So the well-intended states and civil society groups in New York have produced a second draft of the nuclear weapons ban convention. The draft shows real work, and drafters deserve credit for some notable improvements. However, aside from questionable nature of the exercise to begin with, the second draft retains many of the specific problems associated with the first and raises new, serious questions. I remain concerned that the ban risks doing real damage to the nonproliferation and disarmament landscape and hope that further efforts to the draft reduce those to a minimum.  It seems clear the ban is going to happen, so the question is how to ensure it doesn’t do any further damage to the nonproliferation and disarmament landscape.

The biggest concerns in the second draft include:

Article I commits state parties not to “[c]arry out any nuclear weapon test explosion or any other nuclear explosion.” As with the Comprehensive Test Ban Treaty, there may be different interpretations about what this means? Does it include only full-up nuclear weapon tests  with yields, or would it apply to sub-critical tests? Some states such as Egypt have sought to ensure the ban would cover stockpile stewardship efforts and it is not clear on this reading what is in and what is out. What about internal confinement fusion, such as the effort at the National Ignition Facility (NIF)? It is also unclear how the Treaty would verify or enforce such provisions. There is nothing as it relates to verification on testing in the ban draft, either.  Why have it in at all?

Article I also states that parties will prohibit “[a]ny stationing, installation or deployment of any nuclear weapons or other nuclear explosive devices.” This would mean no NATO or US ally that benefits from extended deterrence and who might in a time of conflict want or need to station US weapons on their territory can be a member of the Treaty. This is not a surprise of course, but it does display – as they have said all along – that the drafters’ intent is to delegitimize nuclear possession and operations, including deterrence and reassurance. Of course, this facet of the treaty could be used to put pressure on US extended deterrent allies while having no impact on, for example, Russia’s reliance on tactical nuclear weapons which are stationed close to Europe. It also immediately raises the question about Ukraine. Russia claims the Crimea as Russian territory, but Ukraine and the rest of the world does not recognize their illegal annexation of that areas. Were Russia to deploy nuclear weapons in the Crimea, as they claim is their right, would Ukraine be in non-compliance of the Ban treaty should it sign? This is more of a question than a critique, but one that serious security officials will have to address.

Article 3 tries to address an earlier concern about IAEA safeguards. The initial draft relied on NPT original standards, including the INFCIRC 153 agreement. The new draft seeks to dodge this question by stating that parties will be required to “at a minimum, maintain safeguards obligations undertaken pursuant to international legally binding instruments to which it is party at the time of the entry into force of this Treaty for it, without prejudice to any higher level of standards that it may adopt in the future.”

While this would fall under the do no harm category, it would also mean it does no good. For over 2 decades, responsible states have sought to increase the effectiveness of IAEA safeguards by encouraging the adoption of the Additional Protocol. I believe it reflects poorly on the advocates of the ban to pass on the strongest opportunity to adopt a new global standard for safeguards when they claim to seek a serious step to reduce the nuclear dangers facing the world. How can such a treaty say that anything less than the best, most effective and comprehensive steps to deter any diversion of nuclear materials or facilities is acceptable? The challenge, of course, is that non-nuclear weapon states do not want to adopt any new obligations but are in fact seeking to force nuclear weapon states to do exactly that. It is not offering more for more, but offering nothing for more. This strikes me as the weakest part of the drafting effort to date and I suspect this will be a major liability for the ban should anything less than the AP be adopted in the final text.  Why would the Treaty signatories want to ensure a standard that has been proven inadequate to ensure the IAEA can do their jobs is maintained?

Article 4 of the draft states that “[a]ny State Party that owns, possesses or controls nuclear weapons or other nuclear explosive devices shall immediately remove from operational status its nuclear weapon systems and destroy as soon as possible any nuclear weapons or nuclear explosive devices it owns, possesses or controls. That State Party shall submit, no later than sixty days after the submission of its declaration, a time-bound plan for the verified and irreversible destruction of its nuclear weapons programme to be negotiated with the States Parties or with a competent authority designated by the States Parties.” It further states that the IAEA shall be responsible for verification of the fulfillment of these plans.

This raises massive questions for me. What is operational deployment? Can mobile missiles be put in garrison? Is that enough? What about de-mating ICBM warheads but leaving them otherwise intact and able to redeploy? How to address tactical nuclear weapons or cruise missiles that are not stored near aircraft? These are sticky issues and subject to national definition in most cases. I am not sure anyone would accept the U.S. definition, and I am sure we would not accept, say, Pakistan or Russia’s definitions.

Also, how long can a time-bound plan be and still be time-bound? Can a state lay out a plan for 10 years? Ok, how about two years? Can a state lay out a two-year plan, and then withdraw 3 months short of the deadline and remain in compliance with the agreement?  It is not clear to me that this is a recipe for adding nuclear states to the convention, although it was clearly intended to leave a pathway open to future membership by nuclear weapon states. I have made clear in a variety of fora that when a nuclear weapon state is prepared to disarm, as they eventually should, it will be via an elaborate effort that will involve the highest verification standards and intrusive access by other states and or international inspectors. I would rather have that effort be undertaken as part of the NPT with its near universality than a new convention with lesser credentials, regardless of its good intent.  My opinion is that the Treaty should stick with a simple “get rid of your nuclear weapons and then you can join” approach.  Simpler, less problems, more serious and consistent.  The current plan opens up a big can of nuclear worms.

Moreover, and I admit I may be missing it by not being in New York,but it is not clear that there is any direct requirement for verification of actual disarmament activities under the ban. Article 4 does state that parties who disarm will ask the IAEA to verify the fulfillment of the obligations under the treaty and to further non-diversion of materials. It appears this means that the IAEA would be asked to verify the completeness of a declaration, but not be part of the disarmament activities itself. It is not immediately clear whether the IAEA or anyone else would have a role in verifying dismantlement of materials, nor does this come close to addressing the massive uncertainties associated with verifying dismantlement of nuclear weapons in existing weapon states. For example, what if the material unaccounted for (MUF) in a former nuclear weapon states is 200 kilograms of plutonium? How will those factors be addressed? Who decides if a reactor produced 9000 kilograms of plutonium or 9050? I am not throwing monkey wrenches, but asking how these very challenging issues are to be addressed?  It is also reasonable to ask who will pay for the massive increase in expenses and operations by the IAEA for this task?

To pursue disarmament, states must be prepared to allow intrusive verification, and states who are being asked to take security actions based on those commitments, the verification must be especially stringent. For interim steps over the past few decades, this could be achieved because there were other nuclear options for ensuring security. But as we pursue a nuclear end game, the intrusiveness and requirements go up exponentially. I worry the ban’s efforts here are sadly inadequate.

There are other lesser concerns, but for now I will address only one – withdrawal. Article 18 sets a withdrawal procedure of three months, after notification to the Depository states. This creates a massive risk of escalation should a states seek to use the treaty and its protections to prepare for a nuclear breakout with only three months warning time. Moreover, as the Treaty has no way to ensure that banned weapons-related research might be taking place, a state could produce massive quantities of HEU in metallic form, and produce workable warhead designs and quickly seek to mate the two. While this could take place even today, it seems that such a short withdrawal procedures repeats past mistakes and should be dramatically extended.

I have really benefited from the reporting and insights from my colleagues including ,of course Gaukhar, Andrea, Alicia and Beatrice (the four horsewomen of the ban, anyone).  I am sure they will help me understand where I am just being dense and where, perhaps, I am not.

This article was originally published in Arms Control Wonk