The nuclear negotiations with Iran started to make a lot more sense to me when I realized that a lot of the disagreements are over things that aren't actually that crucial.
In Washington, for example, you hear people insist — insist! — that any final deal make Iran's "breakout time" as long as possible, meaning that an agreement so limits Iran's nuclear infrastructure that if Iranian leaders decide to construct a nuclear weapon, it will take them nine months rather than six, or 12 rather than nine. And American negotiators indeed appear to be holding up talks over this issue. But no one can really articulate why that's so important, what happens in those three additional months. Meanwhile, Iranian politicians in Tehran and Iranian negotiators in Switzerland are making their own bizarre-seeming demands.
To try to understand what was going on, I spoke to Jeffrey Lewis, the director of the East Asia Nonproliferation Program at the Monterey Institute of International Studies (Lewis also runs an excellent arms control blog network and arms control podcast). He explained what's actually important to prevent Iran from building a nuclear bomb, what the negotiators in Switzerland are fighting over, and why those two sets of issues seem to be so different.
A big reason that the conversation around the Iran nuclear deal can feel so confusing is that it often focuses, Lewis says, on the wrong things. By obsessing over things like breakout time and the number of centrifuges Iran is allowed to have, we're totally misunderstanding how and when Iran would try to build a nuclear bomb — and we're also making it harder to reach an agreement.
What follows is a transcript of our conversation — which left me feeling like I really finally understood what was going on with Iran negotiations, and what has to happen to reach a deal — edited and condensed for clarity.
Max Fisher: So the impression I've gotten is that the two main goals of the Iran talks, at their most basic level, are, first, to block Iran's pathways to a nuclear weapon, in other words reduce their ability to make a bomb, and second, to make it tougher for them to build a bomb, so that if someday they wake up and decide that they want one, it will take them longer. Is that wrong?
Jeffrey Lewis: Totally.
Max Fisher: So what's the right way to think about it?
Jeffrey Lewis: I'm way more worried about, like, the covert facility problem. I'm not as worried that they're going to use Natanz [or another known nuclear development site] to break out. I think if the Supreme Leader [Ayatollah Ali Khamenei] wakes up one morning and is really feeling it, they're gonna dig another hole under a mountain someplace.
So, to me, the value of the agreement is not just "does it lengthen the breakout time" but does it make less likely they can build a secret facility.
Max Fisher: The way Americans have talked about the Iran deal is that the ultimate goal is to limit Iran's nuclear infrastructure, so that if Supreme Leader Ali Khamenei wakes up one morning and decides he wants to build a nuclear bomb, it will take him a long time to build it.
It sounds like you're saying that's the wrong way to think about it, because it assumes Khamenei would use publicly declared facilities and equipment to build a bomb, and those facilities would be under pretty heavy monitoring. Rather, he would do it by building new secret facilities somewhere, where he could develop a program unencumbered. So in that thinking, the goal of the nuclear agreement should be to get nuclear inspectors so up in Iran's business that they can't cheat on the agreement without us knowing. But how do you do that?
Jeffrey Lewis: Some of it is through traditional, safeguards stuff. Like giving the [International Atomic Energy Agency, which manages nuclear inspections] access to Iran's centrifuge workshops.
We have this crazy situation right now where the IAEA has basically no access to the places where the centrifuges are made. And so Iranian put those centrifuges on a truck, and if they drive them to [a publicly declared nuclear site such as] Natanz and install them there, then they're safeguarded. But, if they, you know, drive them to some hole in a mountain then, no, they're not safeguarded, we don't see them. The Iranians have provided some limited access to the centrifuge workshops. But [expanding that access] would be a big achievement.
And there are just other things. There is this question of dual use goods [that could be used for peaceful purposes or for a clandestine nuclear program] that they import; there is a whole bunch of stuff sanctions will come off of. And it would be good to have, like, a registry or database of those things, so we can at least check to see if they are importing, say, specialized ball bearings, and make sure those are not going to a facility under a mountain.
What we want is for Khamenei to know, for certain, that if he wakes up and decides to build a bomb, we're going to know about it, and there's going to be a showdown.
And that's not perfect, but this is real life and that's as good as you get. You don't want them to be able to get a bomb the way Pakistan did, which is little by little, in itty-bitty pieces, and by the time we really get around to dealing with it, it's too late.
Max Fisher: It seems like the key things, then, are having an alert system to know exactly what's happening inside of Iran, and then having the enforcement mechanisms in case they cheat.
On the alert system part, you can have lots of inspections and safeguards so that we know if Iran is cheating, and Iran knows that we'll know, so they don't think they can get away with it.
But how do you handle the enforcement, which I feel like I have actually heard very little about in the talks? How do you tell Iran, "If you cheat, not only will we know, but it'll be bad for you?"
Jeffrey Lewis: Because we never do that in treaties or international agreements. No treaty ever, ever has any enforcement mechanisms in it, ever. There has been some talk about putting some automatic enforcement in the NPT [the Treaty on the Nonproliferation of Nuclear Weapons, barring illegal new nuclear programs worldwide], but nobody likes that.
Max Fisher: Why?
Jeffrey Lewis: It's easy to support automatic enforcement mechanisms [on nuclear treaty violations] until somebody points out that the South Koreans had a safeguard violation. And then [people inevitably say], "These things should be done on a case-by-case basis."
So these treaties never really have an enforcement mechanism, which is, from a legal perspective, kind of weird and kind of a bummer but totally understandable in a world of states that jealously guard their sovereignty.
Max Fisher: I wonder if that's why there's been so much political focus here in Washington on the number of centrifuges or the breakout time. Because those are quantifiable, discrete details in the agreement, and you can debate them.
But it's hard to have a debate about, "How do we make sure Khamenei believes we really mean it when we say we'll punish him if he cheats on the agreement?" That threat of enforcement is really important, but it's not something you can debate about how to put in the text, since it's not something that can go in the text at all. So instead of talking about enforcement, we're talking about less important things like the number of centrifuges.
Jeffrey Lewis: Right, I think that's right. So here's the thing about the conversation focusing so much on breakout time [and how to maximize the amount of time it would take Iran to build a nuclear bomb]. I don't think Khamenei wakes up and says, "Ugh, did you see the breakout calculation today? Nine months! That's too much. If we can get it down to about six and a half months, then I'd be willing to go [build a nuclear bomb]."
He won't do that. But since we can't get inside his head and know what he's thinking, then we're going through these bizarre DC wonk thought experiments to satisfy ourselves. "If we make it a year, then he'd have to know that we'd have enough time to respond." But a year is a totally arbitrary amount of time. It's one of these things where we try to take the abstract and unknowable and impressionistic and we try to put a number on it.
Nobody ever specifies what they would do with all this time. And keep in mind, the breakout calculation is for one bomb.
That's why I'm much more worried about covert [nuclear] sites. Because if you're Iran, you wanna do what Pakistan did, which is to get a whole bunch of fissile material [prepared in secret]. So that by the time the world really figures out [that you're running a clandestine weapons program], they're like, "Oh, I think they have an arsenal. How many? I have no idea." And then it's an unsolvable problem.
Max Fisher: Is there a counter-argument that you do actually need the long breakout time, because if Iran does decide to cheat and to break out to a nuclear bomb, it might take us a few months to find out? So the longer that breakout time, the longer we have to find out before it's too late?
Jeffrey Lewis: People will game this out. They'll say, "Well, maybe Iran would start [breakout] the day after the IAEA was there. And they visit every month, so we'd know a month and a half in, but it would be ambiguous they'd stall for three weeks." I get that people want to give themselves some time.
But I don't think it's time that is going to cause a leader of Iran to [decide to break out to a nuclear bomb]. I think it's the thought that they could get away with it. A leader is going to make a gut call: "Can I get away with this?"
I'm trying to imagine, like, the little conference room with the supreme leader where he's looking at the calculations on breakout time. And it says, I don't know, 238 days. And he's going to decide whether or not to build a nuclear bomb based on moving that number. I've just never seen a government that works like that.
I care about whether the IAEA can look at Iran's centrifuge workshops and can make sure that they're getting information about all of the different mining locations, so there's not another source of natural uranium that could be enriched some place else.
So the agreement will have verification provisions. But our confidence in the agreement will also be based on an assessment of our intelligence capabilities. And give the intelligence community credit. I mean, they caught Iran with [secret nuclear sites at] Lavizan, Natanz, and Fordow.
Max Fisher: There also seems to be a lot of fighting over the number of centrifuges Iran is allowed to have — that's something Netanyahu is really upset about — and over how much research and development they're allowed to do. The New York Times reported that a big remaining disagreement is over how much R&D Iran can do in the final five years of the 15-year agreement. How important are those?
Jeffrey Lewis: This is one of the reasons I don't like the focus on breakout time. So, the IR-1 centrifuges that the Iranians have are crappy centrifuges. This stuff is old. We're acting like, "Oh, the Iranians are developing these centrifuges!" This is 1970s stuff.
But the degree of technical improvement on them that is possible is really, really big. It's not like a super-advanced centrifuge would be just 50 percent better — we're talking orders of magnitude more capable. So if I were a really clever Iranian negotiator, I'd argue that 1,500 centrifuges is actually enough, but I'd insist on keeping R&D going. And so then Iran would have fewer centrifuges, but they'd be so much more powerful.
So there's some effort to try to put some, I don't know if I would call them limitations, maybe conditions or constraints on R&D. Like, you want them to only sort of do it at a few places, you want to make sure you know about it. But verification for that is hard.
Whatever number of centrifuges you agree to — I heard 6,000 last — [the important thing is] that they're the old generation, and that you're not letting them deploy new stuff.
Max Fisher: Let me ask about you another issue that's getting a lot of attention, about what to do with Iran's stockpile of low-enriched uranium. The concern is that Iran could decide to take this low-enriched uranium and then start turning it into highly enriched uranium for a bomb.
So the Americans want to solve this by having Iran ship all of their low-enriched uranium abroad, or something like that, and it's become a big sticking point.
Jeffrey Lewis: Oh yeah, this makes me crazy. So, okay, if the Iranians have a bunch of low-enriched uranium sitting around, they could conceivably feed it back in [to enrich it until it was weapons-grade], and that shortens their breakout time, right?
But, why do that? Why take material that's under IAEA safeguard and pop off the seals? And then have the IAEA show up and be like, "Hey, where'd all it go?" That, to me, would be so dumb. Like, if I'm the Iranians, I'm going to try to find a source of uranium that's off the books, and I'm going to do this in secret.
I don't know if you've ever been to an enrichment facility. But there's a room and like all these casks of uranium and they've measured them and they have all these little IAEA tags and seals on them. So if the IAEA shows up and is like "Hey, where's cask 472839?" "Oh, uh, I think it's over here in a closet and oh gosh we can't find it." That's gonna trigger a huge crisis.
Max Fisher: This makes me realize that, in my mind, I've sort of separated out the negotiations in Switzerland from the problem of domestic politics in the US and Iran. The way I, and I think a lot of people, have been thinking about it, is that maybe they'll reach a deal on the technical merits in Switzerland, and then after that you have a separate process of selling the deal politically in Washington and Tehran.
But it sounds like those processes are actually happening at the same time: domestic political pressures in Washington and Tehran are pushing the American and Iranian negotiators to demand unreasonable things, or to insist on provisions that are nice to have don't really matter so much. And that's how domestic politics could make a deal fall apart.
Jeffrey Lewis: Yeah, I think that's precisely the problem. The Iranians are going to go home, and they're going to have to say, "Hey, this doesn't really impinge on our nuclear program at all." We're going to have to come home and say, "This is a deathblow to the Iranian nuclear program."
It's not just the rhetoric of it; it's going to be the concessions that you demanded to prove this. It's what makes me really frustrated, because my guess is we could get a lot more transparency out of the Iranians if we weren't spending our time demanding they do stuff like ship the uranium out of the country.
If I were an Iranian hardliner [who wanted to torpedo any nuclear agreement with the US], that's a really useful symbolic humiliation. "Oh, we had to send all our uranium that we spent billions of dollars to acquire, we had to send it away." That's a great talking point if I'm a hardliner.
Max Fisher: But it's not just the Americans pushing for unreasonable things, or demanding things that are not actually that important but are holding up the talks, right?
Jeffrey Lewis: A lot of the Iranian demands are nutty. They don't need Natanz. [Iranian negotiators have reportedly demanded the right to continue using their hardened facility at Natanz.] Natanz is not big enough, even with 50,000 centrifuges, to be like a useful plant, given the centrifuges they have. It's unclear to me that Iran will ever be a competitive player in the international market for enriching uranium.
The entire Iranian program is like a national airline. It's this incredibly expensive, preposterous thing that makes people feel better about their country. But there is just no reason that they need to do this.
If you took out nationalism and people's testicles, you would look at the centrifuge program and be like, "Ah, this is really expensive and causes us a lot of anxiety, do we really need to be doing this?" And the answer would be, "No."