Bob Woodward’s recent piece in the Washington Post argues that the debacle of the Iraq-WMD case should have made the Bush administration more circumspect about intelligence—and that everyone understood this lesson except the vice president. He offers the Syrian nuclear reactor destroyed by the Israelis in September of 2007 as an example.
At a National Security Council meeting in June of that same year, Cheney urged a U.S. strike on the reactor that U.S. intelligence had discovered the reactor months before. The president declined and would later write in his memoirs that he was dissuaded by the doubts—fully shared by virtually the entire intelligence community—expressed by CIA director Michael Hayden.
Woodward parses Hayden’s warning in an attempt to distinguish between hard facts and judgment. He concludes that while Bush stuck to the facts, Cheney was too willing to go beyond those facts and make unsupported judgments. “A key lesson of the 9/11 decade for presidents and other national security decision makers,” Woodward writes, “is the importance of rigorously testing intelligence evidence: poking holes in it, setting out contradictions, figuring out what may have been overlooked or left out.”
Here, then, are some contradictions in Woodward’s account, as well as something things that he—or his sources—overlook or leave out. Woodward reports that at the June 2007 NSC meeting, Hayden presented four key judgments:
1. There was high confidence that the building in question was in fact a reactor;
2. There was high confidence that Syria and North Korea had been cooperating for ten years on a nuclear reactor program;
3. There was medium confidence that North Korea built the reactor; and
4. There was low confidence that the reactor was part of a weapons program.
It’s not clear how Hayden arrived at the third from the first two. Why only “medium confidence” given a ten-year relationship between the two countries specifically devoted to a reactor program? If all Hayden meant was that he could not be sure, that would be one thing. But intelligence is almost never 100 percent sure about anything, yet the community still routinely states certain assessments with “high confidence” despite a lack of perfect certainty.
Did the intelligence community have evidence that some other country might be helping Syria? If so, who? The cohort of potential helpers is small—the declared nuclear powers and a few others. It’s safe to scratch from the list the Western powers plus Israel, India, and Japan. That leaves Russia, China, Iran, Pakistan, and North Korea. Iran spent decades unsuccessfully trying to start its own reactor with intermittent Russian help and without success until the Russians finally committed. So scratch Tehran off too, as incapable of providing meaningful assistance.
Did the intelligence community have any evidence that Russia, China, or Pakistan was helping Syria with a reactor? If so, that would be significant news. I suppose, if it were true, those who told this story to Woodward might have left that part out. Implicating any one of those three countries in yet another proliferation scheme would greatly complicate U.S. foreign policy. On the other hand, if the intelligence community really wanted to cast doubt on North Korean involvement and it knew that some other country was engaged in nuclear cooperating with Syria, then it would have an interest in saying so.
We don’t know. What we do know from the reported facts is that we have significant reason to believe that North Korea was behind the reactor and little reason to believe it was not.
But even if the Syrians had managed to build the reactor all by themselves (a highly dubious proposition), other problems with the analysis remain. As reported, the bases for point four—low confidence that the reactor was part of a weapons program—are weak. The CIA acknowledged that the reactor could produce plutonium, which can be used to make nuclear weapons. But first it has to be reprocessed from spent reactor fuel rods. The CIA claimed that it could find no evidence of reprocessing “capability at the site or nearby in that region of Syria.” Leave aside that, in Donald Rumsfeld’s words, “absence of evidence is not evidence of absence.” There is no reason why a reprocessing facility has to be located at the site of a reactor. Most such facilities around the world are not.
As for not finding one “in the region,” what about the rest of the country? Spent fuel rods are routinely transported great distances all over the world. And what about the possibility that the rods were intended to be reprocessed in another country—say, in North Korea, which we know has mastered and used the technology? Bilateral reprocessing deals are not uncommon. Granted, one between two rogue states would be more difficult to carry out than an agreement between France and Japan (to cite but one long-distance arrangement that actually exists). But it would appear that the two states got pretty close to clandestinely completing a nuclear reactor. Keeping secret shipments of spent fuel rods would be easy by comparison.
The CIA also, according to Woodward’s sources, cited the lack of any “identifiable means to manufacture uranium fuel.” Fair enough, as far as it goes. But not every nation that operates a nuclear reactor manufactures its own fuel. Uranium for reactors—as opposed to bomb-grade highly enriched uranium (HEU)—is commercially available on an international market. Of course, the trade is tightly controlled and a rogue state like Syria might have trouble finding an honest seller. But it also might have less trouble finding a dishonest one. It takes a great deal of HEU to make a bomb but comparatively little low-enriched uranium to run a reactor. Uranium is also not that expensive. It’s hardly inconceivable that Syria had a plan to import its fuel. Most of the world’s successful nuclear programs have relied on such outside assistance, at least to get started.
In a background briefing with reporters nearly a year after this meeting, a CIA official said (in Woodward’s paraphrase), “there was not much physical evidence the reactor was part of a weapons program.” This is a curious statement. There need not be any physical evidence that a reactor is part of a weapons program and yet it might still be used to make weapons. A reactor, simply by the nature of what it does, produces plutonium that can be used to make bombs (and smaller, more sophisticated and lethal bombs than HEU).
The reactor may be designed in all earnestness and good faith only to generate electricity. Nonetheless it will still produce plutonium as a byproduct. Whether to harvest that plutonium and use it to make weapons is a political decision that need not affect the design of the reactor. The U.S. ceased reprocessing plutonium from commercial reactors in 1977. Before that, some of the cores in our nuclear arsenal were derived from plutonium created in the same reactors powering lamps, refrigerators and televisions in American homes. The reactors kept on operating after the decision to stop reprocessing. While there are special plants (“breeder reactors”) designed to produce plutonium at a faster clip, there is nonetheless no clear line between a reactor intended for a weapons program and one used to generate electricity.
The briefer went on to explain that his “assessment” was that the reactor was “planned to be part of a weapons program,” but he had to distinguish between evidence and an assessment. In his view, the evidence was not strong enough to conclude that the reactor was part of a weapons program. Why then was his own “assessment” that it was? He doesn’t say, and Woodward does not clarify.
It does, however, seem that based on the evidence presented and on some facts not mentioned, the briefer’s assessment was in that respect correct. Consider, for instance, the fact that to this day the Syrians have never protested the Israeli strike—as surely they would have done had the reactor been intended for purely peaceful purposes. Nor did the Syrians inform the IAEA of the plant’s construction as required under the terms of the NPT, to which Syria is a signatory. Iran, by contrast, has always acknowledged its own reactor at Bushehr, which remains under IAEA safeguards.
The most telling and newsworthy details in Woodward’s account however have nothing to do with the reactor itself. He writes that Hayden “later told others that he … intentionally shaped his presentation that way to discourage a preemptive strike … Hayden’s declaration of low confidence was, in effect, his anti-slam dunk.” Then: “At the CIA afterward, the group of specialists who had worked for months on the Syrian reactor issue were pleased they had succeeded in avoiding the overreaching so evident in the Iraq WMD case. So they issued a very limited-circulation memorial coin. One side showed a map of Syria with a star at the site of the former reactor. On the other side the coin said, ‘No core/No war.’”
These are extraordinary revelations. Many Bush administration officials have long maintained that the intelligence community deliberately shaded and slanted its reports in order to produce the policy outcomes that it wanted to see. Intelligence officials have always heatedly denied this. One is tempted to say that Woodward’s reporting presents a slam dunk case that those Bush administration officials were right all along.
In any event, contrary to Woodward’s account, Vice President seems to have considered the facts very carefully. What he looked past was the intelligence community’s shaded assessment of those facts. Based on Woodward’s own reporting, Cheney appears to have been the only person in the room to have based his judgment solely on the known facts and nothing more.
Michael Anton served in National Security positions in the Bush Adminitration.