The ICJ’s Advisory Opinion on Climate Change: Does It Throw a Wrench into the Negotiator’s Toolbox of Diplomatic Problem-Solving Techniques?

The International Court of Justice’s recent Advisory Opinion on Climate Change (AO) has been widely celebrated by many States, non-governmental groups, and commentators as a victory for climate ambition and climate justice.  To the extent there have been criticisms, they have generally been on the basis that the AO does not go far enough.  Like others, we hope that the AO helps galvanize stronger across-the-board action to mitigate and adapt to climate change, which we agree is urgently needed.  Positive effects of the AO might include bolstering both domestic and international litigation efforts, strengthening the hand of proponents of stronger climate action within the UN climate regime, and prompting some Parties to adopt more ambitious NDCs.-

At the same time, we think it is important to consider whether the AO, although not legally binding, may also have unintended chilling effects on the conduct of global climate diplomacy.  Given widely divergent national interests and policies, negotiators have often been able to reach agreement in the UN climate regime only thanks to an arsenal of gap-bridging techniques.  By calling into question the reliability of some of these diplomatic tools, the AO may make it even more challenging to take climate-forward decisions.   

For some, the trade-offs may be worth it; the AO’s expansive interpretation of international law may more than compensate for whatever incremental progress might be lost in an international process they consider inadequate, even in the best of circumstances, for delivering sufficient climate action.  After all, many of those who sought an advisory opinion likely did so, at least in part, out of frustration with diplomacy’s slow pace and consensus-driven compromises.  

We understand that frustration.  Even if it operated perfectly, the UN climate regime could not be left to “do it alone”; the climate challenge is simply too large.  And, given that the regime’s operation is far from perfect, the need for action in other fora and from other actors is even more vital.  

Yet we still see value in the ability of the global climate regime, however constrained, to get nearly the entire world to take decisions that incrementally bolster international and national climate responses, sending important market signals along the way.  Given the current U.S. situation and geopolitics more broadly, the argument for preserving the global climate regime (with improvements) may be even stronger today than ever.  It is with this perspective that we raise some concerns about the AO’s potential unintended consequences.

1. International environmental law, and the climate change regime in particular, have relied upon a large toolbox of techniques to solve problems in negotiations and otherwise bridge gaps between countries with very different interests.

International environmental law has developed an extensive set of techniques to resolve negotiating impasses and enable multilateral agreement.  The climate change regime in particular has had to employ (and sometimes invent) a wide variety of linguistic and other problem-solving tools to bridge gaps between national positions.  Not only has the subject matter of climate change been consistently contentious but the Parties to the regime have never been able to agree on a voting rule, the absence of which requires them in practice to reach consensus or something close to it. Thus, from the early days through to the landmark 2015 Paris Agreement, negotiators have needed to draw heavily from the problem-solving toolbox to accommodate the widely divergent interests of vulnerable small island States, oil-producing States, the United States and China, and the rest of the world.  But for a host of diplomatic problem-solving tools, the Paris Agreement would not likely exist.  

Since 2015, the Parties to the Paris Agreement have continued to employ a wide range of techniques in order to adopt decisions that move the regime forward.  Many of these decisions were very difficult to land, including the various implementing rules, guidelines, and procedures that make up the “Paris rulebook,” the establishment of a “loss and damage” fund that was not previewed in the Agreement itself, and agreement on the need to “transition away” from fossil fuels.  

Nearly two hundred Parties have succeeded in reaching consensus over the years thanks, in large part, to several (somewhat overlapping) tools:

– They have engaged in careful drafting to reflect nuanced approaches.  

– They have struck a balance concerning the extent to which a particular agreement or decision is prescriptive versus permissive, i.e., leaving Parties with national flexibility (so-called “top down” versus “bottom up” approaches).  

– They have designed provisions with a certain margin of linguistic ambiguity, so that Parties can interpret the provisions in different ways.

– They have drafted certain provisions to be non-binding, e.g., to create norms or expectations rather than legal obligations or provisions with legal effect.

These techniques have not only facilitated nearly universal participation in the regime but also enabled Parties to go to the edge of their comfort zones.  People can reasonably debate whether such tools have been over-utilized, i.e., whether the world would arguably be better off with stronger provisions and fewer Parties.  But that goes back to the perceived value – or not – of Paris’s global approach.  

2. The AO potentially compromises key problem-solving tools.

Two examples of contentious topics that have been the subject of highly nuanced treatment in both the Paris Agreement and subsequent decisions of the Parties are the Agreement’s temperature goal and its provisions on nationally determined contributions (NDCs).  In both cases, the AO could have the effect of reducing the perceived reliability of tools that have helped States bridge differences in national positions – potentially diminishing the prospects for further global agreement on stronger climate action.  

(A) The AO’s treatment of the Paris Agreement’s temperature goal

The AO’s approach to the Agreement’s temperature goal, specifically the pronouncement that subsequent decisions of the Parties have made 1.5°C the “primary” temperature goal (para. 224), potentially affects the viability of a key tool of negotiators, namely carefully nuanced drafting.

One of the key ambition-related issues confronting negotiators of the Paris Agreement was how, if at all, to include a reference to 1.5°C in the Agreement’s temperature goal.  Some sought to make 1.5 the temperature goal, while others sought to avoid any mention of it.  

The relevant provision (Article 2.1) provides that the Agreement:

 “…aims to strengthen the global response to the threat of climate change…by [h]olding the increase in the global average temperature to well below 2°C above pre-industrial levels and pursuing efforts to limit the temperature increase to 1.5°C….”  

Thus, the text includes a reference to 1.5°C but the reference is different – and intentionally softer – than the reference to well below 2°C, the former focusing on effort (“pursuing efforts”), in contrast to the latter, which focuses on actual achievement  (“holding”).  

In recent years, the Parties have, in their decisions, incrementally leaned into the importance of a 1.5°C limit.  But they have done so in a careful manner, without any suggestion that they intended – legally – to make 1.5°C the Agreement’s “primary” temperature goal.

In the Glasgow Climate Pact, the decision that emerged from COP 26 in 2021:

– The Parties recognized, as a factual matter, that climate impacts will be much lower at 1.5°C than 2°C and that a 1.5°C limit will require certain greenhouse gas emission reductions (paras. 21-22). 

– The Parties “resolved” to “pursue efforts” to 1.5°C (para. 21).

– At the same time, the Parties “reaffirmed” the Paris temperature goal, repeating the text of Article 2.1 verbatim.

Subsequently, the Parties’ decision in Dubai on the first “global stocktake” under the Paris Agreement continued to lean into 1.5°C.  It included a reference to “1.5°C pathways” in a key provision (para. 28) and “encouraged” Parties to align their next NDC targets with, among other things, “limiting global warming to 1.5°C” (para. 39).  But, again, it repeated Glasgow’s reaffirmation of the exact wording of the Paris Agreement’s temperature goal.

The AO took these references to 1.5°C in the Glasgow and Dubai decisions to constitute “subsequent agreements between the parties regarding the interpretation of the treaty” within the meaning of Article 31(3)(a) of the Vienna Convention on the Law of Treaties and concluded that 1.5°C has now become the “primary” temperature goal of the Paris Agreement (para. 224).

The AO’s legal reasoning concerning “subsequent agreements” at best raises significant questions and at worst may cause at least some Parties to worry about agreeing to decisions in the future that might later be characterized as having unanticipated legal effects.  

In terms of the AO’s legal analysis:

– The Vienna Convention on the Law of Treaties (VCLT)’s provisions on treaty interpretation place primary emphasis on the text.  As noted above, the text of Article 2.1 of the Paris Agreement uses different/stronger language for well-below 2°C (“holding”) than 1.5°C (“pursuing efforts”), suggesting that well-below 2°C is the primary component of the Agreement’s temperature goal.

– The AO states that COP decisions “may constitute subsequent agreements” of the Parties (para. 184, emphasis added), but does not spell out which factors are relevant in determining whether a particular COP decision – or specific provision thereof – is, in fact, a “subsequent agreement” within the meaning of the VCLT.

– According to the International Law Commission, for a COP decision to constitute a subsequent agreement, it must reflect “a common understanding regarding the interpretation of a treaty, which the parties are aware of and accept” (ILC Conclusions on Subsequent Agreements, Conclusion 10(1), emphasis added).  Its intention must be “clear,” as the IMO Sub-Division for Legal Affairs concluded when considering the legal status of a decision of the London Convention COP (ILC Conclusion 11, Commentary, para. 12).

– Nothing in the language of the Glasgow and Dubai decisions indicates an intent – much less a clear intent or common understanding among the Parties to the Paris Agreement – to legally interpret the Agreement’s temperature goal (in contrast, say, to COP decisions adopted as “unified interpretations” or as “definitions”).  

– Moreover, both decisions explicitly “reaffirmed” the Agreement’s temperature goal, quoting the language of Article 2(1), which contrasts “holding” temperature increase to well below 2°C and “pursuing efforts” to 1.5°C.  Although there is no indication that the Parties intended either reaffirmation to constitute a formal interpretation of the Agreement, to the extent this was their intent, their interpretation would appear to be that the Paris Agreement’s temperature goal is unchanged, that it consists of the two components set forth in Article 2.1, and that it does not elevate 1.5°C over well below 2°C.

The point here is not whether the Paris Agreement’s temperature goal should be 1.5°C as a matter of policy; the world is, for the most part, already treating 1.5 as the de facto limit.  The point is that, for the reasons laid out above, not all Parties are likely to agree that their Glasgow and Dubai decisions set a 1.5°C temperature goal as a legal matter. In addition, some may be concerned more broadly at the lack of clarity concerning the ICJ’s elevation of the legal status of certain provisions in COP decisions versus others.  As a result, they may have less confidence that they can count on careful drafting with respect to future efforts to take the regime in more ambitious directions. If, for example, some Parties worry that carefully-worded decisions might be construed as de jure interpretations of the Paris Agreement – and/or given legal effect where none was intended – the AO could have a constraining effect on the ability of the Parties to adopt future decisions.

(B) The AO’s conclusions regarding nationally determined contributions (NDCs)

The ICJ’s pronouncements regarding NDCs could also have potentially negative implications for the future viability of key negotiator tools, as well as for the ambition level of some NDCs moving forward.

One of the most significant issues during the negotiation of the Paris Agreement was the way in which Parties’ emissions commitments would be addressed.  Would they be negotiated?  Would they be legally binding?  Would they be differentiated among Parties?  An important backdrop to the negotiation was the history of previous efforts to reach agreement.  On the one hand, the Kyoto Protocol involved negotiated, legally binding targets; however, these applied only to certain Parties and did not end up covering the two biggest emitters of greenhouse gases – the United States, because it rejected the Kyoto paradigm, and China, because it joined the Protocol but was in a category of Parties exempted from emissions targets.  On the other hand, the later Copenhagen Accord/Cancun agreements did cover all the major emitters, but were thoroughly non-binding and imposed negligible disciplines on national flexibility.  It was clear that Paris would need innovative approaches if it was to get key countries on board and do so with some degree of rigor.   

The resulting NDC package in the Paris Agreement could be the subject of a master class on the negotiator’s problem-solving toolbox.  Specifically, global agreement on a new paradigm for mitigation action was enabled by a painstakingly balanced combination of “top down” and “bottom up” features, a hybrid approach to binding and non-binding provisions, and a bespoke approach to domestic implementation involving some ambiguity.    

In brief, the Agreement is largely “bottom up” when it comes to NDCs. It accords Parties extremely broad latitude to design their respective NDCs based on national circumstances.  It contains neither required nor prohibited NDC features; creates no substantive link between NDCs and the Agreement’s temperature goal; and does not prescribe the timeframe for NDC targets/actions.  NDCs are also non-legally binding.  This flexible, largely “hands off” approach was fundamental to getting key countries on board and for moving past the Kyoto Protocol’s strict division of Parties into two categories.

At the same time, the Agreement would have failed had that approach not been carefully coupled with numerous counter-balancing elements.  For example:

– The Agreement contains several NDC-related legal obligations of a procedural nature, such as requirements to submit NDCs early to leave time for comment and potential revision; to regularly update them; to be clear about their content, including how they contribute to the temperature goal; and to report in detail on their implementation.

– The Agreement sets out certain non-binding provisions relating to the substantive content of NDCs, such as creating an expectation that each Party’s successive NDCs will be progressively more ambitious, reflecting its “highest possible ambition.”

– The Agreement includes a requirement regarding Parties’ domestic implementation of NDCs.  

This last requirement (Article 4.2, second sentence) epitomizes diplomatic problem-solving. It needed to provide some comfort to those States that had sought legally binding targets, while not going so far as to, in effect, make NDCs binding through the backdoor of mandatory domestic implementation.  Per a heavily nuanced provision, Parties have a legally binding obligation to “pursue domestic mitigation measures, with the aim of achieving the objectives of such contributions” (emphasis added).  This is an obligation of conduct, rather than result; refers to NDC “objectives” rather than to NDCs per se; and is ambiguous as to whether the “aim” attaches to the Parties or to the measures.  In short, it leaves room for interpretation and application along a broad continuum.

Since Paris, the Parties have taken various decisions that maintain the careful balance in the original Agreement.  For example:

– With respect to the content of NDCs, the Parties discussed but did not decide on any “common features” of NDCs (Decision 4/CMA.1, paras. 19-20). 

– In terms of a substantive relationship between NDCs and a temperature limit, both the Glasgow and Dubai decisions create such a link (Glasgow between NDCs and the “Paris temperature goal,” Dubai between NDCs and 1.5°C); however, in both cases, the verb is non-binding (“requests,” “encourages”), and the decisions explicitly reiterate that NDCs are “nationally determined” and based on “national circumstances.”

– The Paris Rulebook includes the Agreement’s NDC-related procedural obligations within the scope of the Implementation and Compliance Committee, but not the provision on domestic implementation, thereby avoiding any risk that the Committee might clarify or “enforce” the provision.

In contrast to these carefully calibrated provisions and trade-offs among them, the AO’s conclusions regarding both the content of NDCs and their domestic implementation are quite blunt:  

– The “discretion of Parties in the preparation of their NDCs is limited” (para. 245).   

– The “content of a Party’s NDC must … be capable of making an adequate contribution to the achievement of the [1.5°C] temperature goal” (para. 242).

– Each Party must do its “utmost” to ensure that its NDC represents its highest possible ambition” (para. 270).

– Each Party’s NDC “must become more demanding over time” (para. 241).  

Regarding domestic implementation of NDCs, the AO concluded:

– Parties must “make best efforts” rather than simply “pursue” domestic mitigation measures (para. 253);

– “Parties’ measures must be “stringent” (para. 254); and

– Parties must pursue measures to achieve the “content” of their NDCs (para. 270), rather than the “objective” of their NDCs (the actual language of Article 4(2)), thereby vitiating the distinction between the two words.

The AO’s reasoning regarding the provision on successive NDCs is noteworthy.  Article 4.3 provides that each successive NDC “will represent a progression … and reflect its possible ambition,” in contrast to the provisions of the Paris Agreement clearly intended to create legally binding obligations, which use the verb “shall.”  Although the AO acknowledged that “will” is different from the “prescriptive ‘shall,’” it nevertheless pronounced that “will” is used in Article 4.3  in a “prescriptive sense” (para. 240) and that Article 4.3 creates legal obligations of progression (“a party’s NDC must become more demanding over time” (para. 241)) and highest possible ambition (“a party’s NDCs must reflect its ‘highest possible ambition’ (para. 242) – obligations that can be used to objectively assess the adequacy of NDCs.  In doing so, the AO did not take account of the widespread understanding that “will” is not a verb used to create international legal obligations or of the interpretive principle that, if drafters use different terms (“will” versus “shall”), it should be assumed that those differences matter (the ut res magis valeat quam perat principle).

Similarly, as noted, the AO concluded that NDCs must align with the Paris Agreement’s (purported) 1.5°C temperature goal – even though the Agreement creates no such substantive link and the Parties in their subsequent global stocktake decision only “encouraged” Parties to align their NDCs with that limit.  If the ICJ is correct that the global stocktake decision constitutes a subsequent agreement of the Parties that should be used in interpreting the Paris Agreement, then the decision clearly establishes the Parties’ view that the Agreement only “encourages” them to align their NDCs with the 1.5°C temperature goal; it does not require them to do so.

We do not address whether the AO’s conclusions about the Paris Agreement’s NDC provisions are justified under other rules of treaty interpretation.  However, by straying from the specific text of the Paris Agreement and subsequent decisions, the AO may raise concerns for at least some Parties about how safely they can rely on treaty drafting that appears to provide them with flexibility, to be non-legally binding, or otherwise to provide the needed comfort to agree to a forward-leaning decision – in the climate regime and potentially more broadly.

The AO could also have implications for the more specific content of NDCs, namely their ambition level moving forward.

– On the one hand, the AO says a Party’s NDC must make an “adequate” contribution to a 1.5°C limit and that each Party must do its “utmost” to ensure that its NDC represents its “highest possible ambition” (para. 270).

– On the other hand, it says that, once a Party submits an NDC, it must undertake “best efforts” to achieve its NDC (para. 253), and each subsequent NDC must be more ambitious (para. 270).

For some Parties, the two sets of obligations may point in the same ambitious direction, i.e., encouraging them to both adopt more ambitious NDCs and intensify their domestic implementation.  

However, for other Parties, the second set of obligations may undercut the first – it may discourage Parties from adopting ambitious NDCs in the first instance, out of concern that they may not be able to meet the purported obligations about domestic implementation and progression.  Parties may calculate that it is better to submit a weaker, easier-to-implement NDC – and argue that the NDC is adequate, given the countries’ various circumstances – than to adopt a stronger NDC that they may have difficulty implementing and progressing.   

For Parties that advocated the Agreement’s NDC package of strong but non-binding expectations, one of the drivers was the goal of encouraging as much ambition as possible.  They hoped that Parties would rely on this package to put forward higher levels of NDC ambition than they might otherwise have been in a position to do if the Agreement had created legal obligations to achieve, implement, and progress their NDCs. Removal of the flexibility provided by the text of Article 4 could have the effect of chilling NDC ambition, at least for some Parties.  

3. Other thoughts/questions 

The AO may have  other consequences that we do not discuss in detail here but that merit further consideration.

What will be the effects of the AO on future litigation, at both the international and domestic levels, and how will the specter of increased international litigation affect the negotiations? 

– The AO will likely enhance the prospects for future litigation in some States – for example, by providing support for claims that a State’s mitigation policies violate the duty of due diligence or international human rights law. But such litigation will have a highly variable geography, depending on the degree to which a State’s domestic legal system incorporates international law and its courts view of the ICJ advisory opinions as authoritative, and whether it has accepted the jurisdiction of various international tribunals.  As a result, litigation may have more impact on States that already have comparatively ambitious climate change policies and less impact on the largest emitters, including China and the United States.

– The specter of litigation might also induce some States to adopt more conservative positions in the negotiations to avoid prejudicing their positions in potential future litigation.

What will be the AO’s effect on State behavior more generally?

– The AO could have a number of positive effects on national climate policies, e.g., by helping to shape public opinion; by bolstering domestic climate advocacy groups in domestic political processes; and by prompting at least some States to adopt more ambitious policies in order to align themselves, whether legally or politically, with the AO’s pronouncements about their international law obligations.

– However, in contrast to a decision in a contentious case, the AO is non-binding.  Given that the ICJ is the “principal judicial organ of the United Nations,” its opinions have significant persuasive weight, but they are not determinative about what international law provides.  Therefore, if a State is not persuaded by the Court’s analysis, it might not accept the AO’s conclusions.  Indeed, even in contentious cases, where the Court’s judgments are legally binding, compliance has been uneven.  

What effects will the AO have on other issues in international climate change policy, in particular contributions to the Fund for Responding to Loss and Damage (FRLD)?

– The AO could have uncertain implications for other issues in international climate change policy, such as contributions to the FRLD. 

– On the one hand, it might have encourage greater contributions to the FRLD, out of a conviction of responsibility for damage and/or as a hedge against a future legal proceeding.

– On the other hand, it might chill contributions, given that its suggestion of potential liability/compensation is at odds with the FRLD’s foundational premise, namely,cooperation and facilitation rather than liability or compensation. 

Finally, what will be the effects of the AO on other multilateral environmental negotiations?

– In assessing the costs/benefits of the AO, it is important to also consider its potential impacts on efforts to advance environmental protection in other realms.  
– In this regard, the AO could make it more difficult to reach multilateral agreement.  As discussed above, some of the most widely used problem-solving tools have been called into question, which could affect their future usefulness. Some States may be reluctant to accept vague or ambitious wording, for fear that a court may adopt an interpretation adverse to their interests.   As Judge Nolte noted in his separate opinion, the result may be that States in the future “shy away from accepting new treaty obligations … that could subject them to unpredictable legal consequences” (Declaration of Judge Nolte, para. 32).

Conclusion

The ICJ’s Advisory Opinion on Climate Change has been celebrated as “the start of a new era of climate accountability at the global level,” as a “turning point” for climate justice, and as a “watershed.”  Like others, we believe that more ambitious efforts to reduce greenhouse gas emissions are urgently needed and hope that the AO will contribute to those efforts.  But amidst the general adulation, it is also important to examine the AO’s potentially less positive implications.  

The degree to which the AO will (unintentionally) render less usable time-tested tools of climate diplomacy is uncertain.  The answer may depend, in part, on how much attention various Parties pay to the AO and how worried it makes them, if at all, that the agreement they think they are adopting – diplomatically – may not be the same as the agreement as later interpreted – judicially.  

Again, if one thinks the global climate regime is so weak and hobbled as not to make a substantial difference to the cause, then concerns about the AO’s potential effects on climate diplomacy may seem a price worth paying or even beside the point.  But if one believes, as we do, that the Paris Agreement makes a positive difference and is important to preserve and advance, then the AO may not be an entirely unalloyed advance in the effort to address climate change.