About a decade ago, I started my nuclear policy career at a small British NGO that focuses on verification. My career choice turned out to be a mixed blessing for my social life. Saying that I worked on nuclear weapons was a great icebreaker at parties, but the ensuing conversations would go downhill rapidly after I mentioned the “V” word. For reasons inexplicable to me back then, my fellow party goers just weren’t that interested in finding out how to determine whether states are abiding by their nuclear treaty commitments.

James M. Acton
Acton holds the Jessica T. Mathews Chair and is co-director of the Nuclear Policy Program at the Carnegie Endowment for International Peace.
More >

Over the next few months, I expect to have many conversations with officials, analysts, journalists about the Joint Comprehensive Plan of Action — better known as the JPCOA or the Iran deal. These conversations will be quite different from those at London parties ten years ago; today, all my interlocutors will profess a deep and profound belief in the importance of verification. But, when I start to dig into the details of how the International Atomic Energy Agency will assess Iranian compliance, I’ll see that glazed look again…

Nonetheless, where verification is concerned, the details do matter and we really should be debating the finer points of the Iran deal’s verification provisions. (See: Annex I, Sections L, M, N, O, P, Q, and R — yes, it’s that detailed.) In assessing whether these arrangements are “good enough,” the best place to start is with the following question: If Iran decides to cheat, how would it go about doing so?

Iran’s leadership would have three options and, in deciding between them, it would presumably choose the pathway that maximized its chances of success.

First, Iran could overtly renounce all its nonproliferation commitments chuck out international inspectors, and build the bomb loudly and proudly. The Nuclear Nonproliferation Treaty contains a clause that allows states to withdraw under “extraordinary” circumstances and, even though JPCOA doesn’t have any such provision, there can be no certainty that Iran won’t abrogate it anyway. No verification system can prevent this scenario, but what almost certainly can deter it is the threat of American weaponry hitting Iran before Ayatollah Khamenei can say “death to.”

The second, more likely scenario would be for Iran to use its declared nuclear materials and facilities for bomb-building: the much-discussed “breakout scenario.” Many of the Iran deal’s limits are intended to make breakout much more time consuming than it would currently be — and that’s a good thing. Ultimately, however, breakout still isn’t all that likely. Declared facilities are subject to International Atomic Energy Agency (IAEA) monitoring. As a result, Iran understands it would almost certainly be caught quickly if it attempted breakout.

Iran’s third option would be to build a secret parallel nuclear program dedicated to military purposes — sneak-out. Detecting small clandestine enrichment plants is difficult and Tehran might view sneak-out as it most attractive option. Indeed, Iran has tried to sneak-out before. Repeatedly. It failed to declare three out of the four facilities in which it has enriched uranium (the Kalaye Electric Factory, the Fordow Fuel Enrichment Plant near Qom, and the Pilot Fuel Enrichment Plant at Natanz) in accordance with IAEA rules.

“Anytime, anywhere” access is often advocated as the solution to detecting secret facilities — in fact, Secretary of Energy Ernest Moniz, an MIT physicist and one of the U.S. negotiators, said back in April that the United States expected it. The Iran deal doesn’t provide for it, however, as critics, including Sen. Tom Cotton, have noted (rather gleefully, at that).

So, what access provisions does the deal contain?

It does allow the IAEA to go anywhere — including military sites — if there is evidence of undeclared facilities hosting nuclear activities. But, if Iran declined to grant access, a complicated dispute resolution negotiation process would ensue under which Iran would have to negotiate first with the IAEA and then with the Joint Commission created to oversee implementation of the deal. This process could take up to 24 days. (On day 25, if Iran still refused access, it would be in non-compliance with the agreement and sanctions could be re-introduced.)

Fortunately for the JCPOA, the refrain of an “anytime, anywhere” access may make for a great soundbite, but its utility is overstated by Cotton and other critics of the agreement. An access delay — even one of 24 days — wouldn’t make any material difference to the IAEA’s ability to detect undeclared nuclear activities.

When IAEA inspectors search for undeclared nuclear activities they look for tiny traces of nuclear material on surfaces. Fortunately for them, nuclear material lingers. And, modern detection technology is so amazingly effective that miniscule traces of nuclear material can be detected years after nuclear activities took place. Countries have tried to sanitize facilities completely to remove every last trace of nuclear material. Iran did so at the Kalaye Electric Factory after its secret nuclear program was revealed in 2002. Syria tried the same thing in 2007 after Israel bombed its plutonium production reactor at Al Kibar. In both cases, the IAEA still managed to detect nuclear material. Those findings were critical to persuading the organization’s governing body to make a formal finding of non-compliance against both Iran and Syria.

Perhaps Iran has learned from its past mistakes and could do a better job of cleaning up nuclear material in the future and keeping its program secret. But, what’s clear is that perfect cleanup — if it were possible — would take many months. After just 24 days, the IAEA would have little difficulty detecting the residue from undeclared nuclear activities.

So, here’s the bottom line: The Iran deal doesn’t provide for anytime, anywhere access, but it does facilitate timely access anywhere — and that’s what needed for effective verification.

But wait, as they say on QVC, there’s more!

Not only is anytime, anywhere access not necessary, it’s also not sufficient. In other words, its inclusion might have placated (a few) critics, but would not be enough, by itself, for effective verification. After all, it would be physically impossible for the IAEA to inspect every building where Iran could conceivably be hiding clandestine nuclear activities.

What the IAEA actually needs is some preliminary evidence about where a secret nuclear facility might be lurking. The much-discussed but little-understood Additional Protocol was developed precisely for that purpose and the JCPOA obliges Iran to accept it, first voluntarily and subsequently on a legally binding basis. But, the JCPOA goes beyond the Additional Protocol in two innovative and important ways.

First, IAEA monitoring will be extended to declared yellowcake (the precursor material to the feedstock for enrichment) and to declared centrifuge components. This measure will deter Iran from diverting this material and equipment to a secret program. Iran could, of course, try to acquire yellowcake or centrifuge components secretly instead — but doing so would create more opportunities for detection.

Second, the deal also creates a “Procurement Working Group” to oversee the import of all equipment and material that either is used or could be used for nuclear purposes. The intelligence communities of the United States and its friends spend considerable resources monitoring Iranian imports. If they discover that Iran has obtained any items that should have been declared but weren’t, they will have acquired clear evidence of secret nuclear activities in Iran. They could hand this evidence to the IAEA, which could conduct inspections to investigate further.

All in all, therefore, the JPCOA provides for some impressive verification provisions to guard against sneak-out. That said, no one should be under any illusions. Detecting small, undeclared centrifuge plants is difficult and there is no guarantee of success. But, perfection is the not right metric against which to assess a nonproliferation agreement. The real question is whether sneak-out is more likely with a deal or without one. And here the answer is clear: sneak-out would be much more likely without a deal, because the IAEA’s powers to detect clandestine facilities would be much more limited.

I realize, of course, that all this talk of timely access, yellowcake monitoring, and procurement working groups isn’t exactly headline grabbing (though, fortunately, since I’m now happily married, I have ceased trying to use them in chat-up lines). But, ultimately, it’s complex technical considerations that determine whether the JCPOA’s verification regime will prove effective. And, in the final analysis, the JPCOA does significantly enhance the ability of the IAEA to guard against sneak-out, the most likely pathway for Iran to acquire the Bomb. That’s not the only metric for assessing the deal — but it is a bloody important one.

This article was originally published in Foreign Policy