Jeffrey LewisCovert Site in Iran

The United States, France and Britain have announced that Iran is building a second centrifuge facility at a site near Qom. Laura Rozen has QandA’s and talking points from the White House.

The general reaction is that “this changes everything.” My reaction is: Damn, Mark Fitzpatrick is money.

Also, this shouldn’t change anything. This is the scenario we’ve been warning about all along.

For some time, a few of us – including Josh Pollack and Andreas Persbo – have been arguing (with little success) that the public debate is misguided in its singular focus on breakout scenarios at Natanz. Is Iran 18 months away? How much LEU does it have? These were interesting questions but, to my mind, distractions. Natanz is the most watched site in the world. If the Iranians build a bomb, they will do it someplace else. Like Qom.

Josh Pollack did a wonderful job of tackling these issues in his post, Why Iran’s Clock Keeps Resetting (August 19, 2009) and over at TotalWonkerr, where he noted “One of the shortcomings of breakout lit so far may be its emphasis on on a single site. A hidden site is also a possibility…”

The real risk was always that Iran would construct a covert site other than Natanz. As long as Iran remains under the current safeguards arrangements, I wrote to a colleague this summer, we have “no confidence that Iran is not simply trucking centrifuge components to another location, buried deep under some mountain.”

That was just intended as a colorful way of explaining the problem of covert sites, relative to breakout. I didn’t know that there was in fact another site, actually under a mountain. And, I confess, I am pleasantly surprised that the IC caught the bastards – again.

But the revelation of a covert site ought to put the focus of our policy back on getting more access to the Iranian program to detect and deter the construction of undeclared facilities, rather than attempting to place arbitrary limits on declared facilities.

Worries of a Covert Site Grew After 2007

There were plenty of reasons to worry Iran was working on a second site. I started to worry until late 2007 when I read Nuclear Iran: How close is it? — a little remarked IISS Strategic Comment. The unnamed author suggested in passing that he wouldn’t be surprised if Iran were constructing a second, clandestine site:

The IAEA did not believe Iran’s statements that it did not pursue any work on the P-2 design from 1995 to 2002 and Ahmadinejad’s boast in April 2004 about research on advanced centrifuges belies this official claim. It would not be surprising to learn that Iran has a pilot P-2 plant – Iran has announced that it will ignore the routine safeguards obligation to make early declarations of nuclear facilities.

Well, I was surprised to read that. (When IISS makes a statement about Iran, assume that that the Great Fitzpatrick has spoken. It always makes sense to read Mark very, very closely.) So surprised, in fact, that I emailed some friends.

Then the 2007 NIE came out. If you read it carefully, it made the same argument:

We assess with moderate confidence that Iran probably would use covert facilities—rather than its declared nuclear sites—for the production of highly enriched uranium for a weapon.

There it was clear as a bell: The IC wasn’t worried about what they see at Natanz; they are worried at what they can’t see elsewhere. (I wonder what gave them “moderate confidence.”)

Yet, for some reason, most reporters and policy analysts continued to focus on breakout at Natanz.

I should point out that our friends at ISIS focused on a much more sophisticated breakout scenario in which Iran would divert LEU produced at Natanz to a second, clandestine facility. (We just don’t know yet if Iran intended to divert LEU from Natanz, as ISIS worried, or uranium from earlier in the fuel cycle.) And, to his credit, after I made an irritated comment on the blog about the focus on breakout, Bill Broad called me to ask why I though that was the wrong story. But those are the exceptions.

I don’t know why breakout obsessed people for so long – maybe because wonks could make calculations in kilograms and days. These have a sense of precision, though false, that the abstract possibility of a second site lacked – until today.

Second Site Points to the Need to Improve Monitoring

Maybe, the revelation should change some things — like shifting the international community’s approach from trying to constrain Natanz to the goal of improving monitoring of the entire Iranian nuclear program. This is the conversation that Jackie Shire and I had back in February (around the 2:30 mark):

If we decide that we are more worried about clandestine facilities than breakout, there are profound policy ramifications. Instead of bargaining with Iran to cap or suspend work at Natanz and other declared sites – which will be like playing whack-a-mole – we should have been bargaining about gaining extraordinary access for the international community. That means, among other things, getting the IAEA into Iran’s centrifuge workshops and talking to engineers and technicians. A colleague suggested the “safeguards equivalent of a colonoscopy.” That’s an apt metaphor.

The ability to detect clandestine facilities is essential to enforcing Iran’s compliance with its safeguards obligations. Otherwise, the Iranians will keep digging holes in mountains. Our best chance of preventing an Iranian bomb depends on policymakers in Iran believing that certain steps toward a bomb will be detected and expose the regime to mortal peril.

So far, that has not been our focus – there has been a lot of hullabaloo about forcing Iran to accept “zero” centrifuges. But even if Iran shuttered Natanz and, now, Qom, without better monitoring arrangements than are currently in place, we would have little confidence that Iran wasn’t building yet another clandestine centrifuge site.

Fortunately, we’ve been either that good or that lucky twice in a row now. If the New York Times is to be believed, Iran opened up about the facility largely because it became clear in Tehran that the site’s secrecy had been compromised. The intelligence community deserves praise for catching Iran twice now; let’s hope Tehran’s third time is not a charm.

The way to make sure that Iran can’t move secretly toward a bomb lies in much better access in several areas: The international community has demanded access to the site near Qom, which is the place to start. But it also needs regular, intrusive access to Iran’s centrifuge workshops and other suspect sites. And it needs access to Iran’s personnel, including those who worked in what is believed to have been a clandestine program at Lavizan-Shian in Tehran. Such access is far more important than arbitrary limits at Natanz, which Iran seems unlikely to accept in any event.

It’s time for Olli Heinonen to put on the latex gloves.


  1. Ataune (History)

    Secrecy is not the same as Iran being in breach of its safeguard obligations. I am curious to know what the verdict (or your thought) on the latter issue is.

  2. nick (History)

    Shire’s claim that why build Natanz when you don’t have a reactor is not correct. It will take roughly another 5 years to complete all 54,000 IR1s to produce the annual replacement fuel for a single 1000 MW reactor (roughly 25 tons of LEU per year). This is assuming all the imported parts such as: vacuums, and special oils are readily available, which is doubtful. Even if the accumulation of 25 tons is done before Iran’s Darkhovin (domestically designed) is finished, they can always sell the oxide in the open market, that is transfer the LEU to potential customers overseas in order to allay any concerns, this can be part of the negotiations. Finally, it should be noted that mixing SEU (about 1%) with natural UO2 will increase the neutron budget and make it more efficient to convert uranium to energy for a HWR such as the CANDU design.

  3. mark hibbs

    This revelation should, if we are thinking clearly, change the calculus of how we deal with Iran in the direction you suggest: There is no way that a ban on enrichment in Iran can give us more confidence that there isn’t a secret program. The more we try to snuff out Iran’s centrifuge infrastructure with sanctions, the more we will drive it underground. Some IAEA people who should know say that that is what happened when Israel bombed Osirak. Iran already HAS the technology. They can be slowed down with vigilent export controls, but there is no track record of outside pressure ending a program at THIS stage of development in Iran. South Africa is a terrible example. Without the end of Apartheid rule, the supremacists would never have given it up. Pakistan? I have argued elsewhere that those who argued that the US and others could wade in there and shut down Pakistan’s program anytime after they set up production plants including enrichment plants were informed by an inflated and unrealistic notion about the projection of US power in that region. “No enrichment in Iran” is a non-starter unless you are prepared to go in there and take the place over. Iran has as least as many cards to play as Pakistan did during the Cold War. Invading Iran with the object of forcing regime change is clearly not an option, so it would be more prudent to think about doing things to deter Iran from re-starting the weapons program it looks like they shelved in 2003. It may turn out that the decision by Iran to have shelved that activity then was taken because of Iran’s current assessment of its political risk because the IAEA and member states were in fact penetrating the program. If so, that would argue that getting more access to Iran’s activities is the proper approach, not shutting down every centrifuge in Iran. If anything this new development should underscore that a primary goal of international diplomacy should be to get Iran to implement the AP, and to put the IAEA in the position of understanding as much as possible about Iran’s centrifuge production capabilities. Right now, the IAEA doesn’t have a lot of confidence in what it knows about how many centrifuges Iran can build. If they were to say now, okay, we’ll shut them all down, the IAEA on the basis of its present information would not be able to have confidence that Iran couldn’t continue the program secretly.

  4. archjr (History)

    There is no time like the present for a special inspection of Qum – like, tomorrow. There have only been 2 requested – in Romania and the DPRK – and none since 1993, and the concept was tailor-made for just this situation (see Carlson and Leslie, 1995 – link text).

    The special inspection authority under 153 agreements, if used as it should be, is the sine qua non for safeguards until the AP becomes the norm. I would hope Iran would accept quickly, particularly lest the IAEA become as marginalized as it did under the Agreed Framework with the DPRK. And I hope the big powers will allow this to happen, rather than taking longer to come up with a jerry-rigged inspections regime as occurred with North Korea. Nothing could better underline the NPT’s safeguards requirements and the IAEA’s considerable expertise and unique role.

  5. DZ (History)

    “And, I confess, I am pleasantly surprised that we caught the bastards – again.”

    Can you please specify what you mean by the above statement? who is we? and how can you claim to have caught them when they announced the existence of the proposed plant to the IAEA on the 21st? and what did we catch them doing? making a building that will be used for an enrichment facility… The NPT is clear that Iran or any other country has no obligation to tell anybody anything until they are 6 months away from introducing nuclear materials.

  6. bruno

    @mark hibbs:

    You’re absolutely right that the goal should not be “zero enrichment in Iran”, but that is not what the P5+1 are seeking. They are seeking assurances that Iran does not have military-related activities. The suspension of enrichment is means, not an end-state. The strategic goal is to induce the Iranians to make the choice to give up any short- or medium-term military option, and demonstrate it by implementing the AP and other confidence-building measures as deemed necessary. And this is where there is a slight difference between the US approach (which, so far, seeks zero enrichment on Iranian soil) and the EU approach (which has always been that enrichment on Iranian soil is not a problem in itself as long as trust is there).

  7. LAC (History)

    The NYT story says that this Spring we saw P-1 centrifuges being moved into the Qum site. Is there anything that would make it unlikely they are trying out P-2 technology somewhere else?

  8. Avner Cohen (History)

    I think that the either/or way—either no enrichment or additional protocol and more safegaurds— that this discussion frames the Iranian issue missed the real issue.

    To begin with, one must realize what is the role of the nuclear issue for Iran. The nuclear issue has been for Iran the perfect international subject to allow Iran assert itself. It reflects identity issue of the regime and its leadership. In other words, the nuclear issue, and especially enrichment, is the best card for this Islamic Republic of Iran to assert itself in this world.

    It is not so much the bomb per se that Iran is pursuing but rather the political attention, prestige, influence, respect, etc, that goes with the nuclear card that Iran is after.

    Based on that I do not think Iran would give up enrichment or accept a unique form of tailored supervision. Still, it does not mean necessarily that Iran really wants to be a true nuclear weapons state. They want the aura that goes with that status, and they can do it by becoming a truly threshold weapons state which is viewed by others as a weapons state.

    Furthermore, there would not be “no enrichment” outcome without doing it in some regional context, that is, including Israel. I can think of the region as zone without any dangerous technologies in the region, i.e., enrichment and reprocessing. The extreme difficult of moving towards that is that Israel and present iran cannot communicate. The problem is that the Iranian regime has helped Israel to demonize itself, so I cannot see Israel getting to such a deal with the present regime.

    Contra Mark, I also think that the international community has more power and leverage than he thinks. I do not know of any case that could go nuclear in total defiance of the interantional community. And this applies to Iran as well. Iran CANNOT defy the will of the entire international community.

    Mark, your over pessimism is unwarranted. There is no case that a stae was able to defy the will of the international community on nuclear matters. The SA, PAKI, were not cases of those countries against the international community.

    Ultimately, if united, the will of the international community, especially thorugh the P5, should be much stronger than Iran.

  9. Michael Lieberman (History)

    Interesting post and very much agreed. One question: Several times you state that to get transparency re additional sites we need access to Iranian scientists. I’m just wondering why you think they would be forthcoming and reliable. I’ve worked on several investigations relating to sanctions-busting and bribery where we’ve interviewed people that you wouldn’t say have incentives to be truthful, yet you get enough information (based on assertions, discrepancies, holes etc.) to piece bits together to add to your understanding. So maybe that’s what you have in mind. But it seems that if the Iranians did have additional facilities that they would get any scientists they allowed access to to sow disinformation or be evasive, not reveal what the government wants to hide. So how do you see these interviews working? Thanks,


  10. raghar (History)

    A second and third sites only improve resilience against Israeli attacks.

    Perhaps they should show that successful nueclear test before badmouthing Iran.

  11. mark hibbs

    Bruno: I don’t disagree with you fundamentally. My problem in part is that what you are saying has gotten lost in the day-today frenzy over Iran because it is dominated by a media- and cyberspace-driven fixation about “Does Iran have enough centrifuges operating to get a bomb’s worth assuming they divert it?” As you point out, longer term, that is NOT what we should be focused on. The confidence at the IAEA Department of Safeguards about detecting a future diversion of material from Iran’s production of LEU is in fact quite high. So why is everyone focused on this scenario? I did a piece a couple of years ago where I wrote that, if I were to go into CIA headquarters, I would see a huge map of Iran and a very large number of analysts combing high-resolution aerial photos covering a huge amount of the country looking for anything that resembled a second enrichment plant. They under stood that, with Iran facing external threats, that that was the scenario they had to be most worried about. Unfortunately, there has been some intellectual lazyness implying that if we slam Iran with more sanctions, we will succeed. What Jeffrey points out in his post is that even if the sanctions happen (and conversations I’ve had with people in the US the last 48 hours suggest that they very well may not happen) sanctions will not stop Iran from engaging in secret activities, in fact, sanctions may be counterproductive to that outcome.

    Avner: I hope you are right. But what I have heard the last 48 hours is that there is currently no common ground to get a hard P-5 consensus on this. And getting to sanctions, to reiterate, will in the short run drive Iran’s program further underground. You say the “international community” must act. Well, what do you want the “international community” in fact to DO? Punish Iran? Isolate Iran further? What happens then? Does Iran’s leadership have a conversion experience and see the light, and give it all up? Nowhere have I seen an analysis which carefully calculates that, if these sanctions are slapped on Iran, Iran will cave because the Mullah regime has no sustainable alternative. I agree that there was potentially less unity among the powers in reacting against Pakistan’s nuclear ambitions since the 1980s (I’ve in fact argued this in print as you know, against those who since 2003 have insinuated that the US or the ‘international community’ could ‘take out’ AQ Khan and his network and thereby fix the problem in Pakistan). But don’t forget that Zia replied to his adversaries that Pakistan would “eat grass” to match India. What are Iran’s priorities here? Many of those who say “let’s sanction Iran” aren’t thinking clearly about this. What does the US, Israel, and others have to give Iran for Iran to go the route that Bruno I think correctly argues we want to see Iran embark on.