ongoing, asserting that a violation or harm is necessary but not sufficient.Footnote 25 For example, black Americans may claim reparations because the current oppressive conditions in the US stem from the history of slavery.Footnote 26 In defence of forward-looking accounts, Young argues that backward-looking models ‘naturalise unjust social structures and therefore deflect attention from the collective action that is required to transform them’.Footnote 27 We thus fail to take current inequality seriously when we focus solely on the past. Of course, a fully-fledged defence of either view is not necessary here, as our claimants, climate refugees, have on-going claims. Sinking islands are not a problem of the past, but a problem of the present and the future.
The second distinction is between rights-based and obligations-based accounts. Thompson sketches this debate in terms of claimants and obligation bearers.Footnote 28 Rights-based views rely solely upon the existence of a claimant, arguing that as long as there is a claimant, reparations are owed.Footnote 29 However, an obligations-based view places emphasis on the establishment of responsibility, arguing that reparative claims depend on the existence of a duty-bearer. In mediating this debate, we should note that the rights-based account collapses into a discussion of compensation for unjustified wrongdoing.Footnote 30 That is, in focusing on the rights-bearer, this account fails to distinguish between compensation and reparation. The obligations-based view does just the opposite. In focusing on obligation-bearers, this approach centres on the responsible agent. Of course, the major consequence of adopting an obligations-based theory is that the number of legitimate reparative claims becomes much smaller. However, if we are convinced by the obligations view, then we could only resolve reparative claims when there exists a duty bearer. Reparations may therefore only be due when a responsible party can be identified.
Whilst there has been little discussion of reparations for climate refugees, there has been some consideration of reparations for refugees more generally. In conceptualising asylum as a form of reparative justice, Souter proposes moving from a synchronic to a diachronic approach whereby asylum may be understood as a form of remedy for past wrongdoing.Footnote 31 Under our current normative framework, refugees have ongoing humanitarian claims which generate general duties.Footnote 32 However, some refugees have an important historical element to their claim, namely, their displacement results from the actions of a specific party. For instance, to use Souter's example, the claims of Iraqi refugees are historically linked to the US invasion. Therefore, certain states may be more responsible for these refugees than others.
This diachronic approach draws our attention to the distinction between special and general obligations. Special obligations are those owed to a particular subset of agents. Souter's reparative approach claims that states have special obligations to those refugees whom they have caused harm. These are not obligations generated on the grounds of equity, but the grounds of responsibility. We should remember here the distinction between reparations and compensation. Reparations can only be paid by those who are responsible for the harm in question, whereas compensation can be provided by a third party.
The difficulty lies in linking certain agents to the harm of climate displacement. That is, we need to show that certain agents are responsible for climate change and potentially, by extension, the damage caused to small-island states. Then climate refugees, as we understand them, may have reparative claims against these agents. Here we open up the complex questions of responsibility of climate change or, indeed, whether anyone is responsible at all.
In the following sections, we will take the three key core questions of reparations for climate refugees in turn, namely: who (if anyone) is owed reparations, who (if anyone) should pay, and in what form (if any) should payments be made?
1. Who Is Owed?
Who is owed reparations (if anyone) is usually a complex question for those defending reparative justice, as the moral wrong in question has often occurred in the distant past. There are, therefore, difficulties determining current claimants when the actual victims no longer exist. Usually, the options are narrowed to (a) the direct descendants of the victims or (b) the continuous group that remains affected by the injustice. For instance, in the case of reparations for slavery in the US, the debate concerns whether the claimants are (a) the direct descendants of slaves or (b) black Americans more generally. There is also the additional worry of the non-identity problem, whereby individuals in the present who purport to be claimants of reparations would not have existed if the injustice had not occurred.Footnote 33 A current black American may not have been born if her great grandparents were not enslaved. It is perhaps counterintuitive for an individual to claim reparations for a harm that caused their existence.Footnote 34
For climate refugees, both of these worries seem resolvable. Given that state destruction due to sea-level rise is an injustice of the present and the future, we avoid the difficulties of determining descendants and the non-identity problem. Our claimants exist now and can therefore demand reparations. However, there is still some complexity in answering the question of who is currently owed reparations that mirrors the descendant issue. That is, are reparations owed to individuals affected by land loss or the collective? In reality, either answer would give us a similar list of claimants. Those affected individually by sea-level rise and those affected as part of a group, are largely one and the same.
Both Kolers and Nine argue that an individualistic solution to climate displacement is not satisfactory because it doesn't account for the collective nature of the loss, namely, loss of self-determination and loss of culture.Footnote 35 I likewise consider the claim to be group-based in nature. There are several reasons for this. First, whilst individuals may have atomised claims for reparations based on land and property loss, this is not mutually exclusive with group claims. For instance, a black American may have a two-fold claim to reparations. She may have a claim based upon the loss of land from her ancestors. For example, land was unjustly taken from black families during the Jim Crow era for which individual reparations are owed.Footnote 36 Moreover, she may also have reparative claims couched in her membership of the broader group ‘black Americans’ stemming from the systemic violent oppression of all black individuals in the US. I see no reason why these two claims should be incompatible. Indeed, on a practical level, compensatory mechanisms for development-induced displacement have attempted to address both collective and individual harms. For instance, through individual property relocation in tandem with mass resettlement programs.Footnote 37 Climate refugees, therefore, may have strong individual claims, as well as strong collective reparative claims.
Second, we might hope to focus on collective claims due to the current lack of attention to total land loss. In the limited displacement reparations literature, there is a strong focus on individual claims. Paying sole attention to individual loss misses the important cultural losses faced by the group as a whole. In an attempt to move away from this individualised discussion, my focus will be on group claims.
So, in our case, which group is owed reparations? The answer to this question largely depends on what reparations are owed for. I began by considering the destruction of self-determination due to territory loss. If reparations are owed for this particular harm, then it is the state who should be paid. However, the above discussion concerning land loss as a substantial cultural harm presented damage done to the community, not necessarily the state. Here then, the community at large is owed reparations as it is the primary bearer of this harm. Again, there has been some focus within the literature concerning the loss of political self-determination. Our focus in this paper has been to develop a more novel and context-appropriate conception of reparative justice for climate refugees. Therefore, although climate refugees may be owed reparations for the loss of self-determination, our principal focus will be on reparations for the cultural harms explicated earlier on.
2. Who Should Pay?
As previously stated, the agents who owe reparations are those responsible for the harm. Therefore, in considering who should pay reparations to climate refugees we must determine what it means to be responsible for climate change and potentially, by extension, the displacement that it causes.
We should begin by distinguishing between moral and outcome responsibility. Moral responsibility is often understood as a species of ‘blameworthiness’. You are morally responsible for some consequence if you freely chose to do some action, for which you can therefore be held accountable. Outcome responsibility is instead ‘whether a particular agent can be credited or debited with a particular outcome [and]… exists only when there is a ‘foreseeable connection between … action and … result’.Footnote 38 Souter claims that outcome responsibility is both necessary and sufficient for reparations:
When an agent bears outcome responsibility for causing another agent unjust harm, the first agent bears a special obligation to provide the second agent with the most fitting form of reparation for that harm available.Footnote 39
This reliance on outcome responsibility when determining whether reparations are owed is meant to accommodate the intuition that reparations are owed even when agents are ‘morally innocent’.Footnote 40 Consider the case of a ‘just war’ whereby a state may be justified in invading a foreign country, for instance, to stop genocide. This state could be held outcome responsible for any displacement caused, but perhaps not morally responsible. Similar arguments are made concerning development-induced displacement. Penz rejects the ‘simplistic moralism’ that displacement is always an evil, rendering all development projects unacceptable.Footnote 41 Instead, if such a decision is made in fair circumstances, then the agent(s) responsible may only bear outcome responsibility as opposed to moral culpability.
I have two concerns about this distinction. First, I worry about how this affects the moral function of reparations. This returns us to the discussion of reparation and compensation. I question whether scenarios of ‘just war’ or development-induced displacement fulfil this criterion. Would we truly object to third-party payment in these cases? If not, then they demand compensation, not reparation. Perhaps then, moral responsibility is necessary to generate reparative duties. Reparations have a strong moral function and perhaps only arise in cases of moral responsibility. This discussion is particularly pertinent to our case, as locating moral responsibility for climate change is notoriously difficult. If we find outcome responsibility unacceptable, then we may have hit a wall in our search for reparations for climate refugees.
Souter would respond, I believe, that even if we think that compensation is payable by a third-party in these cases, we would still argue that something is owed by those outcome responsible. This ‘something’ may come in the form of an apology or a statement of why it was believed the action was justified; the responsible party is not entirely free from additional obligations. Perhaps also, if no one could pay compensation, then the party responsible would be the last resort. This confusion, I believe, arises from the fact that cases of extreme moral wrongs, such a slavery and colonialism, appear to generate powerful claims for reparative justice. Cases of outcome responsibility perhaps generate weaker reparative duties, ones which could be easily confused with duties of compensation.
My second concern is that outcome responsibility fails to consider the role of benefit. An agent may not be outcome responsible for some wrong, but she may still have reparative obligations stemming from unjust profit. This concern applies to Souter's own example of Iraq, whereby US and UK oil companies profited greatly from access to Iraqi oil fields.Footnote 42 Souter, I suggest, would reply that their lack of outcome responsibility makes them only liable for compensation. In the following section, I will discuss the role of benefit further, arguing that it may be useful as a secondary consideration when reparations cannot be paid by principal duty-bearers. I agree with Souter, that outcome responsibility bars us from holding beneficiaries accountable, but propose that benefit triggers a secondary duty of repayment. For the time being, outcome responsibility remains our key focus for the task of determining who owes reparations.
I now turn to discuss who is outcome responsible for sea-level rise. The question of responsibility for climate change is extremely contested. Climate change is often described as a ‘perfect moral storm’, comprised of problems of collective action failure, diffused responsibility, and intergenerational justice.Footnote 43 However, having adopted outcome responsibility as our principal framework, we begin in a better position. Our theory need only consider which agents have a direct and foreseeable relationship to the harm in question.
The debate on responsibility for climate change pivots almost entirely around the application of three principles; The Polluter Pays Principle (PPP), the Beneficiary Pays Principle (BPP), and the Able to Pay Principle (APP).Footnote 44 Importantly, this debate pays little attention to the demands of reparative justice and instead focuses on compensation. This section draws out the discussion thus far, concluding that whilst the BPP may be a successful subsidiary mechanism in allocating responsibility, our main focus must be on the PPP.
The PPP holds that ‘emitters of high quantities of greenhouse gases (GHG) are responsible for compensating those negatively affected’.Footnote 45 This is not only intuitively plausible, but has also been affirmed in several international climate agreements, including a directive passed by the European Union.Footnote 46 This principle distinguishes between those agents who make admirable choices on climate change, and those who continue to contribute to GHGs beyond their reasonable level. Thus, the PPP presents a straightforward way of determining responsibility for the payment of reparations to climate refugees. Moreover, from the perspective of reparative justice, this principle also seems to fit our purposes. It has, as its basis, a prerequisite of causal responsibility. To be obligated to pay reparations, you must be, at a bare minimum, causally implicated. The PPP ‘does not seek to forget history’.Footnote 47
Evidently, there is a certain amount of GHG that an individual or state can emit without being morally blameworthy. In an industrialised world, everyone pollutes at least marginally. The moral issue comes into play once people have overstepped their ‘fair share’ of GHG emissions. Of course, what we conceive of as a ‘fair’ quota is debatable. However, from the perspective of outcome responsibility, this moral threshold is not relevant. Here, we are simply considering whether there is a direct and foreseeable harm. Even if this harm is necessary, reparations may still be owed. Therefore, under our current framework, polluters are obligated to pay for all emissions, not only those which are immoral. Indeed, in light of our focus on outcome responsibility, we must also add a foreseeability requirement to the PPP. That is, there was surely a point at which it was not reasonable to assume knowledge of climate harms. Our prerequisite of foreseeability therefore necessitates a cut off point. For instance, Muller et al. suggest holding people responsible for emissions only after 1970.Footnote 48 Polluters can therefore only be held outcome responsible for sea-level rise when this harm is reasonably foreseeable.
There are several objections to the PPP, all of which, I believe, can be rejected. First, Caney suggests that it is difficult or even impossible to say which emissions are causally responsible for which effects.Footnote 49 One unit of GHG is not equal to one unit of sea-level rise. Instead, this process is mediated by complex processes which make us unable to attribute harms to particular emitters. The atmosphere becomes a space of moral alibi given the complex sequence of events between emission and harm.Footnote 50
Neither Caney nor I agree with this criticism. First, we know that GHGs cause climate change. This grey area doesn't allow for agents to claim innocence. Indeed, it is not whether your GHG caused a certain unit of harm that is central, but that your GHG contributed to the aggregate which caused the harmful outcome. Second, no reasonable suggestion for determining reparations depends upon tracing individual emissions. Given that harms occur only in light of aggregate GHG emissions, it is this aggregate that should be divided in terms of proportional responsibility. Most writers endorse this model, calling for the US and China to bear the greatest responsibility in proportion with their historical emissions.Footnote 51
A second objection is that the PPP fails to account for intergenerational emissions.Footnote 52 Climate change is a process that takes place across centuries, which the PPP fails to consider. Of course, not everyone who has contributed to climate change is dead. Though, certainly, the problem of past emissions should worry defendants of the PPP. Baatz believes that this intergenerational problem is enough to reject the PPP in favour of the BPP, to which we now turn our attention.Footnote 53
The BPP asserts that ‘duties of remedial justice [should] be assigned to agents who have profited from activities that impose undeserved disadvantages on other agents’.Footnote 54 This principle draws our focus away from causation and considers profit: free-riding on the immoral actions is itself immoral.Footnote 55 Notably, this focus attributes responsibility to a similar set of agents as the PPP. Those who have benefited from climate change are generally those who pollute the most. The major bonus of this principle is that it overcomes the intergenerational issue. The BPP considers present benefits and therefore need not account for agents who no longer exist. Whilst this success adds to the appeal of the BPP, there are several reasons why we should reject it.
The BPP is less immediately intuitive than the PPP. It is not clear whether passive benefit is enough to establish remedial responsibilities.Footnote 56 It is almost impossible to live without benefiting from climate change. It seems unfair that people who have not consented to the system should be obligated to pay remedial costs because of their secondary role. Such a focus may violate the central intuition of an obligations-based account of reparations: that agents who are not responsible for a given harm cannot be held accountable. As such, the BPP doesn't satisfy the causal prerequisite of outcome responsibility. Therefore, the BPP does not give us a strong account of reparative obligations. A second problem for the BPP is that benefit is not the sole determinant for whether reparations are owed. If no one benefitted from climate change, then would we not think that certain agents were obligated to pay reparations anyway? Once we remove benefit from the equation, the moral intuition in favour of reparations appears to remain.
The benefit of the BPP is that it escapes the intergenerational problem. However, there is an evident way to deal with the intergenerational worry for the PPP: focusing on states as opposed to individuals. As Stilz points out, state responsibility is a guiding principle of international law:
It is a key principle of international law that states (…) can be held responsible for their acts. When a state breaches an international obligation, it has a duty to make reparation for any injury it causes, either by making restitution, paying compensation, or providing some alternative form of satisfaction.Footnote 57
Indeed, states continue to flagrantly ignore climate legislation to the detriment of the planet and do so largely to the benefit of corporations. Some states are clearly more outcome responsible for climate change. To this end, China and the US are often cited as being the biggest polluters as well as the most problematic in terms of overstepping quotas. Such a statist model allows us to differentiate between states in terms of their historic emissions whilst maintaining them as the ultimate bearers of responsibility for policies and bad economic practice.
The state offers an escape route from the intergenerational problem, as it is itself an intergenerational entity existing despite constant shifts in membership. This statist model therefore answers the intergenerational question: as states persist whilst individuals come and go, they can bear outcome responsibility for emissions, both past and present. But perhaps the role of benefit may be a good subsidiary principle for determining the payment of compensation. For this, Baatz suggests a ‘Qualified Beneficiary Pays Principles’ (QBPP) which places the obligations of beneficiaries as secondary to polluters:
QBPP: In case agent A performs action X that harms agent C, agent B is under an obligation to compensate C, if:
(i) C's harm cannot be (fully) addressed by the responsibilities of A*,
(ii) B receives a net benefit from X
*A might be dead, unable or inhuman. Footnote 58
In cases where those outcome responsible for pollution cannot be held accountable, benefit may be a useful secondary principal. This allows us to address the moral intuition that those who benefitted from a harm have some obligations for repayment. However, it also allows us to maintain our central focus on outcome responsibility, which is important from the perspective of reparations.
Therefore, the PPP appears to be the best way forward. It allows us to maintain our focus on outcome responsibility whilst solving the major intergenerational objection in focusing on the state. The role of benefit should be secondary; benefit may be a useful metric for determining who owes compensation, should those with reparative obligations be unable to pay.
3. What is Owed? Money, Immigration, and New Territory
We now turn to our third and final question. In what form should reparations to climate refugees be paid? Both Souter and Miller argue that reparations must take the most ‘appropriate’ form. However, what we consider to be ‘appropriate’ depends upon the goal of reparative justice. Walker distinguishes between two models of reparations: corrective justice and restorative justice.
Reparations as a form of corrective justice are largely legalistic, based on ‘mechanisms of restitution and compensation’ and focusing on proportionality between harm and payment.Footnote 59 Reparations here attempt to offer the ‘correct’ reparations to restore the individual or group to their pre-harmed state. The restorative justice model instead focuses on ‘restoring relationships’ and ‘supposes that it is disregard or violation of acceptable human relationships that stand at the core of [the] agenda, practically and philosophically’.Footnote 60 This account therefore aims to repair the harm caused by centring the experiences and needs of victims.Footnote 61 This model does not focus on an objective standard of redress, but rather aims for reconciliation between both parties, accepting that reparations are successful when the harmed group is satisfied.
This restorative approach is of central importance when considering a form of harm as subjective as the loss of place. Such an approach seems most fitting when harms cannot easily be quantified. But perhaps we cannot adopt a theoretical framework which does not allow for ‘reasonable refusal’. That is, centring the voices of climate refugees cannot go so far as to block reparations that are rejected on unreasonable grounds. Abandoning a reasonableness requirement altogether would allow claimants to demand reparations over and above the harm done. For instance, if climate refugees demanded potential new land that was vastly larger than the land that they had lost, the whole of the US perhaps, and refused to accept any other offer, we would most likely find that unreasonable. Although the preferences of climate refugees are of central importance, some reparations may still be more appropriate than others from the perspective of fairness. We should move forward with clear consideration of the desires of climate refugees, but tread carefully with the idea of ‘reasonableness’ in mind.
Of course, a fundamental question to address is whether anything can truly repair the damage wrought by climate change. De Shalit argues that the crucial functioning of ‘sense of place’ can never be compensated for. Therefore, our central focus must be mitigation.Footnote 62 Perhaps nothing can truly repair the loss of home, just as no amount of money could ever compensate for the death of a loved one. However, I do not believe that this is a reason to dismiss the question of reparations altogether.
For instance, we should remember that the case of reparations for climate refugees in particular impacts upon future as well as current generations. Although this generation of climate refugees may never have their sense of place restored, certain forms of reparation may re-establish a sense of place for future generations. The symbolic and material significance of land is constructed subjectively and can therefore be reconstituted over time. Perhaps this reconstitution is not the same as the original connection, but that does not bar it from being equally meaningful. Climate refugees perhaps cannot be restored a full position of status quo ante, but this does not mean that we owe them nothing. Nor does it mean that the question of what they are owed should be ignored; some reparations may still be more appropriate than others. With this in mind, I now consider three suggested forms of reparations for climate refugees in succession: money, immigration, and land.
3.1 Money
Possibly the most obvious form of reparation is money. To this extent, money provided by the responsible party is used to restore the recipient to a position as near as possible to status quo ante. In this regard, the central success of offering money is its fungibility. Money is the most versatile option available, as recipients can do as they please with it. Moreover, money as a mode of reparation has been offered for some of the gravest moral wrongs in human history. West German reparations for the Holocaust were paid in the form of 3 billion marks. Reparations for slavery and colonialism are also usually considered in the form of money.
However, there are several reasons why we might not think of money as appropriate. First, money seems ‘incommensurable’ with the loss faced by climate refugees.Footnote 63 Given that the point of reparations is to raise the victim to a position that is comparable to a pre-harm level, money does not seem appropriate given the significant cultural harm. Even when looking at the problem of climate refugees as a straightforward self-determination issue, money seems to fall short. The loss is so significant that the offer of money to remedy the harm seems a little hollow. The usual benefit of money is that groups can invest in their shared future. You cannot invest in infrastructure when there is no state, nor in long-term development when the ‘people’ do not remain as a collective. Something additional would be needed for climate refugees to be able to use money collectively in order to benefit from this form of repayment.
A current example from the Australian High Court is worth considering; the case of D1/2018 concerning the native title of indigenous people who were expelled from their land.Footnote 64 This case depends on the Native Title Act (1993) which states that indigenous groups have a right of compensation for the “impairment and extinguishment” of native title rights in a range of circumstances. However, Parliament has not clarified what this compensation amounts to in practice; this is now to be decided by the High Court. Case D1/2018 depends on two earlier decisions, known as the ‘Timber Creek’ decisions, whereby two lower courts issued compensation to the Ngaliwurru and Nungali peoples for the amount of AU$3.3m in August 2016. This has been appealed in D1/2018 on the grounds that these lower courts failed to fully recognise the non-economic loss that is faced by the Ngaliwurru and Nungali peoples. That is, loss of native title cannot simply be compensated on an economic basis—something additional has been lost that cannot be remedied by monetary award. Although this case is ongoing, and the High Court will likely uphold the decisions of the lower courts, this is an interesting example on monetary compensation being viewed as inadequate to remedy the loss of land.
Of course, our restorative focus means that we must listen to the views of those seeking reparations. In our case, climate refugees. We should not assume that the group we are considering will have the same view as the Ngaliwurru and Nungali peoples. Should the community in question ask for reparations in the form of money, and an agreed upon amount can be reached, then reparations should be given in this manner. It may also be presumptuous to assume that the community would want to continue as a collective. If they chose to take this as an opportunity for dispersal, then individualised payments would be more appropriate.
However, we have good reason to think that this is not what the residents of these communities want. Indeed, all calls to the international community have focused on the community at large, with governments planning for mass resettlement as opposed to individual economic compensation.Footnote 65 Thus, money seems neither practical nor normatively acceptable from the perspective of reparative justice for the time being.
3.2 Immigration and Free Movement
We should now consider providing climate refugees with opportunities to migrate. Souter suggests that asylum could function as a form of reparation for displacement whilst, of course, taking into consideration the choices of the refugee.Footnote 66 Miller responds to Souter, claiming that immigration is not generally a good way to compensate refugees; it only occasionally constitutes the most appropriate form of reparations.Footnote 67 Instead, we should attempt to rectify the harm in question. However, Miller notes that environmental destruction may be a particular case for which immigration is necessary. He writes:
At first sight, an even clearer instance of immigration as a form of reparation would be territories that become unable to sustain human life as a result of global warming, where no physical form of repair is possible. … [P]erhaps, in the absence of an agreed scheme for resettling those whose land has become uninhabitable, a right to reparation could be asserted against any state that has contributed significantly to the warming.Footnote 68
Other similar proposals have considered immigration rights for the territorially displaced. Eckersley argues that states have a common responsibility to receive climate refugees on humanitarian grounds, particularly with the loss of self-determination in mind.Footnote 69 Climate refugees therefore have a right to choose their new state, which provides ‘partial compensation for the injustice and trauma of their loss’.Footnote 70 The obligations of states to accept this account depends upon their ability to pay. This, Eckersley argues, maximises total protection for climate refugees. Of course, our reparative focus on the PPP demands that those outcome responsible for climate change are held accountable. Byravan and Rajan propose such a framework, arguing for a policy that ties the number of climate refugees a state is obligated to host to that state's historical emissions, rather than whether the state is best placed to offer assistance.Footnote 71
However, the immigration proposal misses two keys things. First, it does not fully account for the importance of your place. Simply allowing climate refugees to move to new land may not remedy the loss that they have suffered. As choice is central to their immigration proposal, climate refugees could decide where to put down new roots. Although this element of choice is important, it is still not obvious that this would make an immigration account wholly satisfactory. This brings me to my second concern, namely, that immigration as a solution does not consider the collective issue at hand. The above proposals focus on the rights of individuals to immigrate to any state of their choosing. However, allowing individuals to move where they please does not respect the possibility that the community may want to stay together, allowing them to maintain their distinctive culture, history, and right to self-determination.
A resident of Shishmaref, Alaska, an Inuit community that is set to be destroyed by rising seas, notes that ‘collocating or merging with other villages may be cheaper than relocation, but the risk is high that the village's lifestyle and culture will be lost’.Footnote 72 Likewise, the Tuvalu HLD initiative seeks the establishment new land for mass resettlement so that the community can remain together.Footnote 73 Our restorative approach forces us to consider the wishes of the community and individuals at hand. Piecemeal immigration proposals miss something of fundamental importance.
3.3. Land or Territory
A focus on land appears to have several benefits over both money and immigration. First, it considers the importance of collective as opposed to individualised harms. Resettling the group does not focus on individual claims, but instead recognises duties towards the group. Moreover, new land appears more commensurable with the loss. Of course, one could immediately reply that new land will never hold the same material or symbolic significance as the old. Even a perfect replica of a community's land could not remedy its loss.
To this end, de Shalit notes an analogy, claiming that new land is similar to a stepfather replacing a lost father.Footnote 74 Whilst step-parents allow for the same functioning as the ‘real’ parent, they are never a complete replacement. Whilst thought provoking, this analogy overlooks two things. First, given the constructivist nature of the connection to land (and indeed the connection to a parent), there is no reason why a community couldn't develop just as meaningful a relationship with new land. Whilst something is clearly lost when your home is destroyed, something else that is equally meaningful can be redeveloped. Second, it misses the intergenerational nature of the community. Whilst a step-parent may never replace a father, perhaps this irreplaceability would not extend to future generations. A step-grandparent may be equally meaningful to a granddaughter, particularly if she never knew her biological grandfather. Thus, new land may become important to future generations with new connections made further down the line.
Of course, what form this land should come in is still up for debate. The type of land owed depends on the harm for which reparations are given. When considering the loss of political self-determination, perhaps nothing short of new territory will do. Indeed, this is precisely Nine's argument, that ‘environmental refugee states’ should be given new territories to restore them to a position of status quo ante. However, our argument has focused on the distinct cultural loss. Perhaps in this case, ‘autonomous’ land would constitute a sufficient form of reparations. Indeed, such a position is reflected in the recent attempts of sinking islands to buy land. Kiribati has recently bought several small islands in Fiji to which it will move residents when their territory becomes uninhabitable. Whilst Kiribati will not exercise jurisdiction over these new islands, it could at least hopefully preserve cultural heritage.
Although such a model may be preferable to piecemeal immigration or monetary compensation, there remain reasons to be critical. For instance, mass movement programs have been rejected by climate refugees themselves on the grounds that it does not respect their right to cultural heritage. For example models of mass movement without statehood have been rejected by the people of Nauru whose land was nearly destroyed by phosphate mining in the 20th century. UN visits to Nauru in the 1950s determined that long-term resettlement was the only viable option given the community's growing population and the increased demand for phosphate.Footnote 75 It was proposed that the entire population be relocated to an island off the coast of Queensland. However, this model was rejected on the grounds of cultural identity. Nauruans viewed this relocation as an attempt to destroy their ‘strong personal and spiritual relationship with the island’, ignoring ‘the right of the Nauruan people at international law to permanent sovereignty over their natural wealth and resources’.Footnote 76
A further problem with this proposal is how such land would actually be provided. Aside from Nauru's unique case, states are unlikely to offer up land to the many thousands of climate refugees that will be displaced in the coming century. Nine proposes simply ceding land through an international mechanism, such as the UN, from states with the most space.Footnote 77 However, a clear criticism of this model is that it replicates the harm which our reparative theory attempts to remedy; the loss of culturally significant land. Indeed, if land were taken without the consent of its residents, then this model would not respect the harm we are attempting to remedy.
Moreover, the provision of land that just happens to be available doesn't attempt to place climate refugees within a place of their own choosing, or in one that is relevantly similar to the land that was lost. We would find it unacceptable if the people of Tuvalu were relocated to Nevada without consultation, simply because some land happened to be empty. That is, whilst symbolic connections are hard to reconstruct, finding a place where people can at least continue their agricultural practices is more straightforward. The worry here is that, practically speaking, those islands that are relevantly similar to current sinking Islands in the Pacific also tend to be at risk of sea-level rise. Mass movement to these islands is not a long-term solution. Even with these worries, land appears to be the only form of reparation that addresses the collective nature of the harm.
3.4 Floating Islands: The Case of French Polynesia
A new proposal in French Polynesia may solve several issues with the provision of new land. Seeking to mitigate the existential threat of climate change, French Polynesia is working with the ‘Seasteading Institute’, an American non-profit think-tank, to construct new islands. Development of a pilot island off the coast of Tahiti will begin in 2018, with completion anticipated in 2020.
From the perspective of reparations, this model appears somewhat successful. First, we do not face issues of ceding new land, as these islands will be situated within the borders of French Polynesia. Second, this land, in being designed precisely for the community, may herald the maintenance of agricultural practices. We might think here that enabling climate refugees to remain in the same patch of ocean is significant. For instance, Tuvalu currently holds status as Exclusive Economic Zone, which affords it special fishing rights: 46% of Tuvalu's GDP is generated from fisheries, giving its location central importance to both the functioning of the state and the identity of the people.Footnote 78
A significant issue for such a model is its ability to maintain or restore the cultural ties that the community has with its land. This model at least partially deals with this issue in allowing the people to remain in the same geographic space. On our restorative model, it is important that the community is as involved as possible in the development of the new land, which would potentially allow them to imbue cultural significance into the new space.
I wonder, however, whether the cultural significance of home could ever be replaced with a new island. This problem seems to haunt all proposals outlined thus far. A new place will always be a second best option when compared to the old. Instead, we should note that the development of new land could answer the reparative claims of climate refugees more successfully than would appear at first glance.
Conclusion
This paper has developed a preliminary account of reparative justice for climate refugees. We began with a discussion of the literature surrounding the moral claims of climate refugees, noting the focus on compensation as opposed to reparations. Throughout this work, I have attempted to offer an alternative route forward. Our final thoughts should return to the question of possibility. This paper has not painted a straight path in the search for reparations. Many easier routes that we could take, such as seeking compensation, are not satisfactory from the perspective of justice. The task of reparations is a great one, demanding far more collective action than the political reality has made possible thus far. Indeed, ever more, the task of repairing for loss of home seems comparable to the loss of a loved one. Nothing can replace the loss of a sense of place, just as the loss of a parent cannot ever be remedied. This paper has not attempted to claim that reparations are simple. Nor that there is a direct way forward from the perspective of justice. Instead, we should listen to the voices of those most affected and remember that, for the most part, they have not caused the great challenge that they face. In the end, perfect reparations may be impossible. Even so, we still have an obligation to shoot the arrow as close to the target as we can.
