In 2007, U.S. intelligence said Iran had stopped its nuclear weapons program. Analyst policy bias and disinformation from Iranian double agents may explain the mistake.
By Edward Jay Epstein
July 29, 2010
In a stunning departure from a decade of assessments, the 2007 National Intelligence Estimate (NIE) on Iran declared: “We judge with high confidence that in fall 2003, Tehran halted its nuclear weapons program,” including “nuclear weapon design and weaponization work” and covert uranium enrichment. Even more astonishingly, it attributed this change to “increasing international scrutiny and pressure resulting from exposure of Iran’s previously undeclared nuclear work.” In other words, the threat of sanctions had ended that country’s surreptitious effort to obtain nuclear weapons.
This assessment suggested that further action against Iran was unnecessary. Unfortunately, as the Obama administration has now acknowledged, the NIE’s conclusion was dead wrong, costing us precious time in dealing with a serious threat.
The question remains, what caused such a disastrous mistake?
In 2007, there was still much the same mountain of evidence that led U.S. intelligence to conclude in the 2006 NIE with equally “high confidence” that Iran was secretly engaged in a nuclear weapons program. This evidence included verified reports that Iran had experimented with Polonium 210, a key ingredient in the trigger of early-generation nuclear bombs. And documents recovered from a stolen Iranian laptop described its efforts to fit a warhead in the nose cone of its Shahab 3 missile that would detonate at an altitude of 600 meters, which is too high for anything but a nuclear warhead to be effective.
The CIA had learned that Iran had most likely acquired a digital copy of a Chinese nuclear warhead design from the A.Q. Khan network. It also had monitored Iran’s crash program at Natanz to build a nuclear enrichment plant that could house up to 50,000 centrifuges.
Taken individually, these secret activities might have a nonnuclear explanation. For example, Iran claimed the purpose of its Polonium 210 experiments was merely to find a power source for an Iranian spacecraft (though Iran had no known space program at the time). Taken together, however, these efforts added up an inescapable conclusion: Iran was going nuclear.
What helped change this conclusion, in addition to the reorganization of U.S. intelligence following the report of the 9/11 Commission, was the receipt of new secret intelligence from Iran. This intelligence included convincing evidence that the facilities of the weapons-design program (code named “Project 111″) revealed on the stolen laptop had been closed down in 2003. Satellite photographs showed that buildings involved in the program had been bulldozed, communications intercepts indicated that scientists were no longer at the location, and a high-level defector from the Iranian Revolutionary Guard, Ali-Reza Asgari, reported that Project 111 had stopped functioning.
Since the Iranians knew that we knew about Project 111 in 2004—the CIA had released technical drawings from it to the International Atomic Energy Agency (IAEA)—it was not surprising that the Revolutionary Guard, which runs Iran’s nuclear activities, shut it down. The issue was how to interpret the closure. Had the weapons-design work been quietly moved to avoid further scrutiny? Had it been closed because the warhead design had been solved with the acquisition of the digital blueprints of the Chinese nuclear weapon? Or had Iran abandoned its quest for a nuclear weapon?
Deciphering a government’s intentions is no easy task. It is especially difficult in a closed and terrorized society in which the U.S. has no diplomatic relations and little direct access. So it came down to espionage to illuminate the intentions behind the shut-down of Project 111.
Over the years, the CIA had recruited a network of Iranian agents who had, or claimed to have, access to the thinking of Iran’s governing elite. These agents were in a position to cast light on Iranian nuclear intentions, and presumably they provided reports that supported the thesis that Tehran decided to end its nuclear weapons program. In any event, the authors of the 2007 NIE cited secret evidence to support the conclusion in its publicly released summary document that “Tehran’s decision to halt its nuclear weapons program suggests it is less determined to develop nuclear weapons than we have been [previously] judging.”
As we now know, the Revolutionary Guard was secretly completing the construction of new facilities in 2007. For example, at Fordo, 20 miles north of the holy city of Qom, it was reinforcing tunnels leading inside a mountain cavern designed to house a new uranium enrichment plant. This underground facility was only disclosed by Iran to the IAEA in late 2009. Clearly, Tehran was not abandoning its nuclear-weapons program.
What may have misled the CIA was a flaw in its espionage system. James Risen, the New York Times’s national security reporter, explains in his book “State of War” that since the CIA had no embassy base in Iran, it communicated with its agents through state-of-the-art satellite transmissions, which it assumed were invisible to the Iranian security services.
Then, in 2004, a CIA communications officer accidently included data in a satellite transmission to an agent that could be used to identify “virtually every spy the CIA had in Iran.” This disastrous error was compounded, according to Mr. Risen, because the recipient of the transmission turned out to be a double-agent controlled by the Iranian security service.
So the Iranians knew the identity of all the agents that the CIA had arduously maneuvered into positions of access, and the technical methods by which the CIA communicated with them. The agents (or their replacements) in Iran would have little choice but to allow the Iranian Security service to control the information they provided the CIA. If so, the CIA may have been vulnerable to receiving misleading secret intelligence that Tehran had abandoned it nuclear ambitions in 2004
One Iranian agent who supplied information to the CIA is Shahram Amiri, who defected to the U.S. last year and re-defected back to Iran this month. He reportedly provided details about the termination of Project 111 that presumably dovetailed with other information we got from the CIA’s compromised network. Iran now claims Mr. Amiri was a double agent all along.
Whether Iran controlled his secret reports to the CIA will be hotly debated for years to come. But willful blindness on our part should not be ignored. There were high-level people in the newly reorganized U.S. intelligence community who wanted to believe Iran was ending its quest for the bomb, and messages to the CIA from agents inside the country that diplomatic pressure was accomplishing this task fell on receptive ears.
Whether the erroneous conclusions in the 2007 NIE proceeded from Iranian deception or American self-deception, they undercut the case for taking more drastic action against Tehran. To the degree that other countries believed Iran had ended its nuclear program, they had little incentive to join us in imposing further sanctions.
To be sure, Iran could not conceal forever the evidence of its massive increase in uranium enrichment capabilities at Natanz, its missile testing, and its preparation of other underground facilities. In the interim, however, Iran managed to upgrade a large portion of its centrifuges and stockpile enough low-enriched uranium gas to manufacture, if it chose to further process it, the fuel for a nuclear bomb.
The moral of this sad spy story is that espionage is by its very nature a two-way game. Spies that are viewed as “assets” in a closed country can turn out to be very risky liabilities.
Mr. Epstein, an investigative reporter, is currently completing a book on the 9/11 Commission.