In former CIA and NSA director Gen. Michael Hayden’s new memoir, Playing to the Edge: American Intelligence in the Age of Terror, he describes the case of Al Kibar, in which Israeli officials informed the United States in 2007 about a building under construction in Syria that they thought was a nuclear reactor. Hayden writes, “Then we gave the data to a red team, dedicated contrarians, and directed they come up with an alternative explanation. Build an alternative case as to why it’s not a nuclear reactor; why it’s not intended to produce plutonium for a weapon; why North Korea is not involved.” (p. 258)
For the full story of the red teaming of Al Kibar, read this excerpt from my book—based upon interviews with senior Bush administration officials—Red Team: How to Succeed by Thinking Like the Enemy.
Red teaming is not only about using a devil’s advocate to scrutinize and challenge day-to-day operations. For institutions facing a significant decision, red teaming may also be a one-time effort. We can see how a properly administrated red team can help ensure that a crucial decision is the right one by studying the following example found in recent national security decision making.
In April 2007, Israeli national security officials surprised their American counterparts by informing them about a large building under construction at Al Kibar in a valley in the eastern desert of Syria. In oneon- one briefings, the Israeli officials provided dozens of internal and external color photographs dating back to before 2003. The evidence strongly suggested that the building was a nuclear reactor, remarkably similar to the gas-cooled, graphite-moderated reactor in Yongbyon, North Korea. Israeli Prime Minister Ehud Olmert then delivered his request to President George W. Bush: “George, I’m asking you to bomb the compound.”
Senior Bush administration officials were deeply troubled. North Korea had conducted its first nuclear weapons test the previous October using plutonium produced in the Yongbyon reactor. The Israeli briefings reinforced the US intelligence community (IC) assessments of “sustained nuclear cooperation” between North Korea and Syria. Though the IC had been monitoring the construction of a facility that they had described as “enigmatic” since 2005, the new Israeli photographs cast the compound in Al Kibar under a harsh new light. Immediately, a Central Intelligence Agency (CIA)-led task force reevaluated all of the available intelligence related to Al Kibar and North Korea’s nuclear cooperation with Syria. Given the flawed intelligence assessment that resulted in the incorrect conclusion in 2002 about Iraq possessing weapons of mass destruction (WMD), nobody wanted to be wrong again. As Bush told his intelligence chiefs: “Gotta be secret, and gotta be sure.”
The CIA task force reaffirmed the Israeli officials’ claims, but Bush administration officials took extraordinary measures to increase their confidence level. To ensure that they could be nearly certain in their assessment of Al Kibar, they employed devil’s advocate techniques markedly similar to those invented by the Vatican centuries earlier. National Security Advisor Stephen Hadley told IC officials to assemble some of their best analysts to review the data to see if the facility could be anything other than a reactor. The CIA director, General Michael Hayden, was similarly concerned given that “we had a poor record of assessing the WMD programs of countries bordering the Euphrates River.” He noted, “You increase your certainty by widening the circle, but we still had to keep the circle small to keep it a secret.” To do this, the IC employed two red teams that were totally independent from the task force and had not yet been “read in” on the intelligence regarding Al Kibar.
Bush’s intelligence chiefs so thoroughly bought into the concept of red teaming that they issued the two groups opposing goals: one would be commissioned to prove “yes” and the other to prove “no.” The “yes” red team assessment came from a private sector analyst who held a top-secret security clearance and was well known for his proficiency in monitoring nuclear weapons programs. The analyst was not told where the facility was located, but was provided with the Israeli and American internal and overhead imagery of it. The obvious efforts to camouflage the reactor vessel and the spent fuel pools within a building that had nearly an identical footprint to that of the Yongbyon reactor, and the trenches and pipes leading to a nearby water source (the Euphrates) were among several telltale giveaways. Within a few days, the analyst informed the IC officials, “That’s a North Korean reactor.”
Hayden’s “no” red team was composed of senior analysts from the CIA’s Weapons Intelligence, Nonproliferation, and Arms Control Center (WINPAC). This team received the same access to all the available data and intelligence as its counterpart, but was explicitly instructed to reach a hypothesis that the facility in Syria was not a nuclear reactor. “Prove to me that it is something else,” the CIA director told them. Over the course of the following week, the WINPAC group considered whether Al Kibar could contain a chemical weapons production or storage site, or something related to missile or rocket programs. Anything was plausible—they even investigated the possibility that it might be some sort of secretive nonweapons- related vanity project of Syrian President Bashar al-Assad. They also explored whether al-Assad had directed that a mock-up of a reactor be built, simply because he wanted it to be bombed for some reason. Another senior CIA official recalled that they had particular difficulty finding an alternative explanation for the internal photographs of the facility, which not only closely resembled Yongbyon but also even contained what appeared to be North Korean workers. “The alternative hypothesis that they came up with, for which the most evidence unquestionably and markedly lined up behind, was that it was a fake nuclear reactor,” Hayden recalled.
At the weekly Tuesday afternoon meeting in Hadley’s office, a handful of senior officials met to discuss what to do about the purported Syrian reactor. The results of the red-teaming exercises gave officials a high degree of confidence that they had their facts straight. They took comfort in the additional levels of scrutiny that had been applied to the initial intelligence estimates. “It gave us more confidence about the instinct and conclusion of the intelligence community regarding whether it was a reactor. Every other alternative explanation was not plausible,” according to Hadley. Secretary of Defense Robert Gates, who attended all of these meetings, also recalled, “Everybody agreed that we could not find an alternative to this being a nuclear reactor.”
However, even though the Al Kibar compound was all but confirmed to be a nuclear reactor, this did not mean that the United States should accede to Prime Minister Olmert’s request to destroy it. While Hayden could comfortably declare, “That’s a reactor. I have high confidence,” the red teams had notably found no evidence of a facility required to separate spent reactor fuel into bomb-grade plutonium or of weaponization work, which further led him to state, “On [the question whether] it is part of a nuclear weapons program, I have low confidence.” Bush subsequently told Olmert that the United States would not participate in a military attack: “I cannot justify an attack on a sovereign nation unless my intelligence agencies stand up and say it’s a weapons program.”
The two independent intelligence assessments provided Bush administration officials with far greater confidence about what was being constructed in the Syrian desert. They informed Bush’s decision-making calculus, even though his primary concern remained the risks to US interests in the Middle East if he authorized another preemptive attack on a Muslim country. With bombing now off the table, the CIA developed options to covertly sabotage the reactor before it went critical; however, CIA Deputy Director Stephen Kappes told the White House that sabotage had a low likelihood of success. Therefore, Bush chose to pursue diplomatic channels by going public with the intelligence to the United Nations Security Council and International Atomic Energy Agency, in order to pressure Syria to verifiably dismantle the reactor. Before this could happen, four Israeli fighter jets destroyed the suspected reactor at Al Kibar on September 6, 2007, without any resistance from Syria’s air defenses or overt support from the United States.
In this case, the findings of the two devil’s advocates, based on their independent analysis of available intelligence, greatly enhanced the credibility of the intelligence estimates regarding the existence of a nuclear reactor, and enabled Bush to make up his mind on the basis of more complete and vetted information. Ultimately, the president decided to refrain from launching strikes. This was a classic example of red teaming in action—having outsiders test the validity of the intelligence and consider the possibility of alternate hypotheses.