In a recent controversial article in The New Yorker, Seymour Hersh points to growing radicalization within Pakistan’s military that could endanger the security of its nuclear arsenal…

Though the United States has repeatedly stated that it has no intention of seizing Pakistani nuclear weapons or materials and maintains confidence in the security of Pakistan’s arsenal, Hersh reports that the Obama administration has been negotiating “highly sensitive” understandings that would allow specially-trained American units to provide added security for the Pakistani arsenal in a crisis. According to Hersh, the secrecy surrounding these agreements arises from a growing antipathy and history of distrust toward America within Pakistan, which has only intensified as a result of U.S. pressure on the Pakistan Army to take more aggressive action against Taliban enclaves inside Pakistan.

Hersh is right that the security of Pakistan’s nuclear arsenal is cause for concern. Though nuclear safeguards are in place, primarily to keep a confrontation with India from escalating too quickly, these same safeguards could make Pakistan’s arsenal more vulnerable to terrorists. Extensive procedures that require warheads to be stored separate from their triggers and delivery vehicles leave these elements most exposed, as they are constantly being moved and reassembled. Andrew St. Denis noted in August that most of Pakistan’s nuclear facilities lie in and around areas populated by the Pakistani Taliban and al-Qaeda.

Rolf Mowatt-Larssen recently wrote in Arms Control Today about the “lethal proximity between terrorists, extremists, and nuclear weapons insiders” in Pakistan. “Insiders have facilitated terrorist attacks. Suicide bombings have occurred at air force bases that reportedly serve as nuclear weapons storage sites. It is difficult to ignore such trends,” he said.

Michael Krepon, however, points out that Hersh’s record, though relatively solid in other respects, is mediocre at best with regard to Pakistan. Unfortunately, writes Krepon, Hersh’s “sourcing is weak and his conclusions are suspect.” If his story were true, “those ‘specially trained American units’ can now forget about helping Pakistan to secure its arsenal: Public revelation of such an agreement makes it about as palatable within Pakistan as changing that nation’s religious preference.”

Still partial to the argument that our deterrent can’t be maintained without designing and producing new nuclear warheads?  Still think the authors of the 2002 National Academy of Sciences report and the 2006 JASON pit lifetime study belong in the loony bin?  Then you’re probably not going to like the September 2009 JASON study on Life Extension Programs (LEPs), the unclassified version of which was released today.  According to the executive summary:

JASON finds no evidence that accumulation of changes incurred from aging and LEPs have increased risk to certification of today’s deployed nuclear warheads

This finding is a direct consequence of the excellent work of the people in the US nuclear weapons complex supported and informed by the tools and methods developed through the Stockpile Stewardship program. Some aging issues have already been resolved. The others that have been identified can be resolved through LEP approaches similar to those employed to date. To maintain certification, military requirements for some stockpile warheads have been modified. The modifications are the result of improved understanding of original weapon performance, not because of aging or other changes. If desired, all but one of the original major performance requirements could also be met through LEP approaches similar to those employed to date.

Lifetimes of today’s nuclear warheads could be extended for decades, with no anticipated loss in confidence, by using approaches similar to those employed in LEPs to date.

The report discusses details and challenges for each stockpile system.

My advice to the remaining skeptics: It’s one…two…three strikes you’re out at the old…ball…game!

Guest Post by Alex Bollfrass (Click here to read Freezer Burn, Part II)

News broke last week that prosecutors are asking to have William Jefferson (D.-Louisiana) locked up for some 13 years.  You may remember that he stashed about $90,000 in dirty money in his freezer.

If you thought that a bribe to a foreign vice-president was the most unappetizing thing stuffed in with the pie crusts, read Thomas Schelling’s “A world without nuclear weapons?” in the current issue of Daedalus (gated). In his telling, one of the major barriers to a world without nuclear weapons is his refrigeration capacity:

“[E]nough plutonium to make a bomb could be hidden in the freezing compartment of my refrigerator.”

While this observation about the size of the plutonium in a nuclear bomb may be true, it tells us little about the feasibility of nuclear disarmament.  

Schelling’s point is that if sufficient fissile material for a bomb is small enough to fit into a freezer, an agreement to eliminate nuclear weapons could never be verified.  But hiding weapons is only part of a cheater’s problem.  There are additional formidable obstacles that would prompt anyone contemplating reconstituting or developing nuclear weapons in a disarmament regime to think very carefully.

Cheating on a disarmament agreement could happen in one of three ways:

1. Cheating at the outset and retaining undeclared warheads or fissile materials;

2. Clandestinely producing the materials for a nuclear arsenal after disarmament had been accomplished;

3. Diverting declared materials from nuclear energy facilities or from left-over fissile material stocks.

The last method is roughly equivalent to grabbing the money and dashing for the casino door.  The ploy would be immediately uncovered as all such facilities and stockpiles would be safeguarded far more closely than ex-Congressman Jefferson’s “take” in any self-respecting disarmament regime.  Such a cheater’s gain, if any, would be short-lived.  Either the other signatories would undertake collective action with conventional military forces to stop the program, or those with the potential capabilities would rearm themselves, leaving the world no worse off than it is now. A rearmament effort triggered this way would be unfortunate, but unlikely. Furthermore, it would be comparatively stable because the governments would be building a defense-oriented deterrent, as opposed to seeking a coercive advantage.

The middle course of action? Check with this bloke. With on-site challenge inspections and national means of verification, such an action would likely be discovered relatively quickly, just as the Qom facility was uncovered and has every covert weapons program to date.  The world may not have acted on South Africa’s or Israel’s or Pakistan’s covert weapon programs, but the United States and others were certainly aware of what was going on – even without an on-site challenge inspection scheme.  

The first option, Schelling’s path, is only available to the nine countries that already have enough unsafeguarded plutonium or highly enriched uranium to make a bomb: China, France, Great Britain, India, Israel, North Korea, Pakistan, Russia, and the United States.

In practice, the majority of these countries would almost certainly be caught.  Their holdings of materials are modest enough that their declarations of weapons and stocks could be confirmed with great certainty using a combination of forensic verification techniques.  The only two exceptions are the US and Russia, both of which produced massive amounts of fissile material during the Cold War. For these two exceptions, the margin of error that could reasonably be expected from thorough verification procedures amounts to several hundreds of bombs worth of uncertainty. If we agreed to implement nuclear disarmament tomorrow, we or the Russians could presumably squirrel away at least several dozen bombs (or at least the fissile materials to make them).

Will this prove to be nuclear disarmament’s fatal vulnerability – a la Siegfried’s exposed back?

It need not be. That margin of error can be reduced by an increase in transparency, coupled with the healing powers of time. Long before moving to the White House, Steve Fetter wrote “Verifying Nuclear Disarmament” as the Stimson Center’s Occasional Paper #29 in 1996. He identified a remedy, which is worth quoting at length from the paper:

A data exchange could help build confidence between the parties even before the accuracy of the data was verified. An early exchange of data is particularly important because it would force governments to make decisions about compliance with reporting requirements well in advance of possible disarmament agreements. A government that possesses thousands of nuclear weapons and has made no near-term commitment to disarmament is less likely to be suspected of falsifying records or hiding weapons than a country that has few weapons and is obliged to eliminate the remainder, simply because a country with thousands of warheads would have little incentive to cheat.

It is also important to begin verifying declarations as far in advance of a disarmament agreement as possible. As the number of nuclear weapons falls into the hundreds, states would be far more likely to have confidence in a declaration whose accuracy had been verified for years and for tens of thousands of nuclear warheads, than one whose verification had begun only recently and only after thousands of warheads had been dismantled. Failure to verify the dismantling and consolidation of the huge US and Russian nuclear stockpiles could undermine severely the two sides’ confidence in a declaration made later about much smaller numbers of weapons. There is little pressure for warhead-verification today because current and planned stockpiles are so large as to make existing uncertainties unimportant. But unless the nuclear powers begin now to describe and verify their warhead stockpiles, when the need for verification is not pressing, they will have failed to lay a foundation that is strong enough to bear the weight of a disarmament regime.

Data on the history of stockpiles and the operation of warhead-related facilities cannot be verified directly, of course, but it could be checked for internal consistency, and for its consistency with archived intelligence data. If, for example, US satellites had detected the movement of nuclear warheads from a particular Russian facility on a particular date in the past, this could be checked against the records exchanged between the two countries. Indeed, such records should improve the value of archived data by confirming or contradicting past interpretations by intelligence agencies. The fact that countries would not know what intelligence information might be available would act as an incentive to provide complete and accurate data.

As with declarations on delivery vehicles, data on the current status of nuclear warheads would be verified by regular and short-notice inspections of declared facilities, combined with challenge inspections to verify the absence of warheads at other locations. For example, inspectors could count the number of warheads in a particular storage bunker and compare this to the number listed in the data exchange.

If President Obama sees nuclear disarmament as a serious destination, and not just rhetorical decoration for traditional arms control and nonproliferation, these kinds of early transparency measures should be included in the next round of negotiations with the Russians. So should tactical nuclear weapons.

In the most optimistic abolition scenario, ex-Congressman Jefferson would find himself released into a world contemplating the final destruction of the remaining nuclear weapons when he retires as ex-convict Jefferson several decades from now. That is a long time to devise technologies and procedures for verifying US and Russian fissile material stocks. Whether the nuclear-armed countries of that era will have the confidence to take such a leap depends on how seriously we prepare for this possibility in the next few years.

Alex Bollfrass works for the Unblocking the Road to Zero project at the Stimson Center, where he occasionally comes across suspiciously plutonium-like materials in the office refrigerator.

The October-November 2009 issue of Survival includes an excellent back and forth on the merits of a nuclear “no first use” declaratory policy (hat tip: PONI).  The exchange is in response to Scott Sagan’s article supporting such a policy in the June-July 2009 issue.

As the most hackneyed saying in the blogosphere goes, read the whole thing.  And if you’re interested in some further discussion on the arguments of one of Sagan’s interlocutors, well, you know what to do.

Bruno Tertrais raises a number of objections to Sagan’s case, most of which Sagan ably rebuts in his reply.  Yet Sagan did not challenge two points made by Tertrais that I think merit a response.

A “commitment trap”?

In his seminal article on no first use in the Spring 2000 issue of International Security, Sagan wrote:

current nuclear doctrine creates a “commitment trap”: threats to use nuclear weapons in response to a chemical or biological attack are credible, because if CW or BW are used despite such threats, the U.S. president would feel compelled to retaliate with nuclear weapons to maintain his or her international and domestic reputation for honoring commitments.

Sagan reiterated this point in his Survival article, which prompted the following response from Tertrais:

But why would there be such a trap as long as there is no promise of a guaranteed nuclear response?  That is precisely the point of “calculated ambiguity”….I cannot believe, moreover, that an American president would see “reputation” as a reason to take the most dramatic military decision a Western leader has had to take since 1945.

Is Tertrais’ disbelief warranted?  I think the more compelling logic is on the side of Sagan here.  In fact, Tertrais himself appears to concede that reputation is something to be taken seriously in international politics.  

For instance, one of the cudgels he yields against no first use is that it could carry significant “non-proliferation costs”: allies covered by the U.S. nuclear umbrella might doubt the credibility of the U.S. deterrent, thereby leading them to develop their own indigenous nuclear weapons programs. Put in another way, a no first use policy would call into question the U.S. commitment to live up to its reputation as the ultimate guarantor of Japanese security.

If reputation is a reason why the U.S. should not abandon calculated ambiguity, how can Tertrais be so certain that no American president would view reputation as a reason to use nuclear weapons?  These are different cases, to be sure, but in the event of a North Korean chemical or biological attack on Japan, couldn’t the perceived need to make good on a pledge to meet deterrent commitments to Japan put immense pressure on the President to respond with nuclear weapons – even though there was no firm promise to respond in such a way?  

Sagan spoke to this pressure in more detail in his 2000 essay:

The problem here is one of degree: it is unlikely that any U.S. president would respond with nuclear weapons to a very small-scale chemical or biological weapons attack; yet a president’s belief that his or her statements, or those of subordinate officials, had created a commitment in the eyes of allies and future adversaries could tip the balance between nuclear and conventional response in more serious contingencies. Indeed, it is difficult to imagine a more important or valid reason for using nuclear weapons in response to chemical or biological weapons attacks. [emphasis mine].

Would no first use leave the U.S. vulnerable?

According to Tertrais, the answer is yes:

[no first use] would signal those adversaries who would take such a commitment seriously that they could do anything to the United States or its allies without ever facing the risk of a nuclear response, using chemical weapons against our forces on a battlefield, raining down conventional ballistic missiles on our homelands, or launching biological munitions against our populations.

This is a straw man of what a no first use policy is all about.  As Sagan notes in his Survival article, “declaratory policy is not about making ‘promises’ about future restraint; it is about signaling intent and therefore shaping the expectations of allies and adversaries alike, even if some residual uncertainty remains.”

Look, even if the United States were to adopt a no first use policy, its nuclear arsenal is still on the table, so to speak.  This existential reality is something potential adversaries would have to take into account when considering whether to attack the United States with WMD or advanced conventional weapons.  The point is that the more U.S. leaders and government officials loudly and publically suggest that nuclear weapons remain relevant for various contingencies, the more likely it would be to actually use them in the event of such a contingency – because of the commitment trap.