Fate changes no man, unless he changes fate.
—Epigraph on the opening page of a status report prepared by engineers of Project 111, the Iranian military's effort to design a nuclear warhead
BY THE TIME President Bush's national security team gathered in the Situation Room the Thursday before Thanksgiving 2007, the rumor had already raced through the upper reaches of the administration: America's much-maligned spy agencies had hit the jackpot.
With a mix of luck and technological genius, they had finally penetrated the inner sanctum of Iran's nuclear weapons program. For weeks the dialogues, laboratory drawings, and bitter complaints of Iran's weapons engineers had secretly circulated through the headquarters of the CIA and the National Intelligence Council, the small organization charged with putting together classified, consensus “estimates” about the long-term security challenges facing the nation. Now the highlights were crammed into a draft of a 140-page National Intelligence Estimate (NIE) that was stacked in front of every chair in the Situation Room's new, high-tech conference center, where Vice President Cheney, National Security Adviser Stephen Hadley, Secretary of State Condoleezza Rice, and others prepared to pick through it. Though it would never be explicitly discussed that morning, the memories of another NIE—the disastrously wrong-headed one on Iraq in the fall of 2002—was the subtext of their deliberations. No future NIE on weapons of mass destruction could escape from under that cloud.
But this report was different in every respect. It detailed the names of each of the Iranian engineers and program managers, along with excerpts from their deliberations about the nuclear program and speculation about the political travails inside Iran's fractious circle of top leaders. What those exchanges revealed turned out to be so mind-blowing that it threatened to upend Washington's strategy toward Tehran for months, maybe years, to come. The estimate concluded, in short, that while Iran was racing ahead to produce fuel that would give it the capability to build a bomb, it had suspended all of its work on the actual design of a weapon in late 2003. No one knew whether the weapons programs—what the Iranians referred to as “Project 110” to develop a nuclear trigger around a sphere of uranium, and “Project 111” to manufacture a warhead-had been resumed since then. The discovery cut the legs out from under Bush's argument that Iran harbored an active nuclear-weapons program that needed to be stopped immediately.
To those who delved into the report, starting with Robert Gates, the former director of the CIA who was now defense secretary, the intelligence estimate was one of the most imprecisely worded, poorly assembled intelligence documents in memory. Later, Gates would declare that in his whole career in intelligence he had never seen “an NIE that had such an impact on U.S. diplomacy.” He did not mean it as a compliment.
“The irony is it made our effort to strengthen the political and the financial sanctions more difficult because people figured, well, the military option is now off the table,” Gates told me a few months after the estimate was released.
To many of Gates's colleagues on the national security team, it seemed clear that Bush and Cheney were paying the price for twisting the intelligence on Iraq. Either out of a new sense of caution or out of fear that Bush was laying the predicate for war, the authors of the intelligence report had hemmed the president in, leaving Bush little justification for military action unless, as Gates put it, “the Iranians do something stupid.”
The summary opened with a set of “key judgments,” the first section of every National Intelligence Estimate, and sometimes the only pages that top officials read. To this day, those judgments are the only part of the intelligence estimate that has been made public—a decision prompted largely by the realization that once the classified version went to Capitol Hill, the main conclusion would leak instantly. The key judgments were written in a shorthand that emphasized the remarkable new discovery that some powerful Iranian had ordered a halt to the weapons design work. But it failed to say what sophisticated readers instantly understood: Designing the weapon is the easiest step in putting together a nuclear bomb. It could be done relatively quickly later on in the development process, presuming the Iranians had not already purchased a workable design from the Russian nuclear scientists who kept jetting into Tehran after the fall of the Soviet Union, or from the Pakistanis. The hard part of bomb-building is obtaining the fuel—the part of the project that was still speeding along in public view. The omission of that distinction in the NIE summary had to do with its intended audience. “We never wrote this to be read by the general public,” one of the authors of the report told me. “So it is missing a lot of the context.”
But once the key judgments became public, the reaction astounded everyone from President Bush to Michael McConnell, the director of national intelligence, and Michael Hayden, the CIA director. No one was more astounded than the authors of the report inside the National Intelligence Council, who had written the document with the assurance that it would never be made public. Around the world, critics of America's many intelligence failures in Iraq trumpeted the Iran report's conclusion to make the case that even the Americans now had their doubts that Iran was pursuing a bomb. As a consequence, the Germans delayed plans to announce new sanctions; the Russians and the Chinese said they would not vote for stiffer action against Iran.
Readers of the full classified version of the NIE, however, walked away with a very different impression. In their copies, the report contained the first allegations of a complex, covert program by the Iranians to enrich uranium at sites other than the giant facility outside the ancient city of Natanz, where inspectors were counting every gram of nuclear material. The covert enrichment program, too, had been halted, the classified sections of the report concluded.
“I'm not saying we saw centrifuges spinning on the edge of the Caspian Sea,” said one senior intelligence official who was deeply involved in reviewing the intelligence with Bush. “But there was a secret enrichment program too.” That was important, he said, because “none of us believe that they will create weapons-grade fuel at Natanz. What they are producing at Natanz is a body of knowledge there that they can transfer elsewhere.”
Whatever the truth—that Iran wants a bomb, that it wanted a bomb until it realized the cost, or that it simply wants the capability to build a bomb someday should the mullahs decide to take the last step—it is now clear that the effect of the intelligence report was far more detrimental than anyone realized at the time. The NIE's findings, or at least the awkwardly worded declassified version, sent a go-back-to-sleep message around the globe. In the intelligence community's overcaution about not repeating its mistakes in Iraq, analysts may have actually erred the other way, veering toward the kind of mistakes they made when they underestimated the Soviet effort to build a bomb sixty years ago or the pace of the Chinese, Indian, and Pakistani efforts that followed. It may turn out that one of the great post-Iraq paradoxes was that in crying wolf about Iraq, the American intelligence community found itself unable to raise the alarm about Iran.
Certainly the Iranians think so. The country's messianic president, Mahmoud Ahmadinejad, not only celebrated the intelligence findings, he sped up the deployment of new centrifuges, continuing to enrich uranium as fast as possible. The goal seemed clear: Create such a large infrastructure and inventory of nuclear fuel that the rest of the world will conclude it is simply too late in the game to get it back.
The result is that the next administration inherits an Iran newly emboldened to race ahead with its nuclear program and become ever more dominant in the region. By the middle of 2008, other nations that have historically feared Iran—a group of countries led by the Saudis—were nearly apoplectic. They were publicly talking about building up their own nuclear capabilities. And suddenly the world turned upside down: When the Israelis staged a clearly provocative military exercise that simulated a hundred-plane attack on Iran's nuclear facilities, the Saudis issued not one word of protest.
“You know,” one of Bush's top aides said to me in the summer of 2008, after returning from a Middle East trip with his boss, “there are a lot of people in Iran who are afraid we are going to bomb them. And there are a lot of other people in the region who are afraid we aren't.”
IT FELL TO Michael McConnell, the man who oversees the president's daily briefings, to tell Bush about the intelligence breakthrough in the summer of 2007, while the rest of Washington was fleeing for the cooling climate of the mountains and the beach. Just weeks before, a draft of the NIE that was circulating through the intelligence community had read a lot like the previous reports on Iran, though in a fit of Iraq-induced caution it said that the intelligence community now had only “moderate” confidence that Iran was determined to build a weapon, which was down from “high confidence” a few years before. The change had been made because in the interim there had been no new evidence of nuclear work—and no one wanted to repeat the mistake they made with Saddam's program, which analysts asserted must have been progressing because there was no reason it should have stopped.
But now, McConnell told Bush, a team of CIA analysts had come to Hayden with the fruits of an astounding technological breakthrough. After twenty years of watching via spy satellites or relying on international inspectors who were playing a running game of cat-and-mouse with the Iranians, the United States finally had found a way to glean the intentions of Iran's leaders and nuclear engineers.
How they did so ranked among the biggest secrets in Washington. Officials insist there were several sources; they would not have relied on a single source of intelligence for a finding of such magnitude. But clearly a good deal of the success came from the penetration of Iranian computer networks.
For four or five years American spy agencies, led by the code breakers at the National Security Agency in Fort Meade, Maryland, had worked on perfecting a series of technologies that enabled them to tunnel through computer networks—with astounding results. In Iraq, they had successfully bored into the computers of suspected al Qaeda terrorists, in one case even manipulating data to lure someone into a trap. Iran was a far harder target, with much more sophisticated computer security. The country's nuclear designers report to one of the most elite and secretive units of the Iranian Revolutionary Guard Corps.
The details of exactly how the United States got inside the Iranian network are highly classified.1 But the cyber invasion gave Americans access to a treasure trove of reports that detailed, with remarkable specificity, Iran's covert efforts to design a weapon and eventually to make it small enough so that it could fit atop a Shahab-3, an Iranian missile capable of hitting Israel or parts of Europe. It was all there—designs, sketches, and most important in this case, a running dialogue among engineers detailing what was going well and what wasn't.
Hayden was intrigued, but suspicious enough that he immediately sent the whole package of evidence to what is known at CIA headquarters in Langley as the CIC, the Counterintelligence Center. It is a corner of the agency that is staffed by professional paranoids who examine every major new piece of evidence to determine whether it amounts to “strategic deception,” information planted by someone looking to deceive the agency with false data. The analysts came back saying they thought the evidence was genuine, but just in case, they recommended reducing the agency's level of confidence in the trustworthiness of the new finding.
“We didn't do that,” one senior intelligence official told me. “These are guys who wouldn't trust that their mother was telling the truth if she said, ‘You were a beautiful baby.’”
Hayden told McConnell that he needed more time to assess the discovery, and that despite the pressure from Congress to report on what was happening in Iran, a thorough analysis of the credibility of the new information would take until Thanksgiving.
McConnell knew that he would have to handle Bush carefully— and he did. The “good news,” claimed McConnell, was that the intelligence appeared to confirm what Bush had long alleged but could never before prove: Despite their years of denials, the Iranians had constructed a secret military program devoted to solving the mysteries of building a bomb. There could be no other explanation for why engineers were tinkering with warhead designs.
But there was a serious hitch, McConnell warned. The flood of computer data that was still being translated and picked apart left it completely unclear who gave the order to shut down the weapons program. There was no indication of whether the bomb design effort had subsequently been turned back on.
Later, Bush's national security adviser, Stephen J. Hadley, told me he thought that a combination of events in 2003 must have led the Iranians to fear that the heart of their weapons-design program was about to be exposed. If the program was discovered, it would rip the cover off the story Iran was telling about how it was merely pursuing its right to produce reactor fuel. Hadley argued that it was no coincidence that the order to suspend the work came eight months or so after the American invasion of Iraq and just two months after the seizure of nuclear centrifuges headed for Libya. The noose was clearly tightening around Iran's first big supplier of centrifuge technology, the Pakistani nuclear engineer A. Q. Khan. The Iranians knew he was likely to talk and reveal Iran's purchases— if he hadn't already.
Whatever the motivation, the intelligence now in the hands of senior administration officials in Washington revealed that Iranian engineers, like engineers everywhere, were fuming about the idiocy of their technologically clueless masters. (Those masters were never named in the material.) The engineers believed their bosses were making a huge mistake. After years of work and huge sums of money expended, Iranian scientists had finally been making progress toward a Persian bomb, one that would level the playing field with Israel, leap ahead of the Saudis, and help restore Iran to a day of glory and influence it had not enjoyed for centuries. It had taken decades to get this far: The program had run into technological roadblocks, political opposition in Tehran, and covert efforts by the United States and others who, among other things, sabotaged the power supplies for Iran's centrifuges so that the equipment would blow up if it were turned on. But all those setbacks were temporary.
McConnell's warning to Bush in the summer of 2007 could not have come at a more critical moment.2 The capital had been buzzing for a year with speculation about whether Bush and Cheney would decide on one last, big confrontation before they left office—or whether the Israelis, concerned that Bush was too bogged down in Iraq to pay attention to a growing Iranian threat, would execute an attack on Iran themselves.
The left was convinced that Bush and Cheney would attack; some on the right, belatedly realizing that Iran was a far more serious threat than Saddam's Iraq had ever been, were afraid Bush would be paralyzed after the Iraq debacle. Bush himself, some of his friends said, was worried about how history would judge him if he left office with a legacy of invading a country that had no weapons, ignoring North Korea as it built six to eight of them, and leaving his successor to handle an Iran that was on the verge of getting a nuclear option, if not a bomb. Soon the perception of imminent confrontation had taken on a life of its own.
Senator John McCain, in the heat of a presidential primary campaign, had already declared that “there's only one thing worse than the United States exercising a military option. That is a nuclear-armed Iran.”3 That led some to conclude that his threshold for attacking Iran was relatively low—and that his statement could tempt Bush to act before the end of his term. Seymour Hersh, the veteran investigative reporter, fueled speculation about an impending attack with a series of articles in The New Yorker describing the Pentagon's contingency plans to take out Iran's nuclear facilities.4 What Hersh failed to mention is that there are Pentagon plans for scores of contingencies—probably somewhere in the back of a file cabinet, even plans to invade Canada. But in the hothouse atmosphere in Washington, contingencies and intentions were thoughtlessly blurred in the rumor mill—or deliberately blurred in hopes of worrying the Iranians.
With speculation about Bush's objectives swirling, the National Intelligence Estimate on Iran took on an unusual urgency. Not since the intelligence agencies rushed to turn out their disastrous estimate of Iraq's weapons capabilities—also produced in response to a congressional demand—had the prospect of war seemed to be riding on a single political document. The authors of the Iran report knew they needed to demonstrate that the intelligence agencies had learned two lessons from that disaster. The first was that you can start a war—or fail to stop one—by drawing hasty conclusions about what kind of weapons a country may be able to manufacture, given enough time, talent, and political will. But there was a second subtext to the Iran NIE. Many intelligence analysts wanted to show that they could stand up to political pressure and deliver a message the president and his team did not want to hear.5
“This community is consumed with not repeating the mistakes that were made in 2002,” McConnell said to Lawrence Wright of The New Yorker shortly after the National Intelligence Estimate was published.6 Another senior intelligence official who was deeply involved in putting together the NIE was more blunt over dinner one night. “If we screw up on Iran the way we screwed up on Iraq,” this official told me as the administration was coming to grips with the new discovery in Iran, “we're finished.”7
The Israelis were not convinced. They insisted that the CIA was blind to the evidence because it was overcompensating for mistakes made in Iraq. The Israelis complained that the International Atomic Energy Agency (IAEA)—the organization responsible for ensuring that civilian nuclear programs and facilities were not being used as part of a weapons program—was being duped. Iran, the Israelis alleged, was hiding secret nuclear facilities and programs from the international inspectors. But like the Americans, the Israelis had little recent evidence to support their claims.8 Israeli politicians feared that Bush's stand against Iran was weakening and maintained they could not accept the possibility of an Iranian nuclear weapon. They were, after all, in range of Ahmadinejad's missiles and at the receiving end of his apocalyptic threats.
One of Washington's favorite parlor games at the end of the Bush presidency was trying to figure out how Bush and Cheney would react if the Israelis declared one day that they were going to bomb Iran's facilities on their own. By running a huge military exercise in the Mediterranean in early June 2008, in which F-15s and F-16s flew 900 miles to simulate a strike on Natanz, the Israelis were clearly signaling they were prepared to take action. But part of that was bravado. One senior Air Force officer who studied the problem in detail for the administration told me in the spring of 2008 that he doubted the Israelis had the capability to hit the twelve to fourteen sites that would need to be destroyed. “I'm not sure we could do it,” he said. One thing was clear: If it ever came to mounting an attack, whoever struck the blow that crippled Iran's program, Washington would be blamed.
TO BUSH, the whole idea that Iran had suspended its covert weapons design work sounded suspicious. Sure, he told his aides, the Iranians may have stopped work on the most incriminating part of the program, designing the actual weapons. But they could quickly resume the weapons design work. His anger shone through when he was questioned about the intelligence after he landed on his first trip to Israel as president, early in 2008. “That's the CIA's view, it's not necessarily my view,” he told his hosts.9
Even though the final analysis was incomplete, the implication of McConnell's warning was clear: If the Iranian order to suspend designing the weapon proved to be real—a huge “if,” McConnell quickly acknowledged—everything Bush had been saying for years about the Iranians’ actively working to produce a nuclear weapon was technically wrong. The intelligence community still believed that Tehran wanted nuclear weapons and was strategically advancing toward that goal, but technically they had stopped the actual design of a weapon—and probably not resumed it.
“Mike needed to warn the president not to go too far out on a limb before we figured this out,” a senior intelligence official mentioned to me.10
Bush was brittle because he knew that the NIE report undercut the last bit of leverage he had with the Iranians: his ability to credibly threaten military action if they crossed some invisible line in their drive to make a bomb. It was a loss of leverage he could not afford. In 2003, in the glow of “Mission Accomplished,” he had squandered his best moment to strike a deal with the Iranians. He had toppled their greatest enemies. He had installed a Shiadominated government in Baghdad friendly to Tehran. He had put 150,000 American troops in easy reach of Iran's elite Quds Force, which supplied Iraqi militants with next-generation, armor-piercing IEDs, used to horrific effect in the worst moments of anti-American violence in Iraq. It was exactly the “empowerment” of Iran warned about in the intelligence reports sent to Bush in 2002. Whether Bush ever read those warnings is unclear.
Now the next American president faces not only a budget deficit at home and a troop deficit in Afghanistan, but a huge leverage deficit with Iran. Bush left his successor with two grim choices: Accept the reality that Iran will continue producing its own nuclear fuel and live with the clear risk that Tehran could ultimately use it for weapons, or try to force the country to disassemble nuclear production facilities that had blossomed during the Bush years. The first course could send Iran down the path blazed by North Korea, now a nuclear-weapons state. The second could easily lead to economic sanctions, oil embargos, and, perhaps, military confrontation.
EVEN AFTER McConnell's warning, Bush kept talking about Ah-madinejad, Iran's fiery but shaky president, in apocalyptic terms. “We've got a leader in Iran who has announced that he wants to destroy Israel,” Bush said to reporters during a press conference two months before the NIE was published, but after he already knew much of its contents. He told reporters that at a just-completed summit meeting he had said to his counterparts: “If you're interested in avoiding World War III, it seems like you ought to be interested in preventing them from having the knowledge necessary to make a nuclear weapon.”11 What jumped out of his statement was his use of the word “knowledge.” Even if Iran didn't have a weapon, he was arguing, no one can take the risk that they would soon learn how to make one.
And it was pretty clear that the Iranians were learning how. In two years’ time, the country had gone from installing a few gas centrifuges at a test facility at Natanz to putting more than 3,800 centrifuges into operation—a threshold number. If the Iranians could somehow keep that many machines spinning largely uninterrupted for a year, they could produce enough fuel for a nuclear weapon. In early 2008 the machines were still crashing regularly. But international inspectors reported that the Iranian engineers appeared to be solving the problem.
The Iranians tried to explain away the centrifuges. As they liked to point out to anyone who would listen, the Japanese had them, as did many non-nuclear states in Europe, and no one was complaining about their nuclear programs. To the rest of the world, that was beside the point: Iran had hidden its activities for seventeen years, disqualifying them from ordinary treatment.
Of course, if the Iranians got caught running a secret military program to design a weapon, their benign explanation would evaporate. “It would be game-over,” one of the authors of the National Intelligence Estimate told me. Instead, the Iranians developed a strategy that took advantage of what they termed their “right to a civilian nuclear program,”12 which allowed them to develop the capability to make a nuclear weapon while living within the letter of the Nuclear Non-Proliferation Treaty.
The Iranians have thrived on nuclear ambiguity. They understood the reality of post-Cold War power: To attain influence in the Middle East and beyond, the mullahs did not need a ruinously expensive arsenal patterned after the Americans or the Russians. That would simply give the rest of the world a target.
What Iran needed was a convincing capability to build a weapon on short notice. And as McConnell pointed out to Bush, the United States may never know when the Iranians decide to make the leap from a civilian nuclear program to a weapon. As long as the mullahs keep that uncertainty alive, they win.
MIKE MCCONNELL often reminded his colleagues of the caution that he received in the late 1980s and early ‘90s when he worked for Colin Powell. “Tell me what you know, then tell me what you don't know, and only then can you tell me what you think,” McConnell recalled Powell saying. “Always keep those separated.”13 Powell's colleagues—and Powell himself, his critics argue—forgot that maxim in the run-up to Iraq. In the case of Iran, McConnell needed to make sure the intelligence agency relearned Powell's rules.
McConnell had arrived at the National Security Agency just before the Clinton years, as the Internet and e-mail were catching on and were about to revolutionize the world of high-tech eavesdropping. Breaking into computer systems was nothing new for the agency: For years, in country after country, American spies had learned how to insert Trojan programs that collected data without detection. They had learned how to insert devious programs into the VAX computers that the Soviets were buying up through front companies around the world—programs that led to slight miscalculations that were undetectable for all intents and purposes but just big enough to result in fatal errors in missile designs or weaponry.
The rise of the Web changed the nature of espionage: Suddenly engineers around the world were putting their computers on the Internet, offering the NSA a new keyhole. McConnell presided over the agency just as it was making a technological leap to the next big challenge in detection and interception.
But after four years McConnell gave up the job, heading off, as he later put it bluntly, to “become rich.” He became one of corporate America's leading consultants on cyber threats, a lot more profitable pursuit than working in the government. He thought he was done with intelligence work until Bush pulled him back into service in late 2006, putting him in the newest and most difficult job in the spy world, director of national intelligence. There he was to preside over sixteen intelligence agencies that constantly battle for influence and assets.
It was only after he had settled into that job that McConnell fully understood the magnitude of the quiet revolution that had happened in his absence. When he had left the NSA, the main way to determine the status of a nuclear program was by looking through the periscope of spy satellites. From low Earth orbit, the satellite images could identify where the North Koreans were getting ready to conduct a nuclear test or what was being delivered to the Iranian nuclear facility at Natanz. Now, all of a sudden, McConnell and his colleagues were peering through what my colleague Bill Broad called “the world's biggest microscope.”
The microscope was the ability to reach inside computer networks—a gift the world of the Internet had given to the world of spycraft. Inside the operations rooms of the NSA and counterterrorism centers, twentysomethings with cans of Diet Coke spent their days conducting cyber missions into foreign computer systems. Once they were able to crack the computer systems of rogue states, they could see through into the networks—and copy the plans, the intentions, and the frustrations of engineers who had to make it all happen. Unlike with art thieves, they did not need the original paintings: A perfect digital copy was just as valuable.
“It's pretty astounding,” said one former intelligence official who has reviewed similar penetrations of foreign computer systems. “You have to assume that anybody in Iran, or anywhere else, who is computer savvy, knows that we can read e-mail,” he said. “But they assume that most of their communications are needles in the haystack—lost in the huge volume of traffic.” Oftentimes, that is a mistaken conclusion.14
In Iran's case, the effort required far more than simply capturing e-mails. The reports and conversations cited in the classified version of the NIE bore no date or time stamps. It was unclear who exactly was doing the writing and when the key conversations took place. Lurking in the background was the risk that the data being harvested was disinformation, designed to convince Washington that the weapons program had been halted.
So until the analysis was complete, Hayden and McConnell were not about to send raw, unanalyzed intelligence into the Oval Office. Hayden was determined that there would be no “slam-dunk” moments akin to the mistake George Tenet, the former director of the CIA, famously made as he presented Bush with evidence on Iraq's WMD programs. Nor, he vowed, would the intelligence community blind itself to inconsistencies, as many did when they ignored evidence that “Curveball”—the wonderfully named source for much of the intelligence about Iraq's weapons of mass destruction—was a known fabricator.
WHEN THE NIE was ready, Hayden fulfilled his promise to deliver it before Thanksgiving. It happened at one of those meetings that doesn't show up on any public schedule, in which the top members of the national security team gather in the Situation Room for a briefing—without the president. These meetings enable more open debate, including a discussion of what to present to the president. As they trickled into the room, whose low windows look out on West Executive Avenue, the president's advisors were in familiar but still somewhat unaccustomed surroundings: The Situation Room had been reopened just months earlier, after the first full renovation since the time of John F. Kennedy.
During the Cuban Missile Crisis, Kennedy had realized that there was no single place where top officers of the United States government could gather to review up-to-date intelligence and argue over strategy without being seen entering and exiting the Cabinet Room or the Roosevelt Room by the Oval Office. The Situation Room first came into operation toward the end of Kennedy's presidency. But for decades the facility was a disappointment to anyone who stepped inside: It was cramped and filled with paper and the buzzing of fax machines. There was a kitchen with no sink. The place looked much like it did when Lyndon Johnson used to head downstairs into the maze of rooms to micromanage the Vietnam War. Communications with other arms of government were haphazard at best, even a few years ago. There were moments in the early days of the Iraq War when Bush tried to talk to his generals in Baghdad, and the screens went blank, events that one of the president's aides said, with some care, had been known to “prompt a presidential outburst.”15 On the morning of 9/11, the place was not only chaotic, it was dangerously underequipped: There were only two rooms where it was even possible to have a video conference, and the ambient noise was so bad that often participants could not hear what was going on. Even before 9/11 the Bush White House had begun planning to turn the Situation Room into something resembling a modern facility. After 9/11, said Joe Hagin, the deputy White House chief of staff who oversaw the project, “what at first seemed desirable suddenly seemed urgent.”
It took more than a year of renovations to create a Situation Room that finally began to live up to Hollywood's depictions. The main conference room where the national security team gathered that morning had six flat-screen televisions that could pipe in generals from the field, as well as the British prime minister from 10 Downing Street. Bush was so enamored with the technology that he had it installed at Camp David, in a trailer just outside the ranch in Crawford, and on Air Force One.
ON THIS MORNING, every participant had been given the draft entitled “Iran: Nuclear Intentions and Capabilities.” The opening pages explained that like all NIEs this was supposed to reflect the consensus of all sixteen intelligence agencies. But it is rarely that easy.
Every year, roughly a dozen or so NIEs are published, sometimes prompted by a presidential question or demanded by Congress. They range in subject matter from the status of al Qaeda and other terror groups to the possible consequences of global economic meltdowns. Until the Iraq War, few outside the circulation list even knew when an NIE was being prepared.
This one had been wordsmithed by Hayden and a murder board of other experts, and every phrase was chosen carefully. It put the new news first.
“We judge with high confidence,” the first line read, “that in fall 2003, Tehran halted its nuclear weapons program; we also assess with moderate-to-high confidence that Tehran at a minimum is keeping open the option to develop nuclear weapons.”
It went on to portray the Iranians as rational actors who “are guided by a cost-benefit approach rather than a rush to a weapon irrespective of the political, economic and military costs.” It argued that the weapons work had stopped “in response to international pressure”—presumably the invasion of Iraq and the arrival of inspectors. It was impossible to know, the report's summary concluded, whether Iran “currently intends to develop nuclear weapons.”
If this latest report was right, it would mark a huge change not only in the assessment of Iran but in how Washington watched its enemies. Inside the Situation Room, Cheney, Rice, Gates, and Hadley, among others, wasted no time trying to poke holes in the sourcing and the logic. They peppered Vann Van Diepen, the national intelligence officer for weapons of mass destruction who oversaw the writing of the report, with questions about how the network was infiltrated and whether the Iranians could have deliberately allowed a “back door” to their system to remain open, precisely for the purpose of feeding disinformation to Washington. Van Diepen detailed the steps that the “red team”—the counterintel-ligence analysts who had examined the interceptions from Iran to determine if they were deliberate deceptions—had taken to test the proposition that the entire discovery was actually intended for American eyes. Their conclusion, he reported that morning, was that none of what they had in hand was “feed,” the agency's shorthand for information planted to take the Americans off course. “This wasn't a document that the Iranians ever intended for us to see,” one of the intelligence officials who prepared the report concluded.
Most of the president's aides were deeply skeptical. “It wasn't just Cheney who wondered if we were missing something,” one of the officials who took part in the briefing told me later.
What if the Iranians had a second, competing team working on the weapons design and the intelligence agencies missed it? And what if Iran's intentions suddenly changed and the mullahs decided to resume the weapons program? Would we know? Could we be assured that the same technology that got the intelligence agencies into the Iranian networks would continue to yield results, especially once the Iranians figured out what was going on?
Those were possibilities, of course, that Van Diepen could not rule out, and the report was filled with carefully worded hedges. Those who read it closely saw that its authors cautioned that they “do not have sufficient intelligence to judge confidently” whether the Iranians were willing to continue their suspension of work, or whether they have plans “to restart the program” after solving other technical problems down the road. On balance, the authors concluded that sooner or later, the Iranians probably planned to build a bomb.
“In our judgment, only an Iranian political decision to abandon a nuclear weapons objective would plausibly keep Iran from eventually producing nuclear weapons—and such a decision is inherently reversible,” the estimate concluded.
However, that caution was buried far down on the list of “key judgments.” What gave many in the room heartburn was the fact that the document, for the first time, seemed to redefine a nuclear-weapons project. Everyone in the room had read reports on covert nuclear programs for years. They knew that putting together a crude but devastating nuclear bomb of the kind America dropped on Hiroshima is not all that challenging, especially given that so many elements of the design are readily available online. Learning how to enrich uranium—what Hadley, the national security adviser, told reporters later was “the long pole in the tent”—is far more difficult.
Yet a footnote on the bottom of page one of the NIE made it clear that the report considered the enrichment secondary: “For the purposes of this Estimate, by ‘nuclear weapons program’ we mean Iran's nuclear weapon design and weaponization work and covert uranium conversion-related and uranium enrichment related work; we do not mean Iran's declared civil work related to uranium conversion and enrichment.”16
In other words, the report redefined the process. It largely ignored the public production of nuclear fuel that the international inspectors were watching. It focused only on the big discovery—that the Iranians had halted their work on the final step, the physical construction of the weapon.
Two months later, Thomas Fingar, the chairman of the National Intelligence Council, explained the reasoning for the decision this way:
“We were writing something that we fully expected to remain a classified document,” he said, clearly still smarting from the criticism that greeted the effort. “We were writing for officials who were fully familiar with the issue. We were making assumptions about our audience. And it was important to get right to the new element and not make people hunt around or go down forty pages to get to what was new.”17
Fingar had plenty of reason to believe that the document would stay classified. Since taking over as the director of national intelligence, McConnell had been determined to end the practice—born of the Iraq debacle—of making NIEs public. He insisted that unless some secrecy was restored, his analysts would begin pulling their punches, writing with an eye on how their assessments would read on front pages. Just weeks before the Situation Room meeting, McConnell had signed off on an order that said summaries of NIEs would no longer be released; he soon afterward told a conference of intelligence analysts in Washington that he did not expect the conclusions on Iran to ever see the light of day.18
But as soon as Hadley, Rice, and Gates flipped through this NIE, they knew its conclusions were so explosive that it would never stay secret. The document rewrote the main story line of the Iranian nuclear program.
When Bush was briefed, according to people who were in the room, he told McConnell that the conclusion was so dramatic it would have to be made public—despite the director's orders a few weeks before that all intelligence estimates should remain secret. Clearly, the Iraq intelligence failure was weighing on Bush. He told McConnell, “I can't be in a position of saying something publicly that's not true, backed up by your intelligence.” Had Bush uttered that same line to his intelligence chiefs in 2003, before the Iraq invasion, he might have rescued his place in history.
HADLEY KNEW THAT the inevitable headlines—”Iran Halts Nuclear Weapons Work, CIA Concludes”—would derail the sanctions on Iran. He was right. To the rest of the world, it looked as if Bush's rationale for driving the United Nations to impose sanctions on Iran had disappeared overnight. The irony was that the sanctions were rooted in Iran's refusal to suspend the enrichment of uranium—the main action that the Iranians were continuing, even speeding up. But as Bush himself later admitted, the NIE undercut the American assertion that the only reason Iran wanted to enrich was to supply a secret bomb program.
Now that argument was gone. Yet everyone at the White House knew that they could not order the intelligence agencies to rewrite the report for public consumption, with a greater emphasis on the dangers of allowing the Iranians to continue enriching uranium. That would have looked suspiciously like Bush was manipulating intelligence reports. But to leave the NIE in its current, less-than-subtle form would completely undermine the rationale for sanctions.
The problem was resolved with a muddy compromise that made everything worse. The NIE's “key judgments” were released essentially as written, save for a few omissions to protect sources and methods, chiefly the penetration of the computer networks and the direct references to Projects 110 and 111. Stephen Hadley appeared in the White House pressroom to put his own spin on the findings, stressing that the estimate concluded that Iran had, indeed, sought a weapon—and might again someday restart the weaponization program.
But he could not paper over the fact that the Iranians had played the game brilliantly. By suspending the military part of the program—and getting caught suspending it—they had eviscerated five years of efforts by the United States and its allies to build momentum for confronting Tehran. All of a sudden, the Americans seemed to be endorsing the view of the IAEA which had said that while Iran's efforts were deeply suspicious, there was no evidence of an ongoing weapons program.
Mahmoud Ahmadinejad, meanwhile, claimed a “great victory” for Iran. He was entitled to celebrate. The new intelligence estimate relieved the international pressure on Iran—the same pressure that the document itself claimed had successfully forced the country to suspend its weapons ambitions.
The Iranians responded by beginning to install their next generation of centrifuges—the so-called IR-2—a vastly improved model based on the advanced designs that A. Q. Khan had sold them more than a decade before. By early 2008, the first machines were up and running. If they could amass the parts to build these more-stable machines, the Iranians could enrich uranium faster than ever. The justification for developing the IR-2 centrifuge, as with the older models, was that Iran had to be ready to feed its own nuclear power plants—even though the only facility anywhere close to going into operation, at Bushehr, would use Russian fuel that had already been delivered.
Iran's Middle Eastern neighbors—from the Saudis to the Israelis—complained to the Bush administration that the accelerated Iranian uranium enrichment program was Washington's fault. The NIE had emboldened Iran, they said. They could not understand why Bush would let that happen. After all, wasn't this the president who had declared just a few years ago that he would never “tolerate” an Iran that could build nuclear weapons?19
This was true, but he was also a president who had denied, for years, that the invasion of Iraq would empower Iran. Now, even as the U.S. strategy for Iran was falling apart, a few members of the administration began to admit the obvious. Zalmay Khalilzad, the former American ambassador to Iraq and later to the United Nations, for example, told students at Columbia University one Friday afternoon in the winter of 2008 that Iran had emerged as perhaps the biggest winner of the Iraq War.
“It's helped Iran's relative position in the region, because Iraq was a rival of Iran,” Khalilzad explained. “The balance there has disintegrated or weakened.” As the American inability to control the chaos in Iraq grew more evident, so did the ability of Iran's mullahs to frustrate America's interests across the region—from supporting Hezbollah in Lebanon to arming Hamas in the Palestinian territories.
By the end of the Bush administration, the Iranians could imagine the future: a Persian sphere of influence that extended to the western banks of the Euphrates River. For a country that had terribly mismanaged its economy, that had been frozen out of the international community for a quarter-century, it could be the beginning of a revival. Not a new Persian empire, but a Persian influence far beyond its own borders.
That is where a nuclear capability—not even a weapon itself, but the mere ability to make one—could greatly bolster Iran's position. The Iranians had no hope of challenging the American nuclear umbrella, the ultimate defensive shield that Washington extended over its closest European and Asian allies. But Pakistan, India, and North Korea had taught the Iranians a lesson in modern power politics: Washington does not mess with countries that have nuclear arsenals, even small ones. Saddam Hussein never learned that lesson, and one day his American jailers delivered him to a noose. Kim Jong-Il, however, learned it well, and after he set off a nuclear test, the Americans dropped by to negotiate.
Iran's Arab neighbors looked at the NIE and came to a quick conclusion: The Iranians are getting the bomb. The only question is how long it will take, and whether, for the purposes of plausible deniability, the mullahs stop just short of turning the last screws. The Arab reaction was predictable. Suddenly, most of Iran's oil-rich neighbors became keenly interested in “peaceful” civilian nuclear projects. They are too smart to rekindle the arms race. Instead, they are in a capabilities race—figuring out how to assemble all the elements without assembling the bomb. As a senior Egyptian official told me, “Once the Iranians take the next step, everyone else is going to have to go the same route. No one would have a choice.”20