On builders and blockers

ILPI Publications > ILPI-UNIDIR Joint Papers

States have different roles to play to complete the nuclear disarmament puzzle

By Torbjørn Graff Hugo
22 April 2015
    • A focus on ‘building blocks’ invites an analysis of roles and responsibilities for the achievement and maintenance of a world without nuclear weapons—all states need not be involved in all discussions on all aspects of nuclear disarmament.
    • Each building block should be carved out by a critical mass of interested ‘implementers’, in a format that does not allow ‘blockers’ to prevent progress
    • If states can recognize that they must play different roles to complete different parts of the puzzle, they may see progress sooner on building progress towards the total elimination of all nuclear weapons.
ILPI–UNIDIR NPT Review Conference Series Paper No. 4 of 5

Download full article as PDF


According to a 2014 NPT working paper submitted by a group of mainly nuclear umbrella states, ‘a focus on “building blocks” can complement the pursuit of a “step by step” approach’ to nuclear disarmament.[1] Granted, it is not entirely clear how this complementation would take place, especially since the actual building blocks listed in the working paper seem synonymous with ‘steps’ commonly regarded as included in the step-by-step approach. These elements include the negotiation of a fissile material treaty, entry-into-force of the Comprehensive Test-Ban Treaty (CTBT), legally binding security assurances, a return to substantive work in the Conference on Disarmament (CD), strengthening the International Atomic Energy Agency (IAEA) nuclear safeguards system.

Nevertheless, the building block approach is novel. This is partly because the new metaphor is more relaxed in terms of sequencing than the original (at least rhetorically speaking), but mainly because the working paper introduces the possibility of a certain division of labour when it comes to the construction of the different ‘blocks’. Different measures could presumably be designed and implemented by different configurations of states. While the maintenance of a nuclear-weapon-free world will need to be a multilateral endeavour, as argued in the working paper, the mutually reinforcing building blocks necessary for effective disarmament need not be.

What the working paper lacks is a more profound treatment of how the building blocks and the different levels of ‘lateralism’ are linked. For some of the building blocks the answer is self-evident, but for others it is not. It raises two key questions: what determines the best approach to shaping each of the different building blocks? And which groups of states are best placed to carve them out?

The building of blocks

The building blocks, although not explicitly defined in the working paper, can be understood as practical measures expected to bring the world closer to the goal of full nuclear disarmament. Examples of past building blocks range from the IAEA safeguards system to the Nuclear Non-Proliferation Treaty (NPT) itself.

Compared to the steps of the ‘step by step’ approach, the building blocks also invoke a certain creative, problem-solving spirit—implying that the answer to the nuclear disarmament puzzle could be found by shuffling the building blocks around long enough and trying different angles. That also makes the sequencing of the blocks less important, as it is fine to have movement on several fronts at the same time.

Compared to the steps of the ‘step by step’ approach, the building blocks also invoke a certain creative, problem- solving spirit

In the working paper, a distinction is drawn between multilateral building blocks on the one hand, and ‘unilateral, bilateral and plurilateral actions’ on the other.[2] This dichotomy is neither clear-cut nor particularly useful. For example, the list of non-multilateral building blocks includes a call for ‘promoting plurilateral or multilateral nuclear reduction negotiations’ (emphasis added), as well for ‘promoting disarmament and non-proliferation education’.[3] On the multilateral list, by contrast, you will find calls for ‘all States possessing nuclear weapons’ to declare ‘moratoriums on the production of fissile material for nuclear weapons purposes’, which would appear to be a national-level measure in the first instance.[4]

An alternative, and perhaps more productive way to categorize the different building blocks, is by issue. Some blocks focus on ‘nuclear testing’, others on ‘fissile material’, and so on. Based on the list of building blocks presented in the working paper, combined with other practical measures called for in the broader multilateral discourse, a non-exhaustive list of issues could include the following: fissile material, testing, proliferation, security assurances, break-out / cheating, nuclear terrorism, limitation geographical of scope, nuclear-weapon-free zones, disarmament, security doctrines, and prohibition.

For each of these categories, at least one building block has either been implemented or proposed. For example, the category of ‘testing’ includes both the Partial Test-Ban Treaty (PTBT) and the CTBT. It would also cover unilateral measures such as moratoriums declared by the nuclear-weapon states.

Each category can in principle be reduced to a simple idea. Table 1 of this paper lists each of the categories with an accompanying formulation of the core idea, as well as existing building blocks in each category.


ILPI-UNDIR NPT 4 - Table-1

Each building block is on some level designed to solve a problem—or at least part of a problem. As such, the success of a given building block ought to be measured by the effectiveness with which it serves its purpose. Did the NPT stop the proliferation of nuclear weapons? Not completely, but the NPT has been a relatively effective tool in terms of plugging the proverbial hole in the nuclear weapons dike. Some have (gu)estimated that without the NPT, the number of nuclear-armed states in the world would be three or four times higher than it is today.[5]

Identify the builders

How did the NPT achieve this level of success? One important lesson is that it is important to make sure a critical mass of key implementers take part in the shaping of the building block. Some readers may instinctively want to replace the word ‘implementers’ with ‘stakeholders’ in that sentence. But there is an important difference between the two. A stakeholder is someone with a stake in the outcome; an implementer is someone with a concrete role in the effective implementation of the regime. Usually an implementer is of course also a stakeholder, but it is not always the case the other way around.

This does not mean that stakeholders who do not have a concrete role to play in the implementation of a given regime are less interested in achieving results. On the contrary, these ‘non-implementers’ or often the most keen type of regime-builders. Nor does it mean that they are irrelevant. The point is that their role is primarily one of normative consolidation, which is an important element in the universalization process for most regimes. But it is not necessarily a requirement for its construction.

For the success of a given building block—that is, the ability to generate the normative pull necessary to put the regime on a long-term path towards full adherence— ‘implementers’ are arguably much more important to have involved from the beginning than ‘consolidators’.

Consequently, if a certain category of states (e.g. the nuclear-weapon-free zone states) is considered not to have a meaningful role in the implementation of a given building block (e.g. reducing the role of nuclear weapons in military doctrines), it may be more productive if the negotiations on this particular building block were to take place without their participation.

In the context of the NPT, the purpose of the treaty—to prevent proliferation of nuclear weapons—required the participation of both nuclear-armed and non-nuclear-armed states. The basic prophylactic provisions of the NPT stipulate that it is prohibited to transfer and to receive nuclear weapons. Both the states that have nuclear weapons and the states that do not have a role to play therefore in making sure the objective of the treaty is achieved. This core reciprocity means that the treaty would have been much less effective if it had been negotiated only between the nuclear-armed states of the time Likewise, if the non-nuclear-weapon states (NNWS) had negotiated the NPT only among themselves, the treaty would have been considerably weaker. The negotiation process did not, however, require the participation of all NNWS—nor of all the nuclear-armed states (NAS). A critical mass from each group proved to be sufficient.

Avoid the right to veto

A further lesson to be drawn from the list of existing building blocks is that if a critical mass of key implementers is interested in moving forward, a format must be chosen that allows for this to happen. In plain terms: ‘blockers’ should not be allowed to veto attempts at making progress. This means, for example, that as long as the CD continues to equate ‘consensus’ with ‘veto’, the forum may not be conducive to the negotiation of further building blocks. After a deadlock of nearly two decades, even patient observers have begun to despair of the chronic unproductiveness of that body and have turned their attention elsewhere.[6]

For some supporters of the principle of multilateralism in international diplomacy, the United Nations General Assembly (UNGA) is the preferred alternative to negotiations in the CD. With the adoption of the Arms Trade Treaty (ATT) in 2013, the UNGA showed that it was possible to achieve real results while remaining within the symbolic walls of the United Nations system. The process of getting there, however, is not something the co-sponsors of the annual UNGA-resolution on the promotion of multilateralism should speak too boldly about.

This is because, in the end, the decision-making procedures of the ATT process turned out to be as schizophrenic as those of the CTBT: consensus-based until the bitter end, after which a simple majority in the UNGA would suffice. As a precedent, the ATT is in fact highly problematic. In the long run, the logic of the ‘UNGA fall-back option’ means that any negotiation process in practice will be seen to have a simple majority adoption threshold, which is a much weaker basis for developing international law than the two-thirds majority requirement of the Vienna Convention on the Law of Treaties.[7]

The challenge of choosing the most favourable format for the construction of the building blocks is not first and foremost a question of venue, however. Nor is it about the simple versus qualified majority threshold. A two-thirds majority might be preferable to a simple majority in terms of legitimacy, but if the key catalyst for progress is a critical mass of interested states, then the only procedural requirement necessary is to stay clear of any form of veto privileges.

The search for critical mass

Finally, this begs the question of how to define ‘critical mass’. In nuclear physics, critical mass refers to the ‘minimum amount of concentrated fissionable material required to sustain a chain reaction’.[8] In the construction of building blocks, it can be understood as the minimum number of states required to credibly construct a given measure for disarmament. The credibility aspect is central, since that largely determines whether the building block will gain momentum towards full adherence.

Importantly, critical mass in physics can be calculated in advance—critical mass in the crafting of building blocks cannot. It is all a game of perception, until hindsight can decide whether or not it worked. Also, the actual number of states is in many ways less salient than the credibility they can muster. In bilateral arms negotiations between the USA and Russia, two states would constitute a ‘credible’ amount—a critical mass. For the negotiation of the NPT, the number was 18.[9] Could the NPT have been negotiated with only two states? Almost certainly not. Would it have been enough with 10 or 12 states? We will never know. Critical mass is about the perception of credibility, not the counting of states.

In sum, the lessons from some of the existing building blocks suggest that when carving out new practical measures for disarmament, two elements are needed: 1) a critical mass of interested implementers, and 2) a format that does not allow ‘blockers’ to prevent progress.

Who should carve out the building blocks?

With this in mind, how should states approach the list of proposed building blocks? Table 2 shows an overview of proposed practical measures, sorted along the same categories as in Table 1.

ILPI-UNDIR NPT 4 - Table-2


Fissile material

On top of the list is a treaty banning fissile material for weapons, usually referred to as a Fissile Material Cut-off Treaty (FMCT). The idea of a prohibition on the production of fissile material has been on the agenda of the CD for years, but due to a fundamental disagreement between some of the key implementers about the purpose of the building block, it is unclear when negotiations on such a treaty will begin—if at all.

How could this be solved? According to the principles discussed above, the FMCT should first of all be negotiated primarily between the states that have nuclear weapons, since these states are the key implementers of the treaty. For the vast majority of the United Nations member states (or CD members), weapons-grade fissile material is already prohibited under the NPT. And to the extent other states have access to such materials, it is subject to strict IAEA safeguards.

Secondly, if a critical mass of key implementers are interested in moving forward on this treaty, they should consider moving the discussion to a format where this could realistically happen—which in practice would mean taking it out of the CD. Everyone else—the NNWS—should simply step back and let the nuclear-armed states negotiate a treaty on fissile material on a plurilateral basis.[10]

Security assurances

A similar conclusion could be drawn with respect to security assurances. The demand from the NNWS is that the five nuclear-weapon states recognized by the NPT (the NPT5) agree to legal obligations not to use nuclear weapons against NNWS, possibly in the form of a treaty. Since only states with nuclear weapons can meaningfully undertake not to use them against NNWS, one could argue that legally binding security assurances could theoretically be given without NNWS taking part in the negotiations. On the other hand, the concept of security assurances does have a certain element of reciprocity built into it—the guarantees must be extended to someone. Depending on how the instrument is designed, the NNWS may therefore also have a certain role to play as implementers. However, what distinguishes the security assurances from other non-proliferation measures—including the NPT—is that they do not require any legal obligations to be placed on states that do not have nuclear weapons. The NNWS are simply passive recipients of the legal commitments of the NAS. They are clearly ‘consolidators’.

In practical terms, such a treaty would both reduce incentives for proliferation and consolidate the position of the NAS. Considering the non-proliferation effect of such an instrument, it is surprising that the NAS have not already come together and negotiated it. One reason is perhaps that it would require the NPT5 to recognize the nuclear-armed status of the non-NPT members.

De-alerting and dismantlement

The proposed building blocks in the category of ‘de-alerting and dismantlement’ are also primarily a job for the NAS. The actual removal and destruction of nuclear warheads can only be done by the states that have the weapons (though third parties could possibly have an implementer role in verifying the dismantlement, the feasibility of which United Kingdom and Norway have been exploring in recent years). Between the NAS, dismantlement negotiations could either be bilateral (as with the New START) or plurilateral (involving all or most of the NAS). The dismantlement building blocks could even be crafted the way it was done by South Africa, the only country known to have unilaterally eliminated an entire existing nuclear arsenal. For the NNWS, what matters is that the disarmament obligations of the NAS are implemented, not how this is done.

For the NNWS, what matters is that the disarmament obligations of the NAS are implemented, not how this is done.

If all this should be left to the NAS, what is there to do for the NNWS? Should they just sit back and relax? As tempting as that may sound, a number of the categories in Table 2 require all states to play a role. This includes the continued strengthening of the non-proliferation regime, the work to prevent nuclear terrorism (e.g. universalizing the International Convention on the Suppression of Acts of Terrorism, ICSANT), and the further limitation of geographic scope for nuclear weapons (e.g. by promoting work on a treaty on the Prevention of Arms Race in Outer Space, PAROS).

Secondly, the NNWS that are members of nuclear-armed security alliances each have an implementer responsibility when it comes to reducing the role of nuclear weapons in security doctrines (the penultimate category in the tables). The most important building block in this regard would be the amendment of NATO’s strategic concept, which many members apparently interpret to mean that they must refrain from promoting any actual disarmament efforts, since that in practice would undermine their own defence doctrine.

Most importantly, however, there is a set of building blocks that actually require the NNWS more than the NAS. As can be seen from Table 2, this includes three particular categories: breakout / cheating, NWFZs, and prohibition.


The idea of the ‘breakout’ category is that nuclear weapons can only be abolished if states can be confident that no one is cheating. Confidence-building and compliance measures to enforce the prohibition norm are therefore critical. The IAEA safeguards system is key to achieving this, but it has long been argued that there is need for stronger measures than the NPT requires—providing greater assurance about both declared and possible undeclared nuclear activities. Specifically, there have been calls for making the additional protocol (AP) of the IAEA safeguards agreement compulsory. The Additional Protocol, which is a legal document concluded between a state and the IAEA, grants the IAEA complementary inspection authority to that required by the NPT.

Many see this type of assurance against breakout as critical in order to achieve the confidence necessary for the complete elimination of nuclear weapons. Still, more than a third of the NPT states parties have yet to conclude an additional protocol to their IAEA safeguards agreements.[11] One reason for this may be that many NNWS are reluctant to apply additional obligations while they perceive a lack of progress on nuclear disarmament by the NPT nuclear-weapon states. Such a tit-for-tat mentality is unhelpful, however. Ideally, APs should be seen as of unambiguous value irrespective of the actions (or inaction) of the nuclear-armed states.

To make the AP compulsory for all NNWS would probably require negotiation of a legally binding instrument that states would have to ratify. Importantly, the key implementers of such a treaty would be NNWS, and in line with the argument presented in this paper, it should therefore be the responsibility of the NNWS to negotiate and put into place such an agreement. The NAS would in principle not need to participate in the negotiations at all.

Nuclear-weapon-free zones

The second category of building blocks where NNWS are the key implementers is on NWFZs. The purpose of the regional prohibition treaties is to go further than the NPT and commit the states of a particular geographical region to a comprehensive prohibition of nuclear weapons. The NWFZ treaties are not disarmament treaties. They do not contain timelines and reduction targets. In fact, with the sole exception of the African zone (Pelindaba), the existing NWFZ treaties do not even include a requirement to dismantle nuclear stockpiles.

The only reason NWS have had roles at all with regards to these building blocks is because of the curious decision to add protocols on negative security assurances to all the zone treaties. Perhaps the intention was to lure the NWS into a legal commitment on negative security assurance that they otherwise would not have agreed to. The result, however, has been that few of the NWFZ have been recognized by all of the NWS.

For the proposed zone in the Middle East, the implication of all this would be that in order to get a treaty negotiated, a critical mass of key implementers—namely the NNWS in the region—should join together in a format that allows the states most interested in progress to move forward, even if that means leaving some states behind. A WMD-free zone in the Middle East that includes a critical mass of key implementers is clearly better than no zone at all. And contrary to what some may think, it would not simply let the others off the hook. The treaty of Tlatelolco is an important example in this regard. If the states in Latin America and the Caribbean had waited for all the states in the region to be ready, it would have taken more than 30 years to get started on the negotiations of the Tlatelolco treaty.[12]


The third and final building block where NNWS can be considered the key implementers is the one called ‘prohibition’. The core idea is that if nuclear weapons were explicitly prohibited for all states, the incentives to maintain nuclear arsenals would be reduced, which in turn would provide essential impetus for further disarmament measures. The logic is the same as for the NWFZ, namely to negotiate a treaty that goes beyond the obligations of the NPT and the CTBT. But contrary to the NWFZs, a prohibition treaty would be open to all states.

The basic idea for this building block is to put in place a comprehensive prohibition regime for nuclear weapons—or to fill the ‘legal gap’ in the prohibition of all weapons of mass destruction (biological and chemical weapons are already banned). In view of this, such a treaty would most likely not contain any disarmament provisions or timelines. In fact, one could argue that even if a number of NAS decided to join the negotiations, the treaty should still not contain specific disarmament obligations. For that is not the purpose of the building block, and it would require a different set of key implementers. The dismantlement of nuclear weapons should instead be left for the NAS to sort out between themselves—unilaterally, bilaterally or plurilaterally.

The role of the NNWS is simply to create the incentives that will make the NAS decide to move in that direction.

The core idea is that if nuclear weapons were explicitly prohibited for all states, the incentives to maintain nuclear arsenals would be reduced, which in turn would provide essential impetus for further disarmament measures.

While the building blocks presented in this paper are divided according to issue, there is of course nothing that prevents the key implementers from combining across different categories. If, for example, the NAS concluded that fissile material, security assurances, de-alerting, and dismantlement could all be dealt with in one big potpourri of a treaty, that would probably be more efficient than to negotiate three or four separate instruments.

Similarly, if the NNWS decide that ‘breakout / cheating’ and ‘prohibition’ could be mixed together, the resulting instrument would probably benefit from the merger. That would in fact make it a ‘maintenance’ instrument, which eventually—should it one day achieve full adherence—would serve as the new cornerstone of the legal framework regulating nuclear weapons internationally.


The focus on building blocks introduced in NPT Working Paper 23 from 2014 is a welcome contribution to the discussion on how to move the world closer to zero nuclear weapons. As a supplement to the more established step-by-step approach it adds flexibility in terms of sequencing of the different practical measures for disarmament. More importantly, it invites an analysis of the roles and responsibilities associated with the crafting and implementation of the different building blocks, basically suggesting that all states need not necessarily be involved in all discussions on all aspects of nuclear disarmament.

The argument made in this paper is that for each building block a set of ‘key implementers’ can be identified, and that in order to ensure that a building block becomes effective and relevant, you basically need two things: 1) a critical mass of interested implementers, and 2) a format that does not allow ‘blockers’ to prevent progress. The existing building blocks identified in the working paper provide important lessons in this regard, and both the NPT and the CTBT serve to illustrate that in the construction of new building blocks implementers are more important than consolidators. Secondly, the choice of format is key to securing progress.

By analysing the basic idea and purpose of each category of building blocks, it is possible to identify the key implementers for each measure. From the total of eleven categories listed, the NWS are seen as key implementers for five of them (dismantlement, fissile material, security assurances, security doctrines, and testing), while the NNWS are considered key implementers for three categories (breakout/cheating, NWFZ, and prohibition). The rest (proliferation, nuclear terrorism, and limitation of scope) are identified as a shared responsibility among all states. In addition, nuclear umbrella states are singled out as a particular group of key implementers when it comes to reducing the role of nuclear weapons in security doctrines.

This leads to the following conclusion: The sooner states accept that they can play different roles to complete different parts of the nuclear disarmament puzzle, the sooner we may see progress on the construction and implementation of the building blocks needed to nudge us slowly but surely towards a world without any nuclear weapons.


Torbjørn Graff Hugo is project coordinator for the WMD project at the International Law and Policy Institute.


[1] Working Paper 23 of April 15, 2014 (NPT/CONF.2015/PC.III/WP.23), paragraph 2. Submitted by Australia, Belgium, Canada, Colombia, Czech Republic, Denmark, Estonia, Finland, Germany, Hungary, Italy, Japan, Lithuania, Netherlands, Poland, Portugal, Slovakia, Spain, Sweden and Ukraine.

[2] Ibid, para 6.

[3] Ibid, para 6.

[4] Ibid, para 5.

[5] See e.g. R. Timerbaev, ‘What Next for the NPT?’, International Atomic Energy Agency Bulletin 46/2, March 2005: https://www.iaea.org/sites/default/files/publications/magazines/bulletin/bull46-2/46203590407.pdf.

[6] See: http://www.reachingcriticalwill.org/news/latest-news/9559-wilpf-statement-to-the-conference-on-disarmament-on-international-women-s-day-2015.

[7] See: https://treaties.un.org/doc/Publication/UNTS/Volume%201155/volume-1155-I-18232-English.pdf.

[8] See: http://www.nti.org/glossary.

[9] The NPT was negotiated by the Eighteen Nation Disarmament Committee (ENDC), a forerunner to the current CD.

[10] A United Nations-mandated Group of Governmental Experts is to present its report in 2015 containing recommendations on aspects of a treaty banning fissile material for nuclear weapons. What it has to say on this question will be interesting: http://www.unog.ch/80256EE600585943/(httpPages)/B8A3B48A3FB7185EC1257B280045DBE3?OpenDocument.

[11] See: http://www.iaea.org/safeguards/documents/AP_status_list.pdf.

[12] See Beamont and Rubinsky (2012), ‘An Introduction to the Issue of Nuclear Weapons in Latin America and the Caribbean’, International Law and Policy Institute (ILPI): http://nwp.ilpi.org/?p=1851.