- Treaties, conclusion — Treaties, entry into force — Vienna Convention on the Law of Treaties — Treaties, ratification — Universal international organizations
From time immemorial, States have concluded treaties on all possible matters under the sun.1 Today, treaty-making has become an almost indispensable aspect of sovereignty—the ‘outward-looking’ facet of a State and one of the cornerstones of its relations with other States. Questions of treaty formation play a similarly central role in the formation of international law, witnessed most recently during the Copenhagen discussions on ‘formalizing’ an international environmental strategy and ongoing consultations on finalizing a comprehensive convention against terrorism.2
As a process, the term ‘treaty-making’ used herein actually encompasses four different stages: (a) treaty negotiations; (b) the conclusion of the treaty text; (c) expressions of consent to be bound; and (d) entry into force.3 Various legal acts take place at each stage that operate in two parallel and cross-cutting legal orders, namely international law and the domestic law systems of the States involved. This chapter focuses on treaty-making solely under international law. But, it should be understood at the outset that domestic legal orders also regulate these acts, whether in delineating which organ of a State can conclude treaties, how to issue full powers, what domestic steps are necessary for treaty ratification, or whether and how incorporation of the treaty’s provisions into the domestic legal order occurs.4
References(p. 178) At the international level, a certain formalism has always characterized treaty-making. As instruments that by design, signal permanence and stability, treaties require a record. Today, that record is almost universally required to be a written one. Through a series of treaties and practice, States have agreed on specific rules and processes for how they will record treaty commitments.5 The 1969 Vienna Convention on the Law of Treaties (VCLT) in particular, has achieved an almost ‘sacrosanct’ status; what started out as codification of certain customary rules and the progressive development of others, has now all mostly taken on the character of customary norms.6
On the other hand, for all its formality and customary character, treaty-making remains a remarkably flexible process. The field of treaty law is vast and very complex, leaving room for indefinite variations. Tens of thousands of treaties now exist that can be classified along any number of criteria, including the treaty’s content, the presumed or proven intent of the negotiating parties, or the identity of participating parties. For our purposes, the distinction between multilateral (involving three or more parties) and bilateral (involving two parties) treaties reveals the most significant differences in practice (although additional gradations within these categories can, and do, occur).7
Within such a vast and varied terrain, the rules found in the relevant Vienna Conventions are mostly residual in character. They generally only apply in the absence of any other specific agreement between the parties (the absolute prohibition of agreements contrary to jus cogens being a notable exception).8 Otherwise, parties are ‘free to agree on anything’. In recent years, the process has become even more complex. Changes in domestic regimes, particularly the rise of democratic institutions, have led States to different practices and processes of treaty-making. The presence of possible or potential additional treaty-makers—first of all international organizations (IOs), but also entities like liberation movements, and possibly non-governmental organizations—have further complicated the landscape.9 In such a labyrinth, the fundamental rules embodied in the three Vienna Conventions provide useful signposts to those engaged in the often messy work of constructing a treaty.
Taken together, this chapter focuses on all aspects of treaty-making, in the light, inevitably, of the Vienna Conventions, while cognizant of the actual variations from its terms in practice. Given the difficulty of presenting in detail the minutiae of all the procedural and legal aspects, this chapter seeks to provide an overview of References(p. 179) the four essential stages of the treaty-making process: first, the negotiation of the treaty; second, the conclusion of the treaty; third, the expression of consent to be bound by the treaty; and, fourth and finally, the entry into force of the treaty.
A treaty is drawn up by a process of negotiation which may take place either through the diplomatic channel or some other agreed channel, or at meetings of representatives or at an international conference. In the case of treaties negotiated under the auspices of an international organization, the treaty may be drawn up either at an international conference or in some organ of the organization itself.12
The Commission indicated that the process of drawing up the treaty text is an essential preliminary to the legal act of its adoption. At the same time, it recognized that the article’s content, as drafted, was more descriptive than normative. This attracted much adverse comment from governments. Although Waldock attempted to revise it to meet those criticisms, the ILC ultimately decided to drop the article since it lacked normative content.13
On the other hand, if the participants in that negotiation, for whatever reason, do not intend the arrangement which comes out of the negotiation, however formal that arrangement might look, to be an agreement governed by international law, it will not be a ‘treaty’ for the purposes of international law, whatever else it might be, and however much of a ‘commitment’ it might express.14
References(p. 180) Thus, treaty negotiations should be distinguished from those designed to produce a political commitment such as an MOU or a contract governed by the domestic law of one or more of the participating States.15
Second, treaty negotiations must address questions of participation. As detailed in earlier chapters, international law does dictate who can make treaties. States and IOs are without doubt the main subjects of international law having the treaty-making capacity, while there are fewer examples of treaty-making by entities such as occupied States or semi-autonomous territories. Most notably, negotiating States allowed certain territories that did not possess full independence to participate in the Third UN Conference on the Law of the Sea; they had similar roles in the 1974 Geneva Conference entrusted with the preparation of two Protocols to the 1949 Geneva Conventions relating to international humanitarian law.16
In terms of treaty-making by States, they have broad authority, although occasionally there are questions about whether an entity qualifies as a State, a topic that has given rise to long and difficult debates, inevitably affected by political or other considerations.17 Assuming a State exists, international law leaves the State with great discretion on whether and when to make treaties. Moreover, that discretion extends to include who may negotiate a treaty on behalf of the State, although, suffice it here to say that at the national level, it is mainly the Executive Branch that in most national systems has the competence to negotiate.18
The authority of IOs19 to make treaties is more circumspect. It remains an open question whether all IOs have the capacity to conclude treaties and on what basis they could do so.20 If an organization has treaty-making authority, the Executive Head of the organization (Secretary-General, Director General, etc) usually has the capacity to represent the organization in the negotiation and conclusion of treaties. That said, it is not clear if this assignment is considered to be part of the ‘inherent powers’ of the Executive or if some previous authorization by the deliberative/parliamentary body is required. Certainly such authorization is necessary for the expression by the organization of its consent to be bound by the treaty.21 In the UN, however, the Secretary-General (or his representatives) may on certain occasions conclude treaties on behalf of the organization, either on his own initiative or to give effect to resolutions of UN organs even if they did not specifically request him to enter into such an agreement.22
References(p. 181) Conditions for participating in treaties are not set solely by international law; in practice, many negotiations are limited by the negotiating participants themselves. The 1986 Vienna Convention anticipates IOs negotiating an array of treaties. In practice, however, they are mainly involved in concluding bilateral treaties with States or other organizations. The multilateral treaty process is still dominated by States.23 Similarly, a negotiating State may choose to limit the number of other States, IOs, or other entities with which it wishes to negotiate a treaty. Thus, many multilateral treaties themselves contain clauses determining which States may become parties or even giving the possibility to the parties to invite other States to become parties. Examples of various participation clauses are included in Section VI(3)–(4) of this volume.
In deciding with whom to partner in a treaty, the threshold for States is whether to conclude a bilateral or a multilateral treaty. Bilateral treaties are negotiated between the representatives of each State mostly in a formal setting. These negotiations usually begin after representatives of each party receive, exchange, and examine their respective full powers.24 In other cases, negotiations are more informal or take place at various levels, starting with the involvement of mid-level diplomatic agents and going up to that of high level officials or even Cabinet Ministers, Prime Ministers, or Heads of State. Negotiations may also take place in successive ‘rounds’, whether in a formal setting, or in more informal contexts, on the occasion of a meeting or summit.
Bilateral agreements also commonly include agreements between a State and an IO. A 2009 UN–US Agreement concerning the establishment of security for the UN presence in Iraq is a recent example.25 For the UN, such negotiations usually take place between a specific department or office of the Secretariat and representatives of the interested State (usually part of the personnel of its Permanent Mission to the UN). IOs, including the UN, may also negotiate bilateral agreements with other IOs, usually on matters of mutual cooperation.26
Despite the moniker, some bilateral treaty negotiations may involve more than two parties. Especially in the case of treaties between former opponents that purport to end a conflict, third parties may act as intermediaries and play a crucial role in the course of the negotiations. The most well-known example of this phenomenon involved Algeria’s role in negotiating an agreement between the United States and Iran.27
References(p. 182) In the case of multilateral treaties, treaty negotiations usually take place within an international conference, whether convened under the auspices of a standing IO, or in specific ad hoc bodies established to that effect. Elaboration of treaties was among the first functions of early unions created in the nineteenth century, which prepared their own constituent acts.28 The League of Nations prepared the Statute of the Permanent Court of International Justice and other multilateral conventions.29 Article 62(3) of the UN Charter expressly includes the negotiation of treaties among the UN’s functions.
Over the years, the UN General Assembly has elaborated many treaties either through its main Committees or through ad hoc organs. Other organizations like the Council of Europe (COE) or the International Maritime Organization (IMO) are essentially permanent treaty-making mechanisms. Thus, large multilateral conventions will almost always be negotiated within pre-established institutional frameworks. These entities facilitate the process and offer expertise to the increasingly specialized and fragmented field of treaty-making.
Beyond the role of IOs in multilateral treaty-making, the negotiation process itself has evolved. The 1960s witnessed a period of intensive treaty-making on the codification and progressive development of international law driven by the ILC. At that time, the usual pattern started with a set of draft articles prepared by the Commission, which it submitted to the UN General Assembly. A codification conference was then convened in which ‘all States’ were invited to participate.30
But scarcely any large codification conferences have taken place during the last twenty-five years. The one notable exception (although not strictly a codification conference) was the 1998 Rome Conference convened after several years of preparation to conclude the final statute of the International Criminal Court (ICC).31 Otherwise, the last conference of such calibre was the negotiations leading to the conclusion of the 1986 VCLT.
In place of codification conferences, States adopted other processes of negotiation. The UN Conference on the Law of the Sea, convened over a period of several years, was notable both for the variety of techniques used and for ingenious solutions to a wide-range of problems.32 A more recent trend has been the UN General Assembly’s establishment of ad hoc committees or working groups of the References(p. 183) Sixth Committee to negotiate sensitive treaties.33 The UN Commission on International Trade Law (UNCITRAL) and the UN Environment Programme (UNEP) have similarly negotiated treaties as part of their mandates.34 While diplomatic conferences have a more formal organization—including discussion in sub-committees and a drafting committee—UN ad hoc commissions or committees follow a more informal pattern. They may establish working groups open to all members of the commission or committee and which focus on specific thematic areas to be covered by the treaty. The final outcome is submitted to the plenary of the committee (or commission), which endorses or adopts the text, and then submits it to its parent body, whether the UN General Assembly or the Assembly of the specialized agency or other organization.
Negotiations themselves may be conducted in myriad ways. In the UN, sometimes the Secretariat may prepare a first draft of the convention. At other times, negotiations may be conducted informally on the basis of ‘non-papers’,35 working papers, or more formal modalities. In the latter case, each delegation may present (simultaneously or successively) a complete text of the proposed outcome (agreement). Counter proposals are then submitted (formally or informally, orally or in writing) by other delegations. The question of the status of each proposal or counter proposal may be an extremely thorny one that can become politically charged. Procedural rules together with clarifications on the exact status of each proposal at every stage of the negotiating process may acquire great importance.
In terms of procedural rules, diplomatic conferences convened for the purpose of concluding a multilateral treaty mainly follow parliamentary practices. Several attempts to codify such procedures have not, however, produced any conclusive result. States prefer instead to keep open the possibilities of elaborating specifically tailored rules of procedure for any conference based on its specific subject and presumed goal.36 The fact remains, however, that the Rules of Procedure of the UN General Assembly still provide an acceptable model. Tacitly or expressly, treaty-making conferences and related bodies (such as ad hoc committees) have referred to them whenever a procedural problem arises.
The really important aspect of the negotiation process is, of course, the decision-making rules; for example, whether the negotiating participants will require unanimity, consensus, or required majorities to reach a collective decision. Generally, References(p. 184) the recent trend—both in the UN and diplomatic conferences—is to seek the widest possible agreement at every step of the treaty-making process and on any draft provision of the treaty under negotiation. Resort to voting is viewed as divisive or counter-productive on the logical assumption (not always proved in practice) that treaty-making resulting from a consensual process will eventually produce a more widely accepted treaty. At the same time, another recent phenomenon for multilateral treaty negotiations is the ‘package deal’, according to which ‘nothing is agreed until everything is agreed’. Under this approach, any delegation has the right to reserve its position on certain points until it obtains satisfaction on others.37
The conclusion of the treaty has various stages, beginning with the establishment of the text and usually ending with its signature. A treaty text is established by its adoption and/or authentication by duly authorized representatives of the negotiating parties, who are usually identified in IOs through the presentation of credentials or, when an adoption entails signature, through full powers. Adoption is the act by which the negotiating parties express their agreement with the final text of the treaty. Adoption thus formally acknowledges substantive agreement on a text by the negotiating parties. Authentication, in contrast, is more in the nature of a ‘notarial’ certification or witnessing that the very text as finally established and adopted constitutes the ‘authentic’ (genuine, real) text of the treaty. Multiple linguistic versions of a single, multilateral treaty can frequently complicate the authentication process. An adopted or authenticated text may be signed by representatives of the negotiating participants. In addition, participants may adopt a ‘Final Act’, recording the history of the negotiation process in ways that can provide valuable information for the treaty’s later interpretation or application.
A. Full powers
a person is considered as representing a State for the purpose of adopting or authenticating the text of a treaty or for the purpose of expressing the consent of the State to be bound by a treaty if:
(p. 185) Article 2(c) of the VCLT defines the full powers as ‘a document emanating from the competent authority of a State designating a person or persons to represent the State for negotiating, adopting or authenticating the text of a treaty, for expressing the consent of the State to be bound by a treaty, or for accomplishing any other act with respect to a treaty’.39 Full powers must be distinguished from credentials, which are documents authorizing a person merely to participate in a conference.40
Full powers clarify what authority a representative has in relation to formation of the treaty under negotiation. The content of full powers can range from authorization simply to negotiate the treaty, to authorization to negotiate and sign the final text; it may also include signature alone or some other variation. They must usually name the person authorized (authorizing a particular office holder is insufficient), be signed by someone authorized to issue full powers, and mention the date and place of signature, often accompanied by the official seal.
Who can issue full powers? Practice varies from State to State, but it is widely admitted that Heads of State, Heads of Government, and Ministers for Foreign Affairs may sign the full powers document. It is also customary that those officials by their very function do not require full powers, a principle confirmed in Article 7(2) of the VCLT. All other Ministers, as well as Deputy or Vice-Ministers (including Vice-Ministers for Foreign Affairs) do require full powers.41 The VCLT’s only exceptions are for the principle of ‘limited automatic representativeness’ where (a) heads of diplomatic missions (usually having the rank of Ambassadors) can adopt a treaty between the accrediting State and the State to which they are accredited,42 and (b) representatives accredited by States to an IO or to an international conference (or one of its organs) may adopt the text of a treaty in that conference, organization, or organ.43 This function of a diplomatic mission is reinforced in Article 3(1)(c) of the Vienna Convention on Diplomatic Relations (VCDR).44 In UN practice, accredited representatives of States (and only those representatives) are considered as having full powers to adopt the text of treaties concluded within the UN (but not to sign or to express the consent of the State to be bound by the treaty).
References(p. 186) It often happens that full powers to sign are given to other representatives coming from the capital for a particular conference. Despite the overlap in authority, all full powers emanating from the competent authority of a State have equal validity. Indeed, it is not unusual that more than one representative of the same State sign a treaty at the same time, provided, of course, that each has a duly signed full powers.
In terms of procedures, full powers are transmitted to the Secretariat of the Conference or the IO hosting the negotiations through means of a notification or note verbale.45 The Secretariat checks them and reports any problem to the State concerned. In bilateral treaties or treaties negotiated among a small number of States, full powers are produced either at the beginning of the meeting set out for the treaty’s signature, at the beginning of the negotiations, or even earlier through mutual notifications/notes verbales from the foreign ministries concerned.
Finally, it should be noted that in certain contexts in practice, the negotiating participants may agree to dispense with full powers. This frequently occurs in bilateral contexts, and may also particularly characterize certain early and informal stages of multilateral negotiations. Ultimately, negotiating participants need to decide in each case whether or not full powers will be needed.
B. Adoption of a treaty text
At the end of the negotiations, there should be agreement on a treaty text. Treaty texts are generally comprised of certain, standard parts:
(a) a Preamble containing the names of the parties, a summary description of its object and purpose, the names and official designations of the representatives of the parties (plenipotentiaries) and, often, a paragraph stating that the plenipotentiaries have produced their full powers which were found to be in good and due form and then they have agreed upon the following articles;
(b) the various substantive articles of the treaty;
(c) a set of what are known as ‘final clauses’, dealing with matters such as territorial application of the treaty, signature, ratification (acceptance or approval), accession, entry into force, amendments, denunciation (withdrawal) and duration, or (in the case of a multilateral treaty) reservations;
(d) a clause (testimonium) stating: ‘In witness thereof (en foi de quoi) the respective plenipotentiaries have signed the treaty’; and
(e) the location and date.
For bilateral treaties, there is rarely a specific or distinct process of adoption. The parties’ very signature of the final text (which may be ‘simple’ signature acting as a precursor to ratification, or a signature that actually expresses the State’s consent to (p. 187) be bound)46 constitutes the act adopting the text. Depending on the form chosen, ‘adopting’ the treaty in the bilateral context may also take the form of an exchange of letters, which have identical contents. The letters (with the required changes in names, etc) are duly signed by the representatives of each party. They may be exchanged either through diplomatic channels or during a solemn meeting between the representatives of the two parties.
In the case of multilateral treaties, in contrast, adoption constitutes a specific and distinct process for the formal establishment of the text. In practice, adoption actually often comprises two distinct stages. First, the negotiating body (ad hoc committee, commission, etc) will ‘finalize’ the text that represents the negotiations’ result. This ‘finalization’ takes place in accordance with the relevant rules of procedure of that negotiating body, but, in practice, is mostly achieved through consensus.47 What constitutes ‘consensus’ or ‘general agreement’ has occasionally been the subject of some dispute, but generally ‘consensus’ is distinguished from unanimity, and involves the absence of formal objection, including from States that would not vote affirmatively in favour of adoption.48
Second, once the negotiating body has ‘adopted’ the text, it transmits it with a recommendation (included in a draft resolution) to the parent body such as the UN General Assembly. In theory, the General Assembly may still bring changes or modify substantially the prepared text. In the UN, the General Assembly adopts treaties through a resolution to which the treaty is annexed. For example, on 2 December 2004 the UN General Assembly adopted the UN Convention on Jurisdictional Immunities of States and their Property by Resolution 59/38.49 In UN practice, the treaty is very often adopted by consensus, although UN General Assembly rules of procedure do allow for voting if necessary.50
Outside the UN General Assembly, diplomatic conferences that elaborate a treaty may use another method of adoption. The ‘final text’ (usually prepared within the Conference’s Drafting Committee) is adopted by the Conference Plenary. But in the Plenary context, voting may often occur if consensus is unattainable. Thus, the Rome Conference adopted the ICC Statute by a vote of References(p. 188) 120 in favour, 7 against, and 21 States abstaining.51 According to Article 9 of the VCLT, adoption of a treaty text occurs with the consent of all States participating in the negotiations, or by the vote of two-thirds of the States present and voting (unless those two-thirds decide to apply a different rule). In practice, however, different international fora and conferences have established different majorities, or established that the principle of adoption of the text must be mainly by consensus.52
The adoption of a treaty marks the end of the negotiation; it does not mean that the treaty is already legally binding in all respects.53 Adoption instead signals the moment from which certain of the treaty’s provisions apply, namely those on authentication, consent to be bound, reservations, depositary functions, and ‘other matters arising necessarily before the entry into force of the treaty’.54 In certain cases, the adoption of a large, multilateral treaty may also constitute an important element in the formation of a customary rule.55 Depending on the number of adopting States, the treaty may also take on independent political significance.
The next step in establishing the final text of the treaty is authentication. There are various standard methods of authentication, such as initialling, signature ad referendum, signature, or adoption of the Final Act. Article 10 of the VCLT, however, allows negotiating participants to agree on their own procedures for authentication as well.56
Initialling involves participants placing their initials at the bottom of each page of the treaty text. It signals that after having been read, this initialled text is found to correspond to the one agreed upon and, therefore, constitutes the ‘authentic’ or definitive text of the treaty. Initialling is mostly used in bilateral or very restricted multilateral treaties. It is not a method used in UN processes involving a large number of participants.
References(p. 189) In multilateral treaties, the authentication process is sometimes more complicated. The adoption of the treaty by the Plenary of the Conference or the General Assembly constitutes at the same time also the ‘authentication of its text’. Alternatively, the signature of a Final Act (discussed below) may constitute a means of authentication of the treaty text to which the Final Act is usually attached. In the UN, the final text of a treaty is also reproduced in the form of a brochure duly certified (at its last page) by the Legal Counsel as containing the authentic text of the treaty in all languages. Normally, once adopted on the basis of a working document, an ‘authentic’ final text of the treaty is processed on ‘official’ illustrative paper, in all languages in which it has been finalized and which all together constitute the original text of the treaty. At the end of all the linguistic versions, signature pages contain the name of each State having the capacity to become party, again in all official languages. The pages are then bound into a volume, which is the one signed when the treaty is open for signature. If this process is followed, then there is no other stage of ‘authentication’.
D. Multiple linguistic versions of treaties
Treaty texts will often come in multiple languages. Article 35(3) of the VCLT provides that the terms of a treaty are presumed to have the same meaning in each authentic text. Paragraph 4 of that Article provides that comparison of the authentic texts is a part of the process of interpretation.57
In the case of bilateral treaties, the treaty is usually concluded in the languages of the negotiating Parties, which are each equally authentic. In some cases, however, one of the languages may be given priority over the other. Or, these languages may (but need not) be accompanied by a text in a commonly agreed language (most frequently English or French), which often prevails in the event of any divergence of interpretation. Sometimes, a treaty may even be drawn up in a language which is not that of the contracting parties. For example, the 1905 Treaty of Peace between Japan and Russia was drawn up in English and French.58 In multilateral treaties, the text is established in two or more languages. In the UN, most treaties are established in the six official languages (Arabic, Chinese, English, French, Russian, and Spanish) which all constitute the authentic text of the treaty.
Simultaneous drafting of all the authentic texts is today technically possible and even expert bodies such as the ILC or UNCITRAL conduct their work and prepare their texts in all official languages. Usually the text of treaties concluded under UN References(p. 190) auspices is authentic in all its official languages. The number of authentic texts may vary depending on the body adopting them. Signature pages of treaties deposited with the UN Secretary-General are always, however, in all official languages irrespective of the authentic languages of the treaty text itself. This avoids, in particular, any terminological or diplomatic difficulties that might arise from establishing signature pages in non-official languages.
In some cases, participants may request the depository to prepare an authentic text of a treaty subsequent to its adoption on the basis of other existing authentic texts.59 In other cases, the agreement may contain no provisions on the subject. In the latter case, the Secretary-General’s practice has been to consider as authentic the texts in all official languages and to prepare the original accordingly.60 Even then, exceptions sometimes occur. For example, the Convention on the Privileges and Immunities of the UN exists in English and French only; the working languages of the Secretariat.61 Section VI(9) of this Volume contains examples of various clauses dealing with the conclusion of multiple linguistic versions of a treaty text.
The most complex and sometimes thorny question surrounding treaty languages lies in the correction of errors or lack of concordance in the original of a multilateral treaty.62 The VCLT devotes a whole, detailed article (Article 79) to the correction of errors in multilateral treaties by the parties themselves, or, where there is one, through the depositary.63 Since 1964, the Secretary-General has adopted a consistent practice as depositary of communicating proposed corrections not only to signatory States but to all States that participated in the treaty’s elaboration. Objections to the correction of the original must be notified to the depositary within a certain period of time; Article 79(2) of the VCLT provides that the depositary ‘shall specify an appropriate time-limit within which objection to the proposed correction may be raised’. For the UN, that time limit is normally ninety days from the date of the notification of proposed corrections. Any objections References(p. 191) received are also notified to the parties concerned.64 Where a State is neither a signatory nor a contracting party, the Secretary-General communicates the objection to all interested States only for their information; that objection is not considered valid for the purpose of rejecting the correction.
What legal effects flow from objections to proposed corrections by a signatory or contracting party? The topic was not addressed in the VCLT or by the ILC.65 In practice, the Secretary-General has consistently tried to consult with the objecting State so that any corrections are firstly accepted unanimously. This practice, however, should not be construed to mean that the existence of an isolated objection may, by itself, ultimately prevent the corrections from taking effect. At the same time, it is also clear that, while many corrections are technical in nature and relate only to the authentic text of the treaty, some go beyond that and, in effect, constitute a proposal of amendment. In those cases, the Secretary-General should not treat the proposal as a correction, but insist it proceed under the completely different procedures for amendment, which vary depending on the relevant treaty provisions. The line of distinction between a correction and an amendment is sometimes very thin; in such cases the depositary may have a role to play informally. Otherwise, the communication of such a ‘disguised’ amendment to all States could always trigger objections from other States through the usual mechanism of amendments to treaties.
Where, however, there are no objections to the proposed corrections within ninety days, the corrections are deemed adopted. The Secretary-General circulates a procès-verbal of rectification containing the already deemed ‘as accepted’ correction.66 Sometimes and under exceptional circumstances a more simplified procedure has been followed by the Secretary-General: the text of the corrections may simply be communicated to all parties with a request that they make them in any appropriate documentation or regulation.
E. Signature or signature ad referendum
In bilateral treaties, a treaty will normally be signed by each party following the completion of negotiations and the ‘conclusion’ of the treaty (ie adoption and/or authentication). Signature may take place in a more or less formal setting, whether a ceremony vested with political significance or more discreetly. It may occur at the highest possible level (by Heads of State) or at the level of ministers or ambassadors.
Signature takes place on the authentic text of the treaty (which in this case is a duplicate). The parties then ‘exchange’ their signed texts, which differ only to the extent that in the text kept by each party its own name comes first in the title and preambular paragraphs of the treaty that contain the names of the parties (a practice known as the principle of the alternat).
References(p. 192) Most multilateral treaties are usually open for signature (either on the day of their adoption or, for practical reasons associated with the preparation of the text in all its authentic languages, a little later). For both bilateral and multilateral treaties, signature does more than serve as a means of concluding the treaty; it is also the first (but not necessarily the last) step of the expression of a State’s consent to be bound to a treaty.
The present Convention shall be open for signature by all States Members of the United Nations or of any of the specialized agencies, parties to the Statute of the International Court of Justice, and by any other State invited by the General Assembly of the United Nations to become a Party to the Convention, as follows: until 31 October 1961 at the Federal Ministry for Foreign Affairs of Austria and subsequently, until 31 March 1962, at the United Nations Headquarters in New York.69
This type of clause is called the ‘Vienna’ formula, since it was consistently used by all the Vienna Conventions, starting with the VCDR.70 It allows those States that, for political or other reasons, are not members of the UN to participate in the treaty because they can be members of specialized agencies (where the veto71 does not exist concerning admission of new members).
the Secretary-General, in discharging his functions as a depositary of a convention with an ‘all States’ clause, will follow the practice of the Assembly in implementing such a clause and, whenever advisable, will request the opinion of the Assembly before receiving a signature or an instrument of ratification or accession.73
The Secretary-General’s Summary of Practice explains that the ‘practice of the General Assembly’, referred to in this understanding concerns situations where there are unequivocal indications from the Assembly that it considers a particular entity to be a State even if it does not fall within the ‘Vienna formula’.74
Aside from the Vienna and ‘all States’ formulas, there are many other ways in which participation in the treaty via signature can be limited. A treaty may be open for signature only for the States engaged in the negotiation, only for those State members of a particular organization (such as the COE), or only for other States specifically invited to sign the treaty.75 Limitations on signature do not, however, always preclude excluded States from joining the treaty; they may just be limited to using accession as the means to do so.76 Examples of various types of signature clauses are included in Section VI of this Volume.
The legal effects of signature vary widely. In most multilateral treaties, signature does not have any real legal effect except triggering the rather general obligation of the signatory not to act in a way contrary or detrimental to the provisions of the treaty.77 As discussed below, however, in some cases signature may be simultaneously an expression of the consent to be bound if the treaty so provides.
The representative may also sign ad referendum, a signature that needs to be ‘confirmed’ at a later stage. The representative thus signs the treaty in a purely ceremonial or symbolic manner declaring that (for constitutional or other reasons) this signature (for which full powers are still required) will be duly confirmed. Signature ad referendum occurs in both bilateral and multilateral treaties.
References(p. 194) In general, multilateral treaties have separate provisions concerning signature, on the one hand, and ratification, accession, approval, etc, on the other. These provisions will often specify when signature can occur. Some signature clauses leave the treaty open for signature indefinitely as, for example, the one in the International Covenant on Civil and Political Rights.78 In other cases, a signature clause may leave the treaty open for signature for a specified period of time (from a few months to a number of years).79
There is little practice concerning treaty signature by IOs. The head of the organization (Secretary-General or Director-General) has the authority to sign on behalf of the organization after he has been duly authorized by the competent body of the organization (General Assembly, Council, Executive Board, etc). But it is an open question as to what other officials may sign on behalf of the IO. In the UN, the practice has not always been consistent. Generally speaking, any other official who signs a multilateral treaty on behalf of the organization needs full powers signed by the Secretary-General. In the case of bilateral agreements between the UN and a State or another organization, this practice might be more flexible depending on the nature of the agreement. Agreements of limited duration or of technical or specific content have been signed by the head of the relevant department of the Secretariat under whose competence the agreement falls without full powers from the Secretary-General80 and that signature will also express the UN’s consent to be bound.
Finally, a very different type of signature can occur when a treaty text is ‘witnessed’ by the signature of one or more representatives of third States or IOs. Witnessing has no legal effect for the participants or for those engaged in the witnessing; it is done more for its political effect in promoting the agreement or its effects.81 Thus, the Dayton Agreement was witnessed by various heads of State and government, who added their signatures after the parties.82
F. The Final Act
The Final Act consists of a document (usually attached to the main text of the treaty) containing a summary of the proceedings of a conference (dates and place, participants, organization, etc) as well as of the organization of its work. Final Acts References(p. 195) accompany the conclusion of almost any international conference, not just those associated with the adoption of a treaty. But where a treaty is involved, the Final Act serves as the ‘certificate of birth’. It is often signed separately from the treaty text, but doing so has no effect on whether or not a State signs or joins the treaty itself.
The term ‘Final Act’ (Acte Final) is normally used to designate a document which constitutes a formal statement or summary of the proceedings of an international conference recording the result of its deliberations, including, as the case may be, enumerating the treaties or related treaty instruments drawn up together with any resolutions or vœux adopted by the Conference. The signature of an instrument of this nature does not in itself entail any expression of consent to be bound by the treaties or related treaty instruments so enumerated.83
The Final Act is not normally part of the treaty and does not usually contain substantive provisions. They may, however, contain vœux, recommendations, or even resolutions with ‘semi-binding’ force, in the sense that they constitute strong policy orientations or indications.
In some cases, however, the Final Act may constitute what Shabtai Rosenne dubbed ‘informal treaty making’. Resolutions can be annexed to the Final Act that are intended to produce legal consequences.84 Or, they may incorporate resolutions or decisions intended to complement a treaty.85 Accordingly, and in the absence of any specific clause of the treaty to that effect, a case-by-case examination of Final Acts is appropriate to determine whether a Final Act includes precise legal commitments (deriving from or pertaining to the treaty).86 Indeed, such scrutiny is also warranted given that a Final Act may be part of the context associated with the treaty for interpretative purposes.87
For historical and other reasons, the adoption of a treaty or even its signature does not necessarily mean that a State has consented to be bound by the treaty (unless, of course, the contracting parties have agreed that signature has this effect). Instead, States, IOs, and other treaty-making entities must consent to a concluded treaty before they can be legally bound to comport with its substantive contents (and References(p. 196) then, as discussed below, only when the treaty has entered into force). There are various distinct procedures for the expression of such consent:
(a) signature (as a form of expression of the consent to be bound by the treaty);
(b) exchange of instruments constituting a treaty;
(e) formal confirmation; or
As with other aspects of treaty formation, to perform any of these acts, the representative must normally have sufficient authority (whether as a matter of the office held or full powers) to express the State’s consent to be bound to a treaty.
A. Agreements in simplified form: signature as a form of expression of the consent to be bound by the treaty
Article 12(1)(a) of the 1969 and 1986 VCLT states that consent to be bound by a treaty can be expressed by signature when the treaty provides that signature shall have this effect. This is known as a definitive signature. Treaties containing such a provision are generally referred to as ‘simplified agreements’ or agreements in simplified form (Accords en forme simplifiée). The main reason for this simplified procedure is to have the treaty operative as soon as possible in lieu of the cumbersome and lengthy process of ratification.89 Beside this procedural difference, there is not any substantial difference between simplified agreements and other treaties. Suffice it to say that some very important international treaties were in simplified form.90 Today, the practice of simplified agreements is widespread and represents a large proportion of international agreements.
In addition to recognizing the consent to be bound by signature where the treaty so provides, the Vienna Conventions provide that signature may constitute consent where it is otherwise established that the negotiating States agreed on that effect for signature (Article 12(1)(b)), or that intention appears from the full powers, or was expressed during the negotiation (Article 12(1)(c)). In practice, however, these circumstances rarely, if ever, arise; indeed, it seems difficult to contemplate these conditions occurring in the absence of specific provisions of the treaty to such References(p. 197) effect.91 In the absence of any clause, the assumption is that signature also expresses the consent to be bound by the treaty and consequently triggers its entry into force.
For many bilateral treaties, consent to be bound comes through an exchange of diplomatic notes or letters. Usually, one side proposes the terms of an agreement to the other side, accompanied by notice that a positive reply will constitute consent to an international agreement. The reply note (or letter) will usually reproduce verbatim the substantive language of the first note and then indicate the replying State’s agreement to those terms. Thus, the reply note functions to signal the replying State’s consent to be bound, the proposing State’s consent having come in the initial offer to make a treaty. The treaty’s terms will often indicate that completion of this exchange also serves to bring the treaty into force. In some cases, however, the treaty may condition entry into force on a further exchange (usually involving notifications of each side’s completion of any necessary domestic procedures). An exchange of instruments may, on occasion, involve more than two parties, although as Aust notes, doing so may generate technical difficulties.92
Ratification is among the most time-honoured means for a State to consent to be bound by a treaty; it is ‘definitive confirmation of a willingness to be bound’.93 It is a method, however, usually available only to those States that have signed a treaty. Basically, ratification involves a two-faceted process: (a) the introduction of the treaty into the internal legal order of the State through the completion of an ‘instrument of ratification’; and (b) the notification of this act of ratification at the international level through an exchange of such instruments or providing them to a depositary.
Ratification thus has both an internal and an international component. The modalities of the internal process leading to the establishment of an instrument of ratification can vary depending on the State’s internal legislation (including its constitution). It may involve approval by legislative bodies, the Executive, or both. In some States, ratification of certain treaties may also require approval of the people.94 The historic development of ratification and its modern iteration reflect References(p. 198) its utility ‘in the majority of cases as the means of submitting the treaty-making power of the executive to parliamentary control’.95 Thus, in most contemporary democracies, involvement of legislative or parliamentary bodies in the decision to ratify is usually required.96
Once the internal processes necessary for a State to establish an instrument of ratification are complete, ratification requires an international act; either the formal deposit of the instrument of ratification with the depositary, or in the case of bilateral treaties an exchange of the instruments of ratification between the two sides. To satisfy the international process of ratification, an instrument of ratification must usually state solemnly that the State has ratified the treaty; it may also contain any reservations, understandings, or declarations the State wishes to make with respect to the treaty. An instrument of ratification should be precise and state in full the title of the ratified treaty, contain an unambiguous expression of the will to be bound of the government acting on behalf of the State, and undertake to implement its provisions. It has to be signed by the Head of State, the Head of Government or the Minister for Foreign Affairs or, as the case may be, by another organ which has received the full powers. Usually, however, instruments of ratification are signed by the Head of State or the Minister for Foreign Affairs.
In the case of an exchange of instruments of ratification, each party transmits to the other its own instrument duly signed.97 The transmission may take place through diplomatic channels. Or, it may occur in a solemn meeting between the representatives of each party, which may include signature of an additional document—a certificate of exchange or procès-verbal.98 Deposits are made in the capital of the depositary State or the headquarters of an IO serving in that role.
Where a treaty is deposited with the UN, in contrast, a procès-verbal no longer accompanies the deposit of the instrument of ratification. The Secretary-General, in his capacity as depository, announces immediately (usually the next day) the deposit in the Official Journal of the UN and subsequently issues a depositary notification (addressed to all States).99 Consent itself takes effect from the date of exchange or deposit, but is ‘activated’ when the treaty enters into force.
According to the VCLT, ratification is an available option when the treaty expressly so provides or the representative signs the treaty subject to ratification.100References(p. 199) Like definitive signature, moreover, the VCLT also contemplates ratification where it can be an inferred option if it can be otherwise established that the negotiating States so agreed, or that intention is evidenced in the full powers or was expressed during the negotiations. If, however, a treaty is silent on the question of ratification, there is a presumption that ratification is not required.101
What purpose does ratification serve? As McNair explained, ‘the interval between the signature and the ratification of a treaty gives the appropriate departments of Government that have negotiated the treaty an opportunity of studying the advantages and disadvantages involved in the proposed treaty as a whole’.102 It affords to governments a time of reflection and the space to obtain any parliamentary or other approvals. This does not mean that the negotiation may be reopened (although this should not be excluded in bilateral treaties). But logically this ‘reflection’ leads to the possibility of formulating reservations or declarations at the time of the deposit of the instrument of ratification. Moreover, States can use the interval to enact any necessary domestic legislation.
Generally, it should be stressed that there is not any general obligation to ratify an already signed treaty or to apply its provisions.103 At one time, the Permanent Court of International Justice suggested there might be, in exceptional cases, an abuse of right in case of a refusal to ratify.104 But there has been no subsequent practice or jurisprudence to that effect.
The terms ‘acceptance’ or ‘approval’, although they may not be quite identical, are usually used alternatively to designate a form of expression of consent to be bound by a State which does not want to use the ‘ratification’ process. Usually this occurs due to a desire to avoid the internal, domestic legal processes required for ratification. One case where the acceptance of a treaty (instead of its ratification) was politically expedient occurred in 1934 when the US Congress authorized the US President to accept, for the US Government, membership in the International Labour Organization (ILO); at the time, ratification was excluded since the ILO Constitution formed part of the Treaty of Versailles, to which the US Senate had declined to give its advice and consent to ratification.105
At the international level, acceptance and approval are usually treated as equivalent either to ratification or accession.106 Some treaties adopted within the UN system, however, mention only the term ‘acceptance’ in the context of discussing References(p. 200) consent to be bound, without referring to ratification or even accession.107 In that context, ‘acceptance’ is used as a generic term that may be fulfilled by any form of expression of consent to be bound (ie ratification or accession). As the ILC has confirmed: ‘on the international plane “acceptance” is an innovation which is more one of terminology than of method’.108
in the absence of an accepted term, the Commission has confined itself to describing this mechanism [of consent] by the words ‘act of formal confirmation’ … When necessary, international organizations, using a different terminology, can thus establish on an international plane their consent to be bound by a treaty by means of a procedure which is symmetrical with that which applies to States.109
At present, however, there is no generally accepted international designation of such a mechanism. In practice, the process followed for depositing an act of formal confirmation is the same as for an IO’s signature of a treaty, namely, prior authorization by the deliberative body of the IO concerned (General Assembly, Executive Council, etc).110 The UN, the World Health Organization, and the International Maritime Organization, among others, have all deposited just such an instrument to indicate their consent to the 1986 VCLT.
States which have not participated in the negotiation and signature of the treaty may become parties subsequently through accession. At the internal level, the procedure remains the same (it is that of ratification, albeit by another name). At the international level, an instrument of accession differs from one of ratification because the State concerned did not previously sign the treaty. As noted above, treaties usually provide for a certain period during which they can be signed; thereafter they are open only to accession. But even in cases when treaties are permanently open for signature such as the ICCPR, an instrument of accession may References(p. 201) still constitute a method of consenting to a treaty without signing it. In a few cases, such as the 1928 General Act for the Pacific Settlement of Disputes or the Convention on the Privileges and Immunities of the UN, treaties may only authorize consent by accession.111
There is, however, no guaranteed right for States to consent by accession; accession is only available where the treaty so provides or the parties agree to a State’s accession. Thus, accession may be restricted to specific categories of States or subject to specified conditions (eg requiring an acceding State to be party to another treaty).112 Practically speaking, most multilateral treaties include an accession clause. The opinion that accession is possible only after the entry into force of the treaty, moreover, is no longer valid.113 Finally, the international procedures for depositing instruments of accession follow those for instruments of ratification.
For each of the methods of expressing consent to be bound discussed above, Section VI(6) of this Volume contains a set of sample clauses illustrating the precise form and terminology used.
A State, IO, or other treaty-making entity is not necessarily legally bound to a treaty by its consent; the treaty must also have entered into force for it. Where a State consents and the treaty is in force for that State, it is considered a ‘party’ to the treaty. Moreover, the fact that a treaty is in force for some States does not mean that it is in force for all States in general; only those States that have expressed their consent are so bound.
According to Article 24 of the VCLT, a treaty enters into force ‘in such manner and upon such date as it may provide or as the negotiating States may agree’.114 As a general rule, bilateral treaties enter into force upon the completion of the formalities required for their ratification and the exchange of diplomatic notes or notifications informing the other party about such a completion. In the case of simplified agreements, bilateral treaties may enter into force on signature, the date of the last signature if not done simultaneously, or—in case of an exchange of notes—on the replying States’ affirmative acceptance of the proposed treaty. Sometimes the procedure of a procès-verbal of entry into force is followed. Usually, bilateral treaties stipulate that they enter into force on the date (or after a certain period of time following this date) of receipt of the last of any notifications. In case of silence, there References(p. 202) is a presumption that the entry into force takes place on the date of exchange of the notifications.115
In the case of a multilateral treaty, the conditions for entry into force are stated in the treaty itself, usually in its final clauses. As indicated in Article 24(4) of the VCLT, those provisions, whether regulating the number of ratifications or accessions needed for its entry into force or the date of such entry into force, apply from the time of the adoption of its text.116 The treaty may provide that it shall enter into force on a specific date or on the date when certain conditions are met.117 The determination of such conditions is relatively simple in the case of bilateral treaties, but may be more complicated when multilateral treaties are involved. Usually, in such a case, the date and the fulfilment of the conditions of entry into force are determined by the depositary.
There are many variations with regard to the conditions for entry into force of a treaty. The treaty may require ratification, acceptance, or accession by a minimum number of negotiating States (eg two) or any number all the way up to unanimity by all signatories.118 UNCLOS required sixty ratifications, while the VCLT itself required thirty-five.119 Other treaties may add additional conditions, stipulating, for example, that a number of parties should belong to a particular geographic area or region or that certain percentages are reached or exceeded.120 Or, treaties may link entry into force to consent by specific States—the 1996 Comprehensive Test References(p. 203) Ban Treaty requires consent by forty-four named States.121 Commodities agreements such as the 2001 International Cocoa Agreement may stipulate even more complex conditions for entry into force relating, for example, to ratification by certain governments representing a number of exporting and importing members and holding various percentages of votes.122 In some cases, especially agreements that are constituent treaties of IOs, a supplementary condition may be an additional agreement (after the required number of ratifications has been reached) of the contracting parties.123 Examples of entry into force clauses may be found in Section VI(11) of this Volume.
State which has made and maintained a reservation which has been objected to by one or more of the parties to the Convention but not by others, can be regarded as being a party to the Convention if the reservation is compatible with the object and purpose of the Convention, otherwise that State cannot be regarded as being a party to the Convention.124
The opinion, however, did not address what happens if one contracting party considers the reservation of another contrary to the object and purpose of the treaty. Should the reserving State not be counted for the determination of the date of entry into force? Technically, that seems the correct result, but only if the objecting State is correct in its belief that the reservation is inadmissible. Divergent views on the admissibility of reservations thus create the possibility for much confusion on questions of entry into force.
Since 1952, the practice of the UN Secretary-General has been to avoid such confusion by accepting the deposit of all instruments containing reservations or objections.125 Consequently, the Secretary-General simply counts these instruments to determine whether the required number of ratifications for the entry into force of the treaty has been attained.126
The ILC has also opted for a more ‘nuanced’ orientation in its Guide to Practice finalized in 2011; Guideline 4.2.2 provides that, when ‘a treaty has not yet entered into force, the author of a reservation shall be included in the number of contracting States and contracting organizations required for the treaty to enter into force References(p. 204) once the reservation is established’.127 But, concomitantly even the author of an invalid reservation is considered to be a party to the treaty ‘without the benefit of the reservation’, ‘unless [it] has expressed a contrary intention or such an intention is otherwise established’.128 This position differs from the 1951 Advisory opinion, which had seemed to leave the admissibility of reservations to the subjective views of the contracting States; the Guide to Practice seems to opt for a more objective system where the incompatibility of the reservation with the object and purpose of the treaty may exist as such irrespective of the ‘subjective’ views of States parties. On the other hand, the Guide does allow for counting a reserving State within the number of contracting States if no contracting State is opposed to it in a particular case.129
Beyond such theoretical debates, in practice, most final clauses use wording to the effect that the treaty will enter into force [X] days following the deposit of the (Xth) instrument of ratification. It could be argued that this formulation further reflects a formalistic approach, stressing the act of depositing an instrument rather than the effectiveness of the ratification itself. Indeed, this approach is also an indication of a ‘limitation’ in the system of entry into force of treaties, which seems to repose on a tacit ‘fictional’ belief that all instruments are prima facie valid; otherwise this system would crash like a house of cards.
Assuming a depositary will count all deposited instruments, there are still other potential complications that may arise in determining when the treaty enters into force. A depositary may not recognize an instrument of ratification for purposes of entry into force for a variety of reasons such as where the contracting State has disappeared. It is also possible, according to State practice, for a State to withdraw its instrument of ratification before the treaty enters into force.130 These sorts of adjustments may be decisive for the calculation of the number of instruments required for the treaty’s entry into force.
In terms of timing, the calculation of the effective date of initial entry into force is based on the provision of the relevant final clauses of the treaty. If the treaty provides for entry into force ‘on the thirtieth day following [or after] the deposit of the [Xth] instrument of ratification’, the practice of the UN Secretary-General is to have the time run from the day following the deposit of the last required instrument. Thus, in the above example, if the deposit of the last instrument is affected on (p. 205) 15 March, the period of thirty days will begin on 16 March, and the Convention will enter into force on 14 April. In contrast, for entry into force ‘three months after the deposit’, the time runs from the day of the deposit of the last required instrument. Thus, in this last example, if the deposit was affected on 31 March, since there is no corresponding 31 June, the Convention would then enter into force on the last day of June, ie 30 June. Similarly, upon the relevant deposit on 30 November the Convention would enter into force on 28 February (or on 29 February for leap years). But whenever there is a ‘same’ day [X] months later, that day is the day of entry into force.131 Nowadays there is no procès-verbal for entry into force unless the treaty so requires.132
Finally, it is possible for the treaty to stipulate provisional application or even a provisional entry into force.133 For example, the General Agreement on Tariffs and Trade (GATT) included a Protocol of Provisional Application, which provided for provisional application on and after 1 January 1948 (‘provided this Protocol shall have been signed on behalf of all the foregoing Governments not later than 15 November 1947, to apply provisionally on and after 1 January 1948’).134 Usually a date of provisional application or provisional entry into force is stipulated in the treaty itself. The Agreement on Implementation of Part XI of the UNCLOS, however, incorporated implicit consent to the provisional application of that Agreement, merely upon the adoption of the Agreement or its signature.135
This chapter has attempted to summarize in a comprehensive and not too cumbersome manner the very complex question of treaty-making. From an overview of the treaty-law rules as well as practice, three major points seem to emerge.
First, treaty-making is a manifestation of the sovereignty and free will of States and other subjects of international law. Entities such as States participate willingly and freely in this exercise and, consequently, there is a presumption that States can sacrifice a portion of sovereignty that the participation in the treaty implies to the extent that they undertake obligations (as also they may acquire rights) in an eminently (although not perfectly) symmetrical exercise.
Second, treaty-making is characterized by a combination of formality and flexibility. The formalism of some rules (full powers, expressions of consent to be bound) is complemented by the flexibility of others (required majorities for the adoption of texts). In many ways, everything is dependent on the will of the parties; even the strictest rules may be set aside if the parties so wish. Ultimately, however, References(p. 206) there are limits to such flexibility since extreme informality might defeat the security towards which all treaty relations aspire.
Third and finally, treaty-making is and will continue to be an important aspect of international law and international relations. As long as States, IOs, and other entities are willing to enter into contractual relations and mutual undertakings, treaty-making will remain an essential factor in the establishment of such relations and an indispensable mechanism.
What does the future hold? If in the field of bilateral treaties the techniques have not much changed, the same does not seem to be exactly true in the case of multilateral treaties. The era of great codification conferences now appears finished, although some resurgence might not be excluded in the future. IOs now provide more and more frequently for multilateral treaty-making. We might expect future treaty-making to take place in even smaller and less institutionalized settings (eg, the G20, various diplomatic ‘summits’). Treaties themselves may take a more ‘inchoate’, hybrid form (under titles such as declarations, findings, principles, etc), creating thus a sort of legal limbo, a cross between treaties and instruments whose binding force is not clear. Technological development (especially the use of the Internet) may affect the rules of treaty-making as well. None of this should detract, however, from the fundamental premise: the combined will of States or other entities can make treaties. It will be for the lawyer and historian of the future to shed light on any new practices and emerging patterns, to clarify them and, if necessary, to revise and restate the codified rules.
- A Aust, Modern Treaty Law and Practice (2nd edn CUP, Cambridge 2007)
- S Bastid, Les Traités dans la vie internationale: Conclusion et Effets (Economica, Paris 1985)
- C Brölmann, The Institutional Veil in Public International Law: International Organisations and the Law of Treaties (Hart Publishing, Oxford 2007)
- O Corten and P Klein, Les conventions de Vienne sur le droit des traités: commentaires article par article (Bruylant, Brussels 2006)
- O Corten and P Klein, The Vienna Conventions on the Law of Treaties: A Commentary (OUP, New York 2011)
- M Craven, The Decolonization of International Law: State Succession and the Law of Treaties (OUP, Oxford 2007)
- Y Dinstein, ‘The Interaction between Customary International Law and Treaties’ (2006) 322 RcD 243
- M Fitzmaurice and O Elias, Contemporary Issues in the Law of Treaties (Eleven International Publishers, Utrecht 2005)
- RK Gardiner, Treaty Interpretation (OUP, Oxford 2008)
- DB Hollis, MR Blakeslee, and LB Ederington (eds), National Treaty Law and Practice (Martinus Nijhoff, Leiden 2005)
- M Lang, J Schuch, and C Staringer (eds), Tax Treaty Law and EC Law (Kluwer Law International, Alphen aan den Rijn 2007)
- U Linderfalk, On the Interpretation of Treaties: The Modern International Law as Expressed in the 1969 Vienna Convention on the Law of Treaties (Springer, Dordrecht 2007)
- (p. 207) VD Pastuhov, A Guide to the Practice of International Conferences (Carnegie Endowment for International Peace, Washington 1945)
- N Quoc Dinh, P Daillier, and A Pellet, Droit international public (6th edn LGDJ, Paris 1999)
- I Roberts (ed), Satow’s Diplomatic Practice (6th edn OUP, Oxford 2009)
- S Rosenne, Developments in the Law of Treaties 1945–1986 (CUP, Cambridge 1989)
- J Siotis, Essai sur le Secrétariat international (Droz, Geneva 1963)
- UN Legislative Series, Review of the Multilateral Treaty-Making Process, ST/LEG/SGR.13/21
- ME Villiger, Commentary on the 1969 Vienna Convention on the Law of Treaties (Martinus Nijhoff, Leiden 2009)
- R Wolfrum and R Volker (eds), Developments of International Law in Treaty Making (Springer, Berlin 2005)
* The views expressed in this chapter are solely those of the author and do not necessarily reflect the views of the United Nations. The author would also like to thank Mr G Buzzini for his valuable comments.
1 Paul Guggenheim refers to a treaty concluded in 1280 BC between Ramses II of Egypt and Khatisir, King of the Hittites. P Guggenheim, ‘Contribution à l’histoire des sources du droit des gens’ (1954) 85 RcD 54, 55.
2 See UN Doc FCCC/CP/2009/11/Add.1 (containing the Copenhagen Accord and a proposed Amendment to the UN Framework Convention on Climate Change); UN Doc A/C.6/65/L.10 (containing the latest proposals concerning a comprehensive anti-terrorism convention).
4 For a discussion of the role of treaties in domestic law, see Chapter 15.
5 Eg VCLT; Vienna Convention on Succession of States in Respect of Treaties (adopted 23 August 1978, entered into force 6 November 1996) 1946 UNTS 3; Vienna Convention on the Law of Treaties Between States and International Organizations or Between International Organizations (opened for signature 21 March 1986, not yet in force) UN Doc A/CONF.129/15 (‘1986 VCLT’).
8 For a discussion of jus cogens, see Chapter 22, Part III.D, 570 et seq.
9 See eg Wye River Memorandum of 23 October 1998 between Israel and PLO  37 ILM 1251; Agreements concluded in the context of the conflict in Yugoslavia with local authorities or communities (including Croatia and Local Serbian Community)  35 ILM 1.
10 In a substantively legal normative sense. The practice of conferences and diplomatic meetings has been addressed, as far as the technical aspect is concerned, in handbooks, manuals, etc, such as Satow’s Diplomatic Practice or even the still invaluable V Pastuhov, A Guide to the Practice of International Conferences (Carnegie Endowment for International Peace, Washington 1945).
15 For more on this distinction, see Chapter 1, Part II.B, 31 et seq.
16 Final Act, 1125 UNTS 438 ; on treaty-making by other subjects of international law, see also Chapter 5.
17 For example, participation of entities such as the Palestine Liberation Organization (PLO) in the 1974 Geneva Conference and the revolutionary government of South Vietnam gave rise to many difficulties.
20 For a detailed discussion of the treaty-making capacity of IOs, see Chapter 3.
25 Agreement between the United Nations Organization and the Government of the United States of America concerning the establishment of security for the United Nations presence in Iraq (31 December 2008) (2009) UNJY 5.
26 See eg Memorandum of Understanding between the United Nations Industrial Development Organization and the Latin American Energy Organization (16 and 25 February 2009) (2009) UNJY 56; Memorandum of Understanding on Cooperation between the Commission of the African Union and the Technical Secretariat of the Organization for the Prohibition of Chemical Weapons (24 January 2006) (2009) UNJY 78.
27 P Guilland, ‘Le rôle de l’Algérie dans la conclusion de l’accord entre les Etats-Unis et l’Iran’ (1981) AFDI 29. The fourth agreement on a peaceful resolution of the situation in Afghanistan concluded on 14 April 1988, constitutes another variation; it was concluded apparently between two different sets of parties (United States–USSR/Afghanistan–Pakistan).  27 ILM 577.
32 See the lectures by T Koh, ‘The Negotiating Process of UNCLOS III’ and ‘The Art and Science of Chairing Major Inter-governmental Conferences’ <http://untreaty.un.org/cod/avl/ls/Koh_T_LS.html>. (noting the conference’s innovative use of the principle ‘No agreement until final agreement’; the Gentleman’s Agreement on consensus; the emergence of new interest groups of States based on geographical factors, and parallel formal and informal negotiating processes).
33 See UNGA Res 56/93 (28 January 2002) (establishing the Ad Hoc Committee for the purpose of considering the elaboration of an international convention against the reproductive cloning of human beings).
34 See eg UNCITRAL, Convention on the Use of Electronic Communications in International Contracts (adopted 23 November 2005, opened for signature 16 January 2006) UN Doc A/Res/60/21; Convention on Biological Diversity (adopted 5 June 1992, entered into force 29 December 1993) 1760 UNTS 79.
35 ‘Non-papers’ are texts distributed on an informal basis and designed to facilitate the process of negotiating an agreement. It is not a proposal (although it may foreshadow one) and does not engage its author. UNITAR, Glossary of Terms 116.
36 See eg documents establishing the Rules of Procedure for the UN Diplomatic Conference of Plenipotentiaries on the Establishment of an International Criminal Court. UN Doc A/CONF.183/2/Add.2; UN Doc A/CONF.183/3.
37 RY Jennings, ‘Law Making and the Package Deal’, in Mélanges Offerts à Paul Reuter (Pedone, Paris 1981) 347–55; G de Lothaire, ‘Aspects juridiques de la négociation sur un Package deal à la Conférence des Nations Unies sur le droit de la mer’ in Essays in Honor of Erik Castren (Finnish Branch of the International Law Association, Helsinki 1979) 30–45.
39 ‘[A]ny other act with respect to the treaty’ can include authority to terminate or suspend the treaty for a State, to provisionally apply it, or to extend its territorial reach. See UN Office of Legal Affairs, Treaty Handbook (2006) Sales No. E.02.V.2 <http://treaties.un.org>.
40 Credentials are mainly used in multilateral settings but may also arise in bilateral negotiations, especially those involving a more formal framework. In UN practice, credentials to participate in a conference are also considered sufficient for the signature of the Final Act.
43 As the ILC noted, however, the principle of limited automatic representativeness ‘is not considered in practice to extend, without production of full powers, to expressing the consent of the State to be bound by the treaty’:  YBILC, vol II, 193. Thus, the principle’s coverage extends to initialling of a treaty, but not to expressing the State’s consent to be bound by it.
45 Notifications of full powers have, in the past, been acceptable by telegram or facsimile, followed by delivery of the actual full powers. Today, electronic copies (eg a .pdf) may also precede actual delivery.
47 Where the negotiating body is constituted ad hoc, rules of procedure, including the voting rules, are usually among the first items agreed by the negotiating participants. As a general matter, VCLT Art 9 suggests a preference for unanimity in adoption, but allows for it to occur by a two-thirds vote of States present and voting (unless those two-thirds decide to apply a different rule).
50 See UNGA, Rules of Procedure of the General Assembly (2007) UN Doc A/520/Rev 17, Rules 82–93. For examples of treaties adopted by the UN General Assembly without a vote, see Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment (adopted 10 December 1984, entered into force 26 June 1987) 1465 UNTS 85; International Convention for the Suppression of Acts of Nuclear Terrorism (adopted 13 April 2005, entered into force 7 July 2007) 2445 UNTS 89. For examples of treaties adopted by a vote of the General Assembly, see International Covenant on Civil and Political Rights (adopted 16 December 1966, entered into force 23 March 1976) 999 UNTS 171 (ICCPR); Comprehensive Nuclear Test-Ban Treaty (opened for signature 24 September 1996, not yet in force)  35 ILM 1439 (CTBT).
52 During the Third Conference on the Law of the Sea, a declaration incorporating the ‘Gentlemen’s Agreement’ approved by the General Assembly made by the President and endorsed by the Conference was appended to the rules of procedure and provided that: ‘The Conference should make every effort to reach agreement on substantive matters by way of consensus and there should be no voting on such matters until all efforts at consensus have been exhausted.’: ‘167th Plenary Meeting’ Third United Nations Conference on the Law of the Sea (7 April 1982) UN Doc A/CONF.62/SR.167 .
53 A notable exception, discussed in the text accompanying n 135, is the 1994 Agreement relating to the implementation of Part XI of the United Nations Convention on the Law of the Sea of 10 December 1982 (adopted 28 July 1994, entered into force 28 July 1996) 2167 UNTS 3, which was subject to provisional application.
57 The ILC commentary on draft Art 29 (subsequently Art 35) provided that: ‘The plurality of the authentic texts of a treaty is always a material factor in its interpretation, since both or all the texts authoritatively state the terms of the agreement between the parties. But it needs to be stressed that in law there is only one treaty—one set of terms accepted by the parties and one common intention with respect to those terms—even when two authentic texts appear to diverge’:  YBILC, vol II, 225. For a discussion of the interpretative issues associated with plurilingual treaties, see Chapter 19, Part II.C, 490 et seq.
58 The Conclusion of the Russo-Japanese War (Russia-Japan) (5 September 1905), Art XV (‘The present treaty shall be signed in duplicate in both the English and French languages. The texts are in absolute conformity, but in case of a discrepancy in the interpretation the French text shall prevail.’).
59 Such was the case for example with the Chinese text of the International Tropical Timber Agreement (2006)  OJ L262/8, the testimonium of which read: ‘Done at Geneva on the eighteenth day of November, one thousand nine hundred and eighty-three, the texts of this Agreement in the Arabic, English, French, Russian and Spanish languages being equally authentic. The authentic Chinese text of this Agreement shall be established by the depositary and submitted for adoption to all signatories and States and intergovernmental organizations which have acceded to this Agreement.’
60 See eg UNGA Resolution 317 (IV) (21 March 1950) by which the General Assembly adopted the Convention for the Suppression of the Traffic in Persons and of the Exploitation of the Prostitution of Others; Summary of Practice of the Secretary-General as Depositary of Multilateral Treaties, ST/LEG/7/Rev 1, 11–12.
62 The origin of this practice of correction of errors goes back to the Genocide Convention and the request from the Government of China for revision of the Chinese text of the Convention. The Secretary-General submitted a comprehensive memorandum (UN Doc A/2221) on that matter. The question of lack of concordance in the original of a multilateral treaty may even arise during the conference in which the text of the treaty is elaborated. Such was the case during the Rome Conference with regard to the Spanish text.
63 Where there is no depositary, the parties may allow duly authorized representatives to initial a correction; execute and exchange instruments making the correction; execute a corrected text by the same procedure used in the case of the original text; or decide upon some other means of correction. VCLT Art 79(1).
64 Summary of Practice (n 60) 15.
66 Summary of Practice (n 60) 16.
This Convention shall be opened for signature by:
(c) all self-governing associated States which have chosen that status in an act of self-determination supervised and approved by the United Nations in accordance with General Assembly resolution 1514(XV) and which have competence over the matters governed by this Convention, including the competence to enter into treaties in respect of those matters;
(d) all self-governing associated States which, in accordance with their respective instruments of association, have competence over the matters governed by this Convention, including the competence to enter into treaties in respect of those matters;
(e) all territories which enjoy full internal self-government, recognized as such by the United Nations, but have not attained full independence in accordance with General Assembly resolution 1514(XV) and which have competence over the matters governed by this Convention, including the competence to enter into treaties in respect of those matters;
69 VCDR (n 44) Art 48.
72 See eg Convention on Cluster Munitions (adopted 3 December 2008, entered into force 1 August 2010)  48 ILM 354, art 15 (‘This Convention … shall be open for signature at Oslo by all States’).
74 Summary of Practice (n 60) 23.
75 See eg Customs Convention on Containers (adopted 18 May 1956, entered into force 4 August 1959) 338 UNTS 103, Art 12 (‘Countries members of the Economic Commission for Europe and countries admitted to the Commission in a consultative capacity … may become contracting parties to this Convention … (a) by signing it’); see also Charter of the Asian and Pacific Development Centre (adopted 1 April 1982, entered into force 1 July 1983) 1321 UNTS 203, Arts III and XVI; European Anti-Doping Convention (opened for signature 16 November 1989, entered into force 1 March 1990) CETS No 135, Art 14.
76 See Agreement Concerning the Adoption of Uniform Conditions of Approval and Reciprocal Recognition of Approval for Motor Vehicle Equipment and Parts (adopted 20 March 1958, entered into force 20 June 1959) 335 UNTS 218, Art 6.
77 See VCLT Art 18 (signatory state ‘is obliged to refrain from acts which would defeat the object and purpose of a treaty … until it shall have made its intention clear not to become a party to the treaty’). For a more detailed discussion, see Chapter 8 (‘Treaty Signature’).
78 ICCPR (n 50) Art 48.
79 Convention on Consent to Marriage (adopted 10 December 1962, entered into force 9 December 1964) 521 UNTS 231, Art 4 (‘The present Convention shall, until 31 December 1963, be open for signature’); UN Convention Against Corruption (adopted 31 October 2003, entered into force 14 December 2005) 2349 UNTS 41, Art 67 (‘This Convention shall be open to all States for signature from 9 to 11 December 2003 in Merida, Mexico, and thereafter at United Nations Headquarters in New York until 9 December 2005’).
80 An example of this bilateral practice is the Memorandum of Agreement between the United Nations and France for the Provision of Personnel to the United Nations Assistance Mission in Afghanistan of 4 March 2003, signed for the United Nations by the then Under-Secretary-General, Head of the Department of Peacekeeping Operations (2003) UNJY 12.
81 Aust (n 48) 101–2.
84 Rosenne (n 14) 118. For a famous example, see Resolutions I and II contained in Annex I of the Final Act of the 3rd UN Conference on the Law of the Sea.
85 The Final Act of the Rome Conference included in its Annex I resolutions of more than recommendatory nature, establishing, for example, the ICC Preparatory Commission. UN Doc A/CONF.183/13, vol 1, Resolution F.
86 Shabtai Rosenne wrote at length about the origin of the modern forms of Final Acts (‘as a deliberately non-binding instrument’) going back to the Hague Peace Conference of 1899. Rosenne (n 14) 107.
87 For a more detailed discussion of treaty interpretation, see Chapter 19, Part II.A.2, 482 et seq.
88 This is not an exclusive list. VCLT Art 11 authorizes these methods ‘or by any other means if so agreed’. As Aust notes, the ‘other means’ language has given treaty-makers room to utilize a variety of alternative procedures. Aust (n 48) 113.
89 In American practice, simplified agreements are often called ‘executive agreements’ and are concluded under the authority of the President without a role for the Senate; consent to be bound usually occurs via signature or an exchange of instruments. See R Dalton, ‘United States’ in DB Hollis and others (eds), National Treaty Law and Practice (Martinus Nijhoff, Leiden 2005) 780–3.
90 Eg the 1938 Munich Agreement which is silent on the date of its entry into force and is presumed to have entered into force upon signature. Munich Pact (signed and entered into force 29 September 1938) 204 LNTS 378.
91 Nor does the ILC’s commentary illuminate the meaning of these cases. See  YBILC, vol II, 196. With regard to Art 12(1)(b), the ILC simply noted that ‘it is simply a question of demonstrating the intention from the evidence’. But which evidence? Certainly the treaty itself provides the most decisive evidence. Article 12(1)(c) might be more conceivable; however, it would take place only if Art 12(1)(a) also allows it—that is to say, that the treaty provides for this effect of the signature.
92 Aust (n 48) 103 (citing double exchange of letters involving Bahrain, Saudi Arabia, and Qatar).
94 For instance, Art 140 of the Federal Constitution of the Swiss Confederation (18 April 1999) provides for a mandatory referendum in the case of ‘accession to organizations for collective security or to supranational communities’ and Art 141 provides for an optional referendum concerning particularly significant treaties.
96 Hollis (n 18) 33.
97 Earlier practice allowing for notification of the instrument of ratification to other contracting parties is ‘hardly, if ever, done now.’ Aust (n 48) 106.
99 Because in most cases they potentially may become parties to the multilateral treaty. But in cases of closed or restricted treaties, the notification is addressed only to those having the capacity to become parties.
100 VCLT Art 14. An exceptional procedure was included in Art 36(5) of the International Agreement on Olive Oil, 1956, as amended by the Protocol of 3 April 1958, providing that for purposes of entry into force a simple undertaking by a government to endeavour to obtain as speedily as possible in accordance with its constitutional procedure, either ratification or accession would be considered equivalent to ratification or accession. Such an undertaking was to emanate from one of the government authorities competent to sign an instrument of ratification or accession. See Summary of Practice (n 60) 39.
101 Aust (n 48) 104.
105 S Bastid (n 6) 69.
107 This has occurred especially in the case of amendments to treaties—eg Amendments to the Agreement Establishing the African Development Bank (adopted 17 May 1979, entered into force 7 May 1982) 1276 UNTS 501—or treaties constituent of IOs—eg Convention on the International Maritime Organization, as amended (adopted 6 March 1948, entered into force 17 March 1958) 289 UNTS 3. Sometimes, the terms ratification or acceptance are used in an interchangeable manner, such as in the Constitution of the Asia-Pacific Telecommunity (adopted 27 March 1976, entered into force 25 February 1979) 1129 UNTS 3, Art 17.
112 See the International Convention to Facilitate the Importation of Commercial Samples and Advertising Material (adopted 7 November 1952, entered into force 20 November 1955) 221 UNTS 225, Art X, whereby the Convention is open to accession (as well as for signature) by the GATT contracting parties.
114 If the treaty is silent on entry into force—a rarity in the modern era—it may be presumed that it will enter into force when all the negotiating States have consented to be bound (or upon its signature).
117 For example, the Agreement providing for the Provisional Application of the Draft International Customs Conventions on Touring, on Commercial Road Vehicles and on the International Transport of Goods by Road (adopted 16 June 1949) 45 UNTS 149, stipulated in its Art III that it was to enter into force on 1 January 1950. In the absence of other provisions, this Agreement entered into force on that date for those States that, at that date, had accepted to be bound by the Agreement. However, such a clause is unusual in multilateral treaties.
118 The 1949 Geneva Conventions came into force with two ratifications, while the Treaty of Rome required unanimity. Eg Geneva Convention Relative to the Treatment of Prisoners of War (adopted 12 August 1949, entered into force 21 October 1950) 75 UNTS 135, Art 138 (‘The present Convention shall come into force six months after not less than two instruments of ratification have been deposited’); Treaty Establishing the European Economic Community (Treaty of Rome) (adopted 25 March 1957, entered into 1 January 1958) 298 UNTS 3, Art 247 (‘The Treaty shall enter into force on the first day of the month following the deposit of the instrument of ratification by the last signatory State to comply with this formality’).
119 UNCLOS (n 67) Art 316(1) (‘Amendments to this Convention … shall enter into force for the States Parties … on the thirtieth day following the deposit of instruments of ratification or accession by two thirds of the States Parties or by 60 States Parties, whichever is greater’); VCLT Art 84(1) (‘The present Convention shall enter into force on the thirtieth day following the date of deposit of the thirty-fifth instrument of ratification or accession’).
120 Protocol of 28 September 1984 to the 1979 Convention on Long Range Transboundary Air Pollution on Long Term Financing of the Cooperative Programme for Monitoring and Forwarding of the Long Range Transmission of Air Pollutants in Europe (EMEP) (adopted 28 September 1984, entered into force 28 January 1988) 1491 UNTS 167, Art 10 (‘The present Protocol shall enter into force on the ninetieth day following the date on which: a) Instruments of ratification, acceptance, approval or accession have been deposited by at least nineteen States and Organizations referred to in article 8(1) which are within the geographical scope of EMEP’). A second condition was provided for by subpara (b) of Art 10(1): ‘The aggregate of the United Nations assessment rates for such States and organizations exceeds forty per cent.’ Ibid.
121 CTBT (n 50) Art XIV.
127 ILC, ‘Guide to Practice on Reservations to Treaties’ 63rd session (2011) UN Doc A/66/10, forthcoming in  YBILC, vol II(2) (‘Guide to Practice’) (emphasis added). For the meaning of the term ‘established reservation’, see Guideline 4.1 which states that ‘a reservation formulated by a State or an IO is established with regard to a contracting State or a contracting organization if it is permissible and was formulated in accordance with the required form and procedures, and if that contracting State or contracting organization has accepted it’.
130 An interesting question would be if such a withdrawal is possible if the treaty itself does not provide for withdrawal. It is here submitted that, even in such a case, the inferred withdrawal before the entry into force should be allowed because the treaty itself is not yet applicable and, secondly, this would constitute a convenient solution in case such a drastic measure is required—usually for serious reasons—without jeopardizing any legal principles or legal security.
131 See Summary of Practice (n 60) 70.
132 For an earlier example, see Agreement for Facilitating the International Circulation of Visual and Auditory Materials of an Educational, Scientific and Cultural Character (adopted 15 July 1949, entered into force 12 August 1954) 197 UNTS 3.
133 For a discussion of the provisional application of treaties, see Chapter 9.
135 (28 July 1994) 1836 UNTS 3, Art 7. See also Summary of Practice (n 60) 71.