Thoughts on the verification of nuclear disarmament

An interesting paper on some thoughts on the verification of nuclear disarmament has just been released. It was first presented to the CISAC conference on "P-5 Nuclear Doctrines and Article VI" (which Jeffrey Lewis writes about here). I got hold of a copy of this paper many months ago, but was sworn to secrecy by the donor (the entire conference was surrounded by an, in my humble opinion, laughable amount of secrecy). Then again, the nuclear community has a puerile tendency to divide people into tiers, where some deserves to know and others (usually weaker competitors) are left in the dark.

Still, while the paper's origins leave much to desire, its content does not. Especially this sobering conclusion:

A system to monitor the true reduction of nuclear weapons would need to encompass the entire nuclear weapons manufacture capability and deployments of the nuclear weapons of a country. One would need to monitor the number of weapon and/or pits being produced, and the number being destroyed, and have at least a declaration of how many existed at some point in time in the process of disarmament. The declaration would need to cover not only the active stockpile of nuclear weapons but also any and all inactive reserves or stocks of warhead components (namely pits). It would need to ensure that there was no method to produce and insert new weapons into this system through clandestine means. Such a system would entail far more intrusive processes than we have ever negotiated in any of the Arms Control agreements in the past. It may be possible but it would be an unprecedented step in transparency or verification.

But would this really satisfy those who would apply the standard of "effective verification"? The answer is almost certainly not.

I am not entirely certain that I agree with all the conclusions in the paper. But it is a worthwhile read nevertheless.

Hier Berlin!

It is slightly pathetic to spend a Friday evening holed up in a hotel room in Berlin, but I stopped working only an hour or so ago and the rest of the conference participants have scattered into the cold German night. I have been here for two days now, attending a conference on Coercive Arms Control organized by the Institute for Peace research and Security Policy at the University of Hamburg and the Ludwig-Maximilians-Universität Munchen. VERTIC received an invitation to attend last week by former vertical Oliver Meier (who now works part time for the Arms Control Association).

It was interesting, not least since it raised a lot of memories from my days at BASIC, where I co-authored a report on the Proliferation Security Initiative (with Ian Davis) which, believe it or not, still gets cited in serious books. I also wrote a scary note on what the PSI should do in its early days. I wrote:

The movement of goods and personnel on the oceans is largely unregulated and free. Commercial freighters are capable of moving bulky and heavy equipment, such as missiles, as well as large quantities of explosives. It is also possible to hide smaller items of WMD supplies, such as nuclear materials, within a larger cargo. Therefore, proliferants are likely to use the seas to transport their goods. However, the oceans are also an important lifeline of international trade and commerce, and any venture to restrict the free flow of goods should therefore be carefully implemented. Even though states do enjoy a high level of sovereignty over their territorial waters, they should consider that ships do enjoy the right of innocent passage through it. The PSI is by no means dead in the water, but in order to make it steam ahead, participants should at their forthcoming Washington Operational Meeting:

- Consider harmonization of their national regulations in respect to interdiction in their internal and territorial waters;
- Consider what limitations the freedom of navigation on the high seas place on the initiative;
- Emphasise the development of a consensually agreed interdiction framework, preferably in the form of a legally binding and publicly deposited document;
- Recommend to the Portugal plenary meeting scheduled in January 2004, that such an agreed interdiction framework should be in harmony with current international legal authorities, rather than being based on a reinterpretation of customary international law.

Scary since it's actually too accurate and to well written to be written by me.

This short note actually earned us a meeting with a merry gang of PSI implementers at the FCO. Thinking back, it occurs to me that they probably were as clueless as Ian and I was about what, precisely, the PSI was all about. From a position of initial confusion as to what allied warships actually could do on the high seas, the participants then moved into developing consensually agreed interdiction frameworks – namely the ship boarding agreements. I would have expected no less, since the US always have been the staunchest of advocates for freedom of navigation on the high seas.

Today, the PSI is built on a quite flexible legal framework (bilateral boarding agreements) but there still seem to be some legal problems as to how to deal with the cargo itself. I say "seem" because one of the main conclusions of the conference was that the initiative, despite all public diplomacy in its honor, remains relatively opaque. BASIC has done a great job in compiling and analyzing what information is out there, although a lot of conceptual work remains to be done.

Have a look at BASIC's page.

Verifying the Indian nuclear programme

According to an article published in December 2007, representatives from India's Department of Atomic Energy have again travelled to Vienna to continue discussions on a so-called India Specific Safeguards Agreement (the ISSA). Talks on this agreement started with a meeting between Anil Kakodkar (a former BARC director) and Mohamed ElBaradei on 21 November 2007. Back then, both sides were anticipating a painless process: it was simply a matter of applying a "standard procedure" to a largely technical problem. Today, nearly two months and three rounds of talks later, Dr. Kakodar doesn't want to talk about delivery. Now it "is a detailed technical exercise and no template exists for it". So detailed that it isn't possible to say when it will be finalized.

Details on the structure and content of the Indian safeguards agreement are sketchy. You get a different interpretation depending on who you talk to. Many months ago I spoke to a British safeguards official about the pact. He expressed skepticism over India's confident opinion that the military and civilian fuel cycle can be fully separated without any leakage between them. He could not see good language that could ensure a watertight system: "the UK has had about fifty years experience in managing two fuel cycles - and not even our system is perfect".

Others are more confident. The IAEA DG himself has said that, "this is a standard procedure that we usually do. We have four of them with India … I was surprised to see the focus on the safeguards agreement. Safeguards agreement is standard procedure". Obviously, no one should know the safeguards system better than the IAEA Director-General, but I am nevertheless perplexed by what seems to be an off-the-cuff comment. The problem is not whether the IAEA has safeguards agreements with India, they do. The problem is rather that this is the first time a non-site specific safeguards agreement is being concluded with a non-nuclear nuclear-weapon state. In my mind at least, that mouthful complicates things.

Granted, it is not theoretically difficult to envision an agreement which effectively separates India's civilian and military fuel cycles. One simply has to visualize a non-nuclear weapon state within India's territory and set up firewalls between India's nuclear weapon capable side and the rest of the fuel cycle. That way, material accountancy and control is possible within each isolated fuel cycle. Under this scheme, no nuclear material should be allowed to transfer from the civilian to the military sector. One quick fix would simply be to apply INFCIRC/66 on all civilian facilities. Nothing like this is currently in force in India. Presently, the IAEA has issued the following information circulars in respect to India.

• INFCIRC/154: India & the United States (3 September 1971) - expired in August 1993;
• INFCIRC/211: India, Canada & the IAEA (6 November 1974);
• INFCIRC/260: India & IAEA regarding Soviet supplied heavy water (July 1978);
• INFCIRC/360: India & IAEA regarding Soviet supplied reactor (January 1989);
• INFCIRC/433: India & IAEA regarding INFCIRC/154 (May 1994);
• INFCIRC/433/Mod. 1: An amended version of INFCIRC/433.

The 1994 agreement concerned the twin Boiling Water Reactors in operation at the Tarapur Atomic Power Station (TAPS) built by the United States. The US stopped supplying the station after India detonated a nuclear device in 1974. Tarapur ran out of fuel in 1976. But it wasn't empty for long: the French stepped in to supply the facility instead. When the 1971 agreement was about to expire, the United States refused to receive the spent fuel, and also refused to give consent for it to be reprocessed. India attempted to straighten out these legalities by concluding a unilateral safeguards agreement with the IAEA. At the same time, India also seems to have respected the original agreement: the TAPS fuel was unprocessed in 2000, which is the latest reference that I can find (Frontline, 2000).

Last year, the Indians laid out an ambitious timeline with talks reportedly scheduled to be finished by the end of 2007. Obviously, that did not happen. There are several reasons why India might want to rush negotiations with the Agency. First, India may find itself faced with a less friendly US administration after the 2008 elections. Second, there has been an internal tug-of-war between the Indian government and the country's skeptical communist party. On 12 December 2007, for instance, the communists threatened to call a vote in parliament if the safeguards agreement wasn't concluded before the end of the year. Reportedly, this was dismissed by Dr. Kakodkar, who held that the government's negotiators were "doing our best and try to complete the process as soon as possible but everything has to be done satisfactorily as the negotiations are comprehensive and complex".

I suspect that most of the complexity is caused by India's insistence on a withdrawal clause. On 13 August 2007, the Indian Prime-Minister held in a speech to parliament that:

As agreed in the March Separation Plan, India has accepted only IAEA safeguards that will be reflected in an India-specific Safeguards Agreement with the IAEA. We have not consented to any provision that mandates scrutiny of our nuclear weapons programme or any unsafeguarded nuclear facilities. There are explicit provisions in the Agreement that make it clear that this Agreement does not affect our unsafeguarded nuclear facilities and that it will not affect our right to use materials, equipment, information or technology acquired or developed independently. India and the United States have agreed that the implementation of the Agreement will not hinder or otherwise interfere with India’s nuclear activities including our military nuclear facilities. Nothing in the Agreement would impinge on our strategic programme, our three-stage nuclear power programme or our ability to conduct advanced R&D.

And this is the crux: an acceleration or reinvigoration of India’s nuclear weapons programme may cause some states to reconsider their supply commitments. According to Arms Control Association analyst Wade Boese, "reports exist that New Delhi, among other things, wants flexibility to withdraw facilities from safeguards in the event that foreign nuclear supplies are cut off, even if it is the result of renewed Indian nuclear testing". Some sources say that the India-specific safeguards agreement will be based on a web of INFCIRC/66 type agreements, which doesn't allow for withdrawal. This excellent post by Jeffrey Lewis explains it in a bit more detail. One might want to add that, according to article 56 of the Vienna Convention on the Law of Treaties, the absence of an withdrawal clause means that the agreement isn't subject to withdrawal. Once you've signed, you're stuck with the deal, unless, of course, there are grounds for invalidity, such as a fundamental change of circumstances. But the rebus sic stantibus doctrine wouldn't apply here, since the possibility of fuel interruptions have been foreseen (and that's why they're negotiating about it, obviously).

NB. For more information on the separation plan, the Institute for Science and International Security has already published a briefing paper outlining not only all major Indian facilities, but also their role in the Indian fuel cycle.

Update on DPRK disablement

A friend has drawn my attention to a post by Jeffrey Lewis explaining North Korea's water difficulties. The post, as well as the linked documentation, are extremely interesting and well worth a read. I should not be surprised to learn that North Korea's standards are, well, below what's internationally expected. A regular reader may remember an article published by RIA Novosti explaining that Agency monitoring of the facility was delayed by hazardous radiation conditions. IAEA inspectors had to decontaminate certain areas before proceeding with the installation of monitoring equipment and seals.

The same friend also explained that the disablement steps are expected to take between a year and 18 months to reverse. I have heard less optimistic estimates however, which led us to discuss over e-mail whether or not the correct estimate might be 6-18 months, with a year as the convenient median. Who knows?

The information, which I obviously missed, leads me to wonder why the six parties are so keen to repair the storage pond. Reading IAEA-TECDOC-944, I can't help noticing that cooling ponds storing Magnox type spent fuel needs some healthy water circulation, to avoid differences in water temperature. I hope that this is the reason why the six parties have decided to patch the storage pond to a new and improved version. If the reason is to facilitate a long storage of spent fuel, the outcome will strongly resemble the situation that prevailed during the time of the agreed framework. North Korea will have some nice weapons usable plutonium ready for processing, should they be unhappy with the way the world turns.

Still, I suppose that is better than North Korea actively producing nuclear weapons. As my late grandmother, probably remembering the war years, sometimes said: "man får vara tacksam för vad man har" (one has to be grateful for what one has).

DPRK disablement

The people living and working in and around the Yongbyon Nuclear Complex is enjoying a cold and relatively dry day. But the stable weather, presently hovering around eight degrees below zero, will change in the coming days. A low pressure will come rolling in from the Pacific Ocean and it will bring much lower temperatures. By the day after tomorrow, temperatures will have dropped to 13 degrees below zero and altocumulus clouds will have started to assemble at the horizon.

The employees of the Yongbyon Nuclear Complex have other things to worry about than cold weather, however. Lights will start to come on around the sprawling complex as the sun sets around 5.30pm and darkness grips the surrounding landscape and, given the chronic electricity shortages plaguing the country, possibly the site itself.

Despite shortages, white electrical lights are likely to burn in the main building of the five-megawatt electric reactor at the heart of the nuclear compound. Staff there is hard at work. According some reports, such as this note in the Jerusalem Post, the North Koreans started unloading the reactor fuel on or around 17 December 2007. The last fuel rod should be winkled out sometime this coming week, if the past serves as guidance (see my previous post here). In December, I wrote that the process could be completed, and the rods cool enough for transport by the end of March 2008. This estimate seems consistent with recent KCNA reports (which talks of a 100 day process).

However, other press accounts hints that there have been delays. According to a Bloomberg article, the process has been delayed due to some "technical questions about the cooling of the fuel rods". It is difficult to see what, precisely, those problems are. The North Koreans, in turn, have said that they feel "compelled to adjust the tempo of the disablement" since the other parties has not delivered on their pledges. This is likely to change as a promised shipment of Russian heavy oil reaches the DPRK's borders.

I spoke with several representatives from the six-party nations during Mark Smith's excellent Wilton Park Conference late last year, to figure out in detail what disablement steps have been taken and what the various participants felt about them. My impression was that the disablement steps, with the possible exception of the partial dismantlement of the cooling tower, were relatively easy to reverse. As one participant put it, "how long does it take to put things back in place"? Another participant, who had toured the facility, wasn't precisely sure which steps had been taken at the Fuel Fabrication Facility and the Reprocessing Plant since the group was not allowed inside the buildings. The North Koreans had assigned several minders to the group, and security was very tight, with armed guards literally everywhere.

One thing is certain. The fuel will be moved from the reactor building to the adjacent spent fuel pond. What happens next is uncertain. The worst case is that the fuel will remain there for quite some time, so that the North Koreans can use it as a bargaining chip. If so, we will soon see discussion about how to protect the fuel from corrosion – which was a problem during the implementation of the agreed framework.

I asked, in one of the plenary sessions, how the six parties intend to verify the forthcoming North Korean declaration and was greeted with silence. Afterwards, one of the insiders said that the reasons why no one can answer how verification should be conducted, is because verification has not been discussed yet. The six parties are truly implementing the agreement by the seat of their pants.

In other words, expect more twist and turns down the road.