Jump to Content Jump to Main Navigation
The Law of Treaties Beyond the Vienna Convention edited by Cannizzaro, Enzo (17th February 2011)

Part IV Invalidity and Termination of Treaties, 18 Invalidity of Treaties: Anything new in/under the Vienna Conventions?

Joe Verhoeven

From: The Law of Treaties Beyond the Vienna Convention

Edited By: Enzo Cannizzaro

From: Oxford Public International Law (http://opil.ouplaw.com). (c) Oxford University Press, 2015. All Rights Reserved.date: 22 July 2019

Treaties, invalidity, termination, suspension, withdrawal — Vienna Convention on the Law of Treaties

(p. 297) 18  Invalidity of Treaties:

Anything New in/under the Vienna Conventions?

Clearly, the 1969 and 1986 Vienna Conventions on the Law of Treaties1 represented ‘new’ law in the sense that they constituted the first inter-state codification of the rules governing treaties at a universal level. It does not matter in that respect that the International Law Commission had originally decided that this codification should take the form of an ‘expository code’, and not of a treaty.2 Nor does it matter that ‘private’, ie unofficial, attempts at codifying were previously made (Harvard draft, etc) in a more or less successful manner.

Technically, codifying as such entails only the gathering of various provisions governing a given matter in a single instrument, ie a ‘code’. It does not necessarily imply that those rules are pre-existing, and even less that they are of a customary nature. It is surely not disputable that in history many codifications resulted in transforming into written law what were originally practices accepted as law in a given group (nation, society). But codifying still keeps its raison d’être even where it is concerned with rules lacking unequivocal support in the practice of a given group (nation, society). Consequently, as far as international treaties are concerned, the conclusions of the 1969 and 1986 Vienna Conventions in themselves neither imply nor exclude that their provisions are declarative of general law, ie are binding on all states and international organizations, even when they are not officially parties to them. Needless to say, it is somewhat puzzling to be worried about the ‘novel’ character of provisions contained in a treaty adopted more than 40 years ago as regards the states, and 25 years ago as far as international organizations are concerned. But it is still an important issue, until now left largely unanswered … despite blunt assertions of the ‘customary’ character of most of its articles.(p. 298)

There is no doubt that up to now ‘general’ law in international law has mostly been comprised of ‘international custom[s]’, ‘as evidence of a general practice accepted as law’ to use the terms of Article 38 of the International Court of Justice (ICJ) Statute. In particular, it means that establishing the customary character of the provisions of the Vienna Convention relating to the invalidity of treaties would normally require that evidence should be given of the existence of both a ‘general practice’ and a practice ‘accepted as law’. At the time the Vienna Conventions were adopted, such exercise was problematical in many respects. Indeed, the cases where the invalidity of a treaty was disputed remained very exceptional, and very ambiguous. The difficulty was not only to distinguish what respectively pertains to law and politics in a given case but also to define what were the specific legal requirements characterizing invalidity (which appeared uncertain). Case law in particular is almost totally silent in this regard.3

The customary technique is not, however, the only source of ‘general’ law. Other techniques could be used to produce ‘general’ rules. The best known is the technique of general principles of law as referred to in Article 38(1) of the ICJ Statute. Such a principle is undisputable when it establishes that its binding content is accepted as law in the domestic law of most, if not all, of the states forming the so-called international community. Which actually is often difficult to verify adequately … especially since domestic case law should normally be taken into account to determine the exact meaning of a legislative provision. That should explain why such general principles based on domestic practices remain somewhat exceptional. But apart from resting on convergent national practices, general principles should also be claimed on the sole basis of the internal coherence which is to be maintained in a ‘system’ of law, and which should make clear the necessary articulations existing between the various components of what is constitutive of a legal system. Needless to say, the exact content of such principles could, however, be questionable in many respects, at least to the extent that they do not coincide with general principles accepted as law within the meaning of Article 38 of the ICJ Statute. That should explain why many references to general principles, the exact nature of which remains uncertain, are made in the international legal literature so as to validate the declarative character of the provisions of the Vienna Conventions concerning the invalidity of treaties.4

Finally, it should be borne in mind that, apart from the difficulties relating to the technical support (custom, principle, etc) of a rule, the idea of a ‘new’ law also aims simply at the actual content of a provision which had been modified, no matter whether a ‘treaty’, a ‘custom’, or a ‘principle’ is at stake. And domestic (p. 299) practices should not be excluded in that respect, at least as a matter of principle, in assessing such ‘novelty’. Changes in the regulation of contractual relations could indeed be relevant for states’ regulations, even if they were originally concerned with individuals only.

It remains therefore rather surprising that commentators of the 1969 Vienna Convention, without clearly identifying the legal technique justifying their conclusion, generally consider that its provisions are declarative of general law, and consequently express rules having a binding effect even on states which are not parties to them (3). In such a perspective, the conclusion should be clear: if ‘new’ law means law having no previous support in the general practice of states, there was nothing which did basically constitute new law within the provisions of the Vienna Conventions governing the invalidity of treaties, despite the uncertainties relating to its customary basis. Apparently, there is only one important exception in that respect, apart from purely procedural issues: Article 53 of the Conventions affirming the invalidity of a treaty whose provisions are inconsistent with peremptory norms having a jus cogens character, which was forcefully opposed during the negotiations at Vienna by states who eventually refused to become parties to the Conventions. Today, the mood could, however, be changing in this respect. Indeed, some states are apparently more prepared to accept, or at least less displeased with, the principle of an ordre public in international relations.

The issue is not any longer whether or not to decide on the contractual or general (customary) character of the articles of the Vienna Conventions governing the invalidity of treaties, but rather to determine if, and in the affirmative to what extent, those provisions eventually modify the usual content of the traditionally applicable rules, no matter whether they are of a customary or purely conventional origin. ‘New’ law in such a perspective aims at rules whose content has changed. In such a perspective, however, the difficulty could be to determine the exact content of the ‘previous’ rule, ie of the rule by reference to which the ‘innovative’ character of the Vienna provision is to be appreciated. Since international law is at stake, this rule should be an ‘international’ one. But there is, as previously emphasized, almost no practice in that respect. As already pointed out, domestic law which sometimes provides the only credible reference should therefore not as a matter of principle be disregarded on the sole basis of its ‘national’ origin, even if the respective needs of a national and an international community cannot necessarily be considered as similar. Nor should they be considered as necessarily different.

As far as the law on invalidity is concerned, three kinds of issues have to be distinguished. They respectively pertain to the causes, procedure, and effects of invalidity. It has been previously pointed out that traditional practice was very poor on those issues. The Vienna Conventions did not change much from this point of view, apart from the fact that (relatively) clear written provisions now exist. Inevitably, support is consequently to be found in domestic practices to decide on difficulties left unsettled in the Vienna Conventions. Civil and common law (p. 300) traditions should of course both be considered to that end. But the provisions regarding procedure are in any case specific to international law, since there is still no compulsory jurisdiction within the so-called ‘international community’, whatever those terms, still partly controversial, might exactly mean. Provisions concerning the effects of invalidity also partly reflect some structural characteristics of that community, which a priori does disqualify to some extent references made to national practices in constructing them. But grounds of invalidity were directly inspired, at least as far as the defects in consent (‘vices du consentement’) are concerned, by civil law traditions … despite the fact that most of the rapporteurs within the International Law Commission were common law lawyers.

1. Grounds of Invalidity

The very possibility of the invalidity of a legal act is probably axiomatic. How indeed could such an act produce its legal effects if the conditions which are basically required to that end are not satisfied? The difficulty, however, is to agree on the content/nature of such basic conditions, especially in a system still largely rooted in customs, ie in (state) practices as distinguished from mere usages (habits).

Those conditions remained largely uncertain until the adoption of the 1969 Vienna Convention on the Law of (state) Treaties, cases of nullity remaining—as previously emphasized—very much the exception. The basic provision probably was simply that the consent of the state to be bound is null and void when it has not been ‘regularly’ given. What does the term ‘regularly’ concretely intend to convey? Basically, firstly that the consent be expressed by a person duly authorized to that end and, secondly, that it be expressed ‘freely’.5 Failing other express provisions, it should depend upon the peculiarities of each case as to whether both requirements are satisfied or not. This conclusion does not express any ‘political’ choice; it is only evidencing fundamental requirements of a legal transactions theory. That said, the proposition remained largely theoretical till at least the conclusion of the Vienna Conventions—which identified much more specific grounds of invalidity—especially in the absence of a judge before whom the case of invalidity could be brought. Of course, nothing was preventing the parties from agreeing on the validity of ‘their’ treaty. However, the hypothesis remained apparently exceptional. And in any case the very fact that there was an agreement to nullify would have been muddling what is due respectively to the sovereign will of the parties and to the logic of invalidity.

Those hesitations disappeared with the 1969 Vienna Convention, which listed the admissible grounds of nullity in its Articles 46 to 53 and Article 42; (p. 301) these articles make clear that ‘the validity of a treaty or of the consent of a State to be bound … may be impeached only through the application of the present convention’.6 As previously emphasized, these articles to a large extent reproduce in substance a theory of defects in consent (‘vices de consentement’) very well known in civil law countries whose legal regimes are based, directly or indirectly, on the Napoleonic code. The point is worth emphasizing in a context where common law traditions tend to become the usual reference.

According to the Vienna Convention, six grounds of invalidity are admissible: lack of competence (Articles 46 and 47), error (Article 48), fraud (Article 49), corruption (Article 50), coercion (Articles 51 and 52), and violation of a peremptory norm (Article 53). Such a list probably includes all the defects usually admitted in civil law countries as vitiating the consent, but two of them were not (always) taken into consideration: lack of competence and corruption. Does that imply that the provisions of the Vienna Conventions are ‘new’ in that regard? Probably not.

1.1. The issue of competence does inevitably arise when a legal person (personne morale) is concerned, since it necessarily requires the intervention of a physical person to express its consent, whether orally or in writing. And logically the former could only be validly bound by the consent given by the latter if that physical person was ‘competent’ in that regard. National and international laws are faced with the same difficulty. At first sight, there should not be important divergences of view in this respect. The difference, however, is still that all subjects of international law—apart from the reservation concerning the present evolution concerning individuals—are legal persons. Consequently, the problem of competence is necessarily raised when an ‘international’ agreement is entered into, while it remained much more ‘exceptional’ under domestic laws. This probably explains why the lack of competence is not usually analysed as a specific ‘vice de consentement’ in domestic law. As such, it is a problem for the law governing legal persons, and not for the law governing contracts. But there is no difference between national and international law, subject to this remark.

Should there be a difference, it would arise from the fact that, according to Article 46(1), the lack of competence would only result in invalidating the consent when it was ‘manifest and concerned a rule of its internal law [regarding competence to conclude treaties] of fundamental importance’. No such restriction normally applies where contracts, as distinguished from treaties, are concerned, even when the process of their conclusion is as complex as the one applicable to treaties. Any lack of competence could normally be invoked to declare the contract null and void, at least when it was not covered in one way or another by (p. 302) the contracting party. Does that constitute ‘new’ law as far as international law is concerned? At the time when the Vienna Conventions were concluded, the answer would probably have been affirmative, no practice—not even doctrine—really supporting the idea that a state could not invoke the incompetence of its legal representative, at least when it was acting in good faith. This is probably also the position in domestic laws, especially since general rules would normally be substituted for the nullified contract, so as to provide clear legal answers when needed. From that point of view, the situation is quite different as far as international law is concerned, at least as long as the customary process remains the only usable process to produce ‘general’ law. This should explain why only a qualified lack of competence has been accepted as a ground of invalidity in the Vienna Conventions. Does that mean that the restrictions expressed in Article 46 of the Vienna Conventions only apply as a matter of principle in the relations between contracting states? This would be surprising, almost 50 years after the adoption of the 1969 Vienna Convention on the Law of Treaties. Even if practice remains meagre, Article 46 provisions are so widely accepted, at least by legal scholars and other experts, that it would be somewhat amazing if they were characterized as purely conventional, even if intellectual consensus is not enough —thank God!—to create law! Be that as it may, the conclusion is no less artificial than the one affirming that the mere fact of not invoking the invalidity, supposing it to be established, constitutes a kind of acquiescence curing the invalidity.

Should the ‘restrictions’ of competence, aimed at in Article 47 of the Vienna Convention, be distinguished, when they are ignored by the state agent concerned, from the lack of competence aimed at in Article 46? Probably not; in both cases indeed, the state representative acted outside of its area of competence. According to Article 47, however, ignorance of such a restriction could only invalidate the consent expressed when it was ‘specific’—whatever that means exactly—and had been notified to the other contracting parties. The exact purpose of such a provision nevertheless remains uncertain, the difference between both articles being apparently that the restrictions aimed at in Article 47 do not imply a violation of the ‘internal law’ of the state concerned, as mentioned in Article 46. Which remains problematical … that said, Article 47 has until now remained without any practical interest (applications). Consequently, it does not matter whether it is considered as ‘new’ law7 or not.

1.2. Error, fraud, and corruption are admitted as grounds of invalidity according to Articles 48 to 50 of the Vienna Convention, at least when they have, in one way or another, induced the victim state to consent to be bound by the treaty.

Some pre-Vienna Conventions cases of ‘error’ are established in international practice. The customary character of Article 48 should therefore not be disputed, even if some disagreement may exist concerning the exact conditions which (p. 303) must be satisfied to invalidate the consent. Proven cases of fraud and corruption remain exceptional, however. Some commentators are apparently suggesting that they (should) have been rather common business. But such allegations are insufficient to constitute the factual basis of a customary rule. It would therefore be hazardous to affirm the declaratory character of Article 48 and Article 508 if it is allegedly based on the existence of a general practice accepted as law.

Could general principles of law support a conclusion which is not validated by customary techniques? Why not, in principle? Domestic laws, however, do not clearly confirm that the contract or at least most of the contracts reached through corruption are void, even when corruption constitutes a legal offence, at least to the extent that it did not either cause an error or constitute a coercion opening the way to an invalidation. As far as Article 49 is concerned, it is still surprising that it should be entitled ‘fraud’ in the English text and ‘dol’ in the French as if both terms were identical,9 which they are not.10 A ‘dol’ is a kind of qualified ‘fraud’ unknown, however, in legal systems which are not close heirs of the Napoleonic code and which consequently have serious difficulties in understanding the specificity of that qualification.11 The precedent of the 1929 Warsaw Convention for the unification of certain rules related to international carriage by air is well known in that respect. To rule out the limitation of the responsibility normally benefiting the aerial carrier, its Article 25 provides indeed that such a limitation does not apply, according to the French text of the Convention, when ‘le dommage provient de son dol ou d'une faute [ … ] considérée comme équivalente au dol’, which becomes in the English text: ‘if the damage is caused by his wilful misconduct or by such default [ … ] considered to be equivalent to wilful misconduct’.12 But twenty-five years later a specific protocol was needed to explicate the meaning of a term which has remained particularly obscure for the (judicial) authorities of most of the contracting states. In the amending Hague Protocol of 1955, the formula became: ‘if the damages resulted from an act or omission [ … ] done with the intent to cause damage or recklessly and with knowledge that damage will probably result’ (Article 25 A). In such a context, the reappearance of the ‘dol’ in a treaty on the law of treaties between states is rather amazing, especially since the term (p. 304) ‘fraud’ as used in the English text of Article 49 does exist in French as well, and since ‘wilful misconduct’ and not ‘fraud’ was considered as equivalent to ‘dol’ in the 1929 text. That said, the important point is still that the substance of the article is identical in both texts. From this point one should be induced not to pay (too) much attention to the titling of the article and the discrepancy in its translation, even if official titles should normally be taken into account when construing their content.

1.3. The two cases of coercion referred to in the Vienna Conventions are widely accepted today, even if they raised much debate at the time they were discussed, first within the International Law Commission and later at the Vienna Conference.13 The duality of articles is justified by the distinction made between coercion of the state as such and coercion of its representative only. Does it imply that in the absence of a specific provision concerning the state's representative, coercion could not have been taken into account in deciding on the validity of the consent? Clearly not. Having said that, the interest in distinguishing Article 51 from Article 52 still remains that the treaty as such is void when it was generated by the threat or use of force against the state as a whole while its consent to be bound only is ‘without any legal effect’ when its representative only is concerned. Does ‘logic’ or ‘coherence’ support the distinction, in the absence of any consistent practice in that regard? It depends on what ‘logic’ means. At least, it is coherent to deny any result to the use of armed force when it is illegal, ie when it was exercised in violation of the UN Charter. But could the distinction be based on a general principle of law? Probably. With that reservation in mind, nothing should therefore be considered as constituting ‘new’ law as far as coercion is concerned. And that should be regretted to a certain extent.

It is well known that economic coercion was, for instance, not admitted as a ground of invalidity at Vienna, despite the demands of many, mostly developing, countries. Accepting such a ground in the absence of any use of military force, however, appeared much too risky to the majority of the participants at the Vienna Conference, including almost all developed countries. Possibly, such caution was wisdom 50 years ago. Is it still? That is at least doubtful. It is interesting to note that national tribunals traditionally reluctant to take into account economic violence now move towards accepting it. And proposals are officially made to modify the civil code so as to give a legislative foundation to a judicial evolution. It is not that economic pressure is now illegal; it is simply (p. 305) that a contract obtained by abusing the weak position of another party to gain advantages it would not have given had it not been in a situation of dependence or necessity is voidable.14 Contracts under domestic law cannot, of course, be equated with treaties under international law. There is no reason, however, for excluding categorically that treaties obtained through economic coercion might be voidable at the request of the victim state. Everybody knows that at Vienna the compromise solution was to declare in an annex to the Convention that economic violence was illegal but not to mention it explicitly in the articles relating to coercion.15 Possibly that was a wise solution at the time the Vienna Convention was negotiated. But there is nothing to suggest that it could never constitute a ‘violation of the principles of international law embodied in the Charter of the United Nations’, to quote the terms of Article 52 of the Vienna Convention.

1.4. There was, and up to a certain point there still is, much debate about Article 53 of the Vienna Convention, and the so-called jus cogens it was officializing. Indeed, the article did not rest on any explicit practice existing before its adoption, and it was as such fiercely opposed by some state officials or legal scholars. Why? That remains unclear. The justification apparently was that a state is sovereign, whereas a private contracting entity is not, and consequently the validity of what it has agreed never might be impeached by any general rule of international law, customary or not, ‘peremptory’ or not. French authorities, or even scholars, were adamant in that respect.16 This remains rather strange. The jus cogens norm is indeed simply a norm, necessarily general, ie normally of a customary character in the present state of international relations, which cannot be derogated from by a treaty (contract) because it was protecting a ‘general’ interest necessarily surpassing the private (particular) expectations of the contracting parties. As such, that is far from being revolutionary. ‘Ordre public’ is very well known in any domestic (national) system.17 Why should it be outlawed in international law? Of course, few norms could clearly be said to be ‘peremptory’ because of the present state of disorganization of a ‘community’—to use the contemporary terminology (it used to be a ‘family’ before)—whose members still disagree on many, sometimes very important, issues. That should explain why jus cogens has not proven until now to be of a great use, and consequently why its customary support is (p. 306) rather weak, if not non-existent.18 But it is still axiomatic19 in any system of law that private parties may never by contract endanger the essential values and interests of a given ‘community’.20 Customary practice, requiring a certain number of similar cases expressing converging solutions, should not be necessary therefore to validate in international law the specific effects of the violation of the ‘peremptory’ norms aimed at in Article 53 of the Vienna Convention.21

In other words, and contrary to a rather common opinion, jus cogens is anything but ‘new’ law. It could be politically wise to affirm that Article 53 constitutes a ‘progressive development’ rather than a strict ‘codification’ of international law. But so-called progressive development does not mean much, from a legal point of view, at least where ‘axiomatic’ rules are concerned. That said, the concrete usefulness of jus cogens is still limited. And it will remain weak as long as the content of the ‘intérêt général’ within the community of nations remains poor. But the legal machinery of the ‘ordre public’ (jus cogens) exists, no matter whether its utilization remains exceptional or not. Unless that conclusion is proved to be ill-founded, it is still strange that some states apparently persist in refusing to become party to the Vienna Convention because of its Article 53, which is probably the only provision that need not be accepted. Needless to say, determining the utility of jus cogens outside of contractual matters is another story. Whatever the conclusions in that regard may be, and they may sometimes be rather fanciful—as the assertion that jus cogens necessarily entails jurisdiction of the ICJ when its content/existence is disputed—it does not affect Article 53 as such.

If need be, it must be emphasized that this has nothing to do with what is called, according to French terminology, the ‘cause’ of a contract. In many civil law countries at least, a contract is void when its ‘cause’ is proved to be illegal (illicit). What such ‘cause’ exactly entails remains partly (largely?) unclear. The idea, however, is that it is not enough for a contract to be valid that it is expressing the ‘sovereign’ will of the parties. The reason why or the interest explaining why that was actually the will of those parties also needs to be taken into (p. 307) consideration in that regard, so as to be able to invalidate, for instance, the contract whose specific ‘aim’ is clearly illicit. There is (was) much debate in (some) civil law countries on the soundness of this rather obscure requirement, the contemporary practice apparently tending towards dismissing such an approach. And in common law countries the so-called ‘cause’ apparently should not go beyond what ‘consideration’ implies, which means that the undertaking of one party should be the only ‘cause’ of the promise of the other party. Apparently the ‘cause’ of a treaty was never referred to in international (treaty) practice.22 And it is such a confusing notion that it would certainly be preferable not to make any use of it. But in any case ‘cause’ is normally unrelated to ‘ordre public’, which only invalidates an agreement when its content, ie what the parties agreed, is illicit. And it is all the more difficult in international relations to accept it since states at least will probably strongly oppose any control of what they had in a sovereign manner decided, at least beyond what jus cogens provisions require.

1.5. As previously underlined, no other ground of invalidity is in principle admissible according to the Vienna Conventions.

What about ‘capacity’, which traditionally is a basic requirement? There clearly is no problem as far as states are concerned, Article 6 of the 1969 Vienna Convention affirming that ‘every State possesses capacity to conclude treaties’. None should disagree with the conclusion. The only difficulty could be to establish the existence of the state when it remains uncertain. Controversies regarding the validity of treaties concluded by or with unrecognized states are for instance well known, as are the problems relating to federal states when the federated entities are claiming a separate capacity to conclude treaties. There is no problem in this respect as long as those entities are acting as ‘organs’ of the federation in which the international personality of the state is vested. But in that case the treaty is a treaty of the federal state, no more and no less than the one concluded through a federal organ. Subject to that remark, there is for the time being no international capacity of federated entities to conclude treaties, at least none that is clearly accepted. But the 1969 Vienna Convention is not concerned with the problem, since it deals only with ‘state’ treaties.

Does that mean that, apart from states, no other international person is vested with a capacity to conclude agreements under international law? Clearly not, quite the contrary. The very fact of enjoying an (international) personality should necessarily entail the capacity to enter into relations with others, and especially to conclude agreements. Affirming the contrary would be inconsistent with the very notion of being a person, no matter whether international law or any other legal order is concerned. Non-state persons remain, however, (p. 308) exceptional in present international law. International organizations possibly constitute the only undisputed non-state legal persons in present practice, excepting the colonial peoples … which possibly no longer exist. It is consequently not surprising that a separate convention was concluded to determine the legal regime of the treaties concluded by or with international organizations (1986 Vienna Convention). Why, however, should a special regime be necessary in that respect, domestic laws making clear that there is substantially no difference in the regime of contracts depending on whether a legal person (personne morale) or an individual is a party? That remains unclear, apart from affording the opportunity of checking whether a difference could be justified or not, which actually was the case as far as Article 66 was concerned. The interesting point, however, is that according to Article 6 of the 1986 Vienna Convention ‘the capacity of an international organisation to conclude treaties is governed by the rule of that organisation’, which contrasts with Article 6 of the 1969 Convention just quoted. The exact significance of that article is however, unclear. Certainly, the capacity of an international organization to conclude treaties is to be exercised in conformity with the rules of that organization, and firstly with its statutory provisions. But it is not for the organization to decide itself whether it has the capacity to conclude treaties or not, at least in the absence of specific rules of general international law in that respect. It is for the states establishing the international organization to decide whether it enjoys a separate personality or not, which in the affirmative should (necessarily) entail an autonomous capacity to conclude treaties. That said, there is clearly a trend in contemporary practice towards affirming that, like states, every international organization is a separate legal person, ie that it enjoys legal personality. The conclusion is possibly well founded from a statistical point of view. Legally speaking, it is not.

2. Procedure

The procedure to be followed with respect to invalidity is determined in Articles 65 to 68 of the Vienna Conventions. Needless to say, procedure is a realy important issue, since the very (valid) existence of contractual obligations is concerned. And it is all the more important since general rules governing settlement of disputes in international law still remain ‘elementary’, despite the increasing ‘internationalization’ of social issues and the growing number of states and international organizations. A specific chapter being devoted to those procedural provisions in the present book, a few points only will be dealt with here as regards their ‘new’ character. Two different issues are to be distinguished in that respect: the first one is related to the technique which should be used, and the second to the persons entitled to make use of them, especially where judicial settlement is concerned.(p. 309)

2.1. Mechanism

The procedural mechanism is quite simple in its principle: should a party to a treaty consider its consent (or the treaty) as invalid, it must notify its claim to the other party (or parties). If the parties agree on the invalidity, any difficulty is removed. In case they do not, the resulting conflict must be peacefully settled through one of the techniques aimed at in Article 33 of the UN Charter if the dispute remains unresolved during the 12 months following the objection to validity. Where jus cogens is concerned, any one of the parties may, however, unilaterally submit the dispute to the ICJ unless they agree by common consent to submit it to arbitration. Where others grounds of invalidity are concerned, such recourse to a jurisdictional settlement is not imposed, the only obligation of the parties being to submit their dispute to the special conciliation machinery set out in an annex to the Vienna Convention. Needless to say, the parties are bound by the decision of either the ICJ or the arbitral tribunal as far as the dispute concerning jus cogens is concerned, while they remain free not to accept the conciliatory solution.

Clearly, such a device is not very original. When a contracting party to a treaty contests its validity (or at least the validity of its consent), how could it get the benefit of the invalidity without duly informing the other party(ies) of its claim and of the grounds supporting it. Should all contracting parties agree on a common solution to address it, then it does not really matter whether invalidity was concerned or not (excepting in some respects issues relating to jus cogens); it is sufficient that they all agree. But if they do not, a dispute will be substituted for a mere divergence of (legal) views, which necessarily needs to be settled peacefully. Independently of the particular dispute settlement mechanism chosen by the states, judicial settlement—like any other mechanism—will rest on the agreement of the parties. And it is worth emphasizing that, apart from judicial ones, special conciliation techniques are organized so as to facilitate as much as possible the search for the necessary compromise between the conflicting expectations of the parties to the dispute.

The solution is surely interesting. Is it related to customary, as opposed to treaty, law, or should it be considered as creating/constituting ‘new’ law by accepted national or international standards? The question is probably badly posed. The duty of states to remain peaceful in settling a dispute regarding the validity/termination of a treaty need not be validated by a customary rule in order to be binding on them. The issue is one of outlawing resort to armed force, not of deciding on the procedures that must be used to settle disputes concerning the invalidity of conventional provisions. Of course, the outlawing of force has rested up to now on customary prohibitions, at least as long as the UN Charter provisions are not considered as legally binding independently of either a customary or a conventional technique. But neither custom nor treaty is needed, in a system where the unilateral resort to armed force is as a matter of principle (p. 310) prohibited, to validate the conclusion that the settlement of treaty disputes must remain ‘peaceful’. That said, there still could, of course, be an interest in imposing specific obligations in that respect, so as to ensure as far as possible the efficiency of a particular technique.

None would contest in that regard the importance of being quite specific in determining either the techniques which should be used in informing other contracting parties or the time limits that should be complied with and the persons habilitated to communicate official information, etc, as provided for in Articles 65 and 67. But as such, those requirements pertain to treaty and not to customary law, important as they might be. The fact that the contracting parties are sovereign states is not relevant—what counts is that the parties to the agreement—whatever their specific quality—are duly informed of elements which could jeopardize its very existence as a binding legal act.

To a certain extent, the same comment should be made about the possible role of the ICJ, or at least of arbitration. As such, there is no arbitration without the consent of both parties, which in domestic law at least constitutes the most basic difference between arbitration and judicial settlement. But in international law consent is also needed in the latter case, even if it is not for the parties to decide on the administrative, statutory, or procedural issues relating to the functioning of the tribunal. It is quite interesting in that respect to notice that, according to Article 66 of the 1969 Vienna Convention, resorting to the ICJ is seen as the ‘normal’ procedure in case of disputes concerning jus cogens. This should remind us that, in many domestic laws, arbitration is not permitted when ‘ordre public’ (jus cogens) issues are concerned. It should then be clarified that issues concerning public policy (jus cogens) are always difficult, and raise questions the interest of which goes beyond the ‘private’ expectations of the parties. It is fully understandable consequently that the international ‘judge’ (ICJ) be preferred to the arbitrator as far as jus cogens requirements are concerned. But it does not change anything in relation to the fact that judges, just like arbitrators, need to be approved by the states concerned. Article 66 of the Vienna Convention is nothing but a clause by which states agree to accept judicial settlement in matters relating to a dispute raising validity issues involving the jus cogens. Clearly, it is immaterial in that regard whether the state concerned should or should not be a party to the Vienna Convention. Quite the contrary. But it helps to affirm that such litigation should one day be reserved for judicial settlement only, independently from any agreement of the parties … simply because it is necessarily questioning the idea of an ‘intérêt général’, as uncertain as it might still be, whose determination should be kept within the hands of the most qualified servants of a still stammering international ‘community’. Which is not for the immediate future.

Whatever the opinions in that respect may be, it is still important to emphasize that an element of jurisdictional settlement, be it arbitral or judicial, is introduced, which is binding at least on each state party to the 1969 Vienna Convention. A choice is offered between the ICJ and arbitration, but the state could not refuse (p. 311) both. Such a choice is, however, not the classical one, between diplomacy and jurisdictional settlement; both terms of the alternative are jurisdictional: arbitration or judicial settlement, at least as far as jus cogens is concerned. Could it nevertheless be refused? In other words, would it be possible for states, when becoming parties to the Vienna Convention, to make a reservation about Article 66 so as to exclude any intervention of the ICJ? Why not, in principle?23 But accepting such a reservation would result also in greatly weakening the arbitration, since it is, according to Article 66, the only way of avoiding the unilateral submission to the ICJ. In other words, that reservation, even if it is aimed at the ICJ only, will inevitably result in also rendering the recourse to arbitration purely facultative.

That said, the 1969 Vienna Convention remains silent concerning possible reservations to its provisions. Reservations should therefore be admitted as long as they are compatible with the object and purpose of those provisions, according to its Article 19. The provision is indeed considered today, and probably was considered at the time of the adoption of the Vienna Convention, as largely expressing general rules binding on states independently of any formal acceptance on their part.

The question is consequently quite simple: should it be considered incompatible with its object or purpose that a reservation is made with regard to Articles 65 and 66, at least as far as the 1969 Convention is concerned? Clearly no reservation could be made to Article 33 of the UN Charter, as referred to in Article 65(3). But there is nothing to suggest that reservations should be excluded as far as other procedural technicalities are concerned. What about Article 66, which is concerned with methods of settling disputes rather than with procedure as such? Resort to the special conciliation techniques aimed at in Article 66(b) would surely be very useful; but it is still difficult to view it as essential to the object and purpose of the Vienna Convention, so as to invalidate any contradictory reservation. And finally, what about Article 66(a), providing for recourse to the ICJ or to arbitration when the application or interpretation of jus cogens (Articles 53 and 64) is at stake? As far as domestic laws are concerned, there is probably a kind of intimacy between invalidity issues and judicial settlement when the parties are in disagreement in that respect, and especially when ‘ordre public’ (public policy) is concerned. The point was previously emphasized. Is that sufficient to invalidate such a reservation to Article 66(a), as being incompatible with the object and purpose of the codification of the law on treaties? Probably not. Judicial settlement is surely important, especially when very delicate issues are in dispute—as for instance the existence of a kind of ‘intérêt général’ within the community of nations; but it is not as essential as invalidating a refusal to accept (p. 312) a judicial settlement which is clearly not compulsory under present international law. Regrettably, the role of the judge is still so limited in international relations that it would be difficult to assert, de lege lata, the invalidity of a reservation to Article 66 setting aside the ICJ,24 the arbitration itself being only ‘obligatory’ within Article 66 to the extent that states agree not to submit their dispute to the ICJ. Should it be recalled in that respect that reservations concerning its very intervention were admitted by the ICJ in relation to such basic issues as preventing and punishing genocide?25 No doubt, as the case dates back to the immediate post-war period, at the time it was crucial to avoid deepening a fast-growing gap between eastern and western countries. And the attachment to human rights was relatively new, which is not at all to suggest that the ICJ was in any way neglectful at that time of the horror of the crime and the importance of its punishment. Actually, voices within the ICJ have recently been heard affirming, at least implicitly, that the decision could be different, should the issue be brought anew, one way or another, before the Court.26 Would such a change in the case law of the ICJ in relation to genocide endanger the validity of reservations made to Article 66 of the Vienna Convention? The conclusion is doubtful. Failing at least exceptional circumstances, claims related to the validity of a treaty cannot be equated with problems of application of treaties whose specific purpose is to prevent and punish the most serious international crimes.

But objections to such a reservation should be accepted, in accordance with Article 20(4) of the Vienna Convention. Nothing, indeed, could exclude such objections being made in relation to Article 66, the consequences of which are specified in Article 21 of the Convention. The interesting point, however, is that many contracting states obviously are not satisfied with them. And they consequently declare, even in advance, ie before any reservation was made, that they will not be bound by Article 53 if resort to the ICJ or arbitration is not accepted under Article 66.27 This kind of counter-reservation clearly emphasizes the importance of judicial settlement when the validity of a treaty is disputed. And it should also be noticed that no reservation to Article 53 was ever made directly, ie independently from Article 66. This point highlights the fact that the problem lies not in the jus cogens as such, but only in the decisive role conferred on the judge (arbitrator) when there is a disagreement between the parties as to its actual (p. 313) content. That said, the fact is still that jus cogens cannot be refused either directly, ie via a reservation to Article 53,28 or indirectly, ie via a reservation to Article 66. Both reservations should be considered as null and void. Affirming the contrary would be misjudging the very nature of a peremptory norm. In other words, a mechanism whose specific aim is to consolidate a rule by prohibiting unilateral or conventional derogation is the first provision to which no reservation could ever be raised.29

2.2. Locus Standi

Whatever the mechanism that should or could be used to get a decision on validity/nullity might be, it is still to be decided which persons (states) would be entitled to bring a claim in this respect. Politically speaking, any state which feels concerned might raise the case, at its own political risk. But it is clearly another issue to determine who is legally entitled to bring proceedings before a court.

As a far as defaults in consent are concerned, it is normally for the state whose consent was defective to bring the claim. The Vienna Convention is quite clear in that respect as far as lack of competence (Articles 46–47), error (Article 48), fraud (Article 49), corruption (Article 50), and coercion of a representative of a state (Article 51) are concerned. This is elementary law. Indeed, how could a state whose consent was not vitiated assert that such consent to the treaty was invalid? There is nothing ‘new’ in that regard.

If a difficulty exists, it is in relation to Articles 52 and 53. In both cases, the treaty is said to be ‘void’, without any indication as to the state entitled to claim the invalidity, which is somewhat surprising.

As far as Article 52 is concerned, the state which was coerced clearly has a claim. The contrary would be surprising. But should it be inferred from the wording of Article 52 (‘a treaty is void’) that any other contracting party could also bring a claim in invalidity before a judge or an arbitrator, or whatever other body. The conclusion would be sound if that other party was also a victim of the coercion. But if it was not, what could support its assertion that its consent to the treaty should be nullified? A contrary conclusion would be difficult to reconcile with general principles of law regarding the legal consequences of a ‘vice du consentement’.

The difference between Article 52 (coercion by force) and Article 53 (jus cogens) is clearly that what was agreed on by the contracting parties is, as such, contrary to peremptory norms, in such a way that all of them should necessarily be allowed (p. 314) to claim the invalidity of what they agreed in common, and possibly could be blamed for not doing so, ie for maintaining a situation which is incompatible with basic requirements of inter-state relations. The formula ‘a treaty is void’ is quite correct in that respect. Actually, the real issue would be to decide whether a non-contracting state could have a claim in this respect, ie could bring proceedings so as to obtain the judicial finding that a treaty to which it is not a party is null and void. Until now this has been an open issue. Even if the existence of an actio popularis is generally denied,30 it cannot be categorically excluded.31 At least, it is not explicitly ruled out either by the Vienna Convention or by the general (judicial) practice, despite the reluctance sometimes expressed by the ICJ in that regard. It would clearly be better to entrust some common body, within the UN or elsewhere, with the task of provoking judicial control of treaties when serious doubts exist about their compatibility with jus cogens, at least when the contracting parties remain silent in that respect. However, that eventuality, which would open up new possibilities of controlling the respect of basic requirements, is for the time being too ‘new’ to have a reasonable chance of being accepted. And as such it is in any case much more concerned with judicial issues, especially the ‘locus standi’ (intérêt pour agir) requirements than with the law of treaties.

That said, the state, supposing it has a claim, loses its right to invoke ‘a ground for invalidating [ … ] a treaty under article 46 to 50’ if either it has expressly agreed to its validity or ‘must by reason of its conduct be considered as having acquiesced in’ it (Article 45). It has been pointed out that the provision rests on customary practice, which is problematic in the absence of clear cases where the validity of treaties was seriously questioned. But that provision is without doubt expressing general law simply because what is a stake in Article 46 to 50 is only the private interest of the state concerned, which is quite free to take in that respect whatever decision it deems appropriate. It does not matter whether this is explicitly or tacitly expressed, provided it is certain. But it is worth noting that the ‘loss’ aimed at in Article 45 is excluded when coercion or jus cogens (Article 53) is at stake. As far as jus cogens is concerned, the conclusion is easily understandable: how could a state indirectly validate—by losing its right to claim the invalidity—a treaty directly conflicting with peremptory norms of international law? This is pure ‘logic’. There is no need for customary support to validate the conclusion. But the conclusion is much more doubtful as far as coercion is concerned, whether it is exercised on the representative of the state (Article 51) (p. 315) or directly on the state itself (Article 52). In such a perspective the object of the treaty is as such irrelevant, the consent of the state only being concerned. The difficulty does not result from what was agreed on by the parties, but only from the circumstances under which they had agreed. Why should the confirmation of such consent be excluded, either by an explicit declaration or by unambiguous conduct? It has been said that violence radically corrupts any consent. The point surely is relevant for either the treaty itself, or its direct or indirect acceptance under Article 45.32 But if such acceptance was not ‘coerced’ on the basis of either Article 51 or Article 52, why should it be inadmissible? The conclusion is possibly expressing a new and opportune legislative policy; at least for states not parties to the Vienna Convention; it should not, however, constitute binding provisions.

3. Effects

A void treaty is not (validly) binding on its contracting parties. That is the specific effect of invalidity. There should be no disagreement in that regard. But doubts still exist as to the legal meaning of that rule, ie on its exact consequences at law. That is partly due to often rather obscure distinctions between a ‘relative’ and an ‘absolute’ nullity, whose exact significance remains partly unclear.33 Three aspects must be distinguished in that regard.

3.1. Ratione Personae

As such, the invalidity of the treaty only concerns the contracting states—all of them. When the consent of the state only is concerned, the nullity has no consequences for the parties who validly consented to the treaty. This goes without saying, as it goes without saying that the bilateral treaty will not even be binding on the state whose consent was valid, since there is no creditor or debtor with which it could still be engaged.

Nothing in the Vienna Convention suggests that the opposability of such nullity, be it of the treaty itself or of the consent of any of the contracting parties, might be restricted. Opposability should be the rule, since it logically results from the retrospective effect of invalidity. Should exceptions be admitted to a principle of opposability? Why not, in principle? At least national laws make clear that such exceptions might be justified, for instance to protect the possessor in good faith of a property. There is no reason to exclude similar exceptions as a matter of principle in public international law. The Vienna Convention is, however, silent in that respect, which makes it difficult to validate a ‘creative’ interpretation in that regard.(p. 316)

3.2. Ratione Materiae

The ‘logic’ normally is that the treaty ‘disappears’ as a whole when its invalidity is established. It would be inconsistent with its very nature if it were construed as an accumulation of separate agreements whose eventual invalidity had no consequences on the validity of the others. But the will of the parties should be decisive in that respect, as well as in relation to any other contractual issue. If the parties so agreed nothing should therefore prevent the treaty from being dismantled in such a way that some of its provisions disappear and the others survive. Such is only partly the approach followed by the Vienna Convention. The ‘separability’ of treaty provisions indeed appears to be admitted or refused on a mostly ‘objective’ basis, ie as far as the ‘surviving’ clauses either are, according to Article 44, ‘separable from the remainder of the treaty with regard to their application’, or did not constitute ‘an essential basis of the consent of the other party or parties to be bound by the treaty as a whole’, and provided ‘continued performance of the remainder of the treaty would not be unjust’ (para. 3). No separation, however, is permitted where coercion (Articles 51 and 52) or jus cogens (Article 53) are concerned, the decision to accept or refuse separation being left to the state concerned in case of fraud or corruption (Articles 49 and 50).

Up to now, there has been no customary practice clearly supporting either the principle of separability or the circumstances/conditions under which it could be accepted/refused. And it is doubtful that this rather complex regulation of the separability of treaty provisions as far as invalidity is concerned could be vested in a general principle of law, ie on domestic legal practices. Does this mean that the Vienna Convention is formulating ‘new’ law in that regard? Possibly, at least to a certain extent. The separability is, as such, justified by the necessity to safeguard, if possible, part of what was agreed upon by the parties. The concern is particularly important in international law, since most of its rules are of a conventional origin. The role of customary (general) law is indeed limited, especially in the divided world inherited from the cold war—apart from any other ‘social’ considerations— and in a context of growing technicalities which could not easily be regulated by purely customary practices. It is easily understandable in such a context that the parties, and beyond them the international community, are concerned with avoiding the destruction of the whole of what was agreed upon by them, ie to limit the consequences of invalidity to its core effects.34 Actually, the concern is not specific to the international community. In domestic laws also, such a concern has been manifest. But the decisive factor in that regard is apparently the will of the parties, without any other ‘objective’ consideration. Consequently, the nullity of a contractual provision does not result in invalidating the whole of the treaty when that provision was not the fundamental basis for consent. Other factors are usually not taken into consideration in that respect, and, in particular, (p. 317) no reference is made to a requirement of ‘justice’ as mentioned in Article 44(3)(c) of the Vienna Convention, whose terms are actually rather mysterious, or at least permit a variety of conclusions. With that reservation, separability as a principle could probably be considered as a general principle of law, much more at least than as a customary provision. And it should not consequently be considered as a ‘new’ law binding only the contracting parties. But it would still be difficult to back up most of its technical specifications, as resulting from paragraphs 3 to 5, with convincing references to either customary practices or the convergent national solutions which are in theory necessary to support the existence of a general principle of law.

3.3. Ratione Temporis

It is characteristic of the invalidity that it draws consequences from a defect in the treaty existing at the time of its conclusion. It cannot be confused in that respect with termination or suspension, which rests on factual elements, including the will of the parties, appearing after its conclusion.

Consequently, nullity operates ex tunc while termination (suspension) operates ex nunc. In other words, the retrospective effect is specific to the invalidity. There should be no disagreement in this regard. True, largely unclear distinctions were made, during the debates of the International Law Commission, between a treaty which is void ab initio and one which would become voidable at a certain date, as if it suddenly ceased to be valid although it had originally been so. There does not appear to be any firm basis for such a distinction. And Article 64, according to which ‘if a new peremptory norm of general international law emerges, any existing treaty which is in conflict with that norm becomes void and terminates’ is highly critical in that respect, the specific effect of such new norm being clearly to put an end to the application of the treaty without questioning as such its validity at the time the new peremptory norm did not yet exist, or at least was not of a (jus cogens) character.

Strangely enough, the retrospective (retroactive) effect of the invalidity is not expressly mentioned in Article 69 of the Vienna Convention, dealing with the consequences of the invalidity. However, it results to a certain extent from paragraph 2 thereof, the specific raison d’être of which is to limit such retroactivity ‘if acts have nevertheless been performed in reliance on a void treaty. That is a classical difficulty, with which every domestic law has been confronted. According to the Vienna Convention, a party must only ‘establish as far as possible’ the situation which would have existed in its relations with other parties, had such acts not been performed. But such an obligation only exists if the state is so required by a contracting party (Article 69(2)(a)), the situation of other contracting parties remaining unchanged as long as no similar request is made on their part. It does not matter, apparently, whether those acts were performed in good or in bad faith; but if good faith is established, they ‘are not rendered unlawful by reason (p. 318) only of the invalidity’ (Article 69(2)(b)). Such mitigation of the consequences of the invalidity does not apply, however, in case of fraud, corruption, or coercion as far as the party to which it is imputable is concerned. And apparently it does not apply either when the invalidity results from the violation of jus cogens, the parties being only required to ‘eliminate as far as possible the consequences of any act performed in reliance on any provision which conflicts with the peremptory norm’ (Article 71). No request from another contracting party is needed in such a case; nor is any attention paid to good or bad faith, which has the concrete result that bad faith must always be presumed.

Does such a determination of the consequences of the invalidity of a treaty express existing law or does it constitute ‘new’ law? Once again, no clear support exists in a practice almost inexistent in that regard. Could the solutions of the articles nevertheless be generalized, ie be considered as binding even to states not being parties to the Vienna Conventions on the basis of either general principles of law or at least some intrinsic ‘logic’ of law? The difficulty probably is that the provisions of the 1969 Vienna Convention remain largely unclear, apart from certain obvious points, and this makes any attempt to generalize that content problematic. Certainly, any treaty—as a legal act—could only be binding if the basic requirements imposed by law to that end are satisfied. But that does not prejudge the exact consequences resulting from their violation. Article 69 is especially cautious, as is made clear by the repeated reference to ‘as far as possible’. No doubt, ‘à l'impossible, nul n'est [ne saurait être] tenu’. But not many impossibilities actually exist as far as states are concerned. When they are of a legal nature, the adoption of a statute (act) or judgment could indeed remedy most of the ‘impossible’ situations. And the Problem is generally of a purely financial character when it is not a factual one … but the hypothesis cannot be totally excluded. In some cases, it will surely appear that the consequences of an act performed in reliance on a void treaty cannot be ‘eliminated’, which validates the cautious terms of both articles. It is interesting, however, to notice in that regard that Articles 69 and 71 remain silent on the consequences of such impossibility. That is somewhat disturbing. Should it be admitted that there is no consequence, it would be difficult to distinguish invalidity from termination. In domestic law, some financial compensation at least is ordinarily due in that case. No mention, however, is made of such an obligation—or any other obligation—in the Vienna Conventions.

The point as such does not raise a question of responsibility; it is only to find an acceptable substitute, as far as possible, to the impossibility of re-establishing the situation which would have existed had the relevant acts not been performed in reliance on a (later declared) void treaty. That said, it is quite possible that a question of responsibility may be raised in relation to the invalidity (invalidation) of a treaty. It is certainly the case when the ground of invalidity rests on a breach of international law, as for instance in case of coercion under Article 52 or violation of the jus cogens (Article 53). Needless to say, such a violation necessarily entails for its perpetrator a duty to give reparation for the detrimental consequences of (p. 319) its conduct. But invalidity as such is not part of the reparation; it is only a way of limiting the damage for which reparation is due. It does not matter in that respect that, according to Article 69(2)(b) of the Vienna Convention, acts performed in good faith before the invalidity of the treaty was invoked are not ‘rendered unlawful by reason only’ of that invalidity. As a matter of principle indeed, reparation should be due not only for the illegal conduct on which that invalidity is grounded, but also for all its consequences, even those which are not illegal per se, when their detrimental character is established.


1  The present paper is mostly concerned with the 1969 Vienna Convention on treaties between states, but the comments should normally be relevant for the 1986 Convention on treaties with or between international organizations.

2  See I. Sinclair, The Vienna Convention on The Law of Treaties (Manchester: Manchester University Press, 2nd edn, 1984) 3.

3  As emphasized by Mc Nair, there exists ‘little diplomatic practice, and almost no judicial authority’ (The Law of Treaties (Oxford: Clarendon Press, 1961) 207).

4  Cf. ‘the subtle difference between existing customary rules and the rationalization of general principles of law’ aimed at by V. Degan, Sources of International Law (The Hague: Martinus Nijhoff, 1997) 76, or the ‘principles dictated by legal logic’ referred to by O.A. Elias and C.L. Lim, The Paradox of Consensualism in International Law (The Hague: Kluwer Law International, 1998) 218.

5  According to G.F. de Martens, the negotiators of the treaty must have been ‘sufficiently authorized’ (para. 48), and the consent be ‘declared’ (para. 49), ‘free’ (para. 50), and ‘mutual’ (para. 51) (Précis du droit des gens moderne de l'Europe (Göttingen: De Dieterich, 3rd edn, 1820)).

6  For other possible grounds—largely doubtful—see M. Kohen, ‘Article 42’, in O. Corten and P. Klein (eds), Les Conventions de Vienne sur le droit des traités, vol. II (Bruxelles: Bruylant, 2006) 1603 (hereinafter Commentaire).

7  See P. Martin-Bidou, ‘Article 47’, in Commentaire, supra note 6, at 1728.

8  For the affirmation that corruption as a ground of invalidity is ‘une idée absolument neuve’, see J.-P. Cot, ‘Article 50’, in Commentaire, supra note 6, at 1820.

9  The ‘dol’ is rarely mentioned in international doctrine as a ground which could invalidate a legal transaction (a treaty), even among scholars of civil law traditions (contra, for instance P. Guggenheim, Traité de droit international public, (Genève: Librairie de l'Université, 1953) vol. 1, 92–3).

10  For a denial of the existence of fraud (‘dol’) as a ground of invalidity, see Judge Gros as quoted by G. Niyungeko, ‘Article 49’, in Commentaire, supra note 6, at 1781.

11  Cf. the reference to an ‘erreur provoquée’ by E. Wyler, ‘Article 48’, in Commentaire, supra note 6, at 1763.

12  Cf. the declaration of the Philippines at the Vienna Conference, according to which ‘fraud amounts to deceit or wilful misrepresentation’, quoted in M. Villiger, Commentary on the 1969 Vienna Convention on the Law of Treaties (Leiden: Martinus Nijhoff, 2009) 617.

13  The customary character of Article 52 was not that clear at the time the Vienna Convention was adopted. In any case, very (too) old precedents should be of little significance in that respect (see the concordat between Pope Pascal II and the King of England, Henry I, whose conclusion was obtained through imprisonment of the former by the latter, referred to by G. Distefano, ‘Article 51’, in Commentaire, supra note 6, at 1845 (possibly, there was a confusion in that respect between Henri I Beauclerc and Henri V, Emperor of Germany).

14  See the ‘Principes européens du droit des contrats’.

15  See O. Corten, ‘Article 52’, in Commentaire, supra note 6, at 1877–9.

16  Cf. E. Raftopoulos, sniggering at the ‘futile theoretical attempts to identify the jus cogens rules in international law’ (The Inadequacy of the Contractual Analogy in the Law of Treaties (Athens: Alkyon Publishers, 1990) 249).

17  For a rather unclear distinction between jus cogens and ordre public, see, for instance, L. Alexidze, ‘Legal nature of Jus Cogens in Contemporary International Law’, 172 Recueil des Cours (1981-III) 241.

18  See J. Combacau, in J. Combacau and S. Sur, Droit international public (Paris: Montchrestien, 7th edn, 2006) 51–2, for whom Article 53, however, is only expressing a conventional rule whose content remains ‘indiscernible’ (‘insaisissable’).

19  Which has nothing to do with ‘natural law’ (see J. Verhoeven, ‘Considérations sur ce qui est commun (Cours général de droit international public)’, 334 Recueil des Cours (2008) 228 et seq.), contrary to what had been several times affirmed (see, for instance, G. Danilenko, Law-Making in the International Community (Dordrecht: Martinus Nijhoff, 1993) 214 et seq.).

20  It is sometimes suggested that any resort to ‘jus cogens’ as a ground of invalidity should be excluded before the entry into force of the 1969 Vienna Convention, in accordance with the adage ‘pas de nullité sans texte’ (see G. Robledo, ‘Le ius cogens: sa genèse, sa nature, sa fonction’, 172 Recueil des Cours (1981-III) 148). But that adage is meaningless in a customary system.

21  The affirmation, however, is often, at least implicitly, denied (see, for instance, G. Tunkin: ‘actually the recognition of the existence of jus cogens was an element of a new thinking reflecting the realities of a new era in history’—no less!—the ILC being not fully aware ‘that it was a new approach to international law’, ‘when discussing the problem of jus cogens’ (‘Law and Force in Interstate System’, 219 Recueil des Cours (1989-VII) 300).

22  Cf., in another context, J. Klabbers, The Concept of Treaty in International Law (The Hague: Kluwer Law International, 1996) 86–7.

23  It should be even more readily accepted that Article 66 constitutes ‘a formidable deterrent to ratification’, as pointed out by G. Gaja (‘Jus Cogens beyond the Vienna Convention’, 172 Recueil des Cours (1981-III) 289), who is, however, concerned not be seen as ‘too pessimistic’ about the success of the Convention in that regard; see also M. Virally, ‘Panorama du droit international contemporain’, 183 Recueil des Cours (1983-V) 175.

24  ‘Doubts’ were nevertheless often expressed as regards the validity of such reservation, the justifications for which remain unclear (see, for instance, A. Remiro Bretons et al., Derecho internacional (Valencia: Preliminares Editorial Tirant Lo Blanche, 2007) 75).

26  See the Joint Separate Opinion of Judges Higgins, Kooijmans, Elaraby, Owada, and Simma (para. 6 et seq.) in the case concerning Armed activities on the territory of the Congo (new application: 2002) (Democratic Republic of the Congo v Rwanda), Jurisdiction and Admissibility, Judgment of 3 February 2006, ICJ Reports (2006) at 66 et seq.

27  See M. Villiger, Commentary on the 1969 Vienna Convention on the Law of Treaties, supra note 12, at 822–3.

28  Cf. however B. Conforti, ‘Cours général de droit international public’, 212 Recueil des Cours (1988-V) 134.

29  See J. Verhoeven, ‘Jus Cogens and Reservations or “Counter-Reservations” to the Jurisdiction of the ICJ’, in K. Wellens (ed.), International Law in Theory and Practice—Essays in Honour of Eric Suy (The Hague: Martinus Nijhoff, 1998) 195 et seq.; H. Ruiz-Fabri, ‘Article 66’, in Commentaire, supra note 6, at 2410 et seq.

30  Contra, for instance, A. Cassese, International Law (Oxford: Oxford University Press, 2001) at 133, who refers to ‘customary rules corresponding to the Vienna Convention's provisions on invalidity of treaties’ whose affirmation is at least doubtful in the absence of any practice in that respect.

31  As such the actio popularis is not at all ‘illogical’; it is simply ‘dangerous’ in the present state of international relations. But there is no logic in admitting that, beyond the states parties to the treaty allegedly inconsistent with jus cogens, the parties to the 1969 Vienna Convention be admitted as having locus standi in that regard (see A. Orakhelashvili, Peremptory Norms in International Law (Oxford: Oxford University Press, 2006) 515).

32  See M. Kohen, supra note 6, vol. II, at 1680.

33  See generally J. Verhoeven, ‘Article 69’, in Commentaire, supra note 6, vol. III, at 2483 et seq.

34  To that extent, invalidity is still a kind of ‘luxury’ in international law (see J. Verhoeven, ‘Les nullités du droit des gens’, in IHEI, Droit international 1 (Paris: Pedone, 1981) 47).