Allison: U.S. Policy on North Korea ’Abject Failure’; Iran May Be Closer to Nuclear Weapons Than U.S. Thinks

Graham Allison, a leading expert on nuclear terrorism, says the Bush administration policy toward North Korea of "threaten and neglect" has been a failure. He also warns that faulty U.S. intelligence may be underplaying the threat posed by Iran’s nuclear program.

June 20, 2006

To help readers better understand the nuances of foreign policy, CFR staff writers and Consulting Editor Bernard Gwertzman conduct in-depth interviews with a wide range of international experts, as well as newsmakers.

More on:

North Korea


Nonproliferation, Arms Control, and Disarmament

Graham Allison, a leading expert on nuclear terrorism and director of Harvard’s Belfer Center for Science and International Affairs, criticizes the Bush administration’s policies toward both North Korea and Iran. Allison says new reports of North Koreans’ preparation for a missile launch return a spotlight to the "abject failure" of the administration’s policy of "threaten and neglect" toward Pyongyang.

On Iran, Allison says Tehran has more than two known nuclear processing facilities at Natanz and Isfahan, and as a result, may be much closer to producing a nuclear weapon than U.S. intelligence estimates. Allison was assistant secretary of defense in the first Clinton administration, where he coordinated Defense Department policy toward Russia, Ukraine, and the former Soviet Union. He says of Iran, "What about the other covert sites at which they may be seeking to do enrichment? Do we believe that there are such sites? And what is the likelihood? I would say nearly one hundred percent."

North Korea and Iran are very much in the news these days. North Korea is reported to be on the verge of test-launching a missile that could reach the United States, and Iran is considering whether to suspend its nuclear enrichment in return for negotiations on a set of incentives. How good is our intelligence on these two countries?

The best window that people who are not privy to all of the facts have is provided by the [Lawrence] Silberman-[Charles] Robb Commission on Weapons of Mass Destruction intelligence capabilities, which came out with a report last year. The bottom line from this report, which was done by a very distinguished panel of people, was that the poor performance in estimating Iraq’s weapons was not unusual or atypical. It might be comforting to think, "Gee, we blew it, but this was an exception, this was not unusual because all of us make mistakes sometimes." But their conclusion was that the wrong judgment about Iraq’s WMD was not unusual. They don’t mention Iran and North Korea by name, but they say our information about other states of "nuclear interest" was no better.

But on the North Korea issue we’re not talking so much about nuclear arms as about a missile launch, which I guess is easier to observe by satellite.

I think we’re talking about two things in the case of North Korea. First, what the North Koreans’ preparation for a missile launch does is bring the spotlight back to something the administration has assiduously and effectively sought to avoid—because one can’t avoid the conclusion that President Bush’s policy toward North Korea has been an abject failure. Whereas when the administration came to office, the intelligence community assessment was that North Korea had two bombs’ worth of plutonium, its conclusion today, as reflected in the newspapers, is that North Korea has ten bombs’ worth of plutonium and a running production line.

And secondly, with respect to the missile, North Korea has conducted a number of missile tests, but in 1998 began a moratorium on further tests. Why they are preparing—if that’s what they’re doing—for a further test, nobody knows. But what incentive have we given them not to do that? The administration has mostly been following a policy of threaten and neglect, and then, in recent years, repaired to the six-party talks. But in terms of either positives for North Korea to get them not to do things, or minimal negatives to prevent them from doing things, the administration has not been playing a very agile hand.

Who knows what motivates [North Korean leader] Kim Jung-Il? But I would say if we stand back from the game and ask what he seems to want, it includes, one: the survival of his regime—the opposite of Bush’s objective, which is the collapse of his regime. Two, nukes to fill his nuclear arsenal and expand it. And three, to successfully extort payments from the other parties for anything he does, including attending each meeting of the six parties.

Our objective is to prevent an expansion of the nuclear weapons arsenal and change the regime. If you ask who has done better in this game over the past six years, I think the answer is pretty clear. We say, and the press says, "Well, this guy Kim Jung-Il is strange, and, you know, who could understand [his] behavior?" and so forth. But I have compared this to a situation where you had two Martian strategists watching from a distance, and one would say to the other, "Hey, one of these parties to the six-party talks’ behavior is strange." And the other one would say, "Well yes, I noticed that too, which one?"

So you are saying the U.S. behavior is not exactly exemplary.

Well, you’ve got one global superpower, and one Stalinist basket case that starves its people and is staggering to stay alive. And now six years later you’d say, well, he’s done pretty good, even though he may be crazy like a fox.

What about the Iran situation? In an article in Yale Global Online, you’re postulating that U.S. intelligence on Iran may be the opposite of the intelligence we had on Iraq. Instead of overestimating it, the intelligence community may be underestimating Iran’s potential for nuclear weapons.

I argue that if the conclusion of the Silberman-Robb Commission is correct—and I believe it is—that the intelligence about Iran is no better than the intelligence about Iraq, then it follows almost inescapably that we are in some trouble because what we’re trying to do is achieve an outcome, but operating through a very cloudy glass. And the focus of the attention is two facilities at Natanz and Isfahan, which are the declared overt facilities. We’re seeking to have Iran announce that it will essentially suspend the activity there, enter into this set of negotiations, the conclusion of which presumably would be a permanent or long-term suspension of activity at these facilities, and therefore no enrichment there.

So they’ll have to sign up for something that says no enrichment anywhere. What this article does is say, "What else might conceivably be going on here?" It’s not meant to say I know what the answers to the questions are, but it does present some other questions I haven’t seen adequately aired. First, what about the other covert sites at which they may be seeking to do enrichment? Do we believe that there are such sites? And what is the likelihood? I would say nearly one hundred percent. And there, maybe, are the advanced P-2 centrifuges they got from Pakistan because we believe A.Q. Khan [the Pakistani black market nuclear merchant] sold them some P-2 centrifuges [machines crucial to enrichment] and P-2 centrifuge designs. But where are they? The International Atomic Energy Agency (IAEA) says those are so far unaccounted for.

The second major question is, what’s going on at Natanz? The 168 cascades [a series of centrifuges] that we’re trying to get stopped, is that the leading item of Iran’s nuclear program, or could it conceivably be a distraction? In other words, if you were a magician, what you do is get people to look at one hand, while with the other hand you’re putting the rabbit in a hat so you can pull it out.

It’s possible they’re learning in the Natanz facility how to run a centrifuge. That’s hypothesis one, and that may be true, and I hope that’s true, but hypothesis two, which certainly has to be considered, is that they actually are proceeding on two separate tracks, one of which is mainly about keeping our attention and focus, and one of which is about doing their real business.

It’s conceivable there is a separate, secret cascade of centrifuges operating somewhere else. Now the argument could be, "Why in the world would they go through such an elaborate game?" One reason might be because this is a very good way to keep the superpower they fear distracted. I’m not arguing that’s the conclusion. I’m simply saying that hypothesis is not excluded, and I would like to know the intelligence community had some basis for excluding that hypothesis.

But the bottom line is you think the estimate that Iran is five to ten years away from a nuclear weapon was probably inaccurate, that it’s closer to two to five years.

One, we made a judgment about Iraq, and we know what the facts were there. Second, we have a very distinguished panel of quite estimable people who say the party that told us about Iraq, and which is now telling us about Iran and North Korea, doesn’t know much more about the latter than it knew about the former. So this is like a newspaper reporter who has a source. The source tells you something and you print it the first time. Now the basis for your next story comes from the same source and the same person who confirmed it. You would be suspicious, right? Thirdly, it’s quite possible Iran is five to ten years away and that the activity going on at Isfahan and Natanz could continue to take, say, five years.

Alternatively, if there’s a second secret, undisclosed site, this estimate could be greatly exaggerated. First, understand this estimate may not be correct—I wouldn’t take it to the bank. Second, look and see who else has competing estimates. The Israelis have an estimate that says it’s about six months to a year away, in the same way that we seem always to have one that says it’s five to ten years. Some Europeans are now saying it could be a little bit sooner. My impression is the Russians think it may be sooner. So what I would do from that is two-fold. I would first have a Team B-like exercise that would at least look at all these options. And second, I would see what I could learn from the Russians, because I think they’re most likely to know the most about these topics because of their history [with Iran, including building the Bushehr civilian nuclear reactor]. Because of the worsening relations in general between the United States and the Russian government, and because the Russians believe they haven’t gotten anything out of this—they’ve been giving but not getting [anything] in intelligence exchanges—it’s unlikely there’s been a full cooperative effort to share the best information that might be available.

In your article, you suggest Bush should ask Russian President Vladimir Putin to set up a collaborative task force. Would the G-8 Summit in St. Petersburg next month be a good occasion?

Yes, that would be a good time. When he sits down with Putin, Bush could say, "You guys probably know more about this than we do. We want to ask you for a favor." We understand the people we’re asking to cooperate are mostly suspicious of us, and I suspect it will cost something. But if this is the most urgent threat to the United States and we’re thinking about options including war or an attack on the facilities, then I’d be prepared to pay some price to try to clarify the crystal ball.