Joshua PollackBrixey-Williams on the legal implications of the nuclear ban

This is a guest post by BASIC Project Leader Sebastian Brixey-Williams.

The legal implications of the nuclear ban:
separating fact from fiction


The nuclear ban treaty’s draft text is out, and the arms control community is beavering away trying to understand what it all means. I observed the first round of UN negotiations in March, to try to understand how the ban treaty will intersect with the existing nonproliferation regime. But vague anxieties that the process could negatively affect existing disarmament and arms control efforts are widespread. Jon Wolfsthal’s Monday blog, to which this – my first Wonk blog – is a response, articulates well many of the concerns that have been raised around the process. So what’s the truth of it? Could the ban treaty force (the more democratic) nuclear-armed or host states to disarm, even if they haven’t signed it? Could it put the US or NATO at a strategic disadvantage compared to Russia? Could the ban treaty undermine or even replace the NPT and the complex, critical institutional architecture that surrounds it?

It’s worth going back to the basic principles of international law. By and large, international law is separated into treaty law or customary law (Article 38, ICJ Charter), and the distinction is critical. Treaties only bind states that sign them – full stop. Accordingly, a ban treaty (like the NPT) will be unable to create any legal obligations for non-signatories, although the messaging around the ban hasn’t always made this clear. By contrast, customary law is binding on everyone. The fear that the norm of ‘nuclear weapons are illegal’ could jump from treaty law to customary law is what’s keeping some people up.

Treaty norms do sometimes break out and become customary law. But the bar is high. The North Sea Continental Shelf Cases (1969) set out two tests to be met: first, customary law must reflect the general practice of the vast majority of states, and second — and slightly paradoxically — states must have a general sense of an obligation to act this way (known as opinio juris). Particular attention must be given to states “specially affected” by that area of law — in our case, those specially-affected states have traditionally been the nuclear weapon states, although as Nina Sofie Pedersen and I noted of the Marshall Islands Case judgment last year, states that experienced nuclear testing now may be recognized as belonging to that category, too. Persistent objection is a further barrier to that jump from treaty to custom. As long as a state maintains its uninterrupted objection to an emerging legal norm, they cannot be bound by it. This may not be pertinent for North Korea, which did not vote on UN General Assembly resolution A/RES/71/258 which mandated the talks, or India and Pakistan, which abstained, but the remaining nuclear-armed states have opposed the process all the way along for this very reason.

In other words, for the ban treaty’s prohibitions to become customary law, they would need to be faithfully adhered to by probably all nuclear-armed states, test-enduring states, and the vast majority of other states, in the conviction that there is a general obligation to do so. At present, that seems a long way off. So what does that mean for the questions that Jon raised under five headings on Monday, some of which were premised on this treaty-custom jump taking place?


Jon is right that the nuclear balance between NATO and Russia in the European sphere could be altered by member states joining the ban treaty and prohibiting forward deployed nuclear weapons on their territories. In reality, this would not happen overnight and alternative arrangements would be made, such as redistributing weapons among other host states or reductions. It follows that a state party in violation of the ban treaty’s provisions could be subject to a legal ruling from the ICJ, but only if they’ve voluntarily accepted ICJ compulsory jurisdiction over matters relating to nuclear weapons. Most nuclear-armed states have not, and others could submit amendments to their optional clause declarations, though I would oppose this on the basis that it would demonstrate an unwillingness to be held to account for failure to engage meaningfully in disarmament. It is not yet clear whether a state could be member of both the ban treaty and NATO. If NATO membership is interpreted as assisting, encouraging or inducing “in any way, anyone to engage in any activity prohibited” by the treaty – as outlined in draft Article 1(f) — then membership of the ban treaty and NATO membership could be mutually exclusive. Further work will need to be done to determine this.

At present, the Netherlands is the only NATO state and nuclear host state that looks set to sign the eventual treaty, as mandated by the Dutch Parliament (apparently against the preferences of the Dutch Defense Ministry). Norway could follow, which under a Labor government was one of the birthplaces of the Humanitarian Initiative; the incumbent Conservative Party is expected to be defeated by the Labor Party in the September general election. These states’ signatures could cause other Alliance states to sit up with interest, and although most NATO states have opposed the ban treaty to date, the democratic process may deliver each state new opportunities to consider joining.

This contemplation could be detrimental to NATO cohesion, and many in the nuclear policy community may argue against it. A steady drip of NATO members to the ban treaty could also weaken the US’ hand in negotiating future bilateral arms control treaties with Russia. Like it or not, however, the security community around NATO must recognize the sovereign right of NATO states like the Netherlands to write defense and foreign policy and the overriding legitimacy of democratic national legislative bodies. The Netherlands leaving NATO would not mark the end of the Alliance, though it may force its members to ask serious introspective questions about the nature of the pact and entail consideration of alternatives to the Nuclear Sharing approach. These issues demand greater attention and research.


Does replicating the language of the CTBT in the ban treaty draft create an uncertain relationship between the two obligations? Not obviously, to me. It seems unlikely that the nuclear-armed states would sign up to the ban treaty without the CTBT first coming into force, so the ban would be more likely to reaffirm it. Moreover, any nuclear-armed state that signed the ban would very probably arrange for CTBTO involvement in the verification of non-testing, since they are best placed to carry out this task. It may be constructive for participating states to lobby for greater integration between the emerging nuclear ban regime and the CTBT regime (just as they have for the IAEA and NPT) in the June and July negotiations.


Will the ban treaty seek to replace the NPT as the center of the nonproliferation regime? That may be the aim of some campaigners, but this is not the current objective of states negotiating the treaty. The draft’s language clearly reflects this, both in the Preamble, which reaffirms that the NPT is the “cornerstone of the international nuclear non-proliferation regime and an essential foundation for the pursuit of nuclear disarmament” and in Article 19, which exists to confirm that “This Convention does not affect the rights and obligations of the States Parties under the Treaty on the Non-Proliferation of Nuclear Weapons.”

Nevertheless, it’s not the case that the NPT allows for nuclear possession and the ban treaty does not, as Jon slightly reductively suggests. The 1996 ICJ Advisory Opinion was clear that (temporarily) possessing nuclear weapons is lawful, hence the pursuit of a ban to expedite the process, but that indefinite possession is unlawful under the NPT and that states need to redouble their disarmament efforts. So instead of seeing the ban treaty as a threat to the NPT, skeptics could reimagine it as an instrument on the shelf that has been pre-negotiated for the fulfillment of the NPT’s explicit requirements. The NPT regime can still govern the interim disarmament process, while the ban treaty could signal more clearly the end point and govern the disarmed world.

The Role of the International Atomic Energy Agency (IAEA)

Would, as Jon fears, the draft treaty undermine progress on IAEA safeguards? I’m not an expert, so I’ll pose a further question, but I’m not convinced. Article 4 of the draft ban treaty outlines that the IAEA shall have “full access to any location or facility associated with a nuclear weapon programme and shall have the right to request access on a case-by-case basis,” which as ever, would only apply to signatories. From a realist nuclear weapon state perspective alone, would it not be desirable to have all ban treaty signatories adhere to the IAEA’s strongest verification standards?

Entry into Force

The draft ban treaty requires only 40 states to bring it into force, which is a pretty low figure for a treaty that’s aiming to have comprehensive membership. I know from speaking to diplomats at the negotiations that the number is partly a reaction against the CTBT’s high entry into force provision. But I think that’s beside the point. I’d suggest that Jon’s disquietude that “40 countries relatively unaffected by global security or nuclear security considerations could be in a position to set a new normative legal standard that would apply to all states” misses two important aspects of the ban treaty’s raison d’être.

First, it cannot be stressed enough that the normative legal standard set if the treaty is agreed will not be custom, but would rather seek to reinvigorate a long process envisaged by Article VI of the NPT itself. Previously, nuclear-armed states have supported collective actions by non-nuclear-armed states to further these goals, such as in the creation of nuclear-weapon-free-zones, and it’s not entirely clear how this process differs qualitatively. Second, these countries are only relatively unaffected in as far as the countries attacked would be catastrophically affected by a nuclear exchange. The ensuing humanitarian and refugee crises, global economic meltdown, environmental contamination, crop failures, and potentially, solar blackouts would impact the vital interests of all countries. They too are held at risk by nuclear possession, and whether the action is deemed wise or not, they are legally and morally within their rights to seek to change the international security architecture. The entry-into-force provision has been designed to reflect that.

Final thoughts

The ban treaty will not create a worldwide prohibition on nuclear weapons any time soon. But as it looks to be inevitable, the best thing the nuclear-armed states can do now is consider how it can be integrated to strengthen existing disarmament and nonproliferation regimes, and minimize any risk of harm. Appearing to ignore or write off the ban treaty, on the basis that global nuclear disarmament is a pipe dream, only strengthens the criticism from states engaging in the ban treaty that the nuclear weapon states have no intention fulfilling their legal obligations to disarm. This perception is damaging to the cohesion of the international community.

Yet there’s a need for further analysis of this topic. The above is a relatively straightforward approach to the legal implications of the ban treaty, given the expected behavior of different groups of states in and around the process. But an altogether different picture might be painted by reading the ban treaty draft, especially the Preamble, as a political document and a piece of jurisprudence that attempts to nudge and craft interpretations of existing rules and principles applicable to the law of armed conflict, humanitarian law, and human rights law. Future research should elaborate on this approach.


  1. krepon (History)

    You have a strong future in this field.
    Best wishes,

    • Sebastian Brixey-Williams (History)

      These are very kind words, Michael – thank you.
      Best wishes,

  2. Harold Jr (History)

    Regarding safeguards, Article 4 only seems to cover states that have eliminated their nuclear weapons. Is it not a concern that safeguards for all other countries in this draft only requires a CSA with the IAEA? There is no mentioning of the additional protocol. A lot of work has gone into this additional protocol, and many states view it as the current standard, but this draft obliterates that view.

  3. Jonah Speaks (History)

    Problem motivating many ban proponents: Nuclear-armed states are not moving fast enough to negotiate mutual nuclear disarmament.

    Proposed solution of draft ban treaty: Any nuclear-armed state that ratifies the ban treaty must give up its nuclear weapons within a short period of time, regardless of what any other nations may be doing. This is essentially a demand that every nuclear-armed nation undertake unilateral nuclear disarmament. It is a political nonstarter, because it does not seriously address the need for mutual nuclear disarmament.

    Possible alternate solution: Allow nuclear-armed nations that join the ban treaty to condition their giving up nuclear arms on the actions of other nations that also have nuclear arms. Propose within the treaty some alternative method for negotiating global nuclear arms reduction, that helps speed up the process of negotiating global nuclear arms reductions.

    For example, the ban treaty could set up a commission to negotiate faster nuclear arms reductions. This could consist of at least 5 nuclear armed states and some additional number of non-nuclear-armed states. The commission could issue binding rules for nuclear reductions by 2/3 vote, provided a majority of both the nuclear-armed and non-nuclear-armed states are in agreement.

  4. manny (History)

    The big problem is that Nuclear weapon states don’t want to give up their weapon, but they want poor countries to do so, if Saddam or gaddaffi had them they would be alive today. something north knows very well.
    India is correct its a discriminatory treaty.

  5. Grant Christopher (History)

    We were wondering if something with a non-negligible probability: Corbyn becoming PM, would significantly change the UK’s position on the Ban treaty.