1. Introduction
It is an established principle of international customary law that reparation must ‘as far as possible, wipe out all the consequences of the illegal act and reestablish the situation which would, in all probability, have existed if that act had not been committed’.[1] Article 31 of the International Law Commission (ILC) Articles on Responsibility of States for Internationally Wrongful Acts (ARSIWA)[2] incorporates this principle in the following manner: ‘(1). [t]he responsible State is under the obligation to make full reparation for the injury caused by the internationally wrongful act. (2) Injury includes any damage, whether material or moral, caused by the internationally wrongful act of a State’.[3] Hence, there cannot be reparation without a causal link between injury and the internationally wrongful act.
While causality plays a fundamental role in assessing the extent of reparation, in the law of international responsibility the concept remains rather ‘esoteric’. The ARSIWA recognize causality as an integral part of reparation, yet they do not dwell on its specificity and meaning in the context of State responsibility. The commentary to Article 31 provides only a cursory evaluation of the type of causation relevant for allocating injury to the wrongful act, simply asserting that the causal link should not be too remote and may differ depending on the type of primary rule at stake.[4] International case-law does not score any better. Although international courts and tribunals apply causal principles when called to determine reparation for an internationally wrongful act (IWA), they often do so without expressly recognizing it or spelling out the legal assumptions underpinning their reasoning.[5] This has led some scholars to argue that causality is still a rudimentary concept in the law of international responsibility and to call for a more coherent application of this notion by practice.[6]
In a jurisprudential landscape where causal analyses tend to be sparse, the International Court of Justice’s recent decision on reparation in the Armed Activities on the Territory of the Congo (DRC v Uganda) is welcome. The judgement is one of the few instances where the Court has substantially engaged with causality between injury and the internationally wrongful acts against a complex set of facts. To briefly recall, in 2005 the ICJ had issued a judgment on merit affirming Uganda’s responsibility for several IWAs committed between 1998 and 2003 against the Democratic Republic of the Congo (DRC). These included violations of the principles of non-intervention and non-use of force; military intervention and occupation of part of the DRC territory; breaches of the obligations to respect and prevent violations of international human rights and humanitarian law; and breaches of the obligation to abstain from looting, plundering and exploiting the DRC’s natural resources.[7] The decision asserted Uganda’s obligation to provide reparation to the DRC for injury caused by the wrongful acts and called upon the parties to reach an agreement in this regard. Yet, after the failure of the Court-ordered negotiations, in 2015 the DRC requested the ICJ to reopen the proceedings.
In the Armed Activities judgment on reparation, the ICJ had to reappraise facts occurred more than 20 years ago and evaluate causality against a complicated and contradictory set of evidence presented by the parties.[8] These conditions certainly call for some indulgency toward the Court. At the same time, some of the Court’s assertions on causation lend themselves to criticism.
In this contribution I will critically evaluate the ICJ’s approach toward causality throughout the judgment. My argument is that the Court’s reasoning epitomizes the murky terrain surrounding the application of causation principles when allocating injury to the IWA. Specifically, my criticism is not focused on the outcome reached by the Court – ie whether the ICJ erred in recognizing or denying causality between certain damages and Uganda’s wrongful conducts. Rather, I argue that the problem lies in the legal rationale used by the Court to construe causation and justify its findings on the scope of reparation due. As I will show, the ICJ is not entirely consistent in the way it develops its argument: the legal principles identified to sustain or deny the causal link are not always clearly spelled out, and some contradictions emerge from the Court’s reasoning.
The article is structured as follow. Section 2 discusses the standard of causality invoked by the Court to assess the scope of reparation owed by Uganda. Sections 3 lays down the main critical issues that emerge from the Court’s approach toward causality in the decision. Sections 4, 5 and 6 demonstrate the arguments sustained in Sections 3 with different examples. Section 7 concludes the paper.
2. The standard of causality applied by the Court for assessing reparation
Let us begin by identifying the standard of causality used by the Court to establish the connection between the injury and Uganda’s internationally wrongful acts. In the request for reparation, the DRC had claimed that Uganda must provide reparation for any damage that was the consequence of the IWA, ‘whether it resulted directly from its internationally wrongful conduct or was caused by an interrupted chain of events’.[9] Hence, for the DRC, even damages that are either remote[10] or the result of a multiplicity of causes qualify for reparation, unless they ‘would not have occurred had the internationally wrongful act not been committed’.[11] The DRC suggested as a legal criterion that of ‘foreseeability’.[12] However, the ICJ rejected this approach and provided a general statement on the principles of causality applicable to the case:
The Court hence relied on the criterion of the ‘sufficiently direct and certain causal nexus’. It is not surprising that the ICJ invoked this standard, which was first developed in the 2007 Bosnian Genocide case and has been applied consistently in subsequent decisions concerning reparation.
In the 2007 Bosnian Genocide case the Court had to decide on the question of reparation after finding Serbia responsible for breaching the obligations under the Genocide Convention to prevent and punish genocide. The ICJ judges deemed relevant to ascertain whether genocide at Srebrenica would have taken place if Serbian authorities had attempted to prevent it by fulfilling their obligation. Accordingly, the Court affirmed that the applicable standard of causality was ‘a sufficiently direct and certain causal nexus’ between Serbia’s breach of the obligation to prevent genocide and all damage, moral or material, caused by the acts of genocide.[14] Such nexus could be considered established ‘only if the Court were able to conclude from the case as a whole and with a sufficient degree of certainty that the genocide at Srebrenica would in fact have been averted if the Respondent had acted in compliance with its legal obligations’.[15] Eventually, the Court concluded that this was not the case. While Serbia’s omission had certainly contributed to the commission of genocide, it could not be proved that a prompt and regular intervention by this State would have averted the massacre at Srebrenica.
The ICJ employed the ‘sufficiently direct and certain causal nexus’ also in the 2012 Ahmadou Sadio Diallo decision (Republic of Guinea v Democratic Republic of the Congo) and, more recently, in the Certain Activities case (Costa Rica v Nicaragua). Each case differed from the Bosnian Genocide both in terms of factual circumstances and primary rules at stake. In Ahmadou Sadio Diallo, the Court applied the criterion of the ‘sufficiently direct and certain causal nexus’ to establish the scope of reparation ensuing from the DRC’s breach of certain obligations arising out of the International Covenant on Civil and Political Rights.[16] In Certain Activities, the criterion was invoked to appraise the connection between Nicaragua wrongful activities on the border areas with Costa Rica and damages to the Costa Rican environment.[17] Thus, by invoking the same standard in the Armed Activities case, the Court seems to have developed and consolidated a general principle of causality applicable across the board.[18]
In exposing the general principles of causality applicable to the case, the Court also noted that ‘the causal nexus required may vary depending on the primary rule violated and the nature and extent of the injury’.[19] The assertion that the causal standard required to ascertain the link between the IWA and the injury may differ depending on the primary rule reflects the approach of ARSIWA. The commentary to Article 31 does not prescribe a general rule for how causation should be evaluated in determining reparation. The commentary states that international courts and tribunals may employ different criteria – eg ‘directness’ of causes, ‘foreseeability’, ‘proximity’ of the injury[20] – and that preference for one standard over another may depend on different factors, including ‘whether the harm caused was within the ambit of the rule which was breached, having regard to the purpose of that rule’.[21] The ARSIWA indeed acknowledge that ‘the requirement of a causal link is not necessarily the same in relation to every breach of an international obligation’.[22]
Although this approach has drawn criticism,[23] the ARSIWA’s conclusions testify to the Commission’s impossibility during its work on State Responsibility to identify a unique causal criterion applicable to all international obligations.[24] When the question of reparation emerged during the ILC work on State responsibility, the Special Rapporteur at the time, Mr Arangio-Ruiz, had initially invoked the criterion of the ‘uninterrupted causal link’.[25] However, during the debate within the Commission, some members had expressed doubts about this notion, stressing the non-uniformity of causal tests used in international practice.[26] In the end, the Commission concluded that it would not be advisable to qualify the causal link required for reparation given the diversity of primary rules and the lack of consistent application of causal standards in practice.[27] The ILC only agreed that injuries ‘too remote’ or ‘consequential’ to the IWA would be excluded from reparation.[28]
3. Issues in the Court’s approach toward causality in the decision
Having illustrated the principles of causality invoked in the judgement, there are at least two general observations that emerge from the reading of the decision.
To being with, although the Court in the Armed Activities formally subscribed to the idea that causality may differ depending on the primary rule, a careful analysis of the decision shows that the Court’s statement is merely declaratory. There is no substantive application of the principle to the case. The Court relied on the ‘sufficiently direct and certain causal nexus’ in relation to each claim of damage arising out of all violations of international law. While, in principle, the Court is not bound to adhere to more than one standard of causality – and the ARSIWA do not impose that the causal link must change depending on the primary rule – I will show in the next sections that the ICJ’s approach prompts some critical remarks.
First, if there is any difference in the decision in how the Court applied the causal test, this does not depend on the type of primary rule, but on the factual circumstance that certain violations occurred in a territorial area occupied by Uganda. As I will show, this distinction made by the Court between facts occurring in the occupied territory and facts occurring outside of Ituri is problematic because it leads to unacceptable results with regard to the nature of Uganda’s responsibility for IWAs occurred in Ituri.
Second, there are certain passages in the judgment that would have significantly benefitted in terms of principled reasoning if the Court had adhered, substantially and not only in principle, to the understanding that causality may vary depending on the scope and nature of the primary rule. Instead, by failing to truly appreciate this, the Court has left outstanding some of the most important questions concerning causality in reparation.
Finally, and more generally, one may at least query whether the Court’s exclusive use of the ‘sufficiently direct and certain causal nexus’ is commendable given the type of certain internationally wrongful acts involved in the case – like those concerning the violation of the non-use of force. The ‘sufficiently direct and certain causal nexus’ is a standard that significantly restricts the extent of reparation due. Indeed, the amount of compensation finally awarded by the Court was noticeably smaller than what requested by the claimant,[29] since the Court often found that evidence presented by the DRC lacked a sufficiently certain causal nexus between Uganda’s IWAs and damages. Yet, one may cast doubt on whether it is desirable to rely on such a strict standard in the context of damages ensuing from grave breaches of the non-use of force, where the humanitarian, environmental and economic consequences unfold over a long period of time and are usually the result of a multiplicity of concurrent factors.
The second general observation regards the Court’s failure to coherently treat the causal analysis in the context of reparation as an operation imbued with factual as well as legal dimensions.[30] To be clear, the appraisement of the causal link between the injury and wrongful conduct is first and foremost a fact-intensive process; it requires the interpreter to examine facts that are relevant to the case and identify the empirical, scientific or statistic evidence which explain how damage ensued from the State’s wrongful conduct.[31] However, the selection of one or another causal criterion always underlies normative considerations linked to the function of reparation and, more generally, the law of international responsibility.[32] For instance, it may be that, under international law, not all damage ensuing from wrongful conduct falls within the scope of reparation, but only damage that is not ‘too remote’, ‘unforeseeable’ or ‘proximate’.[33] Essentially, causality in the context of reparation presupposes a factual link between injury and wrongful conduct, and that the judge makes normative evaluations on the extent to which the breach of a primary rule of international law should cover reparation for injury. The ARSIWA endorse this understanding of causality. The Commentary to Article 31 observes that ‘[t]he allocation of injury or loss to a wrongful act is, in principle, a legal and not only a historical or causal process’.[34] The commentary goes on noting that: ‘causality in fact is a necessary but not a sufficient condition for reparation. There is a further element, associated with the exclusion of injury that is too “remote” or “consequential” to be the subject of reparation”.[35] As I will show below, the problem with the Armed Activities judgment is that the question of causality is at times depicted as a simple matter of facts, with the Court failing to properly distinguish between factual and legal dimensions of causation.
To prove these arguments and to better understand their reach, the remaining three paragraphs delve into different parts of the decision that touch upon one or more of these general issues.
4. The Court’s distinction between IWAs committed in the Ituri district and IWAs committed outside Ituri
I will first illustrate the consequences of the Court’s decision to distinguish causality depending on the factual circumstance of occupation and not on the type of primary rule under scrutiny.
In the 2005 judgement on merits, the ICJ had established that between 1998 and 2003 Uganda was an occupying power in the Ituri district, an area within the territory of the DRC.[36] In relation to the violations of international law committed by Uganda in the occupied area, the Court held:
Thus, the Court made a distinction in the kind of international obligations violated by Uganda in the occupied territory. Such distinction not only regarded the content of the primary rules falling upon Uganda (ie, obligations concerning the respect of human rights, humanitarian law and the environment) but also their typologies. Accordingly, the Court found that Uganda’s responsibility concerned both violations of negative obligations of result – ie, obligations not to commit acts which violate international human rights and humanitarian law – and violations of positive obligations of due diligence – obligations to prevent third parties from violating human rights and international humanitarian law.[38]
Note that, at the merit stage, the Court had hinted that the typology of the primary rule violated may affect the scope of reparation owed by a State. The argument had been made in relation to a counterclaim submitted by Uganda alleging the DRC’s violation of the obligation under the Vienna Convention on Diplomatic Relations[39] to protect the premises of a diplomatic mission. In that context, the ICJ had deemed immaterial, for the purpose of establishing the responsibility of the DRC, to ascertain who had removed property from Uganda’s diplomatic premises, since ‘The Vienna Convention on Diplomatic Relations not only prohibits any infringement of the inviolability of the mission by the receiving State but also puts the receiving State under an obligation to prevent others – such as armed militia groups – from doing so’.[40] Yet, the Court had suggested that the distinction between conduct actively removing property and conduct failing to prevent removal by third private parties would become relevant at the reparation stage.[41]
The 2022 judgment fails to distinguish between typologies of primary rules and is silent on how the causal inquiry may change when reparation is sought for omissive rather ran commissive conduct.[42] Instead, the Court noted:
Hence, the distinction made by the Court does not regard the typologies of IWAs but whether these acts have or not been committed by Uganda in the occupied territory of Ituri. For IWAs committed outside Ituri, the Court established the criterion of the ‘sufficiently direct and certain causal nexus’ and affirmed that it would fall on the DRC to prove the existence of such link between the alleged injury and the IWA. In relation to IWAs committed by Uganda in the occupied territory, where Uganda exercised effective control, the Court ruled that:
The problem with the Court’s argument is that it reverses the burden of proof and introduces a presumption of causality which builds on the status of Uganda as an occupying Power and the capacity of this State – due to the ‘effective control’ exercised over territory – to prevent wrongful acts by third parties. As noted by judge Yusuf in his Separate opinion to the judgment, this reversal is problematic on two separate grounds. First, this presumption imposes on Uganda to prove a double negative fact, namely that i) the injury alleged by the DRC has not occurred and that ii) even if it occurred, this injury was not causally linked to Uganda’s failure to meet its obligations.[45]
Second, and more fundamentally, the reversal is at odds with the nature of Uganda’s obligations to prevent violations of international humanitarian law and human right law in the Ituri district. These obligations are duties not imposing on Uganda to achieve a result, ie to prevent all violation of international human rights and humanitarian law at all time and all costs.[46] They are due diligence obligations that only impose to take appropriate effort to prevent private parties from committing harmful acts in the Ituri district. A State under a duty to prevent violations of international human rights and humanitarian law is not responsible merely because violations have occurred in a territorial area under its jurisdiction or control. For responsibility to arise, it is necessary to prove that the State failed to display due diligence in preventing such violations.
The shift in the evidentiary burden established by the decision on reparation transforms the nature of international responsibility in the Ituri district from negligence-based to a form of strict liability.[47] Stating that all damage is presumed to be the outcome of Uganda’s wrongful conduct implies that responsibility is engaged, unless proven otherwise, for the sole fact that injury has occurred in a territory under Uganda’s occupation. Yet, responsibility for breaches of preventive and other due diligence obligations is based on negligence, ie, on a claimant’s proof that the respondent failed to adopt measures which, judged reasonably, would have prevented the event.
A more compatible approach with the nature of Uganda’s obligations in Ituri would have acknowledged that reparation included all injury that was the ‘foreseeable consequences’ of the IWA committed by Uganda.[48] This way, the Court would have, on the one hand, adopted a standard of causality less stringent that the ‘sufficiently direct and certain causal nexus’, recognizing the fact that damages ensuing from breaches of preventive obligations are necessary the product of concurrent causes – the omission of the State and the action of private parties. In other words, the DRC would have benefitted from a more flexible standard of causation necessary to prove injury. On the other hand, the foreseeability standard would have also lessened the burden of proof on the part of Uganda. Foreseeability would have required, for Uganda to liberate itself, to prove that a certain injury was not the consequence of its wrongful conduct, having the State displayed the required due diligence.
In conclusion, the Court’s reasoning in relation to Uganda’s violations of international human rights and international humanitarian law is wanting in two ways. First, the Court failed to clarify if the typology of primary rule has a bearing on the scope of reparation due – ie if there is any distinction between damage resulting from State’s commissive conduct and damage resulting from the actions of third private parties when the State had a duty to prevent these actions. Second, by introducing a distinction in the causal nexus based on the factual circumstance of occupation, the Court misconstrued the nature of Uganda’s responsibility for breaches of obligation to prevent violations of human rights and humanitarian law.
5. The Court’s approach toward damages linked to different violations of the prohibition of the use of force
Another example of the Court’s lack of principled reasoning in construing causality regards damages linked to the unlawful use of force. In the 2005 judgement on merit, the Court had found Uganda responsible for several violations of the principle of non-use of force and the principle of non-intervention committed between 1998 and 2003. Specifically, the Court had established that: i) Uganda had conducted unlawful military operations against the DRC in several parts of the country and that, in the province of Kigani, Ugandan troops had also engaged in large-scale fighting against Rwandan forces[49]; ii) outside the occupied territory of Ituri, Uganda had provided unlawful military, logistic, economic and financial support to irregular armed groups using force against the DRC. [50] In relation to ii), the Court had relied on the Nicaragua case to assert that although the conduct of irregular forces was not attributable to Uganda, this State’s responsibility was engaged for the sole fact of having provided these groups with military and financial support.[51]
One problem that emerges in the 2022 judgment regards damages arising out of Uganda’s unlawful support to irregular forces. The parties held different views on this point. For the DRC, reparation should include all damages resulting from the actions of irregular forces receiving unlawful support by Uganda, since ‘this damage could not have been caused without Uganda’s support’.[52] Conversely, Uganda claimed that, since damage ensued from multiple causes – the actions of the irregular forces and Uganda’s assistance and support – not all injury deserved reparation, but only damage that the Applicant could prove ‘it “was suffered as a result of” Uganda’s illegal support’.[53]
In sum, the parties’ disagreement underlined two consequential problems. The first was whether a State’s mere assistance and support to irregular forces implies that injury should be determined differently from situations where the State caused damage by actively conducting unlawful military operations with its own troops. The second and consequential issue was which causal test is adequate when injury results from the concurrent actions of private parties and State’s wrongful conduct.
The Court’s answer proved once again unsatisfactory. On the one hand, the ICJ recognized that conduct consisting of assistance and support to irregular forces amounts to an IWA and, as such, any injury stemming from it entails reparation.[54] It also observed that a State is not exempted from reparation for the sole fact that damage has resulted from concurrent causes.[55] On the other hand, the Court’s reasoning remains obscure in terms of applicable causal principles. For instance, in relation to damages to property outside the area of Ituri, the Court rejected the DRC’s estimation of property damages arguing that ‘by extending the claim to all damage to property that would not have occurred “but for” the unlawful use of force by Uganda, the DRC disregards the fact that the Court decided, in its 2005 Judgment, that armed groups operating outside Ituri were not under the control of Uganda’.[56] This quote implicitly confirms the distinction between actively conducting armed operations against a State and merely providing support to armed groups using force against a State.[57] It corroborates the idea that causality may vary depending on the type of primary rule. Yet, the Court never expressly says this, nor brings this argument to a sound conclusion. In the end, the ICJ dismisses the DRC’s claim without specifying how causality should be construed when the injury stems from the concurrent conduct of the State and the actions of private groups.
The only passage in the text where the Court addresses the problem of a multiplicity of causes regards the evaluation of injury consisting in damage to property and population displacement in the province of Kisangani. Having found that, in this area, displacements and property damages had resulted from the fighting between Ugandan and Rwandan troops, the Court argued that ‘each State is responsible for damage in Kisangani that was caused by its own armed forces acting independently’.[58] However, this assertion does not shed much light on the problem above. The case of damage arising from two complementary and concurrent internationally wrongful acts is indeed different from situations where damage ensues from the actions of private parties facilitated or supported by a State’s wrongful conduct.[59]
6. The Court’s approach toward the issue of macro-economic damages
A last point worth raising concerns the question of macro-economic damages deriving from the illegal use of force. To recall, the DRC had claimed more than five billion dollars for macroeconomic damages caused over several years by the unlawful use of large-scale force by Uganda. Uganda, on its part, maintained that macroeconomic damages resulting from armed conflict are not compensable under international law.
The Court’s justification for rejecting the DRC’s claim deserves attention:
This argument raises criticism as it points to the Court’s failure to acknowledge coherently that causality in the context of reparation is a legal rather than factual process. To begin with, the Court states that it is unnecessary to make normative considerations on the extent to which damages can be allocated in cases of violations of the non-use of force, since, as a matter of facts, the causal nexus between Uganda’s unlawful conduct and the alleged macroeconomic damages remains unproven. However, this logical leap is contradicted by the fact that the Court’s use of the ‘sufficiently direct and certain causal criterion’ to appraise the link between wrongful conduct and macroeconomic damages is itself an operation imbued with normative evaluations. As I argued at the beginning of this article, the selection of certain causal tests is driven by normative choices linked to how judges understand the function of reparation in the law of international responsibility. Hence, when the Court resorts to the ‘sufficiently direct and certain causal nexus’ to appraise the link between Uganda’s conduct and macroeconomic damages, the Court is engaging in an analysis of legal rather than factual causation.[61] This finds confirmation in the paragraph of the decision subsequent to the Court’s above statement:
As this passage amply demonstrates, the Court is not excluding that a causal nexus may exist between Uganda’s use of force and macroeconomic damages. The Court is simply stating that this causal link is not sufficiently direct and certain and, therefore, damages cannot be allocated to Uganda.
My point here is not to say that the Court should have concluded that causality existed, nor that it erred in picking the criterion of the sufficiently direct and certain causal nexus. My argument is that, by selecting the ‘sufficiently direct and certain causal link’, the Court made a normative evaluation on the scope of operation of the primary rule prohibiting the use of force. The problem is that, in the decision, the Court disguises this normative choice as a simple assessment of facts.
The Court could have reached the same conclusion with a more sounded argument recognizing that the appreciation of causality is not necessarily the same for each internationally wrongful act, since it depends on the primary rule, its purpose and ambit of application whether a particular injury falls within the scope of reparation.[63] For instance, the Court could have said that, since the prohibition of the use of force is ‘one of the most fundamental principles of international law’[64], macroeconomic damages qualify as compensable if there is a sufficiently direct and certain causal nexus between them and the wrongful conduct. Or, the Court could have made a different argument, saying that, despite the status of fundamental rule of the legal order, macroeconomic damages resulting from violations of the non-use of force are not compensable because reparation under the law of international responsibility does not admit damages that are too remote. The issue with the Court’s reasoning is that the Court appears to engage in a sort of ‘judicial self-restraint’ with regard to macroeconomic damages compensable under international law, justified by the factual circumstances of the case. However, the choice of the ‘sufficiently direct and certain causal nexus’ already underlines a legal evaluation that the Court makes in relation to damages compensable under the rule of non-use of force.
7. Conclusion
The Armed Activities judgment on reparation is an important contribution to question of causality in the law of international responsibility. For the first time the ICJ engaged with the problem of reparation for damages arising out of armed conflicts, making causal inquiries with complex evidence and fact-finding issues. To determine the scope of reparation for Uganda’s IWAs, the Court relied on the criterion of the ‘sufficiently direct and certain causal nexus’ between injury and wrongful conduct. This way, the Court upheld its previous jurisprudence and confirmed that the ‘sufficiently direct and certain’ criterion is likely going to become the reference standard for future judgments. From this perspective, the decision adds clarity and specificity to the generic requirement of Article 31 of ARSIWA, which incorporates causality into the elements necessary for reparation yet remains silent on how it should be construed.
At the same time, some of the arguments raised by the Court to analyze the causal link prompt some critical observation. First, the Court’s failure to substantially engage with the principle that causality may depend on the primary rules leads to certain ambiguities and contradictions throughout the judgment. On the one hand, the Court formally acknowledges this principle. On the other, it never truly applies it, instead introducing a distinction based on the factual circumstance that certain violations of international law occurred in an occupied territorial area. The problem is that this approach misconstrues the nature of Uganda’s responsibility for violations of international law occurred in the occupied territory, turning it – albeit implicitly – into a form of State’s liability. In the end, the inconsistency between what the Court says and what the Court actually does makes the reasoning on causation appear unprincipled and arbitrary.
Furthermore, while the causal inquiry remains for the most part a fact-intensive exercise, in the context of reparation the process is also imbued with normative considerations. That is, in assessing the scope of the injury, international courts and tribunals resort to legal other than factual causation. The Armed Activities judgment fails to make the distinction between factual and legal causation apparent, and instead overlaps the two concepts in more than one instance. The risk is again to present the reader with an argument on causality which seems occasionally inconsistent and lacking solid justification.
* Post-doctoral fellow, University of Milano-Bicocca.
[1] PCIJ, Case Concerning the Factory at Chorzów (Germany v Poland) (Merits) [1928] PCIJ Rep Series A No 17 at 47.
[2] ILC, ‘Report of the International Law Commission on the work of its fifty-third session (23 April-1 June and 2 July-10 August 2001)’ UN Doc A/56/10 (2001) II/2 YB ILC 26-30 [hereinafter ILC Report 2001] 91.
[3] ILC Report 2001 (n 2) art 31 (emphasis added).
[4] ibid para 10.
[5] See, extensively, V Lanovoy, ‘Causation in the Law of State Responsibility’ (2022) British YB Intl L 1–84; T Demaria, Le lien de causalité et la reparation des dommages en droit international public (Pedone 2021) 1–486.
[6] I Plakokefalos, ‘Causation in the Law of State Responsibility and the Problem of Overdetermination: In Search of Clarity’ (2015) 26 Eur J Intl L 471, 472.
[7] ICJ, Case Concerning Armed Activities in the Territory of the Congo (Democratic Republic of the Congo v Uganda) (Merits) [2005] ICJ Rep 168, para 345 [hereinafter Armed Activities (Merits) case].
[8] See the contribution by A Bufalini in this Zoom-in.
[9] ICJ, Armed Activities in the Territory of the Congo (Democratic Republic of the Congo v Uganda) (Judgment on Reparation) <https://www.icj-cij.org/public/files/case-related/116/116-20220209-JUD-01-00-EN.pdf> 9 February 2022 para 86 (emphasis added) [hereinafter Armed Activities (Reparation) case].
[10] ICJ, Armed Activities (Merits) case (n 7) Memorial of the Democratic Republic of the Congo on the question of reparations, paras 1.07, 1.09.
[11] ICJ, Armed Activities (Reparation) case (n 9) para 86.
[12] ibid.
[13] ibid para 93 (emphasis added).
[14] ICJ, Case Concerning Application of the Convention on the Prevention and Punishment of the Crime of Genocide (Bosnia and Herzegovina v Serbia and Montenegro) (Merits) [2007] ICJ Rep 43 para 462 [hereinafter Bosnian Genocide case] para 462.
[15] ibid (emphasis added).
[16] ICJ, Case Concerning Ahmadou Sadio Diallo (Republic of Guinea v Democratic Republic of the Congo) (Judgment on Reparation) [2012] ICJ Rep 324 para 14.
[17] ICJ, Certain Activities Carried Out by Nicaragua in the Border Area (Costa Rica c Nicaragua) (Judgment on Reparation) [2018] ICJ Rep 15 para 32.
[18] The ICJ is not alone in having developed the standard of ‘directness’ for assessing causality between injury and the wrongful act. See ITLOS, The M/V ‘Virginia G’ (Panama/Guinea-Bissau) (Merit) [2014] ITLOS Rep 4 para 436; ITLOS, The M/V ‘Saiga’ (No 2) (Saint Vincent and Grenadines v Guinea) (Merit) [1999] ITLOS Rep 10 para 172. For a reappraisal of the various standard developed by Courts and Tribunals, see Lanovoy (n 5) 43–60.
[19] ICJ, Armed Activities (Reparation) case (n 9) para 93
[20] ILC Report 2001 (n 2) 92–93 para 10.
[21] ibid.
[22] ibid.
[23] B Stern, ‘The Obligation to Make Reparation’ in J Crawford, A Pellet, S Olleson (eds) The Law of International Responsibility (OUP 2010) 569–570.
[24] See also J Crawford, State Responsibility: The General Part (CUP 2013) 493.
[25] G Arangio-Ruiz, ‘Second Report on State Responsibility’ (1989) II/1 YB ILC 56.
[26] See ILC, ‘Summary Records of the 2168th Meeting’ (1990) I YB ILC para 55; ILC, ‘Summary Records of the 2169th Meeting’ (1990) YB vol I para 2; ILC, ‘Summary Records of the 2170th Meeting’ (1990) vol I; ILC, ‘Summary Records of the 2171st Meeting’ (1990) vol I para 8; ILC, Summary Records of the 2172nd Meeting’ (1990) vol I para 9.
[27] See ILC, ‘Summary Records of the 2662nd Meeting’ (2000) ILC YB vol I 388, in which the Drafting Committee that ‘it had considered a number of suggestions for qualifying that causal link, but in the hand, it had taken the view that, since the requirement of the causal link were not necessarily the same in relation to every breach of international obligation, it would not be prudent or even accurate to use a qualifier’.
[28] See ILC Report 2001 (n 2) 92–93 para 10; J Crawford, ‘Third Report on State Responsibility’ (2000) ILC YB vol II/2 18.
[29] The global sum the ICJ finally awarded was less than 3% of the amount requested by the DRC.
[30] Plakokefalos (n 6) 478–479. On the distinction between factual causality and legal dimensions of causation see J Stapleton, ‘Legal Cause: Cause-in-Fact and the Scope of Liability for Consequences’ (2001) 54 Vanderbilt L Rev 942; A Belvedere, ‘Causalità giuridica?’ (2006) 52 Rivista di diritto civile 7.
[31] A variety of tests have been developed by courts and tribunals to assess factual causality. However, the most widely applied are the but-for-test or conditio sine qua non (a condition is the cause of a given event if and only if the event would not have occurred but for that condition) and the ‘necessary element of a sufficient set’ (NESS) test (a condition is the cause of a given even if and only if it was a necessary element among the other conditions that bring about the event). See for a detailed analysis HLA Hart, T Honore, Causation in the Law (2nd edn OUP 1985) 110–129; for an analysis of the application of the NESS test in international law Demaria (n 5) 180–182, 184–190.
[32] Lanovoy (n 5) 45–47.
[33] ILC Report 2001 (n 2) art 31 para 10.
[34] ibid; Crawford, ‘Third Report on State Responsibility’ (n 28) 19 (emphasis added).
[35] ILC Report 2001 (n 2) art 31 para 10.
[36] ICJ, Armed Activities (Merits) case (n 7) para 178.
[37] ibid para 178-179. See also para 345 in relation to natural resources.
[38] On the distinction between obligations of conduct and obligations of result see A Marchesi, Obblighi di condotta e obblighi di risultato: contributo allo studio degli obblighi internazionali (Giuffré 2003) 1-174. On the nature of obligations to prevent as due diligence obligations see A Ollino, Due Diligence Obligations in International Law (CUP 2022) 2112–117.
[39] Vienna Convention on diplomatic relations (adopted 18 April 1961, entered into force 24 April 1964) 500 UNTS 95.
[40] ICJ, Armed Activities (Merits) case (n 7) para 342.
[41] ibid para 58. At the reparation stage, Uganda decided to waive its counterclaim for compensation, see ICJ, Armed Activities (Reparation) case (n 9) para 396.
[42] On the issue of reparation in cases of breaches of preventive and other due diligence obligations, see D Jacob, ‘Le contenu de la responsabilité de l’Etat négligent’ in S Cassella (ed), Le standard de due diligence et la responsabilité internationale (Pedone 2018) 283.
[43] ICJ, Armed Activities (Reparation) case (n 9) para 95 (emphasis added).
[44] ibid para 77.
[45] ICJ, Armed Activities (Reparation) case (n 9), Separate opinion of Judge Yusuf, paras 14–21.
[46] See ICJ, Bosnian Genocide case (n 14) para 430; ITLOS, Responsibility and Obligations of States Sponsoring Persons and Entities with Respect to Activities in the Area (Request for an Advisory Opinion submitted to the Seabed Dispute Chamber) ITLOS Reports 2011 paras. 110. On the nature of due diligence obligations as obligations of conduct, see R Pisillo-Mazzeschi, ‘The Due Diligence Rule and the Nature of the International Responsibility of States’ (1992) 35 German YB Intl L 41–50; Ollino, Due Diligence Obligations (n 38) 98–108.
[47] On the understanding of responsibility for breaches of due diligence obligations as a form of negligence-based responsibility see S Besson, ‘La Due Diligence en Droit International’ (2020) 409 Recueil des Cours de l’Académie de Droit International 285–293.
[48] On foreseeability as the causal test normally used to determine breaches of due diligence obligations A Ollino, ‘A “Missed” Secondary Rule? Causation in the Breach of Preventing and Due Diligence Obligations’ in G Kajtár, B Çali, M Milanovic (eds) Secondary Rules of Primary Importance (OUP 2022) (forthcoming); on foreseeability as the causal test to be preferred to assess the scope of reparation in the law of international responsibility Lanovoy (n 5) 57–65.
[49] ICJ, Armed Activities (Merits) case (n 7) para 149–153.
[50] ibid para 160-165.
[51] ICJ, Case Concerning Military and Paramilitary Activities in and against Nicaragua (Nicaragua v United States of America) (Merits) [1986] ICJ Rep 14, paras 126–160.
[52] ICJ, Armed Activities (Reparation) case (n 9) para 80.
[53] ibid para 81.
[54] ibid para 83.
[55] ibid para 97.
[56] ibid para 250 (emphasis added).
[57] See ICJ, Bosnian Genocide case (n 14) paras 379–385; ICJ, Case Concerning Military and Paramilitary Activities in and against Nicaragua (n 51) paras 126–160.
[58] ibid para 253.
[59] On the problem of reparation in cases of multiple causes see Demaria (n 5) 172–175; Lanovoy (n 5) 65–74; P d’Argent, ‘Reparation, Cessation, Assurance and Guarantee of Non-Repetition’ in A Nollkaemper, I Plakokefalos (eds) Principles of Shared Responsibility in International Law: A Reappraisal of the State of the Art (CUP 2014) 224–231; B Bollecker-Stern, Le préjudice dans la théorie de la responsabilité internationale (Pedone 1973) 267.
[60] ICJ, Armed Activities (Reparation) case (n 9) para 381 (emphasis added).
[61] ibid para 382.
[62] ibid para 382 (emphasis added).
[63] ILC Report 2001 (n 1) Article 31 para 10.
[64] ICJ, Armed Activities (Reparation) case (n 9) para 65.
