Symposium on the Genocide Convention: Is the Duty to Prevent Genocide an Obligation of Result or an Obligation of Conduct according to the ICJ?

Editor’s note: This is the final post in our blog symposium arising out of the Nottingham International Law and Security Centre conference to mark the 70^th Anniversary of the Genocide Convention. Read the other posts in this symposium here and here.

This post questions the findings of the International Court of Justice (ICJ) in the 2007 Bosnia v. Serbia case, according to which the duty to prevent a genocide is an obligation of conduct that can be assessed only after the occurrence of a genocide. The post first briefly explores the distinction between obligations of conduct and obligations of result on the basis of the International Law Commission (ILC)’s works and judicial practice. The post moves on to emphasise some inconsistencies in the ICJ’s reasoning in relation to the occurrence of a genocide as a prerequisite for the violation of the duty to prevent genocide. Finally, the post advances some possible explanations of the role of the event ‘genocide’ in relation to the duty to prevent genocide.

The 2007 ICJ’s Decision

In the 2007 Bosnia v. Serbia case, the Court for the first time declared that an autonomous obligation of diligent conduct to prevent genocide exists under Article I of the 1948 Genocide Convention (see my reflections here). According to the Court:

It is clear that the obligation in question is one of conduct and not one of result, in the sense that a State cannot be under an obligation to succeed, whatever the circumstances, in preventing the commission of genocide: the obligation of States parties is rather to employ all means reasonably available to them, so as to prevent genocide so far as possible. A State does not incur responsibility simply because the desired result is not achieved; responsibility is however incurred if the State manifestly failed to take all measures to prevent genocide which were within its power, and which might have contributed to preventing the genocide. In this area the notion of “due diligence”, which calls for an assessment in concreto, is of critical importance. (para 430, emphasis added)

The Court went on to affirm that a breach of the duty to prevent genocide can be assessed only after a genocide has occurred. The Court took the view that:

a State can be held responsible for breaching the obligation to prevent genocide only if genocide was actually committed. It is at the time when commission of the prohibited act (genocide or any of the other acts listed in Article III of the Convention) begins that the breach of an obligation of prevention occurs. […] If neither genocide nor any of the other acts listed in Article III of the Convention are ultimately carried out, then a State that omitted to act when it could have done so cannot be held responsible a posteriori, since the event did not happen. (para 431, emphasis added)

However, the view that a genocide must occur before a State’s compliance with the duty to prevent genocide can be assessed ignores the fact that this duty is a due diligence obligation of conduct. This conclusion is supported by the analysis of the evolution of the notion of obligations of conduct.

Obligations of Conduct under International Law

The first institutional effort to differentiate between obligations of conduct and obligations of result in international law was made by Roberto Ago, the ILC’s Special Rapporteur on State Responsibility. According to Ago, in relation to obligations of conduct, ‘a breach by the State of an international obligation specifically calling for it to adopt a particular course of conduct exists simply by virtue of the adoption of a course of conduct different from that specifically required’ (draft Article 20, YBILC 1977, vol. II, part I, 8), whereas ‘a breach of an international obligation requiring the State to achieve a particular result in concreto, but leaving it free to choose at the outset the means of achieving that result, exists if, by the conduct adopted in exercising its freedom of choice, the State has not in fact achieved the internationally required result’ (draft Article 21, YBILC 1977, vol. II, part I, 20).

Although Ago’s distinction was later removed from the ILC’s works on State responsibility, since it pertained to the structure of primary obligations rather than to secondary obligations (YBILC 1999, vol I, paras 7-8), this terminology was considered relevant by some international courts, which constructed this distinction very differently from Ago. Following a domestic law tradition of civil law countries, first envisaged at international level by some continental authors (eg, Reuter, 140ff; Combacau; Dupuy), today ‘obligations of means or conduct’ refer to those obligations requiring States to ‘to employ all means reasonably available to them, so as to prevent [an event] so far as possible’ (Bosnia v. Serbia, para 430). In relation to obligations of conduct, a State must ‘deploy adequate means, to exercise best possible efforts, to do the utmost, to obtain this result’ (ITLOS, Responsibilities and obligations of States with respect to activities in the Area, Advisory Opinion, 2011, para 110).

The Inconsistencies of the ICJ’s 2007 Decision

The current understanding of the concepts of obligations of conduct and obligations of result renders it possible to ascertain some inconsistencies in the decision on the Bosnia v Serbia case.

Indeed, despite considering the duty to prevent genocide an obligation of diligent conduct (para 430) and ‘irrelevant whether the State whose responsibility is in issue claims, or even proves, that even if it had employed all means reasonably at its disposal, they would not have sufficed to prevent the commission of genocide’ (ibid), the ICJ affirmed that the duty to prevent genocide is breached only when a genocide in fact occurs, following Article 14(3) of the DARS (para 431).

This assertion appears in conflict with the fact that the obligation at hand is an obligation of conduct rather than result, where States are responsible only in relation to their best efforts rather than in relation to the occurrence of a certain event.

The Court itself appears to be aware of the possible contradiction. Indeed, the ICJ was quick to specify that:

This obviously does not mean that the obligation to prevent genocide only comes into being when perpetration of genocide commences; that would be absurd, since the whole point of the obligation is to prevent, or attempt to prevent, the occurrence of the act. In fact, a State’s obligation to prevent, and the corresponding duty to act, arise at the instant that the State learns of, or should normally have learned of, the existence of a serious risk that genocide will be committed. (para 431, emphasis added)

This discrepancy has been criticised by some scholars, who considered it to be a ‘slight, if unavoidable, contradiction’ (Milanovic, 687), the result of ‘some conceptual difficulties’ (Gattini, 702), and an ‘unreasonable conclusion’ (Forlati, 200). Other scholars have supported the ICJ’s conclusion, but they had to argue that the duty to prevent genocide is not an obligation of diligent conduct (Crawford, 231-232).

 Conclusions: Possible Explanations of the Role of the Event ‘Genocide’ in the ICJ’s View

If the duty to prevent genocide is an obligation of conduct, the actual occurrence of the event ‘genocide’ cannot be the object of the prohibition embodied in this obligation. It is however difficult to assess which role the event ‘genocide’ has in the Court’s view.

The first possibility is that there is an additional kind of obligation of conduct which at the same time requires the causation of the event, on the basis of the structure of some specific primary rules (cf. Crawford, cit). This was Ago’s original view, according to which obligations of preventions were obligations of negative result (draft Article 23, YBILC 1978, vol. II, part I, 37). However, this conclusion is in conflict with the ICJ’s view that the duty to prevent genocide is an obligation of diligent conduct and it is not in line with the general understanding of States’ obligations regarding duties of prevention (Marchesi, 59).

A second option is that there is a specific secondary rule on state responsibility demanding the occurrence of the event (eg, Article 14(3) DARS). However, one could question the customary basis of such a rule that is clearly in conflict with the idea that obligations of diligent conduct focus on conduct rather than on result (van der Have, 11). Moreover, such an idea could appear as a way to introduce, in the realm of secondary rules, a further element of a wrongful act besides those two listed under Article 2 DARS, which lists a breach of an international law rule and its attribution to a State as the only elements of an international wrongful act (see the exclusion of ‘damage’ from the elements of wrongful act, as explained by Stern). Accordingly, all the elements that are related to the breach of international law pertains to primary rather than secondary norms (Commentary to DARS, 36, paras. 9-10), and, thus, the occurrence of the event ‘genocide’ cannot be justified under a special, uncodified, rule on state responsibility.

In conclusion, there is no legal reason to consider the actual occurrence of genocide as a legal factor that must be proved to assess a breach of the duty to prevent genocide. Rather, the occurrence of the event ‘genocide’ is only the factual reason why a State would start a case against another for a violation of the duty to prevent genocide. This would explain why State practice dealing with breaches of the duty to prevent genocide when a genocide has not occurred is so scarce. However, this conclusion would not preclude an assessment of the responsibility of a State that manifestly failed to intervene to prevent a genocide, irrespective whether or not that genocide occurred.