After the powers and Iran in late 2013 concluded the Joint Plan of Action (JPA), I cautioned that, a year later, when everything else is supposed to be settled, the toughest nut to crack might be what to do about Iran’s nuclear past. Happy talkers who didn’t like that message marginalized it for months. But right now, questions raised by IAEA Director General Yukiya Amano about what he calls “possible military dimensions” (PMD) of Iran’s nuclear program are standing tall between the negotiators and a comprehensive settlement of the crisis.

While the powers and Iran were negotiating the JPA, they and Iran set up a Framework for Cooperation on a parallel track which committed the IAEA and Iran to resolve PMD issues. That  began with confidence-building steps which were supposed to coax Iran to give the IAEA enough data for it to tell its Board of Governors and the U.N. Security Council that things with Iran were working out.

Mark Hibbs
Hibbs is a Germany-based nonresident senior fellow in Carnegie’s Nuclear Policy Program. His areas of expertise are nuclear verification and safeguards, multilateral nuclear trade policy, international nuclear cooperation, and nonproliferation arrangements.
More >

After six months, all the low-hanging fruit was picked. This May, the IAEA and Iran agreed to a third set of “practical measures” under the Framework, with an August deadline. The list included “exchanging information with the Agency with respect to the allegations related to the initiation of high explosives, including the conduct of large scale high explosives experimentation in Iran.”

The last we heard about this from the IAEA was in its 7 November report to the Board, GOV/2014/58:

Iran and the Agency held technical meetings on two separate occasions in Tehran to discuss the two outstanding practical measures agreed in May 2014 in the third step of the Framework for Cooperation. Iran has not provided any explanations that enable the Agency to clarify the outstanding practical measures, nor has it proposed any new practical measures in the next step of the Framework for Cooperation.

Iran responded to that in a comment filed on 1 December to the IAEA Secretariat as GOV/INF 871:

During technical meetings in Tehran on 7 and 8 October 2014 and 2 November 2014 [and regarding high-explosion initiation allegations] Iran… provided detailed explanations on the documents shown by the Agency to Iran and provided pieces of evidence that indicate such documents are fabricated. Those forged documents have no sign to prove that they are of Iranian origin and contrary to such claim; the documents are full of mistakes and contain fake names with specific pronunciations, which only point toward a certain Member of the IAEA as their forger… Indeed, invalidity of Agency’s information or better to say invalidity of information given to the Agency and lack of substantiated evidences at the disposal of the Agency are the major problems on these issues. In continuation of our cooperation with the Agency, we intend to arrange another technical meeting on these two practical measures as soon as we receive specific questions of the Agency with substantiated documents in order to conclude them and once these issues are clarified and closed, we can start considering implementation of new practical measure.

I read Iran’s statement to mean that, unless the IAEA provides Iran “substantiated evidences” that Iran agrees are valid, the Framework for Cooperation is on ice. Iran says Amano’s information on high-explosion initiation is falsified. Finito la musica.

But in mid-November once again it was Iran, not Amano, that moved next. In the IAEA boardroom it told the IAEA Iran would permit inspectors to carry out one managed access at Marivan, one of two sites (the other being Parchin) mentioned in the IAEA November 2011 report concerning explosives allegations. The IAEA through a spokesman thereafter blurted out that it wouldn’t take up Iran’s offer.

AP Article 8

Not without consistency, Gareth Porter and Robert Kelley then went into print taking issue with the IAEA’s decision not to go to Marivan. Porter published an article speculating that the reason the IAEA passed is that it has no evidence for any weapons-related activity–befitting the thesis of his previous magnum opus which claims that the Iranian nuclear crisis was concocted by the U.S. and Israel to confront Iran. Kelley’s contribution instead cited details from the IAEA November 2011 report and lamented that the IAEA chose not to chase them down in Marivan because–as Kelley has opined–these activities would not likely have been carried out at Parchin.

These authors didn’t mention that the IAEA might consider pursuing Iran’s Marivan gambit for another reason: Promoting the implementation of Iran’s Additional Protocol.

Iran’s AP includes Article 8:

‘Nothing is this Protocol shall preclude Iran from offering the Agency access to locations in addition to those referred to in Articles 5 and 9 or from requesting the Agency to conduct verification activities at a particular location. The Agency shall, without delay, make every reasonable effort to act upon such a request.’ 

That language might imply that the IAEA should go to Marivan if Iran invited inspectors to go there.

I talked to one safeguards aficionado about this. He recalled that he was in the room when a Board working committee conceptualized the AP back in 1996. Director-General Hans Blix, he said, “asserted that, if there is a claim that a state has carried out non-compliant activities at a location, the state could voluntarily call upon the Agency to assist in clearing its name by visiting the location on the state’s invitation and reporting what it found. Blix insisted that would be the best way to resolve such claims, and [that such a provision] should be in the AP… That concept was formulated as Article 8.”

Yesterday I asked Blix if that version of events was basically correct. He confirmed to me that it was.

“I remember vaguely a view I had that while any demand by the Agency for a special inspection would be perfectly legal, it would be somewhat dramatic and likely to lead to controversy, so a possibly less difficult path to inspection and facts could be if the state was given an opportunity itself to invite the inspection. I seem to remember that the special inspections we had had before the DPRK were in fact by such invitation. This might be an explanation for Article 8.”

On this basis, the IAEA might argue that, if a visit were held and it cleared up suspicion about Marivan, Iran might also benefit if it permitted the IAEA to see what it wants to see at Parchin and, along the way, have an interview with Mohsen Fakhrizadeh, a scientist and officer in the Iran Revolutionary Guard Corps who is suspected of having guided nuclear weapons work in Iran. So far Iran has not permitted the IAEA to go back to Parchin or meet with Fakhrizadeh.

How Much IAEA Leverage?

So why didn’t the IAEA grab Iran’s bait?

First off, Iran’s AP isn’t in force. Until it is, IAEA verification work in Iran is mandated by Iran’s Comprehensive Safeguards Agreement, the JPA, the Framework for Cooperation, and resolutions passed by the IAEA board and by the Security Council. AP Article 8 doesn’t apply.

Far more significantly, were the IAEA to agree to Iran’s terms for a one-off visit to Marivan, and if the IAEA failed to find anything, Iran would probably shut the door. That happened at Parchin, where the IAEA in 2005 failed to detect non-compliant activity and Iran then barred the IAEA from visiting a specific location it now wants to see.

Cheryl Rofer joins Porter in suspecting that the IAEA would find nothing at Marivan–but also for the reason that Iran could hide evidence which after a decade may have also eroded, not because allegations are groundless.

There may be internal deliberations concerning the IAEA’s authority and priorities. While UNSC resolutions endorse the IAEA’s pursuit of PMD-related activities in Iran, Iran’s CSA (and for that matter the AP) expressly endorse the IAEA’s authority to inspect as deriving from a nexus to nuclear materials. To my knowledge, no allegations have come forth that Iran used nuclear materials in any undeclared activity at Marivan. The IAEA may be more interested in pursuing allegations at Parchin if it has information suggesting that nuclear materials may have been involved in undeclared activities at that site.

The Kelley and Porter articles have resonated among some pundits and trolls who appear to share as an article of faith Iran’s claim that PMD is is a conspiracy of a big power and its allies abusing a multilateral agency to beat up on a recalcitrant adversary. Their argument follows the approach which Russia pressed home during the 2014 IAEA safeguards symposium in Vienna in October. There, Russians asked the IAEA again and again: “How do you know you are not being manipulated by sources giving you third-party information?”

This Russian question, as I said in Moscow last month, is a good question. But the correct answer doesn’t have to be that the IAEA’s use of third-party information must be poisoned by manipulation and bias. Amano knows that the IAEA’s impartiality is under scrutiny; it would therefore be premature to conclude that the IAEA is foolhardy and is being rope-a-doped by a dozen states sharing data alleging that Iran carried out high-explosives work. Lest we forget: None of the people speculating about what happened or didn’t happen at Parchin and Marivan have access to the IAEA’s current inventory of safeguards confidential data on Iran.

In the short term, it looks like Iran has maneuvered Amano into a corner. If the IAEA doesn’t go to Marivan on Iran’s terms, Iran’s spin doctors will claim that the IAEA is not cooperating to resolve PMD allegations. If the IAEA instead goes to Marivan, and finds nothing, Iran will declare the case closed.

Iran’s chess-playing with the IAEA, in parallel with its negotiations with the powers, is ultmately aimed at release of sanctions. If Iran is hiding evidence of activities related to nuclear weapons development, its moves will be designed to protect that knowledge, including at any site where undeclared activities may have taken place. The IAEA might tell Iran: It is up to the IAEA to decide where its safeguards resources are best put to use. If you want us to go to Marivan, then bring Iran’s AP into force now and assure us in advance that this won’t be a one-off visit. But as long as Iran views its cooperation with the IAEA as a bargaining chip in a negotiation with the powers for future benefits, it may not agree.

The IAEA isn’t powerless in this game. It needs to recall that  under UNSC/RES/S/1929 (2010), nuclear sanctions cannot be rescinded without an IAEA statement to the UNSC that “Iran has fully complied with its obligations under the relevant resolutions of the Security Council and met the requirements of the IAEA Board of Governors, as confirmed by the IAEA Board of Governors.”

This article was originally published in Arms Control Wonk.