Shared obligations and shared responsibility in international law

On 2 November 2017, Natasa Nedeski successfully defended her PhD on ‘shared obligations in international law’ at the University of Amsterdam (UvA). I had the honour of being a member of the reading committee. Nedeski happens to be a former bachelor student of us at Utrecht University (UU). She also pursued an LL.M. in Public International Law at UU, and was a junior lecturer at UU. She subsequently joined Prof. André Nollkaemper’s project on shared responsibility, of which this thesis on shared obligations is one of the outcomes.  In this post, I describe and commend the main findings of Nedeski’s thesis.

To what extent states and international organizations have shared obligations is perhaps not easily pictured. However, as states and organizations often pursue joint goals in a cooperative manner, it may not be uncommon for them to share obligations. Nedeski gives such examples as the obligation of states parties to the Nuclear Non-Proliferation Treaty to pursue negotiations on a treaty on general and complete nuclear disarmament, or the obligation of Australia and Nauru to take measures to prevent the inhuman treatment of asylum seekers and refugees held in offshore detention centres on the territory of Nauru. As international cooperation expands, so will the number of shared obligations. The following doctrinal questions then arise, and are central to the thesis: (1) which duty-bearers actually breach the shared obligation, (2) whether these duty-bearers share international responsibility in case of a breach, and (3) who is responsible to provide reparations.

The thesis is a remarkable work of doctrinal categorization and systematization, which elucidates the legal nature and consequences of the under-studied concept of shared obligations. Probably its main contribution is the distinction made between indivisible and divisible shared obligations. As Nedeski writes, an indivisible shared obligation binds multiple states and/or international organizations to achieve a common result, e.g., when a number of states oblige themselves to reduce their combined catches of a fish species by 50 per cent by a given date. Such a shared obligation is indivisible as multiple duty-bearers are jointly bound to achieve a common result, irrespective of the efforts undertaken by individual duty-bearers. In contrast, when states have a divisible shared obligation, each of them is bound to do its own share only. For instance, when in the example of the offshore detention centre given above, Australia commits to its obligations but Nauru does not and torture occurs, Australia may be released from its obligations.

Whether a shared obligation is divisible or indivisible is ultimately a  question of interpretation of the content of that obligation. A number of indicators could however facilitate the classification. Thus, Nedeski argues that if ‘a shared obligation is a positive obligation of conduct or a negative obligation of result, it is inherently divisible’. In her view, a shared obligation can only be indivisible ‘if it is a positive obligation of result, and even then it needs to be determined whether the duty-bearers in question are bound to achieve a common result … or whether each duty-bearer is bound to achieve its own individual result (in which case each duty-bearer is bound only to its own share’. In this light, only a minority of shared obligations are indivisible.

Nedeski then goes on to examine the consequences of breaches of shared obligations for the determination of responsibility for internationally wrongful acts. She argues that breaches of indivisible shared obligations will automatically lead to shared responsibility. In the example given, this means that when multiple states fail to reduce the catches, they will share responsibility. Nedeski submits that there is no such relationship between breaches of divisible shared obligations and shared responsibility. Such breaches could instead result in three possible outcomes: shared responsibility for multiple wrongful acts, e.g., when several states pollute a river; shared responsibility for one wrongful act, e.g., when several states have established an entity which pollutes the river; or responsibility of only one state, e.g., when only one state pollutes the river, whereas the others do not.

Finally, Nedeski also addresses the impact of breaches of shared obligations on the obligation of cessation of wrongful acts and the obligation to make reparation. As far as reparation is concerned, it is relatively self-evident that breaches of indivisible obligations will, in case of damage, result in an indivisible obligation of reparation. This obligation, which entails that each of them may have to provide full reparation for the damage caused, irrespective of their contribution, is comparable to the concept of joint and several liability as it is known in domestic private law. As far as breaches of divisible obligations are concerned, a distinction should again be made between shared responsibility for one or multiple wrongful acts. In the former case, exemplified by several states having established an entity which pollutes a river, the obligation of reparation will be indivisible. In the latter case, exemplified by several states polluting the river, the obligation of reparation will be divisible if a causal link can be established between each wrongful act (each act of polluting) and the damage. Often, however, such a discrete link cannot be established. Nedeski admits that in this situation, the obligation of reparation remains unclear. A possibility is obviously, again, resorting to the concept of joint and several liability, although one has to concede that this puts the duty-bearers at a clear disadvantage.

These considerations may sound relatively abstract, but one should bear in mind that how states exactly frame a shared obligation to reach a common goal – indivisible or divisible – has important consequences for these states’ responsibility. If the shared obligation is indivisible, a duty-bearer may incur responsibility even if it had made a major contribution towards the realization of the common goal, whereas the other duty-bearers failed to take any meaningful action: all duty-bearers will share responsibility if the common goal is not realized. Given these far-reaching consequences, states will not normally be inclined to encumber themselves with such shared obligations, unless they are in a particularly weak negotiating position. It is no surprise that most shared obligations are of a divisible character.

Nedeski’s work, in any event, is bound to inform the on-going debate in the international law of responsibility regarding whether obligations should be part of the law of responsibility. As the characterization of an obligation as ‘shared’ (divisible/indivisible) has undeniable consequences for the determination and content of the duty-bearers’ responsibility, incorporation obligations into the responsibility regime – so far resisted by mainstream lawyers – may make sense. In the meantime, I hope that Nedeski’s clarity of argument will reach the halls of international dispute-settlement bodies that are invited to rule on the responsibility of multiple duty-bearers. In fact, the International Court of Justice recently had an opportunity to do so in a case brought by the Marshall Islands against a number of states parties to the Nuclear Non-Proliferation Treaty which had allegedly failed to pursue negotiations on a treaty on general and complete nuclear disarmament. However, as the Court found that there was no actual dispute between the applicant and the respondents, and declared the case inadmissible. Accordingly, the Court did not touch upon the substantive questions of how to classify the obligations of the various states, and of how to conceive of responsibility.