Get Access to Print and Digital for $23.99.
Subscribe for Full Access
January 2017 Issue [Report]

The Trouble with Defectors

What informants taught an intelligence officer

“Do you know why you have come here, mother?” the man asked, reaching out for a hug. He was seated in a small room of a military barracks in Tehran, surrounded by intelligence-service agents. His mother, who had been invited by the Iranian government, was hopeful that he would be pardoned and released into the custody of his family. “I have come to see you,” she said. The man shook his head. “You have come to say goodbye.” A day or so later, his body was sent to his parents, with rope marks from his execution visible around his neck. His death was announced on August 7 by the Iranian judiciary: “Shahram Amiri was hanged for revealing the country’s top secrets to the enemy.”

Illustrations by Simon Pemberton

Illustrations by Simon Pemberton

Amiri, a thirty-eight-year-old physicist, had become mixed up in the most vexing nonproliferation dispute of our era. For decades, the United States has accused Tehran of pursuing nuclear weapons, a charge that the regime has consistently denied. The Americans have offered only circumstantial evidence, some of it of questionable provenance, while the Iranians have simply been unable to prove their guiltlessness to the satisfaction of the West. Even the agreement that was made in July 2015, which stipulated that Iran limit aspects of its enrichment program and consent to stringent inspections, has done little to quash speculation about its nuclear ambitions.

The United States has been frustrated by the dearth of actionable information coming out of Iran for as long as its spies have trained their sights on the Islamic Republic; one former official called it “the hardest intelligence target there is.” In 1979, having been taken completely off guard by the success of the Islamic Revolution, the CIA had no resources in place to recruit a new pool of informants. During the decade that followed, the agency gradually rebuilt a modest network inside Iran, but it was blown by sloppy tradecraft. More than thirty Iranian contacts were rounded up and executed.

By the late 1990s, the CIA began to encourage Iranians working in nuclear research to defect. As many as five were lured away; not one of them proved to be valuable. This was also the program that later brought Shahram Amiri to the attention of the agency. Agents first approached him during a visit to Frankfurt, Germany, in 2007, and again during a 2009 trip to Vienna. During this time, the CIA claims, Amiri was providing insights into Iran’s nuclear program, especially with regard to a hidden enrichment facility outside the city of Qom, where he occasionally worked. It was Amiri’s disclosures, at least in part, that led the CIA to conclude, in a 2010 National Intelligence Estimate, that Tehran had shut down its nuclear-weapons program in 2003.

Amiri formally defected in 2009, probably to escape being arrested as a spy. He was extensively debriefed by the CIA, and the information he provided underpinned Barack Obama’s denouncement of the Qom facility during that September’s G20 meeting. As defectors go, he seemed perfect: well educated, well spoken, and knowledgeable about a program that the CIA had been trying to crack for decades.

A year later, however, the relationship suddenly derailed. Amiri defected back to Iran, making bizarre claims to the media about having been kidnapped by the Americans. The CIA, which had treated him as a legitimate defector (going so far as to award him $5 million), became concerned that its star source had been a double agent. The CIA appears to have been so desperate to learn what he knew that the normal checks and balances were skirted, or even ignored. If Amiri had been working for Tehran, then everything — including the CIA’s estimate of Iran’s nuclear capability and the agency’s murky relationship with the International Atomic Energy Agency (I.A.E.A.) — had been exposed to Iranian intelligence. That would make the Amiri debacle one of the greatest blows to American spycraft in modern times.

Amiri returned to Iran in July 2010 and was initially feted as a hero. But he soon found himself in deep trouble. Within months, he was arrested and charged with treason. Despite Iranian media reports that Amiri was a double agent, interrogators there soon reached a different conclusion: he had been a willing asset of the CIA. “The Amiri case,” Paul Pillar, a former CIA officer, has observed, “seems to be a story out of the wilderness of mirrors department.”

More than five years passed before Amiri was put to death. While the evolution of his loyalties remains unknown, American intelligence officers have to assume that the CIA’s knowledge and strategy were compromised. And the fact that Amiri’s family is still alive means that he likely cut a deal. The trouble with defectors is that if they seem too good to be true, they probably are.

I was introduced to the complexities of defections a long time ago. In the fall of 1985, while serving as a Marine Corps intelligence officer, I attended the Soviet Military Power Conference, held at the headquarters of the Defense Intelligence Agency in Washington. Since 1981, the DIA had been publishing an annual booklet with the same title. The 1985 edition, with its ominous red cover, was distributed to all who attended the conference, where DIA staff preached to junior officers the gospel of Soviet malevolence.

The highlight of the event came when we entered an auditorium and saw a man onstage wearing a wig, a fake beard and mustache, and makeup designed to alter the angles of his face — either for his own protection or to heighten the Reagan-era theatrics, we couldn’t be sure. He was introduced as Yuri Nosenko, a defector from the KGB. Nosenko proceeded to regale us with tales of the wicked and bellicose Soviet Union, whose details coincidentally matched almost every talking point in the latest edition of Soviet Military Power. This was exciting stuff. For the better part of a week, we had been the recipients of dull presentations from DIA staff. Now we were listening to an actual acolyte of evil, whose indictment included not only the military elite but also the common people, and we soaked it up.

HA032__03HL0-1Fast-forward three years. I was serving as a weapons inspector at a Soviet missile factory outside the city of Votkinsk, in the foothills of the Ural Mountains. I had expected a nation of hard-line Communist automatons. Yet the Soviet people, even in a city closed off from all outside contact, exhibited no xenophobic tendencies; they were as normal, friendly, and desiring of peace as Americans. The same could be said for the staff and directors at the missile factory, Party members all.

Nosenko, in other words, had been dead wrong. In retrospect, I should have been suspicious. He had defected in 1964 — some twenty-one years before his presentation at the conference. He was from the age of Stalin and Khrushchev, and knew nothing about the Brezhnev era, or about the succession of leaders that followed, including Mikhail Gorbachev and his policies of glasnost and perestroika. How could Nosenko possibly have any viable insight into the current state of affairs inside the Soviet Union?

I should have been even more skeptical about my own government’s motivations for showcasing Nosenko. His defection, it turned out, had been marred by concerns that he was actually a double agent, dispatched by the KGB to discredit an earlier defector named Anatoliy Golitsyn. The CIA eventually deemed Nosenko to be legitimate, paid him, and hired him for speaking gigs like the DIA conference. Yet his CIA handler, Tennent “Pete” Bagley, remained convinced that he was a Soviet plant.

Lucky for Nosenko, it almost didn’t matter. The message he was selling during the mid-1980s was very much in vogue; he could be used as a pawn in a propaganda campaign, or presented to unsuspecting audiences as authoritative despite the fact that his information was out of date. But quality control was (and is) an ongoing problem. The trouble with defectors is that even with their obvious shortcomings, they are frequently the only players available in a game that is desperate for new moves.

The word defect is often used quite casually: individuals defect from Google to Facebook, from the Yankees to the Red Sox. But it denotes something much more serious in the realm of intelligence — a disavowal of national allegiance, often accompanied by the disclosure of secrets, which can produce lethal consequences. Technically, a defector can be anyone guilty of such a betrayal. For this reason, people who remain in their countries and pass secrets to foreign intelligence — spies — are often described as having “defected in place.”

But defectors, generally speaking, provide only a single piece of actionable intelligence. For example, a pair of defectors who fled Iraq provided information used by United Nations inspectors in September 1991, to discover evidence of a hidden nuclear-weapons program. Once their information was acted on, the defectors were no longer worth anything. Occasionally, defectors may be used for propaganda, like Nosenko. But for the most part, they are quickly tapped and led off to a new identity and a modest resettlement — a job in some Arizona 7?Eleven, nothing glamorous.

This can be disappointing to defectors, who often hope for plum rewards for their desertion. Take the case of Hussein Kamel, the son-in-law of Saddam Hussein. In the decade before his defection to Jordan, in 1995, Kamel had overseen Iraq’s various W.M.D. programs. During debriefings conducted by the CIA, British intelligence, and the United Nations Special Commission (UNSCOM), he revealed that the bulk of Iraq’s proscribed weapons programs had been either disposed of by inspectors or destroyed by Saddam’s officers. Kamel also provided information about Iraq’s efforts, which he claimed were ongoing, to conceal W.M.D. capabilities.

In exchange for talking, Kamel expected to be treated by foreign officials as the logical successor to Saddam Hussein — he would be returned from exile and cleanse the nation of its corruption and backwardness. But when the world refused to go along with his grandiose design, thanks to his unsavory reputation as a Saddam insider, Kamel quickly became demoralized. Having been assured of his father-in-law’s forgiveness, he went home to Iraq — where he was made to divorce Saddam’s daughter, denounced as a traitor, and shot. Whatever additional information he possessed died with him.

If defectors are plagued by limited utility and questionable motivations, why do intelligence professionals keep using them? The answer is convenience. Gathering intelligence is a difficult business. This is especially true with regard to nonproliferation and disarmament, matters in which even formal agreements cannot be taken at face value. “Trust but verify” — Ronald Reagan’s phrase — has become the weapons inspector’s maxim. But verification requires the acquisition of independent, high-quality sources.

Not surprisingly, analysts seek variety: communications interceptions, aerial imagery, human reporting (i.e., spies). Defectors are typically a last resort, employed when policymakers and the intelligence analysts who support them are operating in an informational vacuum. This is precisely when defectors may be most decisive — even if their revelations are hollow.

Consider the strange case of the North Korean defector Kang Myung-do. In the fall of 1993, North Korea had suspended its cooperation with international weapons inspectors, which gave rise to concern that it was diverting nuclear material from civilian reactors to weapons development. Speculation was rampant — in January 1994, the director of the CIA estimated that North Korea had already built “one or two” weapons. Nobody had a clue as to what was actually transpiring in Pyongyang, but there was talk of war, and the United States deployed additional military assets to the Korean peninsula. Spooked by the prospect of renewed conflict, American and North Korean diplomats agreed to talk, and by early summer, it looked as if a compromise might be reached that would get inspections back on track.

On July 27, just as another round of discussions was about to commence, the South Korean intelligence service held a press conference. There officials presented Kang, the thirty-five-year-old son-in-law of North Korea’s prime minister. Kang announced that North Korea already possessed five nuclear weapons and had plans to develop five more. The New York Times declared that if his story was true, he was likely to be the “most damaging defector ever to escape from North Korea.”

Kang’s claims about his nation’s nuclear arsenal were immediately downplayed by experts in the United States, who noted that five finished nuclear weapons were far beyond Pyongyang’s technical capacity. Kang readily acknowledged that he did not have direct access to information about the nuclear program, but he claimed to associate with the family and friends of those who did. In the end, none of his revelations turned out to be accurate. Still, the South Koreans were happy to parade him in front of news cameras.

Today Kang makes his living as a university professor, teaching about North Korea, and he supplements his income with several paid appearances a week on South Korean TV in which he comments freely on the affairs of his homeland, even though he has not set foot there in twenty years. He is, however, leery of other defectors exploiting their past lives in the media. “There are too many people on TV titled ‘Kim Il-sung’s doctor,’ which I don’t think is true,” he told a Reuters correspondent in 2015. Even Nosenko might have had something to learn from Kang.

Not every defector who makes eye-opening claims is a fraud, and sometimes the information is of tremendous value. Take Izzadin Hassan al-Majid. The death of Hussein Kamel had placed UNSCOM inspectors in a quandary: before being killed by his infuriated father-in-law, Kamel had insisted that the Iraqis were concealing proscribed weapons — and that their protection was overseen by Saddam Hussein’s personal security forces, the most difficult intelligence target in Iraq. Fortunately, another defector claimed to have firsthand knowledge of such activities: Izzadin, a former major in the Special Republican Guard and a cousin of Kamel’s, who had fled the country with him in 1995.

I first met Izzadin in April 1996, at the headquarters of Jordanian intelligence in Amman. The results of that interview showed promise, and a few months later, I requested that the Jordanians set up another meeting. They told me that Izzadin had been placed in the care of the CIA. In April 1997, I walked through the door of an upscale hotel suite near Tysons Corner, Virginia, biding my time while Izzadin’s handlers prepped him in the adjoining bedroom. On a drink cart, there were several empty bottles of whiskey, a pair of cut-glass tumblers, and melting ice cubes. Izzadin had originally fled Iraq with his wife, who was Kamel’s sister, and his children. When Kamel made his ill-fated return to Baghdad, he brought with him Izzadin’s family — all of whom were also executed. I couldn’t imagine the pain and remorse that Izzadin must have felt, and I understood how it could have driven him to seek whatever comforts the CIA might provide.

When Izzadin finally made his appearance, he was red-eyed and disheveled. For a few hours, I was able to coax critical information from him, which led to the discovery of buried missile components and production equipment at a villa he used to own on the outskirts of Baghdad. Intel from additional interviews I conducted that summer eventually compelled the Iraqi government to acknowledge its concealment of proscribed weapons and documents in the early 1990s — something the Iraqis had, until that point, assiduously denied.

These revelations were devastating enough to the Iraqis that Tariq Aziz, the foreign minister, complained to Richard Butler, the executive chairman of UNSCOM. “Mr. Scott Ritter,” Aziz told Butler that December, “relies heavily for his information on a person called Izzadin Hassan.” The minister conceded that Izzadin, as a former army officer, had undoubtedly known some useful things. But he suggested that there was little point in relying on further disclosures: “I think this source of information in particular should stop being used, because it has exhausted itself.”

However self-serving his suggestion may have been, Aziz was right — as a source, Izzadin was finished.* We inspectors now had to find a new Izzadin. But the trouble with defectors is that the good ones aren’t usually around when you need them most, leaving you grasping at anyone with a shred of credibility.

Even if all appropriate measures are taken to safeguard against unreliable defectors, mistakes still happen. In October 1996, I was invited by the Dutch domestic security service to visit their headquarters in The Hague, where I was shown intelligence acquired from an Iraqi defector code-named Fulcrum. His information provided the precise locations and detailed descriptions of W.M.D. facilities inside Iraq, which subsequent inspections proved to be accurate. Fulcrum was clearly a source with real access. When it came time for me to compile a report about Iraq’s concealed nuclear activities for UNSCOM the next year, I gladly incorporated Fulcrum’s material. “It is assessed,” I wrote,

that Iraq has retained critical components relating to the most recent weapons design, which has not to date been turned over to the IAEA. These components may comprise several complete weapons minus the HEU [highly enriched uranium] core.

I also made reference to this information when I gave testimony before the Senate in September 1998, setting off wild speculation that Iraq was in possession of as many as three undeclared nuclear weapons.

A week or so after my testimony, I received a call from a former I.A.E.A. official. We had worked together closely for several years and had remained friends. He invited me to meet him for lunch at a small Italian restaurant in Manhattan. “We need to have a word,” he said. “This talk about ‘loose nukes’ is getting out of hand.”

“You’ve got it wrong, Scott,” he told me over pasta. “The Iraqi nuclear program has been eliminated. There never was a weapon, let alone the two or three you say are bouncing around in the back of a truck somewhere.”

“You’re missing the point,” I replied. “I never said there was a weapon. I said that there was intelligence about the existence of complete weapons minus the fissile core, that this intelligence was from a credible source, and that I had been prevented from investigating this information. Everything I said was true.”

“Just because it is true doesn’t make it accurate,” the official shot back.

He was correct — Iraq did not have any undeclared nuclear weapons. I had not acted out of malice or bad faith in reporting the Fulcrum data. But there was a fundamental problem in relying on a single defector, even one as demonstrably sound as Fulcrum. The trouble with defectors is that even if they’ve been right in the past, they may be wrong now.

The spy lives in a world of intensely focused actions, where one mistake can mean the difference between life and death. There isn’t much time for reflection — that usually comes in the moments before execution. For a defector, however, time is omnipresent, an eternally ticking generator of remorse. Those who forsake their original loyalties are never again fully trusted. Yuri Nosenko knew this: after his defection, a suspicious CIA held him for more than a thousand days, subjecting him to a brutal interrogation. I can only wonder at the shame he felt while speaking at the DIA conference, denouncing his motherland and its citizenry. But it was during a trip to the Netherlands that I had my most profound encounter with the torment that comes with the act of defection.

On April 10, 1997, I found myself sequestered in a Dutch seaside resort. It was empty that time of year, the off-season, and the place served as the safe house for the debriefing of a new defector under Dutch control. Source A, as he was called, was an engineer who had worked in Iraq’s missile program. Earlier that morning, I had been picked up at my hotel by a Dutch officer who drove me to the resort via a circuitous route designed to throw off anyone seeking to tail us. (He told me that Iraqi intelligence officers were active in the Netherlands at that time, looking for defectors.) During our conversation, Source A revealed that the Iraqis were working on a missile that had a range of 650 to 800 kilometers — well beyond the 150 kilometers permitted by U.N. Security Council resolutions.

But his information came with a warning. “Inspectors will have a very hard time finding these components,” he said. “Unless they are being used in a manufacturing or assembly process, they are safeguarded at the homes of the responsible engineer.” I had heard reports of this kind of activity in the past, but never in such detail. “The security officer is responsible for drafting up an inventory list of all components that have been safeguarded in this fashion. One copy is kept by security, one is kept by the engineer, and one is kept by the plant director. If, at any time, a component goes missing or is unaccounted for, the responsible person will be punished by death.”

Source A identified by name an engineer from the plant’s guidance-and-control design section, who he claimed kept thirty to forty specially imported Russian gyroscopes; another, from the assembly section, who kept a large stockpile of British-made gyroscopic parts; and a third, from the engine section, who kept missile components from Romania and Ukraine. Inspired by this information, I pulled out a high-resolution photomosaic map of Baghdad. “Do you know where these people live?” I asked. “Could you show me where?”

His eyebrows furrowed as he concentrated on the map, and his finger traced a series of roads. Finally he stopped, and tapped where he claimed one of the engineers lived. “This is his front yard, and here is his car.” He also pointed out the address of a second engineer, three houses down.

I took note of everything Source A said, making a rough sketch of the neighborhood and streets. He watched me with sudden concern. “You are not thinking of going to their homes, are you?” he asked.

“Of course,” I replied.

He collapsed into his chair, dejected. “You do not understand. These are my friends and colleagues. I do not support what they are doing — but they are not evil people.”

I was confused. “I don’t understand what your problem is,” I said. “This is how we catch the bad guys.”

He bristled at my comment. “These are not bad guys!” he snapped. “These are people, with families!” He stood and placed his finger on the photograph. “If you inspect these homes, you will be condemning these people and their families to death. They are responsible for this material as long as it is in their possession. Saddam’s people will arrest them, torture them, and kill them, together with their wives and children.”

I stood silently, looking at the map, which only moments before had held so much promise. I studied the locations that Source A had pointed out for me. United Nations inspectors would ultimately choose not to conduct a search, citing concerns over the safety of the occupants. In the room with the source, however, I felt dejected. He must have picked up on my mood, and his voice softened. “You see,” he said, “it is not that simple. Nothing in Iraq is simple.”

The trouble with defectors is that nothing ever is.

is the author of Deal of the Century: How Iran Blocked the West’s Road to War, which will be published in March by Clarity Press.

More from

| View All Issues |

June 2004

“An unexpectedly excellent magazine that stands out amid a homogenized media landscape.” —the New York Times
Subscribe now