* Professor, University Paris Ouest, Nanterre-La Défense; Former Member and Former Chairperson, International Law Commission of the United Nations. Email: courriel@alainpellet.eu.
Search for other works by this author on:European Journal of International Law, Volume 24, Issue 4, November 2013, Pages 1061–1097, https://doi.org/10.1093/ejil/cht067
14 December 2013Alain Pellet, The ILC Guide to Practice on Reservations to Treaties: A General Presentation by the Special Rapporteur, European Journal of International Law, Volume 24, Issue 4, November 2013, Pages 1061–1097, https://doi.org/10.1093/ejil/cht067
Navbar Search Filter Mobile Enter search term Search Navbar Search Filter Enter search term SearchThe purpose of this article is to revisit the long saga of the ILC Guide to Practice on Reservations to Treaties, as the Special Rapporteur has lived it for nearly 18 years and 16 reports. In its first part, the article recounts the elaboration procedure, pointing in particular to the elements of innovation and flexibility introduced in the process. The main one is the very type of instrument adopted, namely a Guide to Practice, and not a set of draft Articles that would eventually become a convention. In the second part, the main issues having retained the attention of the ILC, as well as of the other international bodies and of the academic community, are briefly recalled: the question of the unity or diversity of regimes, the permissibility of reservation and the status of the author of an impermissible reservation were among the most debated issues. Finally, the article explains the structure of the Guide to Practice.
On 11 August 2011, the International Law Commission (ILC) adopted a 630-page document 1 entitled ‘Guide to Practice on Reservations to Treaties’. This ILC product is doubly unusual:
The present article, the purpose of which is to introduce this very peculiar instrument, does not claim scientific objectivity. This author has devoted a non-negligible part of his working time for 18 years to preparing reports to the ILC on the topic of ‘Reservations to Treaties’ for which he was appointed the Special Rapporteur in 1994. These reports have been dissected, discussed, and in part rebuffed by the Commission or its Drafting Committee. However, even though, in some cases, the Commission made what I consider to be mistakes and undermined the global consistency of the draft, I must admit that, globally, the end-product has been improved owing to the careful scrutiny of the reports by the Commission or, more precisely, by a handful of able, interested colleagues whose input has been extremely positive, while the great majority seem to have been discouraged by the austere and technical character of the topic 6 and some – one at least 7 – have been extremely obstructive.
With this in mind, I will briefly describe the process which has led to the adoption of this instrument (1), before succinctly presenting some of the main issues and the more or less fortunate solutions adopted by the Commission (2).
The 1969 and 1986 Vienna Conventions on the Law of Treaties 8 devote six provisions to reservations: Article 2(1)(d) gives a definition of reservations for the purpose of the conventions, and Articles 19 to 23 provide general indications as to their legal regime. For its part, the 1978 Vienna Convention on Succession of States in Respect of Treaties confirms the 1969 definition 9 and limits itself to summary regulations concerning the rights and obligations of newly independent states in matters of reservations.
Given the difficult technical issues posed by reservations, their practical importance in international legal life, and the incomplete and sometimes obscure character of the rules embodied in the Vienna Conventions, it appeared at the beginning of the 1990s that the topic of ‘the law and practice relating to reservations to treaties’ would be a good candidate for inclusion in the programme of work of the Commission, as had been recommended by its Planning Group in 1992. 10 On this basis, the ILC decided to include the topic in its agenda in 1993 11 and, the following year, appointed its Special Rapporteur on the topic. 12
With excessive confidence – or recklessness – I then declared that ‘[i]t does not seem unrealistic to think that the Commission would be in a position to adopt an initial set of draft articles, or a first draft to serve as a “guide” … , within three or four years of the subject being included on its agenda and the appointment of a Special Rapporteur’. 13 This was a genuine belief: I thought that I had been assigned a gentle little topic, technical in nature, which could be dealt with within a few years. I rapidly became disillusioned and realized that, as my illustrious predecessors had noted, ‘the subject of reservations to multilateral treaties is one of unusual – in fact baffling – complexity and it would serve no useful purpose to simplify artificially an inherently complex problem’ 14 or, to put it in Reuter’s words, ‘[e]ven in the case of treaties between States, the question of reservations has always been a thorny and controversial issue, and even the provisions of the [1969] Vienna Convention may not have eliminated all these difficulties’; 15 moreover, the topic brings with it an emotional charge at the political level which I had underestimated and which made things even more complicated. 16 Indeed, 18 years seems – and is – too much and I have my share of responsibility for this excessive length; but the difficulty of the topic would certainly have excluded an acceptable outcome in the short period of time I had in mind when we started tackling it.
Besides its relative length, 17 the process which led to the adoption of the rather special instrument constituted by the Guide to Practice (B) is relatively classical, in that the ILC did not fundamentally move away from its usual practice (A).
Once on the agenda of the Commission, the topic was mostly dealt with in the usual way.
The Special Rapporteur first prepared a ‘Preliminary Report’ in which, after having summarized the previous work of the Commission on reservations, he tackled two main topics: (1) the problems left in abeyance, and (2) the possible forms of the results of the work of the Commission on the topic. 18 This second point was somewhat unusual in that, normally, the Commission decides on the final form of its drafts at the very end of its work on a topic. However, in the present case, the Special Rapporteur urged an early decision 19 and insisted on the specificity of the issue in relation to pre-existing treaty rules on the question. An early decision seemed necessary since you do not (or, at least, should not 20 ) draft a draft convention in the same way as you draft guidelines or recommendations.
In Chapter 1 of his preliminary report, the Special Rapporteur detailed the previous work of the Commission on reservations, on the occasion of the preparation of each of the three conventions on the law of treaties of 1969, 1978, and 1986. 21 The present article is certainly not the appropriate place to summarize these rather lengthy developments; however, one thing was striking: the incredibly conservative approach of the Commission. While, the ICJ had promoted a creative methodology to deal with the validity of reservations inspired by the pre-war Pan-American practice, 22 the ILC – whose statutory functions are both ‘the promotion of the progressive development of international law and its codification’, 23 resisted the adoption of the indispensable ‘flexible principle’ for more than a decade. It was not until Sir Humphrey Waldock’s first report, 24 in 1962, that the Commission changed its mind and at last sanctioned the fortunate progress promoted by the Court.
However, the recognition of the flexible approach (retained in the 1969 Convention) was clearly the result of a compromise reached thanks to a great deal of ambiguity. 25 As explained in the first report on reservations to treaties:
The most remarkable of these ambiguities results from the exact role of the ‘criterion’ of the compatibility of the reservation with the object and purpose of the treaty, to which the Convention ‘doctrinally’ pays tribute, but from which it does not draw any clear-cut conclusions. 26
The most perplexing question in this respect is the relationship between Article 19 of the Vienna Convention, 27 which sets out the rules concerning the ‘Formulation of reservations’ – in reality their validity – on the one hand, and Article 20, which concerns ‘Acceptance of and objection to reservations’ – in reality their opposability. 28 If you put the emphasis on the former provision, you will be seen as belonging to the ‘permissibility school’, in contrast to the ‘opposability school’, which focuses on the reactions of other states, as envisaged in Article 20. 29
The first report also highlighted various other ambiguities, lacunae, and shortcomings of the Vienna Conventions, 30 while emphasizing the global success of the reservations regime. 31 Moreover, it was noted that:
the 1969 Vienna Convention is, at one and the same time, the culminating point of a development which began long ago and which consists in facilitating participation in multilateral conventions to the maximum extent while preserving their purpose and their object, and the starting point of a multifaceted and not always consistent practice, which, on the whole, seems to be much more the result of considerations of political expediency based on a case-by-case approach than of firm legal beliefs. 32
Under these circumstances the (challenging) road map of the Commission seemed rather obvious: the future instrument should
With this in mind, the Commission endorsed the conclusions of the Special Rapporteur which constituted, ‘in the view of the Commission, the result of the preliminary study requested by [the] General Assembly’: 33
(b) The Commission should try to adopt a guide to practice in respect of reservations. In accord ance with the Commission’s statute and its usual practice, this guide would take the form of draft articles whose provisions, together with commentaries, would be guidelines for the practice of States and international organizations in respect of reservations; these provisions would, if necessary, be accompanied by model clauses[ 34 ];
(c) The above arrangements shall be interpreted with flexibility and, if the Commission feels that it must depart from them substantially, it would submit new proposals to the General Assembly on the form the results of its work might take;
(d) There is a consensus in the Commission that there should be no change in the relevant provisions of the 1969, 1978 and 1986 Vienna Conventions. 35
These directives were followed during the whole process of the elaboration of the Guide. In spite of accusations of disloyalty, the Special Rapporteur established himself as the watchful guardian of fidelity to the text of the Conventions – which is integrally reproduced in the Guide. 36 And, while not conceived as being eligible to become a binding instrument, 37 the Guide appears in effect as a succession of provisions (called ‘guidelines’ and not ‘Articles’), explained by abundant commentaries introducing the relevant case law, practice, and doctrinal views.
The next stages should logically have been a proposition for the future work of the Commission on the basis of its conclusions adopted in 1995 and a discussion of the definition of reservations. However, the second report on reservations to treaties deals with only the first of those aspects; its first chapter describes the area to be covered by the study of the Commission and its form, and sketches out the general outline of the study 38 – an outline which was globally followed during the subsequent work of the Commission on the topic. 39 The second chapter of the second report was different in nature. It was entitled ‘Unity or diversity of the legal regime for reservations to treaties’ and sub-titled ‘Reservations to human rights treaties’. 40
The Special Rapporteur considered that there were necessity and urgency for the consideration of the question – which had already been raised with some insistence during the debates in the ILC and the Sixth Committee the previous year – by the Commission: 41 in regional contexts, human rights courts had taken positions which could be seen as hardly defensible with regard to the Vienna rules on reservations 42 and, above all, the Human Rights Committee had just adopted, on 2 November 1994, its General Comment number 24 on reservations to the International Covenant on Civil and Political Rights, 43 which had been vigorously opposed by three states – and not insignificant ones. 44
The crux of the issue was whether it was open to a human rights body to assess the validity of reservations to ‘its’ convention and, in the event of an affirmative answer, what the effect of such an assessment was. As explained in the second report:
While it is obviously fundamental for human rights bodies to state their views on the question, the Commission must also make heard the voice of international law in this important domain, and it would be unfortunate for it not to take part in a discussion which is of concern to the Commission above all. 45
Based on this observation, the Special Rapporteur studied the issue of reservations to human rights treaties in the more general context of reservations to ‘normative treaties’ 46 following a two-stage approach:
By way of conclusion, the Special Rapporteur suggested that the Commission could adopt a resolution on reservations to normative multilateral treaties, including human rights treaties, a draft proposal of which was annexed to his second report. Although, on the substance, the propositions made by the Special Rapporteur were rather well received by the members of the ILC, the very idea of a resolution was rejected on the pretext that it was ‘a somewhat unusual procedure, … premature at the present stage of the Commission’s work on the topic … [and that] the text crystallized positions which were not yet entirely clear-cut and which might subsequently be changed’. 50 They were replaced by ‘Preliminary Conclusions of the International Law Commission on Reservations to Normative Multilateral Treaties Including Human Rights Treaties’ 51 adopted by consensus, 52 the status of which was even more uncertain.
Although the principles embodied in these ‘Preliminary Conclusions’ were rather balanced, 53 they received a fairly cold reception, to say the least, from human rights bodies and activists. 54 Until the very last stages of the study, 55 the question of the special regime, vel non, for reservations to human rights treaties remained an object of debates and a source of concern for the ILC and its Special Rapporteur. Various meetings were organized with human rights bodies and the Sub-Commission on Human Rights either individually or globally. 56 Although it was a time- and energy-consuming process, it has undoubtedly facilitated a better mutual understanding and it enabled me to have a better understanding of the issues and to propose what I think were more appropriate and realistic solutions than the ones initially envisaged. 57
This human rights excursion was the main derogation to the usual codification process through the ILC. 58 For the rest, the Special Rapporteur submitted his reports to the Commission, which, as is usual, introduced draft guidelines. 59 The reports were discussed in plenary, which then sent the draft guidelines to the Drafting Committee, 60 which carefully discussed them (and, sometimes, rather deeply modified them, usually for the better, sometimes for the worse). 61 The new drafts were re-discussed in plenary and usually adopted without any change, which enabled the Special Rapporteur to prepare the commentaries on the guidelines for inclusion in the report of the Commission after discussion in the plenary.
Then, and equally as usual, the guidelines and their commentaries were discussed by the Sixth Committee as parts of the Report of the Commission. Leaving aside the too often stereotyped character of the speeches in the Sixth Committee and the lack of preparation of too many delegations, there is clearly something wrong in this cycle of exchanges between the ILC and the Committee. Other than by transforming its drafts in a Penelope’s tapestry, the Commission cannot take into account the remarks made in the Sixth Committee – at least when they are made: it is only on the occasion of the second reading that the ILC can adapt its drafts in view of the states’ remarks. When a study spreads over a long period of time, it would in any case make no sense to change a work in progress every year. This is why I made a point of reviewing, as carefully as possible, all the interventions made by the delegates to the Sixth Committee over the years when we had to prepare the final version of the Guide. And this probably was the most unusual aspect of the preparation of the Guide.
Normally, the ILC’s drafts are subject to two different readings, separated by a one year fallow period during which states can prepare and send in their comments on the global draft adopted on first reading. This has not been the case concerning the Guide to Practice. It was completed at a forced march during the years 2009 and 2010 62 and, in accordance with the General Assembly’s wish, 63 a final version 64 was completed in 2011 – the text being adopted by consensus 65 on 11 August of that year. 66 To achieve this result, the Commission and the Secretariat (including the translators) had to make tremendous efforts and, in this respect too, a somewhat unusual method was followed:
Equally, during the second part of the session, the same Working Group was entrusted with the task of reviewing and finalizing the text of a draft recommendation or conclusions on the reservations dialogue and of a draft recommendation on technical assistance and assistance on the settlement of disputes proposed by the Special Rapporteur in his 17th report. Both texts were adopted by the Commission with some changes but were given different status: the Conclusions on the reservations dialogue constitute an annex to the Guide to Practice; 72 for its part, the Recommendation of the Commission on mechanisms of assistance in relation to reservations to treaties has been included in the Report of the Commission to the General Assembly, 73 but is not part of the Guide.
By contrast with the relatively classical process which led to its adoption, the Guide to Practice on Reservations to Treaties is a very special kind of instrument.
As explained above, the ILC decided at a very early stage of its study of the topic of reservations to treaties on the form the project was to take: instead of drafting an instrument eligible to be transformed into a convention, it was decided as early as 1995 that, subject to a possible change of mind, the Commission would draft a Guide to practice made of guidelines accompanied by commentaries. This carefully chosen terminology made clear from the outset that the Commission was not turning towards a binding instrument. And this was confirmed at the very end of its study when:
At its 3125 th meeting, on 11 August 2011, the Commission decided, in accordance with article 23 of its Statute, to recommend to the General Assembly to take note of the Guide to Practice and ensure its widest possible dissemination. 74
This makes clear that, unlike most products of the ILC, the Guide to Practice has not been designed as the basis for the adoption of a convention; deliberately so; and from the very beginning of the process. If the General Assembly follows the ILC’s recommendation, 75 it will remain what it is now: a soft law instrument mixing, however, hard rules with soft recommendations.
The reasons for this ‘modest approach’ were explained by the Special Rapporteur in his preliminary report:
166. [W]hat should be termed a ‘modest approach’ certainly offers great advantages:
(a) Amendment of the existing provisions would run into considerable technical difficulty: a State party to one of the existing conventions in force, or that might become a party, might very well refuse to accept such amendments as could be adopted; the result would be a dual legal regime of reservations that would be the source of very great difficulty under international law – at the present stage of its development, there is no means of imposing harmonization of the rules in force;
(b) … if the Commission could undertake the task of clarifying the existing provisions, that would at least make it possible to overcome most of the difficulties encountered;
(c) In their statements in the Sixth Committee of the General Assembly in 1993 and 1994, the representatives of States … expressed their support for the existing provisions. Above all, whatever their defects, the rules adopted in 1969 have proved their worth in that, on the one hand, they comply with the objective of flexibility which seems to have the support of States as a whole and, on the other, although their application gives rise to some difficulties, it has never degenerated into a serious dispute and, although, from the standpoint of principle, the protagonists have in some cases remained on opposite sides, they have always been reconciled in practice. 76
These reasons were at the origin of the decision to stick to the existing treaty law as embodied in the three Vienna Conventions and to adopt a non-binding instrument whose aim would only be ‘filling the gaps and . . . removing the ambiguities in the existing rules, but without embarking on their amendment’. 77
The option for a non-binding instrument complying with the three existing Vienna Conventions was never put into question later on. As explained in the Introduction to the Guide:
as the title and the word ‘guidelines’ indicate, it is not a binding instrument but a vade mecum, a ‘toolbox’ in which the negotiators of treaties and those responsible for implementing them should find answers to the practical questions raised by reservations, reactions to reservations and interpretative declarations, on the understanding that, under positive law, these answers may be more or less certain depending on the question, and that the commentaries indicate doubts that may exist as to the certainty or appropriateness of a solution. 78
Therefore, the Guide to Practice has been conceived as a means to assist practitioners, not as a united collection of rules compulsory for them. As a result, the guidelines have very different legal values, from pure recommendations to fully binding rules – not because they appear in the Guide, but because they have acquired (independently of the Conventions and, a fortiori, of the Guide) the status of customary rules. The Introduction to the Guide distinguishes between three levels of obligatoriness for the guidelines:
This last category is particularly significant: it could not have been considered to include in a draft convention a provision according to which, ‘[w]hen providing bodies with the competence to monitor the application of treaties, States or international organizations should specify, where appropriate, the nature and the limits of the competence of such bodies to assess the permissibility of reservations’. 80 It has fully earned its place in a non-binding instrument like the Guide, whose function is to assist the practice (whether administrative, legislative, arbitral, or jurisdictional) without thwarting necessary evolutions, but in guiding them.
The Guide has sometimes been criticized for its length and lack of manageability. There is some truth in this, but adding new ambiguities to the existing ones would not have been of great help. Moreover, in such a technical and controversial topic, clear-cut solutions would have been hopeless. It is nevertheless to be hoped that in a majority of cases, the user will find in the Guide the answers to the questions he or she is confronted with. And it is for this reason that the commentaries form an integral part of the Guide: 81 in a way, the guidelines are only the outline, the table of contents, of the Guide, the core of which is constituted by the commentaries.
This being said, even if the rules stated or proposed in the Guide should be followed in the absence of contrary special norms, none of them is peremptory by nature – which means that all are derogable. In other words, states and international organizations are perfectly welcome to provide for a special and derogatory regime for reservations formulated vis-à-vis a given instrument. 82 This possibility is too often underestimated by the drafters of international conventions, as well as by those criticizing the Vienna regime of reservations, including the human rights bodies and activists.
Before leaving the topic of the (non-)binding force of the Guide to Practice, it is in order to recall that two ‘recommendations’ have been adopted together with the Guide: ‘Conclusions on the Reservations Dialogue’ on the one hand, and a ‘Recommendation of the Commission on Mechanisms of Assistance in Relation to Reservations to Treaties’. 83 As recalled above, the proposals of the Special Rapporteur on these two sub-topics, which were made on an equal footing, 84 were treated differently by the Commission.
Concerning the text on the ‘reservations dialogue’, it must be noted that there exists no definition of this notion which is not a term of art. I used it first in my eighth report to designate a process followed by states (mainly European at the time) by which states ‘inform the reserving State of the reasons why they think the reservation should be withdrawn, clarified or modified. Such communications may be true objections, but often they merely open a dialogue that could lead to an objection but could also result in the modification or withdrawal of the reservation’. 85 As I tried to explain during the 2011 session, this expression alludes to the fact that, independently of the substantive and procedural rules applicable to reservations, contracting states and contracting international organizations could, and in many cases did, engage in an informal dialogue concerning the permissibility, scope, and meaning of reservations or objections to reservations formulated by a contracting state or a contracting organization. Such a dialogue, which can take place before as well as after a reservation was formulated, can take many forms and employ a wide variety of methods. 86 As I stressed, the reservations dialogue had the advantages of preventing positions from becoming fixed, allowing the author of the reservation to explain its reasons, and facilitating better understanding among the parties concerned. 87 The Commission was convinced and decided to attach, as an Annex to the Guide, the text of ‘Conclusions’ on the subject, ending up with a recommendation asking the General Assembly to ‘call upon States and international organizations, as well as monitoring bodies, to initiate and pursue such a reservations dialogue in a pragmatic and transparent manner’. 88
Although partly followed, my proposition to adopt another resolution, this time conceived as a recommendation to the General Assembly concerning technical assist ance and assistance in the settlement of disputes concerning reservations, was less successful. The recommendation is finally more ambiguous than I would have wished in respect of the settlement of disputes and, instead of being included in the Guide itself, it is lost in the Report of the Commission where it has neither a clear status nor any visibility. 89 The idea is that the General Assembly:
The first report of the Special Rapporteur had offered a ‘Brief Inventory of the Problems of the Topic’; 91 it was, indeed, anything but brief; nor was it comprehensive, as further discussions have shown. But it remains a good starting point to evaluate the difficulties of the topic. Some are general in nature, others are more specific – not much easier to solve however. Going into all these difficulties would go far beyond the scope of the present article. But it can be noted that many aspects which seem secondary in an overall perspective are sources of difficulties in the day-to-day practice of legal divisions of ministries of Foreign Affairs or international organizations, and give serious problems to practitioners, whether they are advocates or judges. The Guide to Practice’s ambition is to help them solve these problems.
Besides a general Introduction explaining its object and scope, the Guide to Practice on Reservations to Treaties is comprised of five different parts: 92
Each of these rubrics has raised unequally difficult issues. But it is worth noting that, if the ILC has promoted a single legal regime for all kind of reservations, 94 this was possible only after it had taken a clear position on the necessary unity of the Vienna regime.
As explained above, at the very beginning of the study I deemed it indispensable to discuss a preliminary general issue: are the rules applicable to reservations to treaties, whether codified in Articles 19 to 23 of the 1969 and 1986 Vienna Conventions, or customary, applicable to all treaties, whatever their object? Although the question could be asked for several kinds of treaties, it is raised with particular insistence in respect to human rights treaties. Since the answer to this question conditioned the drafting of several parts of the future Guide to Practice, it was dealt with in the second report on reservations to treaties. 95
In a large part, the issue is artificial, in that even the ‘hardest’ rules contained in the Vienna Conventions are residuary in nature; none is imperative or peremptory. 96 The Vienna regime merely supplements the will of the parties, which can always derogate to them by introducing in their treaty special provisions concerning reservations if they consider that the Vienna regime is inappropriate. And it could happen that a general practice of promoting special rules on reservations concerning certain types of treaties could be at the origin of a new customary regime, specific to those treaties. But, interestingly, this has not happened: no category of treaties – and certainly not human rights treaties – has generated a particular practice concerning reservations clauses. This is probably a sign that the negotiators of the treaties at least find the Vienna regime satisfactory and suitable.
And the issue was discussed at some length during the elaboration of the Vienna Conventions. However, after some rather heated exchanges, 97 the Commission:
decided that there were insufficient reasons for making a distinction between different kinds of multilateral treaties other than to exempt from the general rule those concluded between a small number of States for which the unanimity rule is retained. 98
And indeed these treaties which have to be applied in their entirety and the constituent instruments of international organizations are the only kinds of treaties for which the Vienna Conventions contain partially derogatory rules. 99
Moreover, concerning human rights treaties more specifically, 100 it can be recalled that the flexible regime adopted at Vienna has its (at least immediate) origin in the 1951 Advisory Opinion of the ICJ which was precisely given in relation to the fundamental and pioneering universal human rights instrument: the 1948 Genocide Convention. 101 Moreover, not only did the now universal Vienna regime directly originate from considerations concerning first human rights instruments, but this set of rules (or the most important of them: the compatibility of the reservations with the object and purpose of the treaty) has also expressly been referred to in the reservations provisions of human rights treaties as well as in recommendations of human rights treaty bodies themselves. 102
This is justified: the Vienna regime is well balanced, flexible, and adaptable. It strikes the right balance between the need for universality and the preservation of the integrity of the treaty – a balance which is sought for all kinds of treaties and which inspired both the majority and the minority in the case concerning Reservations to the Genocide Convention. 103 The traditional unanimity rule – according to which a state formulating a reservation could become a party to the treaty only if and when all other parties had accepted the reservation – would be extremely crippling in a world where nearly 200 states very different from one another can be concerned by a treaty. The new rule is well-tailored to the new conditions of international relations: it facilitates the participation in the treaty of all interested states while guaranteeing that the object and purpose of the treaty – that is its core content – will be safeguarded.
And it is difficult to perceive why this would not cope with the ‘special needs’ of human rights treaties – unless one accepts the intellectual terrorism exercised by some human rights extremists. 104 According to the ‘human rightist’ approach, 105 human rights treaties would be characterized by three main traits which would impede the application of the Vienna regime on reservations:
The non-reciprocity objection, usual as it is, is simply nonsensical:
Except for purely ideological reasons, there is no more ground for the allegation that by essence human rights treaties are not open to reservations. In its (most debatable) General Comment No. 24, the Human Rights Committee stated:
In an instrument which articulates very many civil and political rights, each of the many art icles, and indeed their interplay, secures the objectives of the Covenant. The object and purpose of the Covenant is to create legally binding standards for human rights by defining certain civil and political rights and placing them in a framework of obligations which are legally binding for those States which ratify; and to provide an efficacious supervisory machinery for the obligations undertaken. 111
Taken literally, this position would render invalid any general reservation bearing on any one of the rights protected by the Covenant. However, the Committee itself does not go that far and recognizes that reservations may usefully encourage a wider acceptance of the Covenant. 112
It remains that reservations to general human rights treaties raise specific difficulties. But this is caused not by their human rights object but by their global character. This is why the ILC which, in a first move, had envisaged devoting a particular guideline to the specific issues concerning the determination of the object and purpose of ‘general human rights treaties’, 113 realizing that there was no reason to individualize human rights treaties since the same considerations came into play for all treaties containing numerous interdependent rights and obligations, 114 eventually adopted guideline 3.1.5.6, 115 which attempts to strike a particularly delicate balance between these different considerations by combining three elements:
But, again, as the title of this guideline makes clear, these directives are not specific to reservations to human rights treaties; they apply to ‘reservations to treaties containing numerous interdependent rights and obligations’ in general.
Similarly, it is certainly desirable that ‘the compatibility of a reservation with the object and purpose of the Covenant … be established objectively, by reference to legal principles’. 117 But this holds true for all kinds of multilateral treaties, and is by no means limited to human rights treaties. Whereas the existence of monitoring bodies is certainly a peculiarity of human rights treaties, it is neither a necessary element of these instruments, nor an ‘exclusive’ peculiarity, 118 and indeed not an argument to modify the generally applicable reservations regime which bears upon the substantive principles to be applied by the competent authority to assess the validity of the reservation – whether a state, an international organization, a judge, or a monitoring body. The control of the compatibility of a reservation with the object and purpose of the treaty by independent bodies constitutes a guarantee of a more objective assessment of this rather subjective test. Monitoring consequently constitutes clear progress in the application of the Vienna rules, and therefore contributes to ensuring the integrity of the treaty in question by permitting an objective assessment of the compatibility of a given reservation with the object and purpose of the treaty – whether a human rights treaty or not. 119
After lengthy and difficult discussions between its members, between the ILC and the human rights bodies, and within the Sixth Committee, the Commission’s conclusion is unambiguous: the Vienna regime of reservations to treaties, as completed and specified in the Guide to Practice, is unitary and applies to all kind of reservations to all kind of treaties. 120 In so deciding, the 2011 Commission confirmed the sensible solution adopted by the Commission in the 1960s.
Part 1 may look the least problematic since the three Vienna Conventions give a similar definition of reservations. However, it is an important topic since the application (or not) of the reservations regime depends upon it – and the ILC devoted quite a long time to the related issues. 121
The main point probably was to make the distinction between reservations on the one hand and interpretative declarations on the other as ‘operational’ as possible. 122 It was all the more important that the Vienna Conventions do not mention the latter – a quite noticeable lacuna in the Conventions, which the Guide to Practice attempts to fill as far as possible, not only by giving a definition and tools for distinguishing interpretative declarations from reservations, but also by defining the full legal regime of the former. 123
One of the major difficulties was the fate to be reserved to the ‘conditional interpretative declarations’ defined in guideline 1.4. 124 There is no doubt that such unilateral statements do not correspond to the definition of reservations since they do not purport ‘to exclude or to modify the legal effect of certain provisions of the treaty’. 125 Nevertheless, by formulating such a declaration, states commit themselves only conditionally, just as they do when they formulate reservations. Given the dissimilarity in the definitions, I had systematically proposed draft guidelines dealing with the legal regime of these specific interpretation declarations. 126 However, it rather quickly appeared that, in spite of the variance in the definition, conditional interpretative declarations ‘behave’ exactly as reservations; this is why, in 2001, the Commission decided that ‘[s]hould this assimilation be confirmed in regard to the effects of reservations and of conditional interpretative declarations respectively, the Commission is considering the possibility of not including in its draft Guide to Practice guidelines specifically relating to conditional interpretative declarations’. 127 It was only in 2010, when it had become clear that these declarations followed the same legal regime as reservations, 128 that the Commission dropped all the draft guidelines already adopted in this respect and adopted paragraph 2 of guideline 1.4, according to which ‘[c]onditional interpretative declarations are subject to the rules applicable to reservations’.
Concerning the definition of reservations itself, three main points can be made.
First, I had – and from the very beginning of the research – deliberately envisaged clearly distinguishing between the definition of reservations and the issue of their validity. Going even further, I was – and am still – convinced that you can decide whether a reservation is valid or not only if you define the controversial statement as a reservation; in other terms, the definition must cover valid as well as invalid reservations. ‘It is only once a particular instrument has been defined as a reservation (or an interpretative declaration, either simple or conditional) that one can decide whether it is valid, evaluate its legal scope and determine its effect. However, this validity and these effects are not otherwise affected by the definition, which requires only that the relevant rules be applied.’ 129 I must say that, although I considered this point to be self-evident, I had to battle hard against several colleagues who had difficulty in accepting this, for me, elementary, reasoning. 130
It is true – and this is the second issue – that the Vienna definition itself is confusing, since it includes a temporal element which comes closer to a condition for its admissibility than to a definitional component. Nevertheless it has been included in the definition of reservations given in the Vienna Conventions; and this is why I hesitated for a long time on the position to be adopted for confronting the phenomenon of ‘late reservations’. I concede that, a priori, they do not come within the meaning of ‘reservation’ as defined in the Convention. On the other hand, this is a very formal view and it is logical (and, I would think, easily acceptable) to consider the time factor 131 as a condition of validity of a reservation. But this approach does not solve the problem: if the Vienna definition were to be taken literally, all reservations formulated late ought to be considered as invalid and without any effect whatsoever. This might be so in abstract law, but not in real life where examples can easily be found of reservations formulated late and producing all the consequences attached to a valid reservation with the approval of all the parties to the treaty. 132 This is why I have advocated prudent recognition of this fact of the legal life and maintained it, in spite of strong opposition from within 133 and outside 134 the ILC. Finally, the Commission has endorsed – but not in Part 1 on definitions – a series of guidelines on the ‘late formulation of reservations’, 135 which reintroduce the unanimity principle for these reservations. This reasonable solution coincides with the dominant practice and in tegrally preserves the consent principle. It may not be entirely orthodox, but ‘ayatollah’s law’ leads to deadlocks …
The third troubling issue concerning reservations is of the same nature, but less difficult – and it has given rise to fewer controversies. 136 It bears upon ‘across-the-board’ or ‘transversal’ reservations, that is ‘reservations which purport to exclude or to modify the legal effect of certain provisions of a treaty, or of the treaty as a whole with respect to certain specific aspects, in their application to the State or to the international organization which formulates the reservation’. 137 This kind of statement is not mentioned in the Vienna definition; however, ‘[t]he abundance and coherence of the practice of across-the-board reservations (which are not always imprecise and general reservations) and the absence of objections in principle to this type of reservations indicate a practical need that it would be absurd to challenge in the name of abstract legal logic’. 138
As for the rest, Part 1 of the Guide to Practice brings various clarifications to the definitions of reservations and interpretative declarations, the method of discriminating between the two, other unilateral statements, and various alternatives to reservations and interpretative declarations. 139 It also deals with ‘“Reservations” to bilateral treaties’; 140 the fact that the word ‘reservations’ is, unusually, written between inverted commas is telling: such statements, while currently called ‘reservations’, do not constitute reservations within the meaning of the Guide; 141 such a statement ‘appears to be a proposal to amend the treaty in question or an offer to renegotiate it’; 142 if accepted, it ‘becomes an amendment to the treaty’. 143
Of all five parts, part 2 raised the least controversial issues 144 except for the question of the late formulation of reservations. The guidelines comprising it were, however, carefully drafted and commented on, given the considerable practical importance of the issues concerned. There is not much to be said on sections 2.1 (Form and notification of reservations), 2.2 (Confirmation of reservations), or 2.4 (Procedure for interpretative declarations). 145 Section 2.5 gives useful clarifications on the ‘Withdrawal and modification of reservations and interpretative declarations’ on which the Vienna Conventions are largely mute. 146 Besides cautiously encouraging periodical review of the usefulness of reservations, 147 guidelines 2.5.7 to 2.5.11 elucidate the effects of the withdrawal of a reservation, whether full or partial, and the date on which such effects arise. Section 2.9 deals with the Formulation of reactions to interpretative declarations. 148
More interesting, at least from an academic perspective, are sections 2.6 and 2.7 on objections to reservations and their withdrawal or modification and 2.8 on the Formulation of acceptances of reservations.
Guideline 2.6.1 gives a definition of objections to reservations, which is missing in the Vienna Conventions. 149 This definition is, so to speak, the ‘negative’ carbon copy of that of reservations themselves, in that it characterizes an objection – exactly as guideline 1.1 copied from the Vienna Conventions does for reservations – not by its effects by but its ‘purported’ effects. 150 Moreover, the most important guideline, 2.6.2, establishes the right of states and international organizations 151 to formulate an objection ‘irrespective of the permissibility of the reservation’. This is a prominent element of the essentially consensual nature of the law of reservations: states have a right to formulate reservations; the other parties (or future possible parties) have their own right not to be bound to partners which do not accept the negotiated text in its entirety – whatever the reasons, 152 including by opposing the entry into force of the treaty as between the objecting state and the author of the reservation. 153
Nothing special deserves to be discussed in respect of the formulation of acceptances of reservations. 154 The Guide to Practice of course maintains the principle set out in Article 20(5) of the Vienna Conventions, according to which, unless the treaty otherwise provides, an acceptance results from 12 months’ silence kept by another state after the notification of the reservation. 155 The most noticeable clarification resulting from Section 2.8 is given by guideline 2.8.13, according to which ‘[t]he acceptance of a reservation cannot be withdrawn or amended’. 156
The main difficulties are concentrated in Parts 3 and 4 – one of them being specific to the English version, since English-speaking ILC members and the UK delegation in the Sixth Committee obstinately opposed the use of the word ‘validity’ to designate the fact that a reservation could produce its purported effects. 157 Hence the use of the word ‘permissibility’ in the title of Part 3 of the Guide to Practice (corresponding to validité substantielle in the French text) 158 – an expression which I disapprove of, since it seems to support the ‘permissibility school’ (by opposition to the ‘opposability school’), while the Commission and the Special Rapporteur have tried to keep away from any ready-made position. 159
Part 3 of the Guide starts with guideline 3.1 which simply reproduces Article 19 of the 1986 Vienna Convention, so seminal in the law of reservations. 160 This induced the Commission to try to specify as far as possible the impenetrable notion of ‘object and purpose’ which is at the heart of any assessment of the permissibility of reservations. To this end, guideline 3.1.4 gives a general idea of the meaning of the expression, 161 and guideline 3.1.5.1 suggests a method of determining the object and purpose of the treaty. 162 These crucial provisions are completed by guidelines 3.1.5.2 to 3.1.5.7 which give a series of examples bearing upon the most usual difficulties met in making this determination: 163
This article is not the place to comment on each of these points, which are the object of abundant commentaries. 164 However, two questions, not expressly dealt with in the guidelines, deserve some explanations:
I was prepared to meet huge difficulties in respect to the former, but I did not expect tricky discussions on the latter. The religious war did not come where it was anticipated: the exact opposite happened.
I must admit that, when reflecting upon my topic, I was apprehensive of the reactions Islamic states within the Sixth Committee or my Moslem colleagues could have regarding the ‘sharia reservation’, which I would have found it dishonest not to discuss. At the same time, I was (and still am) sincerely convinced that the issue was by no means the sharia by itself but the inacceptable specificities of certain reservations based on the sharia, specificities which can be found also in other reservations having no relation with the sharia or with Islam. Contrary to my fears, this view was endorsed without any difficulty by the Commission and, to my knowledge, did not lead to protests in the Sixth Committee. As the ILC notes in its commentary on guideline 2.1.5.2 in relation to a most objected to reservation: 165
the problem lies not in the fact that Mauritania is invoking a law of religious origin which it applies, [footnote omitted] but, rather that, as Denmark noted, ‘the general reservations with reference to the provisions of Islamic law are of unlimited scope and undefined character’ [Multilateral Treaties . , chap. IV.8.]. 166
The reason why such reservations are not admissible has nothing to do with their religious origin; it lies in the fact that their vagueness makes it impossible ‘to assess [their] compatibility with the object and purpose of the treaty’ 167 and, therefore, deprives the other parties of their right to react (by accepting or objecting to the re servation) with full knowledge of their meaning and scope. 168 The problems raised by reservations based on national law are similar: ‘a reservation is not invalid solely because it aims to preserve the integrity of specific rules of internal law’; 169 but it can be inadmissible either because the author of the reservation invokes its domestic law ‘without identifying the provisions in question or specifying whether they are to be found in its constitution or its civil or criminal code’. 170
As for the admissibility of reservations to treaty provisions reflecting a norm of jus cogens, I had, with hesitation, 171 proposed a draft guideline resting on a different assumption from the guideline relating to reservations to provisions reflecting a customary rule: draft guideline 3.1.9 accepted that the peremptory nature of the norms set out in the provision rendered the reservation impermissible. 172 However, this proposal met with serious objections during the debates in the Drafting Committee where it gave rise to a most passionate debate reflecting the sensitivity of my colleagues on all matters pertaining to jus cogens; this very lengthy discussion provisionally ended with a meaningless ‘compromise solution’ 173 which was abandoned in 2011 in favour of another unfortunate compromise:
This formulation implicitly recognizes that reservations to provisions reflecting a peremptory norm are subject to the same rules as reservations to provisions which reflect customary rules. It would have been simpler and franker to say it expressly, but some measure of hypocrisy sometimes makes consensus easier …
The other sections of Part 3 of the Guide to Practice are devoted to the assessment of permissibility of reservations (3.2), the consequences of the non-permissibility of a reservation (3.3) – a somehow misleading title, in that the main consequences are detailed in Part 4 176 – the permissibility of reactions to reservations (on which the Commission had little to say since it considered that ‘[a]cceptance of a reservation is not subject to any condition of permissibility’ 177 and that only the permissibility of objections with ‘intermediate effects’ was subject to limitations. 178 According to the general scheme of the various parts of the Guide, Part 3 ends with two sections on the permissibility of an interpretative declaration (3.5) and on reactions to such declaration (3.6).
Article 21 of the Vienna Conventions deals with the ‘Legal effects of reservations and objections to reservations’. 179 But, in reality, it says little on these effects, on which it sheds little light, and nothing on the effects of an invalid reservation.
In effect, as results from the first phrase of paragraph 1 of Article 21, it is limited to the legal effects of reservations ‘established with regard to another party in accordance with articles 19, 20 and 23’. Although some members of the ILC rather vigorously opposed sanctioning the concept of ‘established reservations’ in order not to create a new category of reservations, 180 the Commission considered it indispensable to clarify this notion from the outset; this is done in Section 4.1 (guidelines 4.1 to 4.1.3). The general idea is that three conditions must be met:
When a reservation is thus established, it produces the effects described in section 4.2:
Of course, these effects are partly paralysed when another state makes an objection to a valid reservation, as rather extensively developed in Section 4.3 of the Guide to Practice. The important point is the variations of the consequences of an objection, depending on the will of the objecting state. In this respect, without contradicting the Vienna Conventions, the Guide goes beyond their provisions, which envisage only two hypotheses: what is currently named objections with maximum effect (by which a state excludes the entry into force of the treaty between itself and the reserving state) 185 or with ‘normal effect’. 186 Guideline 4.3.7 for its part sanctions, although with caution, the existence of objections with intermediate effects, 187 by which an objecting state purports to exclude the application of ‘[a] provision of the treaty to which the reservation does not relate, but which has a sufficient link with the provisions to which the reservation does relate’. However, ‘in order to restore what could be referred to as the “consensual balance” between the author of the reservation and the author of the objection, 188 paragraph 2 of guideline 4.3.7 treats such an objection as a kind of “counter-reservation” to which the reserving State can make “counter-objections”’. 189
Indirectly guideline 4.3.8 (Right of the author of a valid reservation not to comply with the treaty without the benefit of its reservation) alludes to what is now usually called an objection with ‘super-maximum’ effect (at least purporting to produce such an effect) by stating:
The author of a valid reservation is not required to comply with the provisions of the treaty without the benefit of its reservation.
Thus, the ILC takes a clear-cut position – which was unanimously accepted – 190 on the question of objections ‘whereby the author of the objection affirms that the treaty enters into force in its relations with the author of the reservation without the latter being able to benefit from its reservation’: 191 they do not produce the effect purported by their author when the reservation is valid – such a consequence would be eminently contrary to the principle of consent.
But the question of the effects of objections is a different matter when they react to an invalid reservation. Section 4.5, which probably is the most innovative (since the Vienna Conventions are mute) and the most delicate portion of the Guide, sets out the consequences of an invalid reservation. The principle is that an invalid reservation ‘is null and void’, 192 independently of the reactions of the other contracting states. 193 But this only partially solves the question of the status of the author of an invalid reservation in relation to the treaty. The ILC’s more complete answer is given in guideline 4.5.3, which deserves to be reproduced in full:
Guideline 4.5.3 is the final outcome of a long and painful process, which, in fact, began with the ‘confrontational dialogue’ between the ILC and the human rights bodies following the adoption by the Human Rights Committee of its General Comment No. 24 and by the Commission of its Preliminary Conclusions on Reservations to Normative Multilateral Treaties Including Human Rights Treaties. 194 The HRC had declared that:
The normal consequence of an unacceptable reservation is not that the Covenant will not be in effect at all for a reserving party. Rather, such a reservation will generally be severable, in the sense that the Covenant will be operative for the reserving party without benefit of the reservation. 195
For its part, the Commission had concluded in 1997 that:
in the event of inadmissibility of a reservation, it is the reserving State that has the responsibility for taking action. This action may consist, for example, in the State’s either modifying its reservation so as to eliminate the inadmissibility, or withdrawing its reservation, or forgoing becoming a party to the treaty. 196
Both texts left the way open for less drastic and general solutions and, in effect, guideline 4.5.3 as finally adopted is midway between the two extreme positions: the Commission has maintained that, in conformity with its initial view, it was in principle for the reserving state to express its intention. On the other hand, it has taken an important step towards the position of the human rights bodies in that it accepts the principle of a (rebuttable) presumption in favour of the severability of the reservation (that is of the super-maximum effect of the reservation). However, several remarks can be made:
This shaky solution was probably the ‘least worse possible’ if one wanted to take into account the deep division between states on this quite crucial issue. And, unfortunately, it is unlikely that they will try to find any better compromise solution when the Guide to Practice is discussed again in 2013. 199 In my experience, states are not much inclined ‘naturally’ to compromise when what they consider (often erroneously) as being ‘questions of principle’ are at stake, even when a compromise would clearly be in the common interest. And concerning guideline 4.5.3, the chances that they will not move an iota are all the more plausible that ‘the harm is done’: unless the General Assembly decides to convene a diplomatic conference (which would be absurd) or clearly reject the whole Guide to Practice (which is unlikely since it would be throwing the baby out with the bathwater for one or two guidelines some states disapprove of), the Guide exists; it is published as an official document of the General Assembly and it is probable that delegations in the Sixth Committee will consider that their only means of influence is to stick rigidly to their position in the hope that it will be taken into consideration in the future ‘implementation’ of this non-binding document. And the views expressed by the states in the Sixth Committee or outside should indeed be taken into consideration.
However – and quite unfortunately – the chances are better than not that, instead of trying to define a common (reasonable and consensual) position on the most difficult issues, the delegates in the Sixth Committee will, as usual, give such cacophonic speeches that the message will be inaudible. Therefore whether the General Assembly takes note of the Guide to Practice 200 or not, it will live its own life; practice alone will be the judge of its adaptation to the needs of the international community of states (and international organizations) or whether it is desirable to adapt some of the rules it recommends following, to leave some aside or to adopt or progressively develop others. The non-binding nature of the Guide fits in this process of continuous adaptation.
Report of the ILC on the Work of its 63rd session, GA, Official Records, 66th Session, Supplement No. 10, Addendum 1, Doc. A/66/10/Add. 1, hereinafter: ‘Guide to Practice’ or ‘Guide’). Given its size and the necessity to have it translated into the sixth official languages of the GA, it was issued and posted on the Commission’s Website only in Jan. 2012.
The commentaries are nevertheless of the utmost importance in understanding and interpreting the ILC drafts, but they are not part of them and the drafts in question are, in principle, meant to be transformed into conventions – which is not the case with the Guide.
Guide to Practice, supra note 1, at 34, para. 1.