U.S. Failures in North Korea

Washington must recognize its intelligence and policy missteps on North Korea or prepare for another round of botched talks.

North Korea is a hard target for the intelligence community to penetrate. Satellites and other classified intelligence capabilities, known as “national technical means (NTM),” monitor the North continuously, but the United States lacks assets on the ground. As a result, it faces considerable uncertainty about North Korea’s nuclear and missile programs, economy and internal politics, not to mention its intentions. For instance, U.S. intelligence still does not know for sure how much plutonium the North reprocessed before 1991 or when the North began enriching uranium in quantity—and hence how many nuclear devices it may have.

Uncertainty is not the same as surprise, however. Surprise occurs when North Korea’s behavior does not live up to observers’ expectations. In this sense, while uncertainty is endemic, surprise has been relatively infrequent.

Knowns and Unknowns

The North’s revelation of an enrichment plant—located at Yongbyon and containing two thousand centrifuges, more advanced than most of Iran’s—to Siegfried Hecker and others was a surprise of sorts. That revelation shows the limits of monitoring countries with NTM, which revealed a suspect building at Yongbyon but not its purpose. It also suggests the benefits of Track II diplomacy and other direct contacts for reducing uncertainty.

Yet it was hardly a surprise that the North was on the verge of having an operational enrichment program. Pyongyang had advertised its progress earlier in the year. And a notorious 2002 National Intelligence Estimate had said “the North is constructing a plant that could produce enough weapons grade uranium for two or more nuclear weapons per year when fully operational, which could be a soon as mid-decade.”

The 2002 misestimate was an artifact of an estimating process that may bias analysts to err on the side of expecting a bad outcome too soon rather than too late. Technical experts seem to have suggested the end of the decade, but others in the intelligence community, perhaps knowing of the Bush administration’s eagerness to press the case for scrapping the Agreed Framework, asked if the North could have an operational program sooner. Mid-decade was the worst-case answer. When mid-decade passed without evidence of a plant, the estimate was revised closer to decade’s end. Worst-case assessment may account for another surprise—that no underground nuclear facility was ever found at Kumchang-ni.

Another case illustrates the limits of NTM: the discovery by Israeli intelligence that North Korea was helping Syria to construct a nuclear reactor. The move was far from unanticipated; North Korea long had been known to enjoy close ties to Syria, having sold missile technology to Assad’s regime. U.S. officials had repeatedly said that Pyongyang would sell anything to anyone. But in the end, assessment depends on evidence, not just presupposition.

When surprise does occur, it is important to examine the expectations that produced it, not just North Korea’s behavior. Some expectations are rooted in U.S. preconceptions and beliefs, U.S. domestic politics and South Korean disinformation, all of which shape the context in which assessments are made.

A major source of mistaken assessments is the unexamined presupposition that aggression, blackmail, lying and cheating are inherent in North Korea’s nature—that evildoers do evil. A more sophisticated version of this view is that Pyongyang has always been determined to acquire nuclear weapons. The implication is that negotiations are a waste of time because it can never be dissuaded from nuclear arming.

Yet until last year, the only way for North Korea to make the fissile material it needs for nuclear weapons was to remove spent nuclear fuel from its reactor at Yongbyon and reprocess it to extract plutonium. Yet North Korea stopped reprocessing in 1991—three years before signing the 1994 Agreed Framework—and did not resume it until 2003. Pursuant to a six-party accord reached in February 2007, it shut down its reactor at Yongbyon and has kept it shut until now. In so doing, it has denied itself many bombs’ worth of plutonium. Similarly, the only way for North Korea to perfect ballistic missiles for delivering nuclear warheads is to test them until they work reliably, but it halted test launches from 1999 until 2006 (as it promised William Perry) and has conducted only four sets of medium- and longer-range missile tests in twenty years.

Another unexamined assumption is that North Korea acts out for purely internal reasons. The intelligence community does have some insight into the North’s internal politics—a recent book by U.S. analyst Patrick McEachern, Inside the Red Box, is a rare good example. All too often, however, too much is made of what little is known.