Exclusive: The CIA’s “Operation Merlin,” which involved providing Iran with a flawed design for a nuclear weapon and resulted in an alleged whistleblower going to prison, was the perfect example of creating intelligence in order to justify operations, reports Gareth Porter.
By Gareth Porter
Jeffrey Sterling, the case officer for the CIA’s covert “Operation Merlin,” who was convicted in May 2015 for allegedly revealing details of that operation to James Risen of the New York Times, was released from prison in January after serving more than two years of a 42-month sentence. He had been tried and convicted on the premise that the revelation of the operation had harmed U.S. security.
But the accumulated evidence shows that the intelligence not only did not support the need for Operation Merlin, but that the existence of the CIA’s planned covert operation itself had a profound distorting impact on intelligence assessment of the issue. The very first U.S. national intelligence estimate on the subject in 2001 that Iran had a nuclear weapons program was the result of a heavy-handed intervention by Deputy Director for Operations James L. Pavitt that was arguably more serious than the efforts by Vice-President Dick Cheney to influence the CIA’s 2002 estimate on WMD in Iraq.
The full story of the interaction between the CIA operation and intelligence analysis, shows, moreover, that Pavitt had previously fabricated an alarmist intelligence analysis for the Clinton White House on Iran’s nuclear program in late 1999 in order to get Clinton’s approval for Operation Merlin.
Pavitt Plans Operation Merlin
The story of Operation Merlin and the suppression of crucial intelligence on Iran’s nuclear intentions cannot be understood apart from the close friendship between Pavitt and CIA Director George Tenet. Pavitt’s rise in the Operations Directorate had been so closely linked to his friendship with Tenet that the day after Tenet announced his retirement from the CIA on June 3, 2004, Pavitt announced his own retirement.
Soon after he was assigned to the CIA’s Non-Proliferation Center (NPC) in 1993, Pavitt got the idea of creating a new component within the Directorate of Operations to work solely on proliferation, as former CIA officials recounted for Valerie Plame Wilson’s memoir, Fair Game. Pavitt proposed that the new proliferation division would have the authority not only to collect intelligence but also to carry out covert operations related to proliferation, using its own clandestine case officers working under non-official cover.
Immediately after Tenet was named Deputy Director of the CIA in 1995, Pavitt got the new organization within the operations directorate called the Counter-Proliferation Division, or CPD. Pavitt immediately began the planning for a major operation targeting Iran. According to a CIA cable declassified for the Sterling trial, as early as March 1996, CPD’s “Office of Special Projects” had already devised a scheme to convey to the Iranians a copy of the Russian TBA-486 “fireset” – a system for multiple simultaneous high explosive detonations to set off a nuclear explosion. The trick was that it had built-in flaws that would make it unworkable.
A January 1997 declassified cable described a plan for using a Russian émigré, former Soviet nuclear weapons engineer recruited in 1996 to gain “operational access” to an Iranian “target.” The cable suggested that it would be for the purpose of intelligence on the Iranian nuclear program, in light of the fact that the agency had not issued a finding that Iran was working on nuclear weapons.
But in mid-March 1997 the language used by CPD to describe its proposed covert operation suddenly changed. Another declassified CPD cable from May 1997 said the ultimate goal was “to plant this substantial piece of deception information on the Iranian nuclear weapons program.” That shift in language apparently reflected Tenet’s realization that the CIA would need to justify the proposed covert operation to the White House, as required by legislation.
With his ambitious plan for a covert operation against Iran in his pocket, Pavitt was promoted to Associate Deputy Director of Operations in July 1997. On February 2, 1998, CPD announced to other CIA offices, according to the declassified cable, that a technical team from one of the national laboratories had finished building the detonation device that would include “multiple nested flaws,” including a “final fatal flaw” ensuring “that it will not detonate a nuclear weapon.”
An official statement from the national lab certifying that fact was a legal requirement for the CIA to obtain the official Presidential “finding” for any covert operation required by legislation passed in the wake of the Iran-Contra affair.
Pavitt obtained the letter from the national laboratory in mid-1999, a few weeks after it was announced he would be named Deputy Director of the CIA for Operations.
But that left a final political obstacle to a presidential finding: the official position of the CIA’ s Intelligence Directorate remained that Iran did not have a nuclear weapons program. The language of the CIA’s report to Congress for the first half of 1999, which was delivered to Congress in early 2000, contained formulations that showed signs of having been negotiated between those who believed Iran must have a nuclear weapons program and those who did not.
The report referred to nuclear-related projects that “will help Iran augment its nuclear technology infrastructure, which in turn would be useful in supporting nuclear weapons research and development.” The shift from “will” to “would” clearly suggested that nuclear weapons work was not yet an established fact.
A second sentence said, “expertise and technology gained, along with the commercial channels and contacts established-even from cooperation that appears strictly civilian in nature-could be used to advance Iran’s nuclear weapons research and developmental program.” That seemed to hint that maybe Iran already had such a nuclear weapons program.
That was not sufficient for Tenet and Pavitt to justify a covert nuclear weapons program involving handing over a fake nuclear detonation device. So the dynamic duo came up with another way around that obstacle. A new intelligence assessment, reported in a front page article by James Risen and Judith Miller in the New York Times on January 17, 2000, said the CIA could no longer rule out the possibility that Iran now had the capability to build a bomb – or even that it may have actually succeeded in building one.
Risen and Miller reported that Tenet had begun briefings for Clinton administration officials on the new CIA assessment in December 1999 shortly after the document was completed, citing “several U.S. officials” familiar with it. The Tenet briefings made no mention of any evidence of a bomb-making program, according to the sources cited by the Times. It was based instead on the alleged inability of U.S. intelligence to track adequately Iran’s acquisition of nuclear technology and materials from the black market.
But the new assessment had evidently not come from the Intelligence Directorate. John McLaughlin, then Deputy Director for Intelligence, said in an e-mail response to a query that he did not recall the assessment. And when this writer asked him whether it was possible that he would not remember or would not have known about an intelligence assessment on such a high profile issue, McLaughlin did not respond. Pavitt and Tenet had obviously gone outside the normal procedure for an intelligence assessment in order to get around the problem of lack of support for their thesis from the analysts.
A declassified CIA cable dated November 18, 1999 instructed the Russian émigré to prepare for a possible trip to Vienna in early 2000, indicating that Tenet hoped to get the finding within a few weeks. Clinton apparently did give the necessary finding in early 2000; in the first days of March 2000 the Russian émigré dropped the falsified fireset plans into the mail chute of the Iranian mission to the United Nations in Vienna.
Pavitt Suppresses Unwelcome Iran Nuclear Intelligence
Pavitt’s CPD was also managing a group of covert operatives who recruited spies to provide information on weapons of mass destruction in Iran and Iraq. CPD not only controlled the targeting of the operatives working on those accounts but the distribution of their reports. CPD’s dual role thus represented a serious conflict of interest, because the CPD had a vested interest in an intelligence estimate that showed Iran had an active nuclear weapons program, and it could prevent intelligence analysts from getting information that conflicted with that interest.
That is exactly what happened in 2001. One especially valuable CPD operative, who was fluent in both Farsi and Arabic, had begun recruiting agents to provide intelligence on both Iran and Iraq since 1995. His talents had been recognized by the CPD and by higher levels of the Operations Directorate: by 2001 he had been promised an intelligence medal and a promotion to GS-14 – the second highest pay grade level in the civil service.
But that same year the operative reported very important intelligence on the Iran nuclear issue that would have caused serious problems for Pavitt and CPD and led ultimately to his being taken out of the field and being fired.
In a November 2005 court filing in a lawsuit against Pavitt, the unnamed head of CPD and then CIA Director Porter Goss, the operative, identified only as “Doe” in court records, said that one of his most highly valued “human assets” – the CIA term for recruited spies – had given him very important intelligence in 2001. That information was the subject of three crucial lines of the key paragraph in the operative’s complaint that were redacted at the demand of the CIA. For years “Doe” sought to declassify the language that had been redacted, but the CIA had fought it.
It was assumed in press accounts at the time that the redacted lines were related to Iraq. But the lawyer who handled the lawsuit for “Doe,” Roy Krieger, revealed to this writer in interviews that the redacted lines revealed that the CIA “human asset” in question was an Iranian, and that he had told “Doe” that the Iranian government had no intention of “weaponizing” the uranium that it was planning to enrich.
It was the first intelligence from a “highly-valued” U.S. spy – one who was known to be in a position to know what he claimed to know – on Iran’s intentions regarding nuclear weapons to become available to the U.S. intelligence community. “Doe” reported what the spy had said to his supervisor at CPD, according to the court filing, and the supervisor immediately met with Pavitt and the head of CPD. After that meeting the CPD supervisor ordered “Doe” not to prepare any written report on the matter and assured him that Pavitt and the head of the CPD would personally brief President Bush on the intelligence.
But “Doe” soon learned from his own contacts at CIA headquarters that no such briefing ever took place. And “Doe” was soon instructed to terminate his relationship with the asset. After another incident involving intelligence he had reported on WMD in Iraq that had also conflicted with the line desired by the Bush administration, CIA management took “Doe” out of the field, put him in a headquarters job and denied him the intelligence medal and promotion to GS-14 that he had been promised, according to his court filing. The CIA fired “Doe” without specifying a reason in 2005.
Pavitt did not respond to requests for an interview for this story both at the Scowcroft Group and, after he retired, at his home in McLean, Virginia.
The intervention by Pavitt to prevent the intelligence from Doe’s Iranian asset from circulating within the U.S. government came as the intelligence community was working on the 2001 National Intelligence Estimate (NIE) on the Iranian nuclear program. That NIE concluded that Iran was working on a nuclear weapon, but the finding was far from being clear-cut. Paul Pillar, the CIA’s National Intelligence Officer for the Middle East and North Africa, who was involved in the 2001 NIE, recalled that the intelligence community had no direct evidence of an Iranian nuclear weapons program. “We’re talking about things that are a matter of inference, not direct evidence,” Pillar said in an interview with this writer.
Furthermore he recalls that there was a deep divide in the intelligence community between the technical analysts, who tended to believe that evidence of uranium enrichment was evidence of a weapons program, and the Iran specialists, including Pillar himself, who believed Iran had adopted a “hedging strategy” and had made no decision in favor or a nuclear weapon. The technical analysts at the CIA’s Weapons Intelligence Non-Proliferation and Arms Control (WINPAC), were given the advantage of writing the first draft not only on Iranian technical capabilities but on Iranian intentions – a subject on which it had no real expertise – as well, according to Pillar.
The introduction of the intelligence from a highly credible Iranian intelligence asset indicating no intention to convert its enriched uranium into nuclear weapons would arguably have changed the dynamic of the estimate dramatically. It would have meant that one side could cite hard intelligence from a valued source in support of its position, while the other side could cite only their own predisposition.
Pillar confirmed that no such intelligence report was made available to the analysts for the 2001 NIE. He noted just how rarely the kind of intelligence that had been obtained by “Doe” was available for an intelligence estimate. “Analysts deal with a range of stuff,” he said, “from a tidbit from technical intelligence to the goldmine well-placed source with an absolutely credible account,“ but the latter kind of intelligence “almost never comes up.”
After reading this account of the intelligence obtained by the CPD operative, Pillar said he is not in a position to judge the value of the intelligence from the Iranian asset, but that the information from the CPD Iranian asset “should have been considered by the NIE team in conjunction with other sources of information.”
That led to a series of estimates that assumed Iran had a nuclear weapons program.
In 2004, a large cache of purported Iranian documents showing alleged Iranian research related to nuclear weapons was turned over to German intelligence, which the Bush administration claimed came from the laptop of an Iranian scientist or engineer. But former senior German Foreign Official Karsten Voigt later revealed to this writer that the whole story was a fabrication, because the documents had been given by the Mujahedin-E Khalq, the Iranian opposition group that was known to have publicized anti-Iran information fed to it by Israel’s Mossad.
Those documents led directly to another CIA estimate in 2005 asserting the existence of an Iranian nuclear weapons program, which in turn paved the way for all the subsequent estimates – all of which were adopted despite the absence of new evidence of such a program. The CIA swallowed the ruse repeatedly, because it had already been manipulated by Pavitt.
Operation Merlin is the perfect example of powerful bureaucratic interests running amok and creating the intelligence necessary to justify their operations. The net result is that Jeffrey Sterling was unjustly imprisoned and that the United States has gone down a path of Iran policy that poses serious – and unnecessary – threats to American security.
Gareth Porter is an independent journalist and winner of the 2012 Gellhorn Prize for journalism. He is the author of numerous books, including Manufactured Crisis: The Untold Story of the Iran Nuclear Scare (Just World Books, 2014).