Resilience and Nonideal Justice in Climate Loss and Damage Governance

The decisions in Article 8 of the Paris Agreement (1/CP.21, §52.) entail that climate loss and damage (L&D) is highly unlikely to be addressed through liability for compensation, even though these framings are the most intuitive (Lees 2017). Although developing countries and civil society organizations have requested since the early debates about L&D under the United Framework Convention on Climate Change¹ (UNFCCC) that it be framed in terms of liability and compensation, Decision 1/CP.21, §52. holds “that Article 8 of the Agreement does not involve or provide a basis for any liability or compensation” (Vanhala and Hestbaek 2016). This article investigates the ethical implications of Article 8 despite this decision. It shows why the justification of responsibilities for remedy must not rely exclusively on liability for L&D as is most often employed in climate litigation. I argue that victims of L&D should be seen as entitled not only to compensation for L&D but to achieving levels of capabilities sufficient to become climate resilient.

The article proceeds as follows. First, I explain my approach from nonideal theory and the current political and legal circumstances. This allows the role of liability and compensation in relation to L&D to be clarified. Second, I argue that differentiating remedial responsibilities by relying on liability and backward-looking responsibility for climate change is only one way of assigning responsibilities for remedial action in cases of L&D. Forward-looking reasons of capacity and solidarity also allow these responsibilities to be differentiated. Third, discussing two key challenges to arguing for such forward-looking justifications for remedial responsibilities clarifies that entitlements in L&D governance under the UNFCCC are ideally defined as entitlements to capabilities for appropriately addressing climate threats. Fourth, this article ends by suggesting that entitlements in case of L&D materializing or due to risks of L&D should be understood as entitlements to climate resilience to recover swiftly after a climate threat.

Even though this suggestion might seem more demanding than the starting point of this investigation from nonideal theory, it demands no more than the parties agreed in the Paris Agreement. They agreed that in cases of L&D, the areas of cooperation and facilitation may include enhancing “resilience of communities, livelihoods and ecosystems” (Article 8). Thus, the Paris Agreement could still be criticized as dangerous incrementalism that masks the collective failure of international climate governance rather than promoting effective climate action (Allan 2019). The argument in this article shows, however, that it involves much more than incrementalism, at least with respect to L&D governance under the UNFCCC.

Although in Article 8 of the Paris Agreement, the parties recognize “the importance of averting, minimizing and addressing loss and damage,” paragraph 52 of Decision 1/CP.21 states that Article 8 “does not involve or provide a basis for any liability or compensation.” From a conceptual point of view, this decision is unfavorable because usually any harm caused by an agent affecting other parties is believed to demand remedy by the insulting party: those causing harm are seen as liable for providing compensation for that harm (Weinrib 2002). Compensatory justice holds that those harmed should be made whole again by those that have harmed (Page and Heyward 2016). In climate litigation, this kind of reasoning is very common to oblige emitters of greenhouse gases to pay remedy or change their polluting behavior. Even though important in climate litigation, the current decisions in the Paris Agreement block these kinds of considerations. In what follows, I first explain the methodological starting point of this article from nonideal theory. I then illustrate the usual understanding of liability for compensation. The discussion in this section serves as a foil with which subsequent sections can analyze how to deal with the nonideal situation currently existing in UNFCCC governance.

This article is an exercise in nonideal theory because it takes the unfavorable political decisions under the UNFCCC as its starting point. The distinction between ideal and nonideal theory has been promulgated by John Rawls, who distinguishes two kinds of nonideal challenges to realizing ideal justice in a globalized world (Rawls 1999, part 3). The first concerns challenges in realizing justice under partial compliance. The second involves obstacles to realizing justice due to unfavorable conditions in societies. This investigation starts from unfavorable political decisions by considering paragraph 52 of Decision 1/CP.21 to the Paris Agreement. This decision is unfavorable to realizing ideal justice because, in ideal theory, just compensation entails that those causing harm are also those responsible for remedying that harm. Paragraph 52 of Decision 1/CP.21 can also be understood as expressing a lack of compliance by developed country parties to the Paris Agreement to accept their responsibility for causing L&D. Excluding L&D as a basis for liability and compensation makes it difficult for victims to claim their rights under the UNFCCC through mechanisms like the Warsaw International Mechanism (WIM), as is currently the case.

These unfavorable circumstances apply only under the UNFCCC and not in other legal contexts at domestic and international levels outside the United Nations (UN) framework. In these legal contexts, liability for climate change is not excluded and therefore plays an important role in climate litigation. Since the new millennium, the number of lawsuits in relation to climate change relying on liability has grown immensely, especially in the United States (Setzer and Higham 2021; Setzer and Vanhala 2019). While until 2005, there were no more than 20 climate litigation cases globally, their number had grown to nearly 180 by 2020. Only a few of these cases have been accepted as valid before courts, but successful cases often rely on liability for climate change and present their claims as compensation for L&D (Beauregard et al. 2021; Huggel et al. 2016). Even when lawsuits concern inaction in reducing emissions, claimants often rely on current or expected L&D or human rights violations as damages due to climate change, as happened after Hurricane Katrina in the United States or in the case of the climate seniors taking the Swiss government to the European Court of Human Rights for not doing enough to protect their health (Maljean-Dubois 2019; Toussaint 2020).

As in UNFCCC negotiations, to define what is a case of L&D litigation is highly contentious and underlies issues of definition and political strategies (Vanhala and Hestbaek 2016). For example, it is often argued that only cases dealing with L&D after a climate threat has happened are actually cases of L&D litigation. However, actual litigation cases often pursue the prevention of the repetition of past harms and corresponding action as compensation. These claimants do not rely on L&D that has already materialized but pursue forward-looking claims. Furthermore, evidence that attributes singular extreme weather events and slow-onset events to anthropogenic climate change is growing, but human emissions are seldom the only cause of L&D and their magnitude. Higher vulnerability and exposure to climate threats also depend on other confounding factors: social and economic conditions. Whether such confounding factors should become relevant in climate litigation is an open question. To overcome these definitional issues, Toussaint (2020) suggests focusing on the objectives of claimants to identify explicit L&D lawsuits. L&D litigation cases can be identified as such only if claimants explicitly seek compensation for L&D by invoking liability for climate change. As we will see, a similar strategy is helpful to define L&D entitlements in UNFCCC governance.

Liability, or strict liability, in tort law ascribes responsibility for compensation on the basis that the plaintiff has intentionally, negligently, or recklessly caused harm (Couto 2018; Gardner 2015). Establishing such a connection serves to assign corrective duties to make good the harm inflicted. The harming party owes redress to its victim to at least restore the situation of the victim before the harm. This understanding of liability corresponds to the classical understanding of compensatory justice (O. O’Neill 1987; J. O’Neill 2017). According to compensatory justice, causing harm establishes a special relationship between the agent causing the harm and its victim. This relationship justifies special responsibilities of remedy because the party causing harm inflicted an injustice on the harmed party. Making good this injustice means making the victim “whole again” (Goodin 1989; Page and Heyward 2016). The harming party must restore the situation ex ante the harm. What is relevant to defining compensatory entitlements is what the harming party has done and the situation of the victim before the threat. What needs to be restored is this situation. If this is achieved, the harming party has done its due.

Taking compensatory justice as the ideal in cases of L&D means, in ideal theory, that climate lawsuits and mechanisms like the WIM under the UNFCCC must ensure that emitting parties compensate victims by restoring the situation ex ante negative climate impacts. In cases of climate litigation, this means that plaintiffs are entitled to demand remedy from emitting parties for L&D already materialized or expected. In the ideal, this means that domestic and international legal institutions ensure that plaintiffs for L&D are accepted before the various courts (Hinteregger 2017). In cases of the WIM, ideal theory demands that those facing potential L&D receive financial and other nonfinancial support to address incurred and expected negative climate impacts (Robinson and Carlson 2021). In both climate litigation and under the UNFCCC, victims of climate harm would be compensated by those who contribute more to climate change in proportion to their contributions.

However, even in the ideal, climate litigation is not a guarantee of appropriate compensation for all victims of L&D. In climate litigation, ideal theory demands that all domestic and international legal institutions become ready to accept climate claimants as legal parties before the court. This is especially important if claims are cross-boundary and concern either communities residing in various countries or claimants in one country suing plaintiffs in other countries for L&D in their region. Inuit people would profit from a change, according to the first revision of domestic institutions, because their claims are not bound to only one country but concern both Canada and the United States (Heyward 2014). Allowing claimants to sue plaintiffs in other countries could help normalize cases like Lliuya v. RWE, in which a farmer from Huaraz in Peru assisted by Germanwatch sued Germany’s largest electricity producer for contributing to climate change and thereby increasing the risk of a glacier lake outburst flood from Lake Palcacocha (Huggel et al. 2020). However, even if successful, such lawsuits would be limited in their scope, only helping those victims able to sue developed countries or their agents but not all victims of L&D. All victims not able to do so would be left behind.

This is why, even if climate litigation becomes more successful, compensatory mechanisms under the UNFCCC would still be needed to provide support for victims of L&D that cannot afford climate litigation. Because the Paris Agreement blocks liability for compensation, it is currently not possible to pursue this route under the UNFCCC. In what follows, I investigate how to deal with these nonideal circumstances under the UNFCCC to understand how to approach the ideal of compensatory justice in L&D governance without invoking liability for compensation. This is another way of understanding the relation between ideal and nonideal theory. There is an ideal to be achieved, in our case, compensatory justice for L&D in climate governance, but to achieve it, a theory of transitional justice becomes necessary (Sen 2011; Valentini 2012). Such a theory defines how best to achieve the ideal end state. In L&D governance under the current situation in the UNFCCC, such a theory means defining how to ensure that those facing L&D receive their due as well as possible. In other words, such a theory means defining who must provide support for those facing climate threats because the Paris Agreement does not allow these responsibilities to be differentiated by relying on liability and compensation.

The previous section explained how liability and compensation play a key role in climate litigation outside of the UNFCC. Assigning compensatory duties based on liability means invoking compensatory justice. Compensatory justice typically relies on backward-looking responsibility. According to this traditional understanding of responsibility, those who have caused harm by their action should make good that harm. However, under the UNFCCC, using backward-looking considerations to differentiate responsibilities is currently blocked. In this section, I argue that under current nonideal circumstances, forward-looking reasons could serve to differentiate responsibilities in L&D governance under the UNFCCC instead. These ways of assigning responsibilities for L&D all rely on various reasons of capacity and solidarity but not on liability. Hence, a forward-looking understanding of responsibility can serve L&D governance well, despite the decisions in the Paris Agreement.

Responsibility traditionally includes at least two components (Bayertz 1995; Wallimann-Helmer 2019a): one is the agent assigned responsibility; the other is a certain outcome for which the agent is ascribed responsibility. These two components establish a relationship between an agent and a certain state of affairs, for which some kind of causality is usually considered to be crucial. Whether such a causal relation must be direct or can be established more indirectly depends on the circumstances of ascribing responsibility (Miller 2007). Some argue that the causal chain must be strong; others believe that social relations and institutions establish relations between agents and outcomes significant enough to ascribe responsibility (Hattori 2021; Young 2006). And still others believe that benefiting from actions by others that contribute to an outcome is enough to assign responsibility (Page 2012). These assignments of responsibility are morally neutral and could be understood as descriptions of relations between agents and outcomes. These descriptions become morally relevant as soon as the agent is asked to justify why and for what reasons the outcome occurred. If actions or their outcomes are evaluated as negative or have negative side effects, such a relation becomes particularly morally significant. Whether an agent is blameworthy for such negative outcomes depends on whether some moral principle or norm defines the action or the outcome as morally wrong.

All these considerations can directly be transferred to L&D. Emitters, especially industrialized countries, contribute to climate change, and climate change leads to L&D. Probabilistic event attribution aims at showing how the frequency and intensity of extreme weather events like the El Niño phenomenon and slow-onset events like sea level rise or the thawing of permafrost can be attributed to anthropogenic climate change (Huggel et al. 2016; James et al. 2019; James et al. 2014). This research aims to attribute L&D to anthropogenic climate change through cascading effects from changes detected in human and natural systems to changes in the climate system. Most recent detection and attribution research claims that climate litigation would be more successful if it relied on the best scientific evidence available (Stuart-Smith et al. 2021). This evidence would even allow single extreme weather events, such as Hurricane Katrina, to be assigned to the individual contributions of states. These obvious ways of establishing causal relations between emitters as responsibility bearers and L&D make it most plausible to assign remedial responsibilities for L&D for reasons of liability, whether in climate litigation or under UNFCCC governance rules.

Such kinds of attributions of responsibilities for L&D are backward-looking. They ascribe responsibility for causal contribution to climate change or more indirect contributions due to participation in social and institutional structures. These kinds of attributions of responsibility can be termed reasons for outcome responsibility (Miller 2007; Wallimann-Helmer 2019b). Outcome responsibility concerns responsibility for recent action already realized or for action in the more distant past, hence invoking historical responsibility. Similar to liability, outcome responsibility is most often seen as a core reason for becoming responsible for remedy. Among such backward-looking assignments of responsibilities, the polluter-pays and beneficiary-pays principles (PPP and BPP, respectively) are the most prominent principles in climate ethics (Gardiner et al. 2010). According to versions of the PPP, all agents are responsible for remedy in proportion to their polluting activities or other contributions to climate change. According to the BPP, they are responsible in proportion to the benefits they received and receive due to climate change or actions contributing to climate change, irrespective of how much they themselves contributed directly.

However, there are at least two challenges showing why assigning responsibilities beyond backward-looking responsibility and liability should play a role in L&D governance. First, extreme weather events due to climate change already happen now and tend to threaten human rights. Some slow-onset events, such as sea level rise, are already to be expected and will lead to existential losses, such as the disappearance of low island states, with detrimental moral and legal implications (Armstrong and Corbett 2021). When L&D has already materialized or will surely materialize, it is more important to help victims in urgent need immediately than to assign responsibilities according to various contributions to climate change (Wallimann-Helmer et al. 2019). Although current conditions under the UNFCCC allow remedial responsibilities not to be assigned according to liability, victims of L&D in need of assistance need support immediately. Second, purely backward-looking responsibility is limited in scope (Wallimann-Helmer 2015). It covers only anthropogenic climate change but not natural climate variability, which also contributes to climate threats. Hence only part of an urgent situation would be covered by relying solely on remedial responsibility assigned due to the contribution to climate change. In such a situation, one might wonder who should help those in need of assistance for the threats they face due to natural climate variability. Such ascriptions of remedial responsibilities demand other justifications beyond liability and compensation.

These two challenges show why backward-looking responsibility, liability, and compensation should not be taken to cover the whole challenge of L&D. Further reasons for differentiating responsibilities to help communities facing climate threats become crucial. These reasons are all forward-looking; they do not differentiate remedial responsibilities by liability for compensation but for reasons of capacity and solidarity. First, in a situation of urgency, humanitarian reasons justify helping those in desperate need as quickly as possible. What gains priority for differentiating responsibilities are reasons of efficiency. What is relevant in such situations is that those in need of help are helped immediately by those who can do so most efficiently. For differentiating responsibilities, it is differing capacities of agents, such as scientific knowledge or managerial experience, shared cultural ties, or geographical proximity, that become most crucial (Wallimann-Helmer 2016). Second, many communities facing L&D, such as inhabitants of low island developing states and Indigenous peoples, are often highly vulnerable to changing climatic conditions due to no fault or choice of their own. This situation demands redress for reasons of solidarity with those worse off that goes beyond compensatory entitlements (Kolers 2014). Doing only what is owed according to compensatory justice but nothing beyond would mean leaving these victims desperate and vulnerable to the negative effects of climate change (Wallimann-Helmer et al. 2019). They would receive help only to the extent that their desperate situation is anthropogenically induced. They would not be entitled to any help for natural climate variability. Both these reasons seem to justify extending the scope of governance under the UNFCCC beyond changes in climate conditions attributable “directly or indirectly to human activity.”²

In one way or another, such forward-looking arguments justifying remedial responsibility provide reasons of capacity and solidarity to support those facing the threats of L&D. Reasons of capacity are relevant because those in need of assistance need immediate help from those best able to provide it. What is key for differentiating responsibilities are efficiency considerations and not derivations from contributions to threats. This is especially crucial if risks of climate threats can be described as emergencies (Wallimann-Helmer 2016). Humanitarian reasons express solidarity with those in emergency situations. Reasons of solidarity define the entitlements of those undeservingly disadvantaged and in that way are an expression of solidarity with the disadvantaged (Meyer and Roser 2010; Meyer and Sanklecha 2017). They focus on the needs of victims and assign responsibility without backward-looking considerations. There are victims below a certain sufficiency threshold, and someone must help. According to the capabilities approach, everyone is entitled to a sufficient level of well-being, expressed in a list of capabilities that everyone must be able to pursue (Schlosberg 2012; Sen 2011).

These considerations should not be taken as implying that liability and backward-looking responsibility should no longer play any role in L&D governance. In legal contexts outside of the UNFCCC, liability surely will and must play its role in climate litigation. Under the UNFCCC, liability will also have to play its role once the current nonideal situation changes and liability for L&D is accepted as binding by the parties to the UNFCCC. However, these considerations invite further reasons to assign responsibilities for “averting, minimizing and addressing loss and damage” (Article 8). Forward-looking considerations of capacity and solidarity are not excluded under the Paris Agreement. Hence, even though paragraph 52 of Decision 1/CP.21 excludes liability and compensation, remedial responsibilities can still be assigned. Furthermore, due to the urgency of addressing L&D for all communities concerned, especially after extreme weather events and because of the limited scope of liability and climate litigation, forward-looking considerations will have to play a role in any case. However, in addition to leading outside the scope of the UNFCCC, relying on forward-looking reasons to differentiate L&D responsibilities poses two other serious challenges, to which we turn first. Their discussion helps to address the scope challenge in the next section.

The previous section argues that responsibility for remedial action in cases of L&D need not be justified by relying solely on liability and backward-looking responsibility, and such an approach is not appropriate in all occasions. Forward-looking reasons of capacity and solidarity provide further arguments for assigning remedial responsibilities. In the current nonideal circumstances, these reasons could unfold their full validity in L&D governance under the UNFCCC because liability as justification for compensation is blocked. However, this suggestion comes with two serious challenges.³ First, there is a risk that relying on forward-looking responsibility allows the status quo under the UNFCCC to be justified with no additional funding for L&D. Second, forward-looking reasons do not establish the crucial relation between climate change and the L&D to be handled. Discussing these two challenges helps to understand how to define L&D entitlements of those in need of assistance under current nonideal circumstances in UNFCCC governance. This section shows why L&D entitlements have a function leading beyond mere remedy of L&D incurred and hence beyond purely compensatory justice. The goal should be to enable victims of L&D to become able to address all negative climate impacts appropriately.

The first, most fundamental challenge is well known in all areas of UN climate governance (Brown et al. 2010; Michaelowa et al. 2019). It concerns the lack of additionality, that is, the lack of additional support for L&D governance beyond support already provided through other channels (Robinson et al. 2021; Shawoo et al. 2021). Justifying remedial responsibilities for L&D by relying on forward-looking reasons like capacity and solidarity risks justifying the status quo: parties avoid providing additional support for L&D governance by relying on existing humanitarian or foreign aid. No additional funding is dedicated to L&D governance because funding for reasons of capacity and solidarity is already available. In cases requiring humanitarian aid, action is taken by those capable of helping victims out of their desperate situations. Foreign aid provided by developed countries is intended to help developing countries to reach levels of economic development from which they can independently foster appropriate living conditions for their citizens. This challenge stems from the fact that justifications for humanitarian and foreign aid usually rely on reasons of capacity and solidarity. These similarities between arguments allow L&D finance to be easily subsumed into other international finance in solidarity with the disadvantaged and global poor if capacities are available. Relying on forward-looking reasons of capacity and solidarity to justify remedial responsibilities risks bolstering this reluctance to provide additional finance for L&D under UNFCCC governance.

This challenge can be overcome by a closer look at the function of compensation for harm. Classical thinking about harm and responsibility for remedy involves duties for the harming party to restore the situation ex ante the harm. Similarly, in cases of L&D, compensation is most commonly considered as the duty of heavy emitters to rectify the L&D incurred and make the victims whole again. However, it would be possible for any institutional mechanism to restore the status quo ante the harm without relying on the agent causing it. Such schemes already exist. The UN Declaration of Basic Principles of Justice for Victims of Crime and Abuse of Power holds that states should compensate victims if rectification is not fully available from the offender.⁴ Similarly, no-fault legal systems, as in New Zealand, provide compensation for negligently caused harm without requiring claimants to prove that any agent was guilty of causing that harm (Holm et al. 2021). Such schemes assign responsibilities for remedy to the state or relevant institutions to help victims unexpectedly and undeservingly in detrimental situations. In both these cases, reasons of capacity and solidarity define responsibilities to compensation for harm independently of who has caused it. What is relevant to justifying entitlements is the kind of harm incurred but not who has caused the harm in question.

Such schemes narrow the function of remedial action to the goal of making whole again the victims of specific harms. However, such a goal need not mean additional finance or nonfinancial support. Such action could easily be subsumed under humanitarian or foreign aid for victims of undeserved disadvantage. Humanitarian aid would be helping restore the status quo ante after extreme weather events. Foreign aid could be defined as action to deal with slow-onset events of whatever kind leading to L&D. Subsuming compensation under humanitarian or foreign aid without providing additional finance or support is possible because the definition of compensation as making good for L&D avoids the crucial relation between the harm incurred and climate change. L&D can occur in whatever circumstances, and capacity or solidarity can serve as reasons to help everyone in detrimental conditions and in need of support. Hence a definition of the function of compensation in L&D governance that only relies on restoring the status quo ante a climate threat is too general to help overcome the additionality problem.

This leads to the second of the challenges mentioned earlier. The lack of any relation between forward-looking reasons for remedial action and climate change is problematic because making good whatever harm has been suffered for reasons of capacity and solidarity misdescribes the problem at hand (García-Portela 2023). It describes losses and damages from climate change as disadvantages that can occur for any reason. But compensation in L&D governance under the UNFCCC must make good L&D occurring for a specific reason: changing climatic conditions. This is why a definition of L&D entitlements under the UNFCCC must express a specific relation to climate change. Relying on causality is the most obvious way and the one most often invoked by developing countries (James et al. 2019; Maljean-Dubois 2019). Like the PPP and BPP demand, emitters or beneficiaries from climate change should compensate for what they have caused by contributing to climate change or benefitting from it. However, forward-looking reasons to assign responsibilities cannot rely on emitters and the harm they cause to establish the relationship between climate change and harm. Such reasons rely on the capacities and solidarity of agents to provide support due to their advantage. These characteristics of agents have no or only an indirect connection with climate change and could, at least in principle, also have occurred without any liability for L&D.

However, the goal of L&D action under the UNFCCC must not only be to restore the status quo ex ante a threat or to handle the severe negative impacts of extreme weather events once they have materialized. Its function is not only to compensate for L&D. The crucial function of this kind of action is to enhance understanding, action, and support for victims of L&D stemming from climate change. L&D governance is there not only to make good the harms incurred but to enable victims to manage these challenges from a specific source (Mechler et al. 2019; Mechler and Deubelli 2021). This action aims at managing L&D due to climate change that cannot or will not be avoided. Consequently, the goal of L&D action becomes both to minimize the harm occurring and to prepare to redress the harm materializing. This means that L&D governance under the UNFCCC should not only restore the situation ex ante negative climate impacts but enable victims of L&D to address these negative impacts before and after they have occurred. Being able to address climate threats implies possessing a sufficient level of capabilities for doing so. Whether this is given once the situation ex ante an impact is restored depends on the capabilities a community or society possessed before the climate risks increased. It depends on what knowledge and financial and other resources they possess and how well they can codetermine L&D action (Holland 2017; Schlosberg 2012).

Victims of L&D in many cases do not possess such capabilities to a sufficient extent (Intergovernmental Panel on Climate Change 2022; van der Geest and Warner 2020). This is exactly why they need support in L&D action. In cases of economic and noneconomic damage, they often lack financial and other nonfinancial resources to replace the capabilities damaged. In case of economic and noneconomic losses, by contrast, victims often lack the capabilities to transform livelihoods to maintain levels of capabilities similar to those they enjoyed before the climate impact. And very often, both aspects are combined. For example, low island developing states even risk losing their access to deep sea resources because their land above sea level is at risk of vanishing (Armstrong and Corbett 2021; Robinson et al. 2021). In consequence, “averting, minimizing and addressing loss and damage associated with the adverse effects of climate change” (Article 8) can become much more demanding than what liability and compensation would imply. The next section suggests how to define these more demanding entitlements.

According to the argument of the previous section, Article 8 of the Paris Agreement is more demanding than it first appears because respecting the importance of L&D to enhancing “understanding, action and support” can go far beyond their mere remedy and compensation (Article 8). Accordingly, under the UNFCCC, the goals of enhancing “averting, minimizing and addressing loss and damage associated with the adverse effects of climate change” mean more than mere liability for compensation (Article 8). In what follows, I argue that under the current nonideal circumstances in UNFCCC governance, these goals should be defined as entitlements to capabilities for resilience to climate threats. To distinguish L&D action from climate adaptation, I suggest two different conceptualizations of resilience: L&D resilience is the capacity of a system to recover swiftly after L&D materializes; adaptation resilience is the capacity of a system to withstand climate disruption. Even though such a definition of L&D entitlements might appear much costlier than mere compensation for L&D and most probably leads beyond the initial scope of the UNFCCC, I claim this to be an appropriate expression of what the parties to the Paris Agreement promised when agreeing to enhance understanding, action, and support in relation to L&D.

In Article 8 of the Paris Agreement, resilience is mentioned as one of the “areas of cooperation and facilitation.” In its usual meaning, the resilience of a community or system describes its capacity to handle disturbances of any sort (Thorén and Olsson 2018). A resilient system can be described in at least two ways, and its resilience concerns various factors (Ferguson 2019; Joakim et al. 2015). First, a system is resilient if it can absorb or withstand disturbances. In climate governance, this would be what is most often understood as capabilities to adaptation: the capacity of a system to avoid L&D or at least reduce its risk to an acceptable level (Dow et al. 2013). A community that successfully adapts to changing climatic conditions is able to avoid severe L&D or reduce its risks accordingly. Second, a system is resilient if it can recover swiftly after L&D has materialized. This kind of resilience can be identified with L&D action because L&D governance deals with negative climate impacts that cannot or have not been avoided (Verheyen and Roderick 2008). A community resilient in this way is able to handle the negative impacts of L&D even though they materialize with severe consequences to the functioning of the system.

Resilience is usually understood as an empirical concept describing the capacity of a system to handle disruptions. Resilience can become a normative concept if it is defined as the entitlements of communities to a certain threshold of capabilities (Thorén and Olsson 2018). Accordingly, in the case of climate adaptation, resilience represents the entitlement of communities to be able to withstand climate threats. With L&D, resilience represents an entitlement to the capabilities to swiftly recover from L&D materializing after a threat. Both kinds of entitlement can be understood as thresholds of sufficientarian justice because they define minimal entitlements to a certain level of resilience. For climate adaptation, this level can be defined by the capabilities needed for a system to withstand disruptions (Schlosberg 2012). In contrast, L&D entitlements can be defined as the capabilities a system needs to recover swiftly from L&D that cannot or have not been avoided (Wallimann-Helmer et al. 2019). In justice theory and the capabilities approach, such sufficiency thresholds are usually understood as unconditional entitlements of individuals or communities below which they should not fall, independently of how their situation was caused and whether they are responsible for their miserable situation (Casal 2007; Herlitz 2019). The same could hold in climate governance: adaptation and L&D define entitlements to two different thresholds of capabilities to resilience below which no community should fall. Such unconditional thresholds would define entitlements regardless of whether communities are vulnerable due to human activities or natural climate variability, and regardless of how much they themselves have contributed to climate change.

Defining L&D entitlements with sufficiency thresholds of resilience places L&D action quite close to climate adaptation. As Toussaint (2020) suggested for climate litigation, what becomes crucial to distinguish between the two areas of climate policy is the goal of action to be taken. In L&D action, the goal is to deal with the negative impacts of climate change that cannot or will not be avoided. In contrast, the goal of climate adaptation is to take action to keep risks of L&D at an acceptable level or even to avoid such consequences entirely. Hence what becomes crucial to distinguishing between L&D entitlements and adaptation entitlements are the different goals ascribed to each (Wallimann-Helmer 2015). Whether these entitlements can be defined independently of the communities concerned is debated (Wallimann-Helmer et al. 2019). Some argue that various measures can be defined as measures for either L&D or adaptation. Others argue that it must be the communities that assess whether specific measures reduce the risks of L&D sufficiently for them to count as action avoiding L&D (e.g., Holland 2017). However, both define thresholds of entitlements that demand support if victims do not possess these capabilities.

Defining L&D entitlements in such a way as sufficiency thresholds of resilience has the advantage of securing the relation between climate change and the goal of L&D governance. In the case of L&D, resilience is defined as the capabilities of a community to handle climate disruptions leading to L&D that cannot or will not be avoided. Such an entitlement to a threshold of resilience not only means that victims should be compensated for the L&D they suffer; it also means that L&D governance should provide resources to develop the capabilities for dealing with L&D due to climate change before they have occurred (Wallimann-Helmer 2015). Resilience for L&D enables communities to be ready to minimize the consequences of extreme weather events with, for instance, early warning systems and insurance schemes. L&D resilience also covers the development of capabilities to recover after extreme weather events and to establish the infrastructure necessary. If potential victims of L&D lack the capabilities to recover swiftly after negative climate impacts, they are entitled to support. If communities were not resilient before a climate threat materialized, then they are entitled to more than mere compensation. They are entitled to all the support needed for their community to become resilient to manage L&D materializing.

This leads to a final challenge. If L&D entitlements are defined by thresholds of resilience, the implications of Article 8 of the Paris Agreement may become much more demanding and may imply expanding the scope of the UNFCCC. Parties to the agreement not only owe support to ensure compensation for L&D incurred but are likely also required to foster capabilities beyond mere restoration of circumstances ex ante a climate impact. Furthermore, because liability is blocked as a reason for assigning responsibilities, but capacity and solidarity are not, these implications might mean heavier burdens for at least some countries than if they were obliged to pay compensation only according to their contribution to climate change. All this might make it unlikely that such demands will be accepted under the UNFCCC and hence contradict my starting point from nonideal theory.⁵ However, whereas relying on compensation based on liability is blocked by paragraph 52 of Decision 1/CP.21 of the Paris Agreement, resilience is explicitly mentioned as an area of cooperation and facilitation in L&D governance (Article 8). Hence, demanding that the parties to the Paris Agreement provide support for reaching certain thresholds of resilience means merely fulfilling this part of the agreement by both enhancing understanding of resilience and providing action and support. Such a demand simply means taking the signing parties at their word, even if the scope of governance under the UNFCCC has to be expanded beyond anthropogenic climate impacts as a consequence.

In this article, I argue that Article 8 of the Paris Agreement has much more demanding implications than first appearance may suggest. This is the case despite paragraph 52 of Decision 1/CP.21, which blocks liability and compensation as the most plausible rationales for differentiating remedial responsibilities in L&D governance. Even though important in legal contexts outside of the UNFCCC, liability for compensation currently cannot play this role in L&D governance under the UNFCCC. I show that reasons of capacity and solidarity can be employed to differentiate remedial responsibilities in L&D governance. However, arguing from capacity and solidarity to differentiate L&D responsibilities also means going beyond mere compensation for L&D as defining L&D entitlements. Under current, nonideal political circumstances, I suggest thresholds of capabilities sufficient to reach climate resilience as the definition for L&D entitlements under the UNFCCC. Resilience not only secures the link between climate change and L&D entitlements; it also ensures that those in need of assistance receive the support they need to handle any L&D they may face.

I thank Laura García-Portela, Fergus Green, Simon Milligan, and Olivia Serdeczny for very helpful feedback on earlier drafts of this article. I am also grateful to the organizers and participants of the special issue workshop held at the LSE in June 2022 for a very stimulating exchange.