Talk on the FMCT

Since I'm not very good at updating my own blog, I've decided to post my talk to the Moscow Nonproliferation Conference. This was a very good meeting, which featured many known faces in the community. I also enjoyed the presentations of my co-panelists, especially Annette Schaper's broad and objective presentation of her work, spanning more than a decade, and R. Rajaraman's no-nonsense assessment of where South Asian countries stand in the debate.

Introduction
Yesterday, we heard many speakers stress the importance of the FMCT. We know that many hundreds of tons of weapons-usable fissile material remain in storage in the United States and Russia, and large quantities also remain in China, France and the United Kingdom. All five nuclear weapon states have ended production within a non-binding moratorium. As we heard in Professor Rajaraman's excellent presentation, however, fissile material production is continuing in South Asia. There, India and Pakistan are trying to define how much material they think they would need to preserve their own deterrence. Officials in those countries often view Fissile Material Control Measures with suspicion. Many in that region argue that too much transparency in material holdings could expose military weaknesses. They also say that an imbalance in stocks could tilt the brittle military balance on the subcontinent. It is likely that India and Pakistan would choose to produce a comfortable cushion of material before signing any control regime.

A ban on production would not reduce already worked-up stocks of material. It would simply preserve the status quo, where it exists, in fissile material balances. The ban would have a limited relevance for general nuclear disarmament. This is not to say that a ban would be pointless. After all, parties cannot produce new material. Stockpiles can only legitimately go down.

The case for a verified treaty
Most analysts, academics and states agree that it is preferable for this treaty to have a verification regime. A minority stubbornly argue that verification is not necessary. They may say that the moratorium has been around for two decades, without a single breach. They may say that verification is impossible. Failing all other arguments, they may say that it is too expensive.

This argument understates the verification regimes significance for some non-nuclear weapon states. For them, verification measures would bring the nuclear-weapon states' obligations closer to those of non-nuclear weapon states, thus making the nuclear non-proliferation regime less discriminatory. In addition, those who oppose a verified ban on production assumes the nuclear weapon states trust each other well enough to forgo verification of compliance, and this is arguably an overstatement.

What should the verification regime do? In a nutshell, the objective is to settle whether a party is compliant with a defined duty that it has undertaken. The certainty with which the verification regime can make this determination depends on the clarity and precision of the undertaking itself. It will be easier to verify a ban on activities – such as producing highly enriched uranium or processing spent fuel – than it will be to verify intent.

From that perspective, a ban on fissile material production for weapons purposes looks unappealing. After all, what is to stop a state from saying, no matter how unreasonable it sounds, that it is producing highly enriched uranium for peaceful purposes? Perhaps the state will argue that it needs weapons grade material to, say, fuel a research reactor?

By contrast, it would be easier to decide whether a state is producing banned materials. The treaty could then allow for certain exceptions to the rule, provided certain conditions are fulfilled and that the exception is carried out responsibly, and in complete transparency. I suspect, given the vastness of military stockpiles, that such exceptions would be rare.

The verification regime is a product of the scope of the treaty, not the other way around. The fundamental question whatto verify needs to be addressed before questions such as where to verify, when to verify, how to verify and who should verify can be answered.

Less than a month ago, Pakistani Ambassador Zamir Akram said, and I quote: "It is imperative that we have a-priori knowledge of what we are seeking to negotiate – an FMCT which is a non-proliferation measure or an FMT which is a disarmament initiative. Once this is clear, then it will be possible to identify positions on relevant issues of definition, scope, stocks, verification and entry into force". End of quote. This is certainly a logical statement, and I agree fully with it. However, I fail to see his questions cannot be solved through negotiations.

Full versus limited scope safeguards
Having said that, I believe that most building blocks to a cut-off regime is in existence already. The safeguards system can be adapted for the task, and the International Atomic Energy Agency is high up on the shortlist of organizations that may be tasked to do the job. But what task would that be? As implied earlier, this is still up for debate.

Some argue the solution is as simple as applying full scope safeguards on the nuclear weapon states. This involves placing verification measures on most nuclear facilities in the country, essentially tracking nuclear material flows from uranium conversion to final disposition. This is beyond doubt a workable solution. The unresolved question is whether it is a necessary one?

After all, while the parties to the cut-off would assume a commitment similar to that for non-nuclear weapon states under the NPT, several important differences remain. A fissile material cut-off would not alter the nuclear weapon states' right to manufacture, store and deploy nuclear arms. Large amounts of legitimate fissile material will be present on the territories of the state parties. There is little point in checking this pool of material, since the nuclear weapon state is free to make use of it as it sees fit.

In addition, is a full scope approach economically viable? It is possible to safeguard reactors, spent fuel ponds, conversion activities, heavy water production and fuel fabrication facilities. And, it is possible to build a picture of the total production capacity of an individual state (through declarations on associated infrastructure such as uranium mining and milling). This is likely to be a human and capital-intensive exercise. By including nuclear weapon states, with their large nuclear fuel cycles, conservative budget estimates suggest a doubling, or perhaps even a tripling, of the present IAEA safeguards budget.

In addition, given that no one knows how much material has been produced in the nuclear weapon states, it is likely that full scope safeguards would yield little utility on the margin.

In my mind, it therefore makes more sense to focus the verification effort on nuclear material flows directly relevant for weapons production. That is uranium enrichment, certain conversion activities and spent fuel reprocessing. Flows of freshly produced weapons usable material, such as highly enriched uranium, would still need to be monitored. But since stocks are so large, it is unlikely that such material will be produced. It also makes sense to start the accountancy work from a new baseline.

I think it is granted that such an approach will not be as robust as full-scope safeguards. In addition, I do not argue that fissile material declared in excess of military needs should not be safeguarded. On the contrary, it is possible to consider such declarations as new production of weapons usable material, and thus bring them into the system of accountancy and control.

The case for challenge inspections
One big problem will always remain. Undeclared nuclear fuel cycles cannot be detected by routine inspections, and the use of remote sensors able to detect various plant signatures on the territory of state parties are likely to be unpractical. This problem can be rectified through requiring safeguards on larger swathes of the nuclear fuel cycle. That is, opting for a full-scope regime. It can also be solved by introducing a challenge or special inspection system to handle suspicions of non-compliance.

On the one hand, while the full scope regime will incur increased costs, it will increase the robustness of the system. A challenge inspection regime, on the other hand, has weaknesses. Normally, some evidence would be required pointing towards non-compliance to get the necessary political support for such inspections. This evidence alone creates a presumption of guilt. Related to that, if the challenge inspection fails to produce any evidence because of the absence of undeclared activities, it is possible that the requesting state will not be convinced by the outcome.

If so, expect classic logical fallacies. Such as a state is non-compliant because it hasn't proven its compliance. Or that a state is compliant because no non-compliance has been proven. These matters cannot be solved technically. They can only be solved through process, through thinking about the burden of proof.

Conclusion
Few weapon states would consider a treaty that encompasses cuts in military stocks. Rather, the idea is to codify the twenty-year-old fissile material production moratorium between the five NPT nuclear weapon states. The idea is also to introduce a prohibition on production in de-facto nuclear weapon states in South Asia and the Far East.

The NPT nuclear weapon states would have little incentive to place their stocks on the negotiation table. After all, it was the build-up of stocks that made them consider the treaty in the first place. It is telling, from that perspective, that smaller nuclear weapon states, who do not consider themselves to be sufficiently stocked, places emphasis on the build-up of fissile material.

To quote Ambassador Akram: "While we do not wish to compete in terms of weapon for weapon as our strategic doctrine is based on minimum credible deterrence, we must ensure that the asymmetry does not erode the credibility of our deterrence". End of quote. Pakistan's position, in a nutshell, is that either it stocks up or India comes down.

Past production of fissile material is also significant from another angle. Given the large uncertainties in historical production in some weapon states, it will be near impossible to establish baseline inventories of nuclear material. There will not be any meaningful way, consequently, to record changes in state inventories of fissile material. This means that a full scope verification regime will yield few benefits on the margin. Simply looking at carefully defined materials, compounds and processes could be enough to assure the nuclear weapon states that no militarily significant production of fissile material is occurring. This low-assurance verification scheme will by no means be foolproof, but given the object and purpose of the proposed treaty, it may be viewed as enough.

In the Conference on Disarmament, the better is still the enemy of the good. And since it shows few signs of starting work, solutions must be found through other means. In the words of US State Department Special Advisor Robert Einhorn: "If the others balk, the five should proceed without them, as befitting their responsibility to exercise leadership as the only NPT nuclear weapon states". End of quote. And one way of showing leadership is to expand the moratoria through other means, perhaps through a Fissile Material Control Initiative, as proposed by Dr Einhorn.

This matter is too important, too essential to international security, to fall victim to the unachievable vision of fissile material control nirvana.

I thank you for your attention.

Trust & Verify No. 127

I cannot believe that we're already at issue number 127 of this publication. Some of you might know that I a couple of year instructed a gang of unhappy interns to scan all back issues, from issue number one. Once I find a host, I'll see if I cannot put everything on-line. For now, though, this issue deals with very technical issues: infrasound monitoring and advanced photoshopping.

It's also littered with references to friends of mine. Before you accuse me of plugging friends in the Centre's publication, know that I did not draft any of the articles in verification watch. You guys just happened to be there when I reviewed the drafts - a quite pleasant surprise.

Also, by no means did I want you to have a 'restive' season, or even a 'festive' season. I'd like you all to have a restful season. Come 24 December, yours truly will switch off completely. Back on 4 January 2010.

Putting the CTBT into practice

As a director, you're never quite have the time to write about the things you like, such as wonkish observations on the state of arms control. And many of you visiting this half-dormant blog also knows that I seem to prioritize my contributions to the Arms Control Wonk. However, once in a while, good stuff happens. Such as VERTIC's new brief on the CTBT, written by a couple of really capable guys from down under.



Enjoy VERTIC Brief 10, Putting the CTBT into practice. And tomorrow, we'll put out the 127th edition of Trust & Verify.

Do have a restful season, dear friends.