The application of consensus in multilateral disarmament processes.

Consensus decision-making is a well established practice in nuclear disarmament negotiations, the General Assembly having enshrined its centrality in the final document of the first special session of the General Assembly devoted to disarmament. Consensus has long been viewed as essential to producing effective and implementable agreements in the nuclear field, given the unique responsibility of nuclear-weapon States to disarm. At the same time, profound and persistent divergences of views — especially the entrenched division between nuclear “haves” and “have-nots” — create unique obstacles to consensus, notably on matters directly related to the elimination of nuclear weapons and the Middle East. These challenges are perhaps most visible in the long-standing paralysis in the Conference on Disarmament, which has not carried out substantive negotiations for more than 25 years. Although some substantive discussions have taken place in subsidiary bodies and thematic meetings convened under various presidencies, agreement on a multilateral negotiating mandate has remained elusive. Members have prioritized different objectives — including a fissile material cut-off treaty, a comprehensive nuclear weapons convention, and legally binding negative security assurances — making a balanced programme of work unattainable. Owing to the consensus requirement, a single State’s opposition to launching negotiations on an issue that otherwise enjoys broad support can be sufficient to prolong the stalemate (see quote 2).

 Moreover, the consensus rule can enable States to “hide behind” the one delegation that breaks consensus, creating the impression of isolated opposition even when a wider group may share similar views. The last substantive negotiation in the Conference on Disarmament resulted in the Comprehensive NuclearTest-Ban Treaty, adopted in 1996. However, the adoption itself took place through the General Assembly by vote, underscoring the challenges associated with consensus decision-making in the Conference. After determining that the Treaty’s text received near-universal support, Australia submitted it to the President of the General Assembly, together with a request to resume the fiftieth session to consider it and a draft resolution calling for its adoption. Resolution 50/245 was subsequently adopted, with India voting against. In a letter A/50/1032 to the President of the General Assembly before the vote, the Russian Federation argued that this approach departed from established practice for elaborating multilateral agreements in the Conference on Disarmament. It nonetheless agreed, citing the “exceptional nature of the situation” surrounding the Treaty, while emphasizing that the General Assembly’s decision must be without prejudice to the Conference’s rules of procedure and established practice and must not set a precedent. Similarly, in introducing the draft resolution, Australia stressed that the submission reflected “particular and exceptional circumstances” that should not set a precedent. Many States echoed this point while expressing support for the Conference on Disarmament’s operating procedures. The constraining effects of consensus can also be seen in relation to the Treaty on the Non-Proliferation of Nuclear Weapons (NPT), often described as the “cornerstone” of the nuclear disarmament and non-proliferation regime. In the NPT review process, disagreement by nuclear-weapon States — given their special status under the Treaty — could arguably render an outcome unimplementable and therefore ineffective, even if all other States parties supported it. Still, the prerogative to block consensus is not exclusive to nuclear-weapon States; non-nuclear-weapon States have also refused to join the consensus. Although the rules of procedure of NPT Review Conferences permit voting, this option has never been used to adopt decisions. Under those rules, if all efforts to achieve consensus on a substantive matter have been exhausted, the President — following a 48-hour deferment to facilitate consensus agreement — may conduct a vote. Decisions would then be taken by a two-thirds majority of representatives present and voting, provided that the majority includes at least a majority of the States participating in the Conference.  

The rules distinguish between matters of substance and matters of procedure: procedural decisions can be taken by immediate vote, whereas all efforts must first be exhausted to attain consensus on substantive matters. What constitutes “exhausting all efforts to reach consensus” is not defined, leaving significant room for interpretation and affording the presiding officer discretion in determining when the threshold has been met. In practice, several Review Conferences — including the most recent two — have failed to adopt a substantive outcome (see quote 3).

Since the first Review Conference in 1975, the established practice has been to pursue a consensus final document. The 1995 decision on the NPT’s indefinite extension offered the clearest indication that majority decision-making might be possible. Nevertheless, while the NPT review rules provide a voting fallback and thus do not entrench a single-State de facto “veto” as in the Conference on Disarmament, the prevailing view among States has been that a non-consensual NPT outcome would be politically weak. Even as voting has not been pursued within the NPT framework, efforts to increase pressure on nuclear weapon States to meet their article VI obligations have gained traction outside that process — and, by extension, outside the established practice of consensus (see quotes 4 and 5). 

These efforts culminated with the 2017 adoption, by vote, of the Treaty on the Prohibition of Nuclear Weapons (TPNW), with 122 votes in favour, 1 against and 1 abstention. The TPNW negotiations were intentionally pursued outside consensus-based forums and under procedures that permitted voting, marking a significant shift in nuclear disarmament diplomacy. The General Assembly resolution mandating the negotiating conference (A/Res/71/258) did not specify decision-making procedures, meaning the General Assembly’s rules applied. The Conference subsequently adopted its own rules, explicitly allowing for voting.A noteworthy pattern, however, is that instruments adopted by vote are often implemented through processes that revert to consensus, whether through formal rules or established practice. This has been evident in the meetings of States parties to the TPNW. In the three Meetings of States Parties held since the Treaty’s adoption, all decisions have been adopted by consensus, even though the rules of procedure permit voting. Negotiations on the rules of procedure for the first Meetingof States Parties were contentious: some States sought to avoid any situation in which a single State could block outcomes, while others strongly preferred to retain an option that would effectively allow a veto. The adopted rules provide that if the President determines that all efforts to reach consensus have been exhausted — or to ensure that a decision can be reached before the end of the session — decisions on matters of substance may be taken by a two-thirds majority of the States parties present and voting. This approach broadly mirrors NPT practice, but without the waiting period. Although TPNW States parties are often considered broadly like-minded, differences remain — for example, on the proposal for a victims’ assistance trust fund. Even so, consensus continues to be the preferred approach and the standard towards which States parties generally strive. Nonetheless, the negotiation and adoption of the TPNW — criticized by the nuclear-weapon and nuclear-aligned States as lacking practicality — marked a significant turning point, both in relation to consensus decision-making and in the empowerment of non-nuclear weapon States within the wider nuclear disarmament debate.

 

Multilateral negotiations on conventional weapons differ from the nuclear field in one fundamental respect: all States possess conventional weapons and ammunition of various types. As a result, there is no inherent hierarchy comparable to that between nuclear-weapon and non-nuclear-weapon States, even if related discussions sometimes refer to “militarily significant” or “militarily advanced” States. Across conventional weapons discussions, consensus decision-making remains central, reflecting a rationale similar to that in nuclear disarmament: in matters of international security, consensus is often viewed as necessary to confer legitimacy and ensure buy-in. While regulation is frequently the objective, States have also pursued the prohibition of certain conventional weapons based on their inhumane or indiscriminate effects. The Programme of Action to Prevent,Combat and Eradicate the Illicit Trade inSmall Arms and Light Weapons in All ItsAspects is an instructive example. Like the NPT, its rules of procedure for follow-up meetings allow voting, yet voting was not used for many years. In 2018, however, sharp divergences over the inclusion of ammunition broke the long-standing practice of consensus and introduced a precedent for voting, although States subsequently returned to consensus adoption of outcome documents. In 2001, the United Nations Conference on the Illicit Trade in Small Arms and Light Weapons in All Its Aspects, initially envisaged as negotiating a legally binding instrument, adopted by consensus a political framework comprising normative commitments at national, regional and global levels. At the first Review Conference in 2006, disagreement over language related to civilian possession of small arms and light weapons and transfers to non-State actors prevented the adoption of an outcome document. Even then, the option of voting was not pursued. After the second Review Conference adopted a substantive outcome document in 2012, States again faced difficult negotiations in 2018, culminating in a “unanimous vote” on the document as a whole. Explicit references to small-arms ammunition — and, to a lesser extent, provisions concerning unauthorized recipients and terrorists — were among the contentious issues. Because two States objected to the paragraphs referencing ammunition, those paragraphs were put to a vote, marking a departure from the long-held practice of consensus decision-making on substantive matters under the Programme of Action. Although the outcome document was ultimately adopted as a whole by a unanimous recorded vote, with 98 votes in favour to none against, the fact that contested paragraphs were put to a vote meant it was not adopted by consensus in the strictest sense. Similar dynamics emerged in the Arms Trade Treaty process. Although negotiations took place under procedures requiring consensus, States ultimately adopted the Treaty through a vote in the United Nations General Assembly after consensus proved unattainable. Decision-making procedures were a point of contention from the outset, with many States insisting that substantive decisions be taken by consensus. States first convened to negotiate the Treaty in July 2012 and reconvened in March 2013 but were unable to agree on a final text by consensus. At the final diplomatic conference, three States lodged formal objections, prompting debate over whether overwhelming support could be interpreted as “consensus” despite explicit opposition by a small number of delegations. Some States strongly rejected that interpretation. The President ruled that consensus had not been achieved and that the Treaty could therefore not be adopted in that form. Supporters subsequently brought the text to the General Assembly plenary, where it was adopted by vote.

At the first Conference of States Parties to the Arms Trade Treaty, held in Cancún in August 2015, States parties debated rules of procedure that were ultimately adopted by consensus. The discussions revealed divergent views on whether consensus should be treated as an absolute or whether a voting fallback should exist. Many delegations expressed concern about the unsuccessful attempts at consensus decision-making during the treaty negotiations. Under the adopted rules, States would strive for consensus on matters of substance and those with financial implications, but if all efforts fail, decisions could be taken by a two-thirds majority of States parties present and voting. For procedural matters, the rules similarly prioritize consensus but allow decisions by a simple majority if consensus cannot be reached.26 Although not subsidiary to the General Assembly, the Convention on Prohibitions or Restrictions on the Use of Certain Conventional Weapons Which May Be Deemed to Be Excessively Injurious or to Have Indiscriminate Effects (CCW) is also illustrative. Its consensus based decision-making has facilitated participation by a wide range of States, including those sometimes characterized as militarily significant, but it has also made substantive progress difficult. For example, discussions on lethal autonomous weapons systems within the CCW Group of Governmental Experts have included engagement by States sceptical of the need for new international rules. At the same time, the difficulties of consensus and the CCW’s careful balancing of military and humanitarian considerations have contributed to the pursuit of arms control instruments outside the United Nations, including prohibitions on cluster munitions and anti-personnel landmines. Delays in advancing a CCW protocol on cluster munitions helped catalyse a group of like-minded States and the launch of the Oslo Process (see box 5). Similarly, when the first Review Conference of the CCW, in 1980, failed to reach consensus on comprehensive measures addressinganti-personnel landmines, Canada announced that it would host the

 International Strategy Conference “Towardsa Global Ban on Anti-personnel Mines”, helping set the stage for what became the Ottawa Process. More recently, negotiations on the Global Framework for Through-life Conventional Ammunition Management demonstrated novel approaches to protecting progress in the face of opposition by a very small number of States. The Russian Federation and Belarus dissociated themselves from the Global Framework and the recommendations of its Open-ended Working Group, arguing that the outcome document was insufficiently balanced and did not adequately reflect their priorities — particularly regarding the characterization of the Global Framework as a mutual understanding on guiding principles and voluntary guidelines. The Russian Federation also objected to references to non-universal instruments and to gender considerations (see quote 6). An initial Russian proposal to record in the procedural report that there was no consensus on the outcome document met strong opposition and was ultimately withdrawn after lengthy negotiations. The procedural report, therefore, stated that the Working Group adopted its final report “without a vote” rather than “by consensus”.