This is not because our intelligence agencies are incompetent—although sometimes they are—but because the task is exceptionally hard.
In Iran We Trust? If Tehran breaks its promises,
we’re unlikely to know.
President Obama is rushing to implement the six-month interim agreement with the Islamic Republic of Iran that went into effect last week. Together with five other world powers, he is now working to negotiate a long-term agreement aimed at keeping Iran from developing a nuclear bomb. He regards his opening to Iran as a signature achievement of his presidency and has proudly declared that diplomacy opened a path to “a future in which we can verify that Iran’s nuclear program is peaceful and that it cannot build a nuclear weapon.”
If we assume that negotiations do not collapse and some sort of long-term accord is struck, there will still be thorny questions. A preeminent one concerns Iranian compliance. How much confidence can we have that the ayatollahs will not press ahead with their nuclear program in clandestine facilities, as they have done in the past? And if they do press ahead, how much confidence can we have that our intelligence agencies will catch them?
Obama’s faith that “we can verify” Iranian compliance glides over the fact that the U.S. track record in unmasking covert nuclear programs is checkered at best. This is not because our intelligence agencies are incompetent—although sometimes they are—but because the task is exceptionally hard. Just last week, a three-year study by a Pentagon subunit, the Defense Science Board, concluded that U.S. intelligence agencies “are not yet organized or fully equipped” to detect when foreign powers are constructing nuclear weapons or adding to existing arsenals. What is more, their ability to find “small nuclear enterprises designed to produce, store, and deploy only a small number of weapons” is “either inadequate, or more often, [does] not exist.”
Past intelligence lapses in the nuclear realm go back to the dawn of the atomic age and include a failure to foresee the first Soviet A-bomb test in 1949, the first Soviet H-bomb test in 1953, and the first Indian nuclear test in 1974. After the first Gulf war, the U.S. intelligence community was astonished to learn that Iraq was only months away from putting the final screw in a nuclear device. In the run-up to the second Gulf war, the CIA blundered in the opposite direction, declaring with high confidence—“a slam dunk” in CIA director George Tenet’s notorious phrase—that Saddam Hussein was developing nuclear weapons. He was not. More recently, North Korea constructed a uranium enrichment facility that, despite intense scrutiny by American intelligence, went unnoticed until the North itself chose to reveal it.
The case of Syria is especially pertinent to our efforts to monitor Iran.
By the late 1990s, U.S. intelligence detected glimmerings that Syria might be embarking on some sort of nuclear project. But the agency had trouble making sense of the evidence it was gathering. It perceived that North Korea was helping Syria with a joint venture involving North Korean nuclear experts, but as a senior U.S. intelligence official explained in a briefing, we “had no details on the nature or location of the cooperative projects.” By 2003, U.S. intelligence had concluded that the activity involved work at sites “probably within Syria,” but they “didn’t know exactly where.” The fog of intelligence had set in: “We had this body of evidence, kind of almost like a cloud of, boy, there is something going on here but we can’t get a whole lot of precision about it.”
By 2005, the United States had made more progress in determining what was transpiring. Satellite photos revealed a “large unidentified building under construction” set in a canyon in eastern Syria near the Euphrates River at a juncture called al Kibar. But American intelligence analysts could not say much more. All they had was images of a structure that was “externally complete,” but it was “hard to figure out, looking at that building, what its purpose is.”
One problem was that “it certainly didn’t have any observable, externally observable characteristics that would say, oh, yeah, you got yourself a nuclear reactor here—things like a massive electrical-supply system, massive ventilation, and most importantly a cooling system.” Another problem was that though the structure closely resembled North Korea’s plutonium reactor at Yongbyon, America’s highly skilled photo-interpreters could not connect the dots between the two facilities. The oversight was not their fault; the Syrians had erected curtain walls and a false roof to disguise the building’s shape and conceal typical features of a reactor. The multibillion-dollar, ultra-high-tech tools of U.S. intelligence were foiled by one of the most low-cost and ancient techniques of warfare: camouflage.
Only in 2007, just as the reactor was ready to be loaded with uranium fuel, did U.S. intelligence conclude that Syria had built a gas-cooled, graphite-moderated reactor. It reached this judgment not by dint of its own collections efforts but thanks to incontrovertible evidence provided by Israel: photographs of the building’s interior. Under our eyes but without our seeing, the Syrians had come breathtakingly close to possessing an operational generator of the nuclear bomb ingredient plutonium.
“This was a significant failure on the part of U.S. intelligence agencies,” writes former defense secretary Robert Gates in his new memoir. Gates notes that “Syria for years had been a high-priority intelligence target for the United States” and that “early detection of a large nuclear reactor under construction in a place like Syria is supposedly the kind of intelligence collection that the United States does superbly well.” The failure clearly shook Gates and led him to ask President Bush: “How can we have any confidence at all in the estimates of the scope of the North Korean, Iranian, or other possible programs?”
That was the right question to ask in 2007 and it remains the right question to ask about Iran today.
It is especially significant that the CIA was undaunted by its own lapse. After Israel presented the United States with photographs of the interior of the building at al Kibar, the CIA told President Bush that while it now had high confidence that the structure was a nuclear reactor, it still had low confidence that Syria was engaged in a project to develop nuclear weapons. The reason for the low confidence estimate: It had scoured Syria and not been able to locate or identify any other components of a Syrian nuclear program. This was not a conclusion without consequences. In the wake of the WMD intelligence fiasco that precipitated the second Gulf war, President Bush was reluctant to strike the Syrian reactor without a rock-solid CIA judgment behind him. Israel was not so reluctant. It destroyed the reactor in an air raid on September 6, 2007.
What does all this mean for our dealings with Tehran? “With respect to Iran, the Syrian episode reminds us of the ability of states to obtain nuclear capability covertly,” is what U.S. intelligence itself has said about its own failure. But President Obama does not appear to take the reminder all that seriously. Even if inspectors were free to roam Iran at will, the ability of American intelligence to monitor a country whose territory is nearly 10 times larger than Syria’s would be in doubt. But under the preliminary agreement with Iran struck by President Obama in November, International Atomic Energy Agency inspectors are not free to roam at will; it appears they will be confined to those nuclear facilities that the IAEA already knows about.
In any longer term agreement with Iran, far-reaching and highly intrusive verification provisions are going to remain crucial. But even in the unlikely event that the United States and its negotiating partners persuade Iran to grant inspectors unlimited access to all potential nuclear sites on its territory, our ability to detect violations will still be limited. It may be difficult to conceal a large structure like a nuclear reactor from the lenses of American satellites (although Syria found it easy enough for a time). It is far easier to conceal facilities housing centrifuges for uranium enrichment, which can be underground and do not require the kinds of cooling facilities that reactors demand. The leaders of our spy agencies may boast of the kinds of intelligence collection that they have been reputed to do “superbly well.” But history shows that their tools are limited and their record spotty.
For more than 20 years, Iran has violated IAEA safeguard agreements, developed covert nuclear facilities, and sought to mislead the West about the scope and pace of its activities. As the American people weigh the value of an agreement with a regime that has a consistent record of cheating on international accords—not to mention lying, inciting hatred, terrorizing, and murdering—they would do well to understand that if the agreement is violated, we may not find out until it is far too late to rectify our oversight, for at that point, Iran will already have achieved its terrible goal.
Gabriel Schoenfeld, a senior fellow at the Hudson Institute, is the author of Necessary Secrets: National Security, the Media, and the Rule of Law and, most recently, A Bad Day on the Romney Campaign: An Insider’s Account.