The Oxford Guide to Treaties Symposium: Validity of Reservations Revisited

[Marko Milanovic is Lecturer at the University of Nottingham School of Law]

I am grateful to Duncan for inviting me to contribute to this conversation, inspired by his important new book. Let me continue where David Stewart and Harold Koh left off, namely with the issue of the validity of reservations and the innovations in that regard in the ILC’s freshly minted Guide to Practice. I would disagee with Professor Koh that the approach ultimately adopted by the ILC ‘makes little sense and smacks of unfairness’. On the contrary – the ILC has not only brought much needed clarity to the Vienna Convention regime, but has succeeded in reconciling seemingly irreconcilable doctrinal and ideological positions, in a compromise that I hope most governments will in the end find palatable. Let me try to explain why.

When one reads the (rather paltry) Articles 19-22 VCLT, particularly in light of the ICJ’s Reservations to the Genocide Convention opinion, one cannot avoid the impression that the process of determining whether a reservation was invalid as being contrary to the object and purpose of a treaty was meant to be more or less inter-subjective: each state should determine for itself whether a given reservation was compatible with the treaty’s object and purpose, and if it was not it should make an objection to that effect. But such an inter-subjective approach looks remarkably unappealing from the perspective of major multilateral normative treaties, particularly in the human rights context. The rights of individuals, so the reasoning among many human rights lawyers went, should not depend on the existence of objections, vel non, by third states, especially when reciprocity of state obligations has little place in the human rights context and when for a variety of reasons states routinely fail to object to reservations even when there are perfectly good reasons to do so. While objections to reservations would be probative, they could not be dispositive. It would indeed primarily be upon courts or treaty bodies to determine whether a reservation is compatible with the object and purpose of the human rights treaty, while the consequence of invalidity would normally not only be the nullity of the reservation, but also its severability, so that the reserving state would remain bound by the human rights treaty without the benefit of its reservation.

Many governments were less than pleased with what they saw as a power-grab by human rights bodies and a usurpation of their sovereign prerogatives. The ILC, being the bastion of international law orthodoxy, was no more pleased, nor was Alain Pellet as its Special Rapporteur. How could international law survive as a coherent, unified system if more of its branches followed the human rights example and asserted that because they were special they needed special rules, rather than the outdated Vienna framework. If that was true for human rights, why would it not be true for trade, the environment, or whatever other topic people became strongly devoted to. Fragmentation had to be resisted, and Pellet and the ILC could not, would not accept the ‘human rights are special’ argument. No true international lawyer, even a gentle, human rights-loving one, could accept its basic ideological premise.

But, during the (very) lengthy process of working on the Guide, the ‘generalists’ and the ‘human rightists’ actually talked to one another, including at a series of meetings organized in Geneva between the ILC and human rights treaty bodies. Rather than harden, their respective positions evolved. While from the generalist perspective the specialty claim could never be accepted, there was still room for compromise. Perhaps it was the general regime itself that could be so interpreted – or adjusted – to accommodate the concerns of the other side, and this time not just for the benefit of human rights. And so we have now the Guide, in which Pellet so very cleverly succeeded in reconciling positions that before seemed irreconcilable. He and the ILC did so by making a series of crucial conceptual moves.

First, according to the Guide, Article 19 VCLT should be regarded as laying down objective criteria for the validity of reservations, as inter-subjectivity is a recipe for chaos. Secondly, Articles 20-23 VCLT only deal with those reservations which are objectively valid under Article 19; they do not mention nor pertain to reservations which are in fact invalid. Thirdly, while states may object to reservations that they consider invalid, this is merely persuasive evidence of invalidity. In fact, objections only have real legal effect if they are made against reservations which are objectively valid; the objecting state may object for any reason whatsoever, simply because it does not want to accept the modified treaty bargain that the reserving state is offering. Fourthly, while the VCLT does not say what are the consequences of an invalid reservation, the only sensible option is to accept that such a reservation is null and void. Fifthly, however, saying that an invalid reservation is a nullity does not resolve the issue of the reserving state’s status as a party to the treaty. That will depend on the intention of the reserving state, which has a choice – either stay on as a party to the treaty without the benefit of the invalid reservation, or say that it no longer considers itself bound by the treaty, while acting under a rebuttable presumption that the reserving state intends to remain a party. (See gudelines 4.3 and 4.5).

Whether this is really the Vienna regime, ‘Vienna-plus,’ or something else entirely will, I imagine, be the object of some debate. But what seems to be beyond debate is that the Guide’s approach to the invalidity of reservation accommodates most of the human rights-inspired critique of Vienna without giving any ground to the idea of specialty. This is a general regime applying to all treaties, but it still moves from the inter-subjective approach in which state objections are the only thing that matters, it treats invalid reservations as a nullity, and it allows them to be severed. Yet they can only be severed if the reserving state does not actively oppose its continued status as a party to the treaty, with the presumption of severability operating on the unstated assumption that passivity is frequently the politically most likely reaction by states, and that the retention of the state within the treaty regime but without the reservations would be a probable outcome.

The Guide thus achieves a compromise that avoids fragmenting international law, accommodates concerns arising from the context of normative multilateral treaties, including human rights ones, yet still gives ultimate say on the matter to the principle of state consent. For instance, all the United States would need to do to avoid the severability of any of its reservations to the ICCPR, on the assumption that some of them in fact are objectively invalid, would be to say (as it has done so several times already) that it considers them an integral component of its consent to the entire treaty. That does not seem to be too onerous a burden, and is indeed a far cry from the approach articulated by the Human Rights Committee in its General Comment No. 24.

There is, in short, much to be commended in this anti-fragmentationist yes still progressive compromise, but it of course remains to be seen whether both governments and human rights institutions will appreciate it for what it is. I certainly hope they do.