I’ve been writing the same blog post for four days. Time to just publish it.

Thanks to McClatchey’s Warren Strobel and a trio of articles by the New York Times’ Helene Cooper
(1, 2 and 3), we now know that North Korea has declared that it produced 37 kilograms of plutonium — and denies that it reprocessed plutonium in any significant amount before 2003.

I hope they are lying.

This is precisely the same dispute that led to the first crisis in 1992 — did North Korea clandestinely withdraw some or all of the fuel roads in 1989 and reprocess them secretly?

North Korea has admitted to three reprocessing campaigns. The question is whether the 1990 campaign resulted closer to 90 grams of plutonium (North Korea’s claim) or 9 kilograms (the upper bound of the US estimate).

  • Campaign #1, March-May 1990 The North Koreans claimed in their May 1992 IAEA declaration to have separated 62 grams from about 90 grams produced in 86 broken fuel rods. The IAEA found many discrepancies in the North Korean declaration, suggesting that North Korea had reprocessed more batches of plutonium over a longer period of time (1989-1991) than declared. It was, as an IAEA official told David Albright, as if “North Korea had presented the IAEA with a pair of gloves but one was red and the other was green. The IAEA now had to look for the missing red and green gloves.” The IC judged that North Korea had secretly unloaded the Yongbyon reactor in 1989 and reprocessed enough plutonium for “one, possibly two” nuclear devices — reportedly corresponding to an estimate of 8-9 kilograms of plutonium.
  • Campaign #2, January-June 2003. In 1994, North Korea unloaded 8,000 fuel rods. These were canned and placed under IAEA surveillance until 2003 when, after the collapse of the Agreed Framework, North Korea reprocessed them. Albright estimated the amount of plutonium recovered at 20-28 kilograms.
  • Campaign #3, April-August 2005 (Dates are approximate.) North Korea unloaded another 8,000 fuel rods in spring 2005 and reprocessed them over the summer. Albright estimated the amount of plutonium at 14-17 kilograms.

These numbers are pretty rough — I make it 40-60 kg of Pu if there were three “kilogram-sized” campaigns, 30-50 if there were two. That, coincidentally, is the basically the range reported by Glenn Kessler, who wrote “The new U.S. estimate is expected to be between 35-40 and 50-60 kilograms.”

So the bottom line — as Warren Strobel reports — is that North Korea still denies a clandestine reprocessing campaign in 1990:

Thousands of pages of nuclear documents submitted by North Korea earlier this month cast doubt on a U.S. intelligence estimate of how much weapons-grade plutonium the secretive communist country has been able to amass, U.S. officials and a leading private analyst said Wednesday.

An initial review of the documents, they said, provides no evidence that communist North Korea covertly extracted plutonium, which can be used to make nuclear weapons, from its reactor complex at Yongbyon before 1992.

[snip]

David Albright, a former United Nations nuclear inspector who consults frequently with the U.S. government, said the reactor records turned over by North Korea are “consistent with what they’ve said.”

The CIA’s contention that Pyongyang extracted plutonium prior to 1992 “is not supported in the record,” said Albright, the president of the Washington-based Institute for Science and International Security. The trove of documents “is internally consistent, and to forge it would be tremendously difficult,” he added.

So here we are — it’s fifteen years later, Bob Gallucci has shorter hair and Dan Poneman has shaved off his ‘stache. North Korea has produced at least 37 additional kilograms of plutonium and exploded a nuclear device. Yet we are still arguing over the May 1992 declaration.

If — and I say if — the North Korean declaration is accurate, than the Bush Administration has a perfect matched pair of catastrophic blunders based on overconfidence in worst-case intelligence analyses. The decision to collapse the Agreed Framework in 2002 would go down as a fateful error that allowed North Korea to go from zero to many nuclear weapons.

The IC had reason, and still has reason, to be suspicious. It is worth noting the reasons many still suspect that North Korea reprocessed more plutonium than it declared in 1990.

  • The IAEA uncovered a number of discrepancies about the number of batches and timing of North Korea’s reprocessing activities. “For many,” David Albright wrote, “the sheer number of anomalies was perhaps the strongest evidence of undeclared plutonium separation.”
  • US officials later revealed that the IC observed North Korea shutting down the reactor for 70 days in 1989, at which time some or all of the fuel might have been unloaded.
  • North Korea constructed, camouflaged and refused IAEA access to what the CIA believed was an undeclared waste site at Yongbyon.

These are obviously suspicious indications — but they don’t amount to proof that, before 2002, North Korea had enough plutonium for a bomb. And, as Stephen Engelberg and Michael Gordon reported in 1993, INR dissented from the “one, possibly two” judgment.

The important point, however, isn’t whether North Korea did or did not. We can’t control that and may, in fact, never know. What we can control is how we, particularly our elected leaders, use intelligence in the service of the protecting the national interest.

The lesson in all this has something to do with how policymakers use intelligence in a complex, uncertain world. The President “gets paid the big bucks” — as my Dad would say — to make the tough decisions that take into account the possibility that intelligence estimates are in error one way or the other. After all, analysts are human beings, not wizards with crystal balls (although David Albright has been known to use the software Crystal Ball). As President who blames his intelligence community is, at least in my opinion, like the poor workman who blames his tools.

The Agreed Framework and the containment of Iraq were not optimal policies, but the Clinton Administration chose them because they balanced the risks of our worst-case intelligence judgments with the downsides of intelligence failure. In other words the policies were robust to dramatic intelligence failures.

In the case of North Korea, for example, the Administration made a choice to focus on preventing North Korean access to additional plutonium, rather than emphasizing past production. As Joel Wit, Dan Poneman and Bob Gallucci write in Going Critical, the Agreed Framework was about protecting the future more than uncovering the past:

[W]hen the president, vice-president, and principals considered the matter, they decided that they would attach the highest priority to stopping North Korea from obtaining any additional plutonium. All agreed, in essence, that it was more urgent to protect the present and the future than to unravel the past, by pinning down how much plutonium North Korea had indeed separated in its earlier reprocessing campaign.