Loss and Damage, Climate Victims, and International Climate Law: Looking Back, Looking Forward

Abstract

After more than three decades of negotiations, the international response to climate change under the United Nations Framework Convention on Climate Change (UNFCCC) appears to have come full circle. At COP27, parties to the UNFCCC agreed to establish a multilateral fund to address loss and damage from global temperature rise, an idea that was initially put forward by the Alliance of Small Island States (AOSIS) in the early 1990s. Employing a historical critique, which draws upon archival and doctrinal research and interviews with key informants who participated in the early days of the Intergovernmental Negotiating Committee for a Framework Convention on Climate Change, this article examines the AOSIS proposal in its wider historical context, and provides reflections for the renewed endeavour to negotiate a multilateral fund on loss and damage, in particular with a view to achieving justice for climate victims.

There should be established, as an integral part of the Framework

Convention on Climate Change, an International Climate Fund to

finance measures to counter the adverse consequences of climate

change, and a separate International Insurance Pool to provide

financial insurance against the consequences of sea level rise. The

resources of the International Insurance Pool should be used to

compensate the most vulnerable small island and low-lying coastal

developing countries for loss and damage resulting from sea level rise.

Vanuatu on behalf of the Alliance of Small Island States,

17 Dec. 1991, paras 1(1) and 1(5) ¹

The Conference of the Parties and the Conference of the Parties

serving as the meeting of the Parties to the Paris Agreement, …

Decide to establish new funding arrangements for assisting

developing countries that are particularly vulnerable to the

adverse effects of climate change, in responding to loss and

damage, including with a focus on addressing loss and damage

by providing and assisting in mobilizing new and additional

resources, and that these new arrangements complement and

include sources, funds, processes and initiatives under and

outside the Convention and the Paris Agreement.

Also decide, in the context of establishing the new funding

arrangements referred to in paragraph 2 above, to establish

a fund for responding to loss and damage whose mandate

includes a focus on addressing loss and damage.

UNFCCC, Decision 2/CP.27, adopted 20 Nov. 2022, paras 2 and 3 ²

1. Introduction

After more than three decades of multilateral negotiations, the international response to climate change under the United Nations Framework Convention on Climate Change (UNFCCC)³ appears to have come full circle: 30 years of failing to solve the problem it is mandated to address, failing to keep global greenhouse gas (GHG) emissions at a level deemed not dangerous for humans and the planet;⁴ 30 years of stalling on meaningful action and support to address the most adverse consequences of climate change, now long inevitable and so frequent and severe that merely helping people to adapt is becoming no longer an option;⁵ 30 years of undermining the demands for justice by climate victims from the most vulnerable communities who bear the brunt of adverse climate impacts. The world is now in an era of loss and damage – an ambiguous policy term,⁶ which essentially denotes that after three decades of failing to contain the problem, we are now dealing with the consequences where questions of reparative justice, long subsumed but never forgotten, are coming back to the fore.

Those old enough to remember, and those eager enough to research, will recall that back in 1991, when states convened under the Intergovernmental Negotiating Committee to negotiate what would become the UNFCCC, Vanuatu, on behalf of the Alliance of Small Island States (AOSIS), tabled a proposal for an International Insurance Pool to address climate loss and damage, specifically sea-level rise.⁷ While unsuccessful, this proposal has often been credited as the birthing moment of loss and damage under the climate regime.⁸ Over 30 years later, at the 27^th meeting of the Conference of the Parties (COP27) held in Sharm el-Sheikh (Egypt) at the end of 2022, UNFCCC parties breathed new life into the idea by agreeing to establish funding arrangements, including an international fund, to address loss and damage.⁹

This article takes a closer look at the content, rationale, and negotiation context of the AOSIS proposal in order to explore what can be learnt from it for the development of the new loss and damage fund. For this purpose, I employ a historical critique, which draws upon archival and doctrinal research, secondary literature, and interviews with key informants who participated in the INC meetings.¹⁰ This historical approach accepts Orford's invitation for international legal scholars to historicize international law with a clear purpose and relevance for political engagement.¹¹ Rather than conducting neoformalist, technical historiography, Orford calls on international legal scholars to ‘take responsibility for actively constructing accounts of the law's past when we argue about law in the present’.¹²

Legal scholars, policy analysts, and civil society organizations are racing to provide their views on what the new loss and damage fund should look like and how it should operate. This article offers a brief respite, an invitation to pause and reflect, to look backward as we look forward. Connecting the present with the past allows us to contextualize key aspects and arguments in the ongoing policy discussions related to loss and damage finance. More specifically, the INC context has been chosen as it represents a critical moment in the formation of the UNFCCC regime. The AOSIS proposal was selected both because it is the most cited point of origin of loss and damage and because of its topicality for the new fund. Heeding Orford's warnings, the aim is not to attempt an impossibly neutral or objective historiography of loss and damage devoid of purpose. Rather, the analysis presented in this article provides an opportunity to take stock of how far the international community has come in addressing loss and damage. It provides several reflections for the renewed endeavour to negotiate a multilateral fund, in particular with a view to achieving justice for climate victims.

The remainder of this article is structured as follows. Section 2 provides an overview of the present state of the international policy response to loss and damage. Section 3 examines the AOSIS proposal in its wider historical context, and provides reflections for the renewed endeavour to negotiate a multilateral fund on loss and damage, in particular, with a view to achieving justice for climate victims. Finally, Section 4 offers some conclusions.

2. The Loss and Damage Policy Response: Where Do We Stand?

Before exploring how we originally came to be speaking of loss and damage under the UNFCCC, it is useful to understand where we are in terms of loss and damage policy today. To construe the present moment of international loss and damage policy I will illuminate its content, structures, and current sites of contestation. To date, UNFCCC parties have not agreed on an official definition of loss and damage, leaving ambiguity as to its precise scope and delimitation from adaptation. While some understand loss and damage as relating to adverse impacts of climate change that are beyond the limits of adaptation¹³ or the residual impacts of climate change, others hold the view that it falls within the scope of adaptation.¹⁴ Furthermore, loss and damage is included in the Paris Agreement¹⁵ through Article 8, but subject to a COP decision that excludes liability and compensation as flowing from this provision (under the infamous paragraph 51).¹⁶

Institutionally, the mandate to address loss and damage at the multilateral level lies chiefly under the umbrella of the UNFCCC regime. It is articulated primarily through the Warsaw International Mechanism (WIM), a technical sub-process established under the Convention in 2014.¹⁷ The WIM has three overarching mandates: (i) to enhance knowledge and understanding; (ii) to strengthen dialogue and coordination; and (iii) to enhance action and support, including finance, technology and capacity building.¹⁸ It is governed by a regionally balanced Executive Committee (Excom), which meets twice a year with participation from observers, and reports annually to the COP. The WIM has been subject to two reviews (2016 and 2019) with a third planned in 2024, and has started to implement its five-year workplan for 2023–2027.¹⁹

In addition to the WIM, several sessions related to its work have taken place under the UNFCCC subsidiary bodies, including the Suva Expert Dialogue in 2017. This dialogue was intended to identify sources of and mobilize financial support for vulnerable countries suffering loss and damage. At COP25, the Santiago Network for averting, minimizing, and addressing loss and damage associated with the adverse effects of climate change was created as a platform to facilitate access to technical assistance for vulnerable developing countries to address loss and damage.²⁰ At COP26, momentum was kept up through the first Glasgow Dialogue, bringing together parties and stakeholders to discuss funding arrangements, although many developing countries and observers were disappointed with the lack of concrete outcomes from the dialogue.²¹ Moreover, a proposal from 134 developing countries convening under the G77 and China to establish a loss and damage finance facility at COP26 was rejected.²²

Between the Glasgow (Scotland) and Sharm El-Sheikh COPs, a broad coalition of developing countries and civil society heavily advocated the inclusion of loss and damage finance as a formal item on the agenda of the COP and CMA. Following a proposal from the G77 and China during SB56 in June 2022, lengthy formal and informal consultations in the interim period and long nights in the days preceding COP27, parties agreed on the first day of the Conference to formally adopt the following sub-item under Agenda item 8f: ‘Matters relating to funding arrangements responding to loss and damage associated with the adverse effects of climate change, including a focus on addressing loss and damage’.²³ Its adoption was subject to several caveats, read out by COP27 President Shoukry when introducing the text in plenary. The most relevant of these for the purposes of this analysis is that the agenda item would be ‘based on cooperation and facilitation and [would] not involve liability or compensation’.²⁴ This caveat demonstrated that the spirit of paragraph 51 remains influential nearly a decade since its adoption.

The major historical achievement, in policy terms, was that at COP27 industrialized and developing countries agreed to establish ‘new funding arrangements’ to assist developing countries that are particularly vulnerable to the adverse effects of climate change in addressing loss and damage, and ‘a fund’ to support the most vulnerable countries already incurring loss and damage.²⁵ The text of the decision – in particular, the references to both funding arrangements and a fund, and the repetitive language around the ‘focus on addressing loss and damage’ – may seem overkill. However, they are the products of lengthy negotiations preceding COP27 and reflect a compromise between demands for concrete language on the new fund and considering other options besides a fund to finance support for loss and damage.²⁶ The COP established and mandated a Transitional Committee to provide recommendations on the operationalization of the new funding arrangements, including the fund, for consideration and adoption at COP28. The details of the fund are yet to be negotiated, including the sources of funding, access and pay-out modalities, and the governance of the fund. This process is expected to culminate in an outcome in 2024. UNFCCC parties are also expected to agree on a new collective quantified goal on climate finance prior to 2025, from a floor of US$ 100 billion per year, taking into account the needs and priorities of developing countries.²⁷ Several developing countries are advocating a sub-goal on loss and damage finance. Whether this will be politically feasible and what relationship such a sub-goal would have with the new loss and damage fund remains to be seen.

3. Unpacking the AOSIS Proposal

Most scholarship on the topic of climate loss and damage traces the origins of the concept back to the early 1990s when the UNFCCC regime was being negotiated under the auspices of an Intergovernmental Negotiating Committee (INC). The majority of literature reviewed identifies a proposal submitted by AOSIS ahead of the fourth INC meeting, held in December 1991 in Geneva (Switzerland), as the birthing moment of loss and damage under the UNFCCC.²⁸ In this proposal AOSIS suggested creating an International Insurance Pool, a funding mechanism paid for by industrialized countries to compensate small islands and low-lying coastal developing countries for loss and damage resulting specifically from sea-level rise. The proposal was modelled on the nuclear liability conventions and funds,²⁹ and was further inspired by the oil spills regime.³⁰ While it survived into the draft consolidated texts at INC4 and INC5, following strong resistance from industrialized countries it was omitted from the final text of the Convention adopted in 1992, alongside other detailed annexes.³¹

The idea for an International Insurance Pool should be read in conjunction with a proposal made by Vanuatu on behalf of AOSIS, in June 1991 before INC2, to create a Climate Fund financed by developed countries ‘to compensate developing countries (i) in situations where selecting the least climate sensitive development option involves incurring additional expense; and (ii) where insurance is not available for damage resulting from climate change’.³² That proposal emerged as the AOSIS response to competing ideas among countries about the scope and purpose of a possible climate fund. Where some delegations preferred the fund to focus on supporting developing nations’ mitigation or adaptation efforts, the AOSIS fund proposal clearly focuses on compensating damage. Ultimately, like the Insurance Pool, the proposal failed to gain traction, primarily because of heavy opposition from industrialized countries.

The drafting of the AOSIS Insurance Pool proposal has been credited largely to maritime and insurance lawyer Michael Wilford, who at the time worked as consultant for the Foundation of International Environmental Law and Development (FIELD), a public interest organization that provided pro bono legal expertise and represented AOSIS in the negotiations.³³ Although it was not possible to interview Wilford for this research, his written work at the time provides unique insights into the thinking behind and the framing of the proposal.³⁴ In 1993, Wilford contributed a chapter on insuring against sea-level rise in a book by Hayes and Smith³⁵ and an article in Environment,³⁶ in which he essentially lays out the rationale behind the International Insurance Pool (which is also appended to the chapter).

Several elements of the AOSIS proposal stand out that are particularly relevant to the present effort to develop a loss and damage fund. In the following I offer six reflections to explore what can be learned from the 1991 proposal for the development of the new fund. These reflections take a closer look at the content and rationale of the proposal, situating it, where relevant, in its wider historical context. Ultimately, these reflections offer an opportunity to take stock of how far the international community has come in addressing loss and damage, and may inform the renewed endeavour to negotiate a multilateral fund.

3.1. Loss and Damage: A Reality, not a Distant Threat

The Wilford pieces demonstrate that even in the early 1990s loss and damage was not some abstract intellectual term or mere placeholder for future harm, but was based on scientific projections and on the lived experiences of natural disasters, which even at the time had seen an increase in frequency and severity.³⁷ This is reinforced by the fact that small islands had early access to scientific information on climate change in the late 1980s and successfully secured their participation in the Intergovernmental Panel on Climate Change (IPCC) process towards the First Assessment Report released in 1990. However, historical data comparable with that of industrialized countries was not widely available for most developing countries.³⁸ Consequently, in developing the proposal, the AOSIS team relied heavily on projections and estimates from the IPCC and the insurance industry.³⁹

For the purposes of the proposal, the AOSIS strategy was to focus on sea-level rise and its associated impacts, such as storm surges and inundations, which, according to Wilford was the ‘more easily measurable indicator’.⁴⁰ This is not to say that AOSIS turned a blind eye to other types of climate impact in 1991. Wilford, for example, suggested that a similar, parallel compensation scheme could be developed for countries most vulnerable to desertification and droughts.⁴¹ Similarly, a key informant working with AOSIS at the time noted that sea-level rise ‘was the most clearly articulated consequence at that time. And obviously it goes with other physical consequences. We were particularly focused on it because of who we were working with, and … it provided quite a good visual image of vulnerability’.⁴² The fund would only pay out ten years after the Framework Convention was adopted and only when both the rate and absolute level of mean sea-level rise reached agreed figures. According to Linnerooth-Bayer, Mace and Verheyen, this period was chosen to allow time for attribution science to advance sufficiently to successfully assess such claims in the future.⁴³ It was estimated that ‘no claim might arise for several decades, even on a “business as usual” scenario’.⁴⁴ While these circumstances applied to sea-level rise, the same would not have worked for other, more immediate types of loss and damage already being experienced by climate victims at the time. The focus on sea-level rise also allowed the proposal to be forward-looking in its calculation of damage in that the valuation of assets was to take place ex ante before an actual claim against the Insurance Pool would arise.⁴⁵ Here, the proposal distinguished between retreat (such as abandoning an area), accommodation (such as elevating buildings, shifting to flood-resistant crops), and protection (such as seawalls and vegetation). The first two would be treated as losses that could be covered by the Insurance Pool, whereas protection was deemed a matter of adaptation and thus to be covered under the proposed Climate Fund.⁴⁶ Thus, quite early on a distinction was drawn between adaptation and loss and damage, which was understood to exceed the limits of adaptation.

Fast-forward 30 years and both the science and policy work on loss and damage under the UNFCCC have advanced significantly. We now have a more solid understanding of the different types of loss and damage and their adverse impacts. On the policy side, there is still disagreement on the formal political definition of loss and damage, which could hamper the operationalization of the new fund.⁴⁷ The funding arrangements, including the fund to be operationalized in 2024, are expected to encompass the whole suite of extreme weather events and slow-onset events, with many calling for the inclusion of non-economic loss and damage (discussed further at Section 3.5).

Crucially, attribution science has evolved to a great extent, with probabilistic ‘extreme event attribution’ (EEA) making it possible to draw connections that advocates could only dream of in 1991.⁴⁸ Of course, attribution science is no silver bullet and recent studies have highlighted the importance of assessing vulnerabilities before attributing damage to climate change.⁴⁹ This is because pre-existing vulnerabilities (and their root causes) are key determinants for the scale and types of loss and damage experienced. Attributing damage to climate change alone risks producing inadequate responses that fail to consider underlying causes and contributes to the diffusion of responsibility.⁵⁰ Moreover, EEA has been met with mixed levels of acceptance among policymakers within the UNFCCC, which dampen prospects of relying on this approach for the new funding arrangements.⁵¹ After all, proponents of EEA would need to overcome the same political hurdles that have made the task of attributing responsibility so difficult over the past 30-plus years.⁵² Jackson and co-authors further highlight the danger of linking EEA to loss and damage, which could lead to attribution evidence becoming a ‘prerequisite for finance or compensation’ for such losses.⁵³ Even as the first loss and damage-related cases succeed (see Section 3.5 for a discussion of the Torres Strait Islanders petition), the role and limits of attribution science in such cases remain subject to ongoing research.

3.2. Private Finance and Insurance Cannot Be a Substitute for Public Funding

The AOSIS proposal did not envisage private/commercial insurance as a multilateral solution for loss and damage.⁵⁴ Wilford argued that the inevitable consequence of sea-level rise and the increased likelihood of associated adverse impacts such as inundation (will) constitute a new normal that would render traditional risk insurance approaches unsuitable, and potentially unprofitable for insurance and reinsurance companies.⁵⁵ This is understandable as, compared with sudden-onset events, sea-level rise is both gradual and predictable, and thus anathema to insurable risks. The proposal leaves a gap in the door by encouraging claimant countries seeking to claim against the International Insurance Pool to arrange commercial insurance on their own. At the same time, the proposal makes it clear that loss and damage to commercially insured property would not be recoverable through the Pool.⁵⁶

Reflecting broadly on the insurability of losses, Wilford concludes that ‘any state compensation regime will have to be internationally funded’.⁵⁷ The AOSIS proposal thus envisaged public funding from industrialized countries as the primary vehicle to deliver loss and damage finance. The proposal specified that the financial burden ‘shall be distributed in an equitable manner amongst the industrialized developed countries’.⁵⁸ Their contribution would be levied at 50% based on gross national product and 50% based on total carbon dioxide (CO₂) emissions in the year prior to their contribution. This formula was modelled on the 1963 Brussels Supplementary Convention on Third Party Liability in the field of Nuclear Energy.⁵⁹ Although the AOSIS proposal specifically called on industrialized countries to pay, in line with a polluter-pays approach, it did not consider their historical emissions as part of calculating their contribution – the base year being that prior to the year of contribution. Neither Wilford nor other AOSIS texts provide an explicit reason for considering only (then) present emissions levels. It is tempting to assume that this was a strategic choice, allowing AOSIS to side-step controversial discussions around apportioning historic responsibility. More plausibly, the proposal simply follows the approach of the International Oil Pollution Compensation Fund, which calculated annual contributions by contracting states based on oil quantities received in the calendar year preceding an incident (provided that state was a party at that date).⁶⁰

Against the AOSIS approach, the modern discourse on loss and damage finance has increasingly shifted to exploring private finance and private insurance models as part of the multilateral solution, downplaying the role of public finance.⁶¹ Proponents of index-based insurance argue that it provides an important tool for managing risks proactively rather than reacting ex post, reducing unpredictable financial costs on the insured, thereby providing a safety net to absorb shocks through timely and reliable payouts.⁶² However, they also acknowledge the limits of insurance in dealing with high-frequency extreme weather events as well as high-certainty, foreseeable slow-onset events (such as loss of glaciers).⁶³ Another problem with insurance is that payouts tend to be vastly insufficient to cover the direct costs of loss and damage (let alone indirect costs). For example, the government of Antigua and Barbuda received US$ 6.79 million from the Caribbean Catastrophe Risk Insurance Facility, equivalent to only 3% of recovery costs in Barbuda resulting from Hurricane Irma; while Malawi received US$ 8.1 million from the Africa Risk Capacity scheme (equivalent to 2.2% of US$ 365.9 million in estimated economic losses).⁶⁴

In developing its recommendations to operationalize loss and damage funding arrangements, including the new fund, by COP28, the Transitional Committee has a clear mandate to consider innovative sources of finance.⁶⁵ A recent study found that several Committee members from both the global north and south recommended use of innovative sources such as taxes and levies.⁶⁶ Similarly, a number of civil society organizations called on parties to include innovative sources of funding to cover the likely shortfall in public finance.⁶⁷ Some of the concrete ideas proposed, which are based on the polluter-pays and/or beneficiary-pays principle, include levies on air passengers and on emissions from international shipping, redirected fossil fuel subsidies, as well as debt cancellation.⁶⁸ Other innovative sources include cross-border carbon adjustments and a levy on financial transactions. Another idea tabled is a global loss and damage tax, requiring oil and gas companies to contribute a share of their profits to the fund.⁶⁹

In principle, support from any source should be welcomed to help in remediating the problem created by industrialized countries and exacerbated by market-driven approaches. However, an over-reliance on private finance and private insurance marks a further step in the diffusion of responsibility for loss and damage. Insurance approaches, in particular, have been criticized for passing on the costs of addressing loss and damage to climate victims. This essentially requires affected communities to pay a premium they cannot afford; or, in the words of Avinash Persaud, Special Envoy to the Prime Minister of Barbados on Climate Finance, ‘it is victim pays, just in instalments’.⁷⁰ This view has been expressed time and time again by civil society organizations participating as observers in the UNFCCC. In their recent submissions to the Transitional Committee, for example, they categorically state that ‘private sector solutions that prioritise profit making and require vulnerable people to pay premiums, such as insurance, have no place in addressing loss and damage in an equitable way’.⁷¹

One proposed solution to this dilemma would be to subsidize premiums, which appears to be the preferred mode of some industrialized countries to tackle loss and damage finance. In their review of active index-based insurance schemes, Nordlander, Pill and Martinez Romera found that private insurance cannot replace finance as a primary tool in responding to loss and damage.⁷² They further argue that private insurance runs counter to equity principles of the UNFCCC and might distract from sourcing new and additional finance for loss and damage. Recent initiatives such as the Global Shield (formerly InsuResilience Partnership), launched by the G7 and the V20 (Vulnerable Twenty) countries during COP27⁷³ and its Solutions Platform launched in June 2023,⁷⁴ are being billed as a one-stop shop for climate risk finance and insurance solutions. The Global Shield enjoys widespread support among industrialized countries as the preferred vehicle for providing financial support for loss and damage – with 65% of the US$ 300 million in voluntary pledges being committed towards the Shield.⁷⁵ Established outside the UNFCCC, the initiative is not bound by equity principles of the Convention – such as the polluter-pays principle and common but differentiated responsibilities and respective capabilities (CBDR-RC) – and is not subject to oversight and review by the COP. From the perspective of climate victims, the initiative may be criticized as yet another link in a long chain of delay and distraction tactics by industrialized countries to shield themselves from responsibility for the real costs of loss and damage.⁷⁶

3.3. Liability Remains Off-Limits

While the AOSIS proposal was modelled on conventions relating to liability for nuclear and oil spills, the proposal itself did not expressly seek to establish liability for loss and damage.⁷⁷ This is important as traditionally liability and compensation regimes have been designed to allow affected countries to claim the costs of responding to a major oil spill or nuclear disaster from a strictly liable public or private entity, thus enabling them to avoid the burden of proving fault and collecting compensation.⁷⁸ Both the oil spill and nuclear conventions are multi-tier regimes, meaning they impose strict liability backed by further tiers of compensation to cover losses and damage that exceed agreed limits of liability. The AOSIS proposal essentially adopts the Insurance Pool tier, which acts as a collective loss-sharing arrangement without imposing liability.⁷⁹ This would have been a strategic choice given the strong opposition from the US and other industrialized countries to agreeing to any form of liability and compensation mechanism forming part of the climate regime. In his explanatory texts, Wilford explicitly rules out relying on ‘ordinary legal criteria of liability or responsibility’ under the Convention because of the difficulty of attributing causality when it comes to loss and damage from climate change. Rather, he argues, any loss and damage mechanism, such as that proposed by AOSIS, would have to be based on a broader criterion of responsibility.⁸⁰ The answer, according to Wilford, thus lies in the very nature of the proposal, namely risk pooling. Conveniently, rather than finger pointing, risk pooling diffuses responsibility for climate harm among a broader category of industrialized countries that provide funding for the Pool – and can operate even without an express acknowledgement of responsibility. As will be explored below, this diffusion and obscuration of responsibility has important ramifications for the prospect of justice for climate victims.

It would be wrong, however, to assume that the idea of liability and compensation was prima facie off the table in the early negotiation phase. By design, the INC was convened as a process to give expression to the full range of committee members’ expectations for the design and content of the new climate agreement. Over the course of the various INC sessions, it became clear which elements would stick and which were unlikely to garner enough backing to survive into the final text for adoption in Rio. At the very first INC session, the UNEP Executive Director at the time, Mostafa Tolba, emphasized that decision-making processes under the Convention ‘must be based on equity between North and South’ and that the negotiations should address pivotal issues such as ‘liability and compensation’.⁸¹ Several delegations highlighted the relevance of the polluter-pays principle as an appropriate legal framework in this regard.⁸² The reliance, in particular, by vulnerable countries on the polluter-pays principle and their attempts to integrate provisions on liability and compensation into the climate regime can be seen as part of a broader strategy to ground moral justice and equity arguments in the language and practice of international law. The support that AOSIS received from legal consultants, including those working for CIEL/FIELD, appears to have been a major contributing factor to this trend.⁸³ As one informant noted:

We were well-qualified lawyers from respected institutions and we weren't intimidated by other lawyers. … A lot of the bullying that gets done by larger powers against smaller powers gets done through having more people, and by having people who can talk in a very authoritative way about the law. And none of that made much difference to me. … Also it was a new space. Why should someone older know more than us.⁸⁴

In the same spirit, AOSIS argued for the inclusion of a disclaimer to indicate that the treaty would be ‘without prejudice to the existing rights under international law, including rules governing international liability for damage to people, property and the environment’.⁸⁵ While it did not make it into the final text adopted in Rio, several small island states reiterated this reservation in declarations upon signing the Convention.

Throughout the INC, industrialized countries maintained their strong opposition to accommodating any notion of responsibility, liability or compensation, and remained unswayed both by moral and formalistic legal arguments.⁸⁶ With the primary goal of securing participation of the US in the future climate regime as the then largest GHG emitter, developing countries were slowly forced to soften their stance on responsibility. A G77 preparatory meeting was held in Kuala Lumpur (Malaysia) in early 1992, ahead of the meeting of the Extended Bureau, where G77 countries led by India, China and Brazil agreed to moderate their position around ‘guilt’ and ‘historical responsibility’.⁸⁷ This led to the dropping of a paragraph that had included the polluter-pays principle and compensation obligations on the North.⁸⁸ In essence, the United States (US) was willing to compromise by allowing the ‘Principles’ section of the (then) draft Rio Declaration to be used as a basis for the Principles section of the Convention – despite denouncing it as non-binding soft law. Far from a clean copy-and-paste, the 27 draft Rio principles were reduced to a mere five. Importantly, references to ‘different contributions to global environmental degradation’ under Principle 7 – which implied responsibility of industrialized countries – were removed, while the aspect of common but differentiated responsibilities (CBDR) survived. Until the very last days of negotiations, the US continued to oppose any implications of historical responsibility.⁸⁹ As a direct result of this effort, even the surviving reference in the Convention that calls on developed countries to take the lead in combating climate change and its adverse effects was left without its causal corollary: that developed countries would do so because of their ‘large share of global emissions of greenhouse gases’.⁹⁰ Consequently, the matter was transformed from one of moral responsibility into a narrative based exclusively on capacity and goodwill. As one key informant noted:

The concept of responsibility is actually in the Convention, or is more in the guise of industrialised country leadership. So a positive rather than a negative, which again lip service was paid to. But there was a strong determination to avoid anything to do with liability which we see up until now.⁹¹

What could be interpreted as a classic case of constructive ambiguity that helped to broker the compromise to save the treaty's adoption amounted to a gutting defeat from the perspective of climate victims.⁹²

Fast-forward 30 years and the political picture remains largely unchanged. Most developed countries continue to oppose any references to liability and compensation. At COP21, at which the Paris Agreement was adopted, this opposition culminated in the now infamous paragraph 51 of the accompanying COP decision. Paragraph 51 expressly excludes liability and compensation in relation to Article 8 of the treaty. Negotiators remain divided over the question of whether the new funding arrangements and fund should be governed by the UNFCCC COP or the Paris Agreement's CMA or both. This seemingly technical question has important ramifications, since putting the fund exclusively under the direction and control of the CMA would potentially subject it to the exclusion clause under paragraph 51 while also limiting the applicability of Convention principles such as CBDR-RC.⁹³

3.4. Climate Reparations Appear Out of Reach

Neither the AOSIS proposal nor Wilford's explanatory texts on the proposal refer to how the legacy of imperialism contributed to the vulnerability of climate victims.⁹⁴ It could be argued that directly referencing colonialism was deemed too politically sensitive for a nascent multilateral process based on decision making by consensus. Rather, the AOSIS strategy, as was that of several other vulnerable countries, was to couch it in the more politically palatable terms of ‘historical responsibility’ (which remains the subject of debate⁹⁵). At the second INC session, Vanuatu suggested, for example, that ‘responsibility for the problem lies historically with industrialized countries’.⁹⁶ This reference survived into final text of the Convention, which recognizes that the ‘largest share of historical and [then] current global emissions of greenhouse gases has originated in developed countries’, and calls on parties to protect the climate system ‘on the basis of equity and in accordance with their common but differentiated responsibilities and respective capabilities’, with developed countries taking the lead.⁹⁷ The distinction is significant, because historical responsibility for historical emissions or atmospheric appropriation is only one side of the coin. The other crucial aspect involves acknowledging how imperialist expansion and its subjugation of peoples and exploitation of natural resources left generations of affected communities vulnerable to the ravages of climate change, and rendered them susceptible to loss and damage in the first place. Somewhat less explicitly, several INC delegations invoked arguments over unequal international economic relations following decolonization and criticizing sovereign debt.⁹⁸ Weinger argues that by subsuming stronger arguments into the more palatable term of historical responsibility, the consensus process of the INC negotiations effectively ‘sanitized’ and ‘sterilized’ more overt calls for responsibility and compensation for ‘environmental damage’.⁹⁹

Today's climate justice advocates appear to be significantly more vocal both about the role of colonialism as one of the root causes of the climate crisis, and the ways in which false solutions to the crisis may constitute a form of ‘carbon colonialism’.¹⁰⁰ They identify as false solutions carbon trading and carbon offsetting strategies, including pledges towards ‘net zero’ emissions, geoengineering, bioenergy and carbon and capture and storage (BECCS), and nature-based solutions (NbS), arguing that these amount to a distraction and, worse, serve as an excuse ‘allowing for continued emissions and profit generation from fossil fuel extraction’.¹⁰¹ Reflecting on his participation at COP27, Chief Ninawa Huni Kui, hereditary Chief and elected President of the Huni Kui People of Acre in the Amazonas region in Brazil, remarked that ‘the vast majority of the discussions reproduce colonial patterns of unsustainable economic growth, ecological destruction and Indigenous dispossession that have been responsible for climate destabilization in the first place’.¹⁰² In an impactful speech at the Leaders Summit at COP27, Mia Mottley, Prime Minister of Barbados, aptly stressed that ‘this world looks still too much like it did when it was part of an imperialistic empire. … we believe that the multilateral development banks have to reform. Yes, it is time for us to revisit Bretton Woods’.¹⁰³

The role of colonialism in exacerbating vulnerability has recently found recognition in multilateral settings outside the UNFCCC. In its sixth and most recent assessment report, the IPCC refers explicitly to ‘historical and ongoing patterns of inequity such as colonialism’ as a key factor that is exacerbating vulnerability to climate change. Similarly, Tendayi Achiume (UN Special Rapporteur on contemporary forms of racism, racial discrimination, xenophobia and related intolerance) recently highlighted how ‘the global South and colonially designated non-white regions of the world … are most affected and least able to mitigate and survive global ecological crisis, in significant part owing to the colonial processes that caused historical emissions in the first place’.¹⁰⁴ Reflecting on the UNFCCC response to loss and damage since the adoption of the Paris Agreement, the Special Rapporteur observed a ‘transition away from confronting historical responsibility’.¹⁰⁵

Even if the UNFCCC parties succeed in operationalizing funding arrangements, including the new loss and damage fund, it is extremely unlikely that industrialized countries would agree to basing any of its provisions on compensation or reparation terms (let alone liability). Though somewhat symbolic, the adoption of paragraph 51 marked the final nail in the coffin on the admissibility of liability aspirations (at least formally, under Article 8 of the Paris Agreement).¹⁰⁶ When pressed about potential obligations arising from the new loss and damage fund during a recent appearance before Congress, US Climate Envoy John Kerry stated unequivocally that the US would not pay climate reparations.¹⁰⁷ Paradoxically, the further the UNFCCC policy response to loss and damage ‘progressed’ by setting up sub-process after sub-process, the further it moved away from the prospect of compensation or reparations. Opposition from industrialized countries, including the US, has led some advocates to adopt a more moderated stance that reframes loss and damage finance in terms of international solidarity. Avinash Persaud, Special Envoy on Climate Finance to the Prime Minister Mottley of Barbados, cautioned against an ‘unhelpful conflation’, suggesting that ‘reparations imply payment for past deeds’ whereas ‘the loss and damage fund finances a resilient recovery’.¹⁰⁸

Contrast this with the demand for climate reparations stipulated by international legal scholar Burkett in 2009. According to Burkett, to be successful, climate reparations must satisfy three elements: (i) an apology, which is essentially an acknowledgement of wrongdoing; (ii) a compensatory award (monetary or other) to give real or symbolic weight to that apology; and (iii) a guarantee by the perpetrator of non-repetition of the offending act.¹⁰⁹ Climate reparations can be understood as a deliberative process of healing that looks backward to address past harm and forward by addressing the ongoing injustices stemming from that past harm. Importantly, climate reparations entail a paradigmatic shift as they are based on moral argument and place the focus on climate victims.¹¹⁰ Building on Burkett's work, Riley-Case and Dehm highlight the potential of climate reparations to offer very different, and potentially transformative outcomes that allow us to think outside the box of existing legal doctrines.¹¹¹ I would further argue that climate reparations are particularly seductive as a concept because they allow one to break free from the confines of reified legal processes and institutions. At the same time, they ground legal responses to climate change in the lived experiences of climate victims. Only by engaging meaningfully with the material reality of loss and damage on the ground can we as international lawyers shift from treating the symptoms to addressing the root causes of the injustices that our legal tools have been hitherto unable to rectify. This may appear somewhat utopian, but I align myself with Riley-Case and Dehm, who remark that it is ‘no less utopian than expecting an International Court of Justice opinion or judgment to meaningfully address the problem at hand’.¹¹²

There is, of course, a legitimate question to be asked: Does there need to be a reference to the colonial past and to compensation or reparations? Would the new fund/funding arrangements not serve the same purpose without referring to either? Arguably, whether by any other name, a fund without an admission of wrongdoing remains exactly that – a fund. The moral reasons of why polluters should pay into the fund would thus be obscured and risk being co-opted into a narrative of greater capacity, goodwill, international solidarity or, more specifically, humanitarian relief. Setting aside their framing, it is doubtful that the scale of contributions into the new loss and damage fund will be sufficient to address loss and damage globally.¹¹³ The low track record on adaptation finance, complex access modalities under the Adaptation Fund and Green Climate Fund, as well as the unfulfilled US$ 100 billion per year target for mitigation finance by 2020, have sown mistrust among vulnerable developing countries that a new loss and damage fund will deliver.¹¹⁴ Some may fear a repeat in the negotiations of the new collective quantified climate finance goal to be agreed under the UNFCCC next year (‘prior to 2025’). This is further reinforced by the fact that, at the time of writing, most loss and damage finance pledges have gone to the Global Shield. Although there is the possibility that the Global Shield will find recognition among the funding arrangements to be developed, for the time being, its governance arrangements and relationship with the UNFCCC remain uncertain.

3.5. Still No Solution for Non-Economic Loss and Damage

The AOSIS proposal struggled to account for what in the modern discourse is known as non-economic losses (NELs). NELs essentially denote losses that cannot be valued in purely economic terms, ranging from loss of cultures, traditional knowledge, personal livelihoods and territory to loss of life, among many others.¹¹⁵ The proposal acknowledged that such losses can ‘in no way’ be adequately reflected in traditional insurance definitions, giving the examples of loss of homeland and development potential.¹¹⁶ AOSIS proposed that any ‘non-marketed interests shall be valued on the basis of formulae to be agreed’ and that countries should be allowed to consider also the existence and option value of their assets.¹¹⁷ While this would allow for a more realistic valuation of assets to be compensated under the proposed Insurance Pool, it still did not provide a satisfactory answer to valuing NELs. Here the proposal deferred to individual negotiations between the Authority – a body administering claims against the Pool – and the claiming country. This notwithstanding, the proposal was quite forward-thinking for including both human and ecological loss and damage.¹¹⁸

Shifting to the modern discourse on loss and damage, a comprehensive approach to dealing with NELs has yet to be found. Addressing NELs continues to be problematic, in particular, because of the many ways in which they affect underlying complex socio-ecological systems.¹¹⁹ Furthermore, given their non-monetary value, NELs continue to be a blind spot for insurance approaches. That said, recent studies found that in some local contexts, affected communities do not distinguish between economic and non-economic loss and damage.¹²⁰ Solutions for NELs would also need to account for losses resulting from policies and measures aimed at addressing climate impacts that lead to, for example, loss of cultural heritage, territory, language, and social structures resulting from forest conservation or disaster displacement measures.¹²¹ Jackson and co-authors, moreover, highlight that while the concept of NELs includes biodiversity and ecosystem services, these are still framed in anthropocentric terms rather than based on their intrinsic value.¹²² As noted earlier, the Transitional Committee is expected to cover both economic and non-economic losses in its recommendations. It could thus be expected that the financing instruments recommended by the Committee for specific types of NEL may differ significantly from those for economic loss and damage.¹²³ It is likely that some types of NEL will fall outside the scope of loss and damage finance.

The inability of the climate regime to adequately address NELs may inspire further legal actions for loss and damage against national governments.¹²⁴ In a world first, the Indigenous inhabitants of the Torres Strait Islands recently succeeded in their petition against Australia before the UN Human Rights Committee for inaction on climate change.¹²⁵ The complainants argued successfully that insufficient action by the Australian government to mitigate and adapt to climate change (such as by reducing emissions and upgrading seawalls) violated their right to culture and right to be free from arbitrary interference with privacy, family, and home under the International Covenant on Civil and Political Rights.¹²⁶ In its decision the Committee required Australia, among others, to compensate the complainants adequately for harm suffered, engage in meaningful consultations with the complainants’ communities to assess their needs, and implement measures to secure their communities’ safe existence on their islands.¹²⁷ What is interesting is that the compensation requested by the Committee was limited to actual harm incurred, excluding prospective harm.¹²⁸ The Committee further noted that the state party is obligated to prevent similar violations in the future, which, while not specified, implies mitigation.¹²⁹ In its response the Australian government interpreted the compensation aspect as part of ‘remedial recommendations [that] are aimed at adaptation measures’.¹³⁰ Further, the government points to its close collaboration with the Torres Strait Islander communities as the appropriate remedy, not addressing further the Committee's request for compensation. Pending further developments, the petition thus marks a missed opportunity to explore different types of compensation for and valuation of NELs in human rights cases. That said, the case provides a strong signal that even in the absence of UNFCCC policy advances on NELs, the courts stand ready to award compensation for such losses.

3.6. Loss and Damage as the Defining Issue of the Climate Regime

In the early 1990s, when AOSIS tabled the Insurance Pool proposal, loss and damage was not even on the map. Yet, as noted earlier, when the concept was introduced, certain types of extreme weather event were already being felt and climate science provided relevant projections of future harm. Policymakers and INC negotiators were thus fully aware of the potentially devastating impacts of climate change. It was clear that something needed to be done to limit future loss and damage. Imagine a world where we thought we could still prevent climate harm and help people in adapting to warming – which appears inconceivable today. Conceptually, there is a strong case to be made that the language around stabilizing GHG concentrations at a non-dangerous level under Article 2, the core objective of the Convention, was ultimately concerned with future loss and damage.¹³¹

The developments in loss and damage policy at COP27 have opened the door for a meaningful engagement with loss and damage beyond dialogues and technical work. Now that loss and damage finance has formally landed on the UNFCCC agenda and a new loss and damage fund has been agreed, the discourse on this critical issue appears finally to be gaining political traction. However, the room for political ambiguity is dwindling. The development of the new fund and funding arrangements will put the multilateral climate regime to its biggest challenge yet. That is because there is a bitter irony in relying on the UNFCCC process to address the complex problematic of loss and damage, which is reminiscent of a Catch 22: if parties were to address loss and damage in earnest, they would essentially admit that the climate regime has failed its principal objective of stabilizing GHG emissions at a non-dangerous level, thus undermining its raison d’être and, by extension, its legitimacy. Yet, if parties continue to defer responsibility and meaningful support for loss and damage, the regime is very likely to result in failure.¹³² As one key informant relevantly observed:

The loss and damage case can never be settled inside the climate negotiations. But it also can never be dismissed. It is a continuous claim on the whole process, and it can only be partially resolved through things that are never fully explicit. … It is as if there is a kind of constant play being enacted where the moral cause is acknowledged but never formally settled and manifests itself in either guilt driven processes or some side arrangements on finance, or insurance or some other accommodations that is less than full and proper accountability for past harms.¹³³

Given the difficulty of achieving climate justice within the framework of the negotiations, climate victims are having to resort to other means to enforce their claims. The rise of transnational climate solidarity movements and a surge in domestic, regional, and transnational climate litigation¹³⁴ are prominent examples of this effort. They are an important reminder that the UNFCCC does not operate in a vacuum. The victorious Torres Strait Islanders petition and the successful bid by Vanuatu for an advisory opinion from the International Court of Justice (ICJ) are testimony that climate-vulnerable countries and affected communities have not given up their faith in the international legal system. A strong advisory opinion from the ICJ that clarifies states’ obligations to protect the climate and spells out their legal consequences for climate-vulnerable states and affected persons (present and future) could bolster legal mobilization efforts in domestic and regional courts. That said, judging from the limited impact of successful climate litigation on international climate policy to date,¹³⁵ it is doubtful that a non-binding advisory opinion could sway industrialized countries to admit liability or frame loss and damage finance as compensation under the UNFCCC.

4. Conclusions

Early attempts by AOSIS to introduce the concept of loss and damage into the nascent multilateral climate regime via an International Insurance Pool hit a brick wall. All too often, the 1991 AOSIS proposal is glossed over as the point of origin of loss and damage under the UNFCCC. However, as the historical analysis presented in this article has demonstrated, there is a wealth of relevant insights to be gleaned from the proposal's content, rationale, and its broader historical context. There is value in looking back as we look forward. The reflections provided in this article are by no means exhaustive, but aim to enrich and contextualize academic and policy discussions regarding the development of the new loss and damage fund.

At a broader level, the analysis reveals how normative and legal arguments by climate-vulnerable countries for liability, compensation, and for more meaningful engagement with loss and damage were suppressed during the INC meetings, a practice which continued under the subsequent 30 years of climate negotiations under the UNFCCC. Three decades on, the key demand of climate victims, for industrialized states to bear liability for loss and damage, remains off the table. Instead, as Dehm reflects, ‘responsibility is displaced to those who are most vulnerable to the impacts of climate change who are increasingly called on to take responsibility for adapting to, preparing for and reducing the risk of climate related harms’.¹³⁶ Similarly, demands for climate reparations remain unheard and there is no evidence of engagement with the colonial past. The prospects are low that the UNFCCC will coalesce around a consensual framing of the new funding arrangements and the new loss and damage fund as compensation.

There are legitimate fears that if the loss and damage fund does materialize, it will be insufficient. Financial assistance disbursed would be neither comparable with nor adequate to satisfy the demand for compensation – or reparations, for that matter. Moreover, it would be very likely to be framed as a matter of greater capacity, international solidarity, charity, and humanitarian relief rather than moral responsibility born out of a history of colonialism and excessive resource exploitation. This would imply, in the words of Burkett, that ‘the developed world would need not meaningfully confront the suffering of the climate vulnerable, nor understand how its current systems have produced such an uneven state of affairs’.¹³⁷

In an attempt to fill this gap, climate victims have launched legal initiatives in a variety of fora. As the article has noted, these come with their own limitations. However, they provide important spaces for climate victims to be heard. Such spaces allow climate victims to share their lived experiences of loss and damage, and require perpetrators to engage meaningfully with these accounts. These are spaces that, as I have argued elsewhere,¹³⁸ remain foreclosed in the multilateral climate negotiations. With loss and damage now firmly on the agenda, the key question going forward will be whether the UNFCCC has the potential to transform its approach to this critical issue, for both its legitimacy and identity hang in the balance.