The Sergeant Who Became Fatemah Zahra

2026-05-16

The Anatomy of a Successful Recruitment

On May 14, 2026, the Federal Bureau of Investigation announced a two-hundred-thousand-dollar reward for information leading to the apprehension of Monica Elfriede Witt, a former counterintelligence officer in the United States Air Force, who, in August of 2013, boarded a plane in Dubai bound for Tehran and, just before takeoff, sent her Iranian handler a message: Coming home. The indictment, returned in February of 2019, describes what happened next. The psychology of espionage allows us to see what happened before.

Discussion of this article — on the LinkedIn post; the link is accessible to members of THE INTELLIGENCE COMMUNITY group.

Three Sentences Worth Reading Twice

In an e-mail exchange from July of 2013—after the F.B.I. had already warned her, in person, that she was being targeted for recruitment by a foreign intelligence service—Witt wrote to her Iranian friend that if all else failed, she just might go public with a program and do what Snowden had done. A week later, frustrated with the laborious pace of Iranian vetting, she added: “I am starting to get frustrated by the level of Iranian paranoia.” And on August 28th, at the airport in Dubai, she signed off with a sentence that ought to be carved into the masthead of every counterintelligence manual: “I’m signing off and heading out! Coming home” (par. 24 of the indictment).

Home meant Tehran. Home meant a country that a woman born in El Paso, Texas, had first laid eyes on in her mid-thirties, at a propaganda conference underwritten by the Islamic Revolutionary Guard Corps. Home meant the place from which Witt would begin assembling dossiers on her former colleagues in Air Force counterintelligence—the same colleagues with whom, fifteen years earlier, she had taken an oath to “support and defend the Constitution of the United States against all enemies, foreign and domestic.”

The question that springs to mind at once is: how. The question worth asking only after some reflection is: why. And it is the second that turns the story of Monica Witt into something more than another entry in the counterintelligence chronicle.

A Career That Would Be Hard to Call a Failure

Witt’s life up to 2008 reads like a brief in favor of discipline. She enlisted in the United States Air Force in August of 1997, fresh out of high school. She was sent to the Defense Language Institute, in Monterey, where she spent fourteen months learning Persian, and from 1999 to 2003 she was deployed abroad on signals-intelligence missions directed at America’s adversaries. From November of 2003 to March of 2008, she served as a Special Agent with the Air Force Office of Special Investigations—an investigator and counterintelligence officer. She held a TOP SECRET/SCI clearance, with access to Special Access Programs identified only by cryptonyms, and to the true names of American assets working under cover (par. 7–8 of the indictment).

Thirteen non-disclosure agreements signed by her hand are in the case file. The most arresting language comes from a document dated November, 1999, in which Witt acknowledged that the unauthorized disclosure of classified information might constitute a violation of federal statutes, including the provision of Section 794 of Title 18 of the United States Code. Fourteen years later, that same provision would supply the foundation of her indictment.

Seen from the outside, her career bears no obvious mark of grievance. She was promoted. She received training. She worked in the very field she had chosen. After leaving active duty in March of 2008, she was hired as a government contractor, remaining the desk officer for the same Special Access Programs she had handled as an agent. Only in August of 2010, in circumstances that have never been fully described in public, did she part ways with the American intelligence community.

Two years later, she was in Tehran.

The Three Elements

Witt’s case ought to be read through a framework that the U.S. intelligence community has refined over four decades of research into betrayal. The most coherent version of this framework is the work of Dr. Ursula M. Wilder, published in 2017 in Studies in Intelligence under the title “The Psychology of Espionage.” Drawing on Project Slammer—a twenty-year C.I.A. study, begun in 1983, of more than forty incarcerated spies—Wilder advances a thesis that sounds banal and is, in fact, foundational. Three elements set the conditions under which a person enters the business of espionage: personality dysfunction, a state of crisis, and the ease of opportunity.

None of these elements is sufficient on its own. Most people with pathological traits never commit treason. Most people who endure profound life crises never commit treason. Most people with access to classified information never commit treason. Betrayal emerges at the intersection of three vectors. This is why it is rare, and why it is so hard to predict. According to the annual reports of the Office of the Director of National Intelligence on security clearances, more than 1.2 million top-secret clearances were active in the United States in the middle of the past decade. The average annual number of arrests for espionage since the mid-nineteen-eighties remains in the single digits. The base rate is, in statistical terms, vanishingly low—the proper lesson for anyone tempted to design early-warning systems. The false-positive rate of any system calibrated finely enough to catch Witt would destroy the careers of tens of thousands of innocent officers.

It is worth noting at the outset that Wilder writes in a tradition whose first generation—rooted in psychoanalysis, with narcissism as the master key to the traitor’s personality—has been rejected by most contemporary researchers as empirically unsupported. Randy Borum, in his 2004 monograph Psychology of Terrorism, summarizes the consensus of the post-Crenshaw, post-Horgan, post-Silke generation in a single line: there is no “spy personality” and no “terrorist personality,” and the search for a psychological profile as a predictor is an epistemological error. Borum quotes Andrew Silke, who in 1998 surveyed the entire literature and noticed a striking pattern. Once researchers stopped finding psychopathology in terrorists, a new trend emerged: it was now asserted that terrorists possessed traits of pathological personality but did not, in fact, meet the criteria for any specific disorder. This development effectively tainted terrorists with a pathology aura, without offering any way to easily test or refute the accusations. When the facts ceased to fit the model, in other words, the model adjusted, and became unfalsifiable. It is a diagnostic observation worth keeping in mind whenever one reads popular analyses of “the traitor’s mind.”

Wilder’s model retains its operational value precisely because it is not a model of personality but a model of configuration. The three elements are not enduring traits. They are states that must converge. A person with a narcissistic-immature personality whose professional life is stable will not betray. A person in deep crisis with no access to information of value will not betray. A person with access and a small inner fracture whom no one is hunting will not betray. The triangle must close in a single biographical moment, and that moment is the window a hostile service tries to recognize and to exploit.

Read against Wilder’s framework, the indictment permits one to reconstruct how the three elements converged in Witt’s case.

Personality Dysfunction

Wilder identifies three principal types of personality pathology found among spies: psychopathy, narcissism, and immaturity. Witt fits none of them cleanly, which is itself operationally informative. The indictment nevertheless reveals a consistent pattern that Wilder would call a narcissistic-immature blend with a strong fantasy component.

The pattern is visible in an exchange with Individual A in October of 2012. Individual A writes that Witt ought to thank the Secretary of Defense for having trained her so well. Witt replies, in a jocular, almost drawing-room key, that she loved the work and is endeavoring to put the training she received to good use instead of evil (par. 17 of the indictment). She is writing this to a woman whose actual professional function is the recruitment of American citizens for the Iranian services. Wilder writes that immature adults “spend an inordinate amount of time daydreaming, deliberately calling to mind ideas that stimulate pleasant or exciting emotions.” Here is daydreaming in mid-performance. Witt is not only casting herself as the heroine of a new story—she is broadcasting that fantasy outward, breezily, to someone whom any rudimentary counterintelligence training should have flagged for her as an operational threat. In that single line, one sees how thoroughly her professional intuition had falsified itself, the intuition that had served her well for a decade.

The line also contains, in miniature, what psychologists call narcissistic moral reconstruction. The state that trained her is implicitly evil. She, using the same training, becomes the good. The logic of this maneuver is deepened over the next eight months: in June of 2013, Witt writes to Individual A that if all else fails, perhaps she will simply go public with one of the programs, and do like Snowden. Wilder describes narcissism as a condition in which the inner world is “built around fantasies about their remarkable personal abilities, charisma, beauty, and prospects.” The Snowden comparison—made by a woman who possessed no organized ideological narrative, no intermediary journalist, no public plan—reveals the chasm between self-estimate and actual means. Wilder calls this the “ascendancy of fantasy over reality.” The decision to become Snowden has been made before any concrete act of disclosure. The role comes first, the script second. It is the exact inversion of the order in which healthy people understand their own affairs.

The third indicator is a deficit in long-range planning. In July of 2013, Witt was weighing, in parallel, Russia as a route to WikiLeaks, Turkey as a country of asylum, the Iranian embassy in Kabul, Dushanbe, Dubai. She wrote that she thought she could quietly slip across the border to Russia, if they would help her, and then contact WikiLeaks without revealing her location. She also wrote that she was beginning to chafe at the level of Iranian paranoia. The frustration is itself diagnostic. Any professional intelligence service offered a source with Witt’s level of access will vet that source with ferocity. Witt expected immediate acceptance as a manifestation of her own worth. Wilder would speak here of an inability to hold the other side’s perspective—an egocentrism that obscures the obvious fact that the recruiter must also be evaluating whether he is being recruited. It is a small observation with weighty consequences: Witt, a professional counterintelligence officer, in the decisive weeks of her life ceased to think counterintelligence-wise about her own situation. Only one side of the table was visible to her: her own.

Crisis

Crisis, Wilder writes, is the second necessary condition. Espionage must be triggered by a crisis and by the person’s judgment that illegal conduct offers either a solution or an escape. The precipitating crisis may be plain to outside observers—a divorce, a lost job, a bankruptcy. It may also be private and invisible.

Witt did not emerge from rock bottom. Her career, until 2008, was, by external measures, a success. But that is the heart of Wilder’s typology: certain crises—the ones most destructive to counterintelligence work—are invisible. They consist of a chronic dissonance between self-image and perceived recognition, between expected reward and ordinary status delivered. Pollard, Nicholson, Hanssen—the spies Wilder analyzes appeared, to outside observers, like successful people, or at least like people functioning steadily. Their collapses unfolded in a private theatre of disappointment.

“States of crisis often produce thought patterns that degrade situational assessment and behavior,” Wilder writes. “A person in such a state of mind may experience tunnel vision, in which attention is fixed on the current crisis. This fixation on the present can erode the capacity for long-term planning and the anticipation of lasting consequences.” This is exactly the state revealed by Witt’s correspondence in the summer of 2013. A woman whose profession had been, for a decade, the long-range planning of operations against hostile services was, in the decisive weeks of her own life, behaving like someone packing a suitcase two hours before her flight.

The indictment notes that, from March of 2008 to August of 2010, Witt served as a desk officer for a Special Access Program in the capacity of a contractor, after which her employment with the intelligence community came to an end. What occurred between August, 2010, and February, 2012—when she surfaced at the Hollywoodism conference in Tehran—the indictment describes only in general terms. The contextual elements fill in the gap. By 2012, Witt was a woman more than ten years into intense military and counterintelligence service, without an active clearance, without current employment in the field for which she had been trained, with fluency in Farsi and specialized knowledge of operations against Iran that she could no longer legally deploy. This is the classic configuration Wilder identifies, citing Edward Lee Howard and Jonathan Pollard: the loss of a professional role combined with a sense of having been treated unjustly by one’s institution.

There is, here, a grim institutional paradox that cannot be evaded. The competencies that make a counterintelligence officer valuable to the employing state are precisely the competencies that make her valuable to the adversary state. Language. Knowledge of identities. Familiarity with procedures. On the employing side, the shock absorber for that value is a long-term career. When the career ends, the shock absorber disappears, and the value remains—just no longer on the same side. Wilder adds an observation whose significance cannot be overstated: prudent risk mitigation in cases of termination or forced resignation should include, where possible, safeguarding the dignity of the individual, so as to inhibit feelings of vengefulness and resentment. For former counterintelligence officers who carry knowledge of asset identities and special-program architectures, this problem is not merely a matter of personnel management. It is a strategic matter. A state that handles its departing officers badly is not committing a breach of professional etiquette. It is drafting a job offer for its adversary.

Opportunity

The third element—ease of opportunity—was not, in Witt’s case, found by her. It was delivered to her. This is where her case diverges qualitatively from the classic Cold War cases of Walker or Hanssen, who walked in on their own. Walker showed up at the Soviet embassy with documents in his briefcase. Hanssen sent a letter. Witt did not have to do anything. It was enough that she attended a conference whose invitation had come to her. She was spotted, assessed, and brought in by a specialized apparatus of the Iranian service.

Wilder describes the mechanism with a precision worth quoting at length. Agents spotting a vulnerable person may insinuate themselves into the situation and find ways to exacerbate the personal crisis, ripening the target’s vulnerability to recruitment. Handlers then continue to manipulate the recruited asset’s vulnerabilities to maintain the person’s long-term engagement in espionage. Every element of this description finds its counterpart in the Witt file, to which we will return in the next section. The final clause of Wilder’s quotation—about maintaining long-term engagement—concerns a phase to which I will return at the end of this essay, and which constitutes the proper playing field of counterintelligence in 2026.

The Path Is Shared; the Outcome Is Contingent

The principal argument of this essay requires careful framing. The literature—from Borum, through Shaw and Sellers’s Critical Pathway to Insider Risk, to Slammer and Wilder—converges on one observation: the path that leads to betrayal is more consistent than its destination. What does that mean?

A person who passes through a sequence—undeserved injury at the hands of an institution; a narcissistic wounding that reframes the injury as a fundamental ethical conflict; social isolation that strips away the friction of dissenting voices; a cognitive opening produced by some discrete event, which renders the previously unthinkable suddenly coherent—arrives at a state that Wiktorowicz calls radicalization, that Wilder calls dispositional crisis, and that the older clinical literature called disorganized identity. The state is shared. What the person does with it is not.

The same person may plant a bomb, hand documents to a journalist, join a cult, enlist in a terrorist organization, or board a plane to Tehran. Which path she takes is determined less by the underlying psychopathology than by who reaches her first and what vocabulary they offer. Ana Montes, the most famous Cuban spy inside America’s Defense Intelligence Agency, met her Cuban handler in New York, introduced by a classmate at SAIS. Witt met New Horizon at a conference in Tehran. The path was not determined by psychology. It was determined by the offer that arrived during the cognitive opening. The radicalization scholar Quintan Wiktorowicz wrote about Salafist recruiters in London in analogous terms: they identified young men in a phase of seeking and offered them a ready-made architecture of meaning. A different offer, in a different city, the same mechanics.

Borum, writing about an entirely different phenomenon, reinforces the observation. He cites Horgan and Taylor on the point that conscious decisions to become a terrorist are rare; most involvement in terrorism is the product of gradual exposure and socialization toward extreme behavior. An analogous logic organizes the espionage literature. Witt did not decide to become an Iranian agent in the sense that people make financial or marital decisions. There was no moment at which she sat down with a sheet of paper and listed the pros and cons. She traversed a sequence of commitments, each of which seemed, in the moment, like another small step. What looks, from the perspective of the indictment, like a strategic decision was, from the psychological perspective, a series of micro-decisions, each of which encumbered the next.

The distinction is of fundamental operational consequence. It means that the prevention of betrayal cannot focus on detecting psychological predispositions as such, because the statistical base is too wide. It must focus on recognizing moments of cognitive opening and on blocking the arrival of recruitment offers during those moments. To put it another way: counterintelligence works when it intervenes in time and space, not in psychology. The door must be shut before someone else knocks, because once someone has knocked and a conversation has begun, any external intervention is folded into that conversation as one of its elements.

The point can be pressed too far, however. Ana Montes was not neutral material waiting for a recruiter to give her direction. She was identified during her student years precisely because she had already articulated a position on American policy in Latin America. The recruiter found a cognitive opening and a preëxisting ideological structure. The same, in a weaker form, holds for Witt. Her February, 2012, conversion, broadcast on Iranian television, may, in some sense, have been authentic. This is precisely the hypothesis that is most awkward for counterintelligence, because a sincere religious conversion cannot be read as an indicator of hostile engagement in the way that signs of living beyond one’s means—the kinds of signals polygraphers are trained to spot—can be. The Iranians did not need to manufacture an ideology. They needed to recognize one forming and to channel it. This is the harder version of the radicalization problem, because it strips out of the equation the classic counterintelligence warning: the spy who does it for money. Witt was not doing it for money. She was doing it, in all likelihood, for meaning—and meaning, unlike cash, leaves no trace on a bank account.

The Tradecraft

Read as a manual of recruitment, the indictment permits a rare reconstruction of the tradecraft executed by the Islamic Revolutionary Guard Corps. It bears noting, and the observation usually gets lost in popular commentary on Iranian propaganda, that the quality of this tradecraft is high. In places, higher than Western services typically attribute to Middle Eastern adversaries.

Spotting Through Exposure

The Hollywoodism conference in Tehran, hosted by the New Horizon Organization, is a textbook spotting operation built around an exposure event. The choreography of such operations has not changed since the nineteen-thirties, when the Comintern used antiwar congresses to the same end. Gather the West’s disenchanted under an ideological banner broad enough to attract people at varying degrees of radicalization. Record the speeches and the identities of the attendees. Observe which attendees return for subsequent events. Escalate commitment through fresh invitations, fresh recordings, fresh public declarations.

Witt appeared in February of 2012. She delivered recorded remarks as an American military veteran criticizing the U.S. government. She took part in a televised conversion to Islam and adopted the name Fatemah Zahra—the name of the daughter of the Prophet, a figure of exceptional standing in Shia tradition. She returned in February of 2013. The pattern of escalation is straight out of the manual, but its psychological power deserves a separate accounting, without which one cannot grasp why it works.

Each public commitment performs two functions in parallel, in two different places. Externally, toward the Iranian side, it serves as proof of loyalty, as operational material, and as a reserve instrument of blackmail. This function is understood by every handler. Internally, in Witt’s own psyche, it functions as an act of what Albert Bandura would call moral disengagement—a mechanism known in social psychology as moral disengagement. Each successive declaration alters the actor’s self-conception in such a way that the prior declarations become retrospectively coherent, and withdrawal from them begins to cost more than continuation. This is the mechanism that Bandura identified while studying how soldiers, doctors, and bureaucrats come to accept participation in atrocity: not through a single moment of moral surrender, but through a series of small displacements, each of which supplies a justification for the previous one. A conversion broadcast on state television is not merely a public event. It is the moment at which the actor’s inner narrative about herself ceases to admit the possibility of return—because return would require admitting that the person before the conversion was disoriented, naïve, or manipulated. The more commitments accumulate, the more expensive this diagnosis becomes.

This is, strictly speaking, the mechanism the Corps exploited systematically. Not through coercion. Through the provision of opportunities for public commitment.

The Role of the Spotter-Assessor

Individual A occupied the role that Wilder calls the spotter and assessor, and that the Russian tradition calls служба наводки. She holds dual American-Iranian citizenship, which gives her lawful access to the United States. She has an operational pretext for contact with Witt: the propaganda film of June, 2012. She keeps the channel open for fourteen months, supplying Iran during that period with a continuous stream of information about Witt’s psychological state, her ideological evolution, the alternatives she is weighing. The prosecution observes that Individual A “took actions consistent with the role of a spotter and assessor on behalf of Iranian intelligence services”—a sentence that ought to serve as an epigraph for everything that followed.

The asymmetry of the relationship deserves separate attention. Wilder analyzes the 1985 case of Sharon Scranage, a former C.I.A. employee in Accra who was recruited by a Ghanaian intelligence officer exploiting a romantic relationship and a sense of isolation. The mechanism is the same as Witt’s: one party offers emotional and logistical support, while the other furnishes information, initially believing she is doing so out of friendship rather than out of betrayal. The threshold between helping a friend and transferring intelligence material erodes gradually, in small acts, each of which, taken alone, still looks like kindness. Witt’s correspondence with Individual A reads like an indictment only because we know how it ends. Read in real time, in the summer of 2012, it might have looked, to her, like a deepening friendship with someone who understood her disappointments.

When the decisive moment arrived, in August of 2013, Individual A forwarded the validation package to Iran in roughly nine minutes from the time of its receipt. The nine minutes are not a technical curiosity. They are the signature of the tradecraft. A transmission so swift means that the destination address was known in advance, that the format was anticipated, that the instructions on priority had already been issued. This is the infrastructure of a prepared operation, not the spontaneous reaction of a woman whose friend has unexpectedly written to her. Every counterintelligence officer reading the indictment pauses over those nine minutes, because that is where, in a temporal detail, the full depth of Iranian preparation reveals itself.

Verification, Not Praise

The most striking, and most counterintuitive, element of the tradecraft is the way the Corps treated Witt in July of 2013. Individual A wrote on July 1st that her Iranian interlocutors had grown suspicious, because on the one hand Witt was saying she had no money, and on the other she was traveling from country to country. Iran did not welcome Witt with enthusiasm. Iran vetted Witt with such severity that Witt herself began to waver.

This is the proper standard of practice for handling a prospective source at the level of access she offered. It is worth lingering here, because much of the popular confusion gathers around this point. The cinematic notion suggests that a hostile service receiving a volunteer from American counterintelligence greets him with open arms, offering wine and money. The professional reality is precisely the opposite. Any handler who welcomed Witt at once and with enthusiasm would have exposed his service to the risk of accepting a plant. The history of intelligence operations is littered with catastrophes produced by excessively warm receptions. American counterintelligence in the nineteen-eighties lived off Russian enthusiasm for certain volunteers. The Corps had absorbed that lesson, and its suspicion was professional, not xenophobic.

The money—cash in Dubai, an airport escort, an apartment and a computer on arrival—appeared only once Iran’s own verification had convinced the Corps that Witt was who she said she was. Wilder writes that a good handler presents himself not only as willing to reward espionage but also as capable of protecting his agent. Managing the source’s composure during transit is a rational element of that capability. A source rattled at the border is a source who will answer the wrong question put to her by the wrong passport-control officer.

The Integration of HUMINT with the Cyber Layer

What distinguishes the operation against Witt from classic Cold War operations is the integration of human intelligence with cyber intelligence within a single campaign, in real time, against the same set of targets. The indictment names, alongside Witt, four Iranian nationals: Mojtaba Masoumpour, Behzad Mesri, Hossein Parvar, and Mohamad Paryar. They are employees or associates of the same Tehran-based company, of which Mesri remained chief executive, and which, on behalf of the Corps, conducted cyber operations against American targets.

The modus operandi laid out in the indictment is a textbook example of modern spear phishing combined with social engineering. The Bella Wood account, created in January of 2015, sent a message containing a seemingly innocuous electronic New Year’s card to an American officer deployed in Kabul, working on a Department of Defense network. The link was meant to launch malware on his work computer. A Facebook account impersonating one of the American officers under his real name, with his photographs and personal details, sent a friend request to a second officer; a third, taking the impersonator for the real person, recommended the account in a private group of operatives. The circle of targets overlaps with the circle of people whose identities Witt knew from her service. The human intelligence supplied by Witt—real names, photographs, social ties, biographical detail—became the operational material for the cyber layer.

Among the charges against the cyber conspirators is aggravated identity theft. It signals that the Iranian operation was no longer the sum of individual offenses, but an institutionalized process of industrial attack production. The indictment, with stoic calm, notes that Mesri’s company “operated in many respects like a typical business or organization, with regular salaries paid, working hours established, tasks assigned, and supervisors and managers employed.” This normality is, in fact, the most disturbing sentence in the entire document.

The implication is worth stating plainly, because it gets lost in many analyses. The betrayal of a spy in the digital era does not end with the handing over of documents. The transmitted information becomes training data for a multi-year operational campaign, which evolves technologically long after the spy has stopped supplying fresh material. Witt delivered her information in 2013. The cyber campaign built on that information ran at least until 2019, when the Department of Justice filed charges, and probably longer. That is a seven-year, probably longer, operational composition from a single human source. The old arithmetic of counterintelligence damage—measured in the number of documents transmitted—is inadequate for this era. One must count in years of campaign, not in pages of reports.

The Anatomy of a Warning That Did Not Work

In May of 2012, three months after Witt’s first visit to Tehran, F.B.I. agents warned her that she was being targeted for recruitment by Iranian intelligence services. Witt replied that if she ever returned to Iran she would refuse to provide any information about her work at A.F.O.S.I. The sentence, read at a remove, is remarkable in two ways. First, it contains the assumption that she will return. Second, it contains the assumption that she will be asked. Neither premise should have been accepted as a neutral promise made by a former counterintelligence officer to the F.B.I. agents sitting across the table from her. They were not a promise. They were a forecast.

Thirteen months later, Witt wrote to Individual A that she had walked into the Iranian embassy in Kabul and told all. A defensive declaration had become a voluntary disclosure in the span of a single year.

This is the moment at which counterintelligence, as a discipline, must look at itself critically. The lesson is not that Witt was warned too gently, or that the warning was poorly phrased. The lesson is that the warning was delivered at a moment when its function had already ceased to operate. Put more bluntly: the warning produced the exact opposite of its intended effect.

Wilder writes that recruitment proceeds as a sequence of irreversible commitments. Each successive commitment narrows the space of exit. The first visit to Tehran was a commitment. The recorded appearance as a veteran criticizing the U.S. government was a commitment. The televised conversion ceremony was a commitment. Each of these acts was registered both externally, by the Iranian side, and internally, in Witt’s own psyche, as an act that could not be undone without the loss of her own self-narrative. This is precisely the double mechanism I described in the previous section: external blackmail and internal moral disengagement operating in synchrony.

The F.B.I. warning of May, 2012, delivered three months after the first visit, arrived after two of those commitments. In Borum’s framework, the cognitive opening had already occurred, and the radicalization path was already active. More important still, taking seriously Horgan and Taylor’s observation about the gradualism of the process, the very architecture of the warning was inadequate to its object. A warning presupposes a decision that can be reversed. There was no decision; there was a trajectory. A trajectory cannot be reversed in the way a decision can be reversed, because a trajectory has no single point at which it was made. It consists of dozens of small displacements, each of which, in the moment of execution, looked like nothing at all.

The warning did not function as an intervention. It functioned as a stressor. More than that, it functioned as a gift to the opposing side. The other side—Individual A and her handlers—almost certainly registered the warning as information about the state of pressure on Witt, and used that information to deepen her emotional dependency. One can imagine the conversation Individual A conducted with Witt after the warning was disclosed: full, no doubt, of sympathetic concern, full of inquiries after her well-being, full of signals that Iran was prepared to protect those whom their own government regarded as suspect. Warnings delivered at the wrong moment in the recruitment cycle are metabolized by the adversary and folded into the operation. They become an element of the narrative about the American state’s persecution of a sincere person in pursuit of truth.

The implication is uncomfortable. If counterintelligence warnings work only when they are delivered before the first irreversible commitment, then detection must be significantly earlier, significantly more sensitive, significantly more integrated with patterns-of-contact analysis. That, in turn, requires a level of monitoring that Western legal systems do not sanction with respect to persons who have committed no offense. The dilemma is structural, not operational. A democratic state that wishes to warn effectively against recruitment must begin monitoring before any criminal act has been committed; a state that monitors persons before any criminal act has been committed ceases to be a democratic state in the sense in which it was. There is no elegance to the solution—only the choice of which price to pay.

The Epistemic Limits of Counterintelligence

The careful reader will have noticed that the entire foregoing analysis rests on post-hoc observation. This is the proper place to formulate a methodological warning that Wilder herself does not avoid. Experts on espionage have at their disposal only the spies who were caught. Those who escaped detection are, by definition, absent from the sample. Every model of three elements, of four indicators, of five phases, is a model of caught fish—not a model of the population.

The consequence of this limitation is graver than is commonly supposed. Suppose there exists a type of spy personality that is never detected, because it is too careful, too disciplined, too operationally adept to fall to routine vetting. Suppose this type constitutes some percentage of the spy population. In the research sample on which Wilder relies, this type does not exist. The conclusions drawn from this sample not only fail to describe this type—they actively obscure it, by suggesting that spies share the characteristics of the caught. All of our knowledge about the psychology of betrayal is, in a sense, knowledge about the psychology of incompetent betrayal. This is an uncomfortable lesson, one that cannot be set aside in any analysis of the Witt case: the type she represents may be a subspecies, not a species.

The second limitation is more fundamental. The patterns visible in cases like Witt’s are visible only after the fact. The indicators that matter most are precisely those that, while in progress, look like ordinary human life. A religious conversion. A divorce. A foreign friendship. A trip to a conference. None of these is, on its own, a counterintelligence event, and treating it as one would generate a false-positive rate that would destroy the institution counterintelligence is meant to serve. The signal exists in the sequence, not in the individual elements. Sequences resolve into signals only after the source is already on the plane.

Borum adds to this problem an observation whose weight cannot be overstated. The predictors of general violence, well documented in criminology and forensic psychology, do not map onto the predictors of betrayal or terrorism. Risk factors, as Borum puts it, citing his own research, operate differently at different ages, in different groups, and for different types of behavior. Most people with a full slate of general-violence risk factors will never commit treason. Many known terrorists, including some field leaders of the September 11th attacks, possessed few of the classic risk factors. From this follows a consequence that must be stated sharply: counterintelligence cannot borrow its predictive instruments from neighboring disciplines. The predictors of deviance are not the predictors of betrayal. Embedding them in an early-warning system would not improve detection. It would generate orderly noise, in which the signal becomes harder to find, not easier.

Philip Tetlock’s work on judgment under uncertainty applies directly. The base rates are too low. The signal-to-noise ratio is too weak. The cost of a false positive is borne by the innocent officer whose career is quietly damaged by a flag that should never have been raised. The system tolerates false negatives—until one of them becomes Witt, and then everyone studies the file and concludes that the signs were obvious. They were not. They were obvious in the configuration they ultimately assumed, which is a different epistemic claim altogether. The human brain, in possession of knowledge about an outcome, automatically reorganizes memory so that the signals leading to that outcome appear more visible than they were. Daniel Kahneman called this hindsight bias. Every post-mortem investigation in counterintelligence cases is, in part, an investigation and, in part, a demonstration of that bias.

The practical implication should be stated without ornament. Counterintelligence cannot promise to predict betrayal as a statistical capability. It can promise only one thing: the swift recognition of a pattern when it emerges, and a swift operational adaptation to that recognition. What Witt took with her to Iran was lost the moment she first walked into the Iranian embassy in Kabul. The value of studying her case lies not in any prospect of preventing the next Witt. It lies in the prospect of recognizing the next one faster—and, as I will argue in what follows, of opening a door on the other side of the cycle. The Witt case is not a warning about what the system missed. It is a map of a second operation that the system can now undertake.

The Symmetry of Instruments

Recruitment is a sequence of irreversible commitments. The first appearance in Tehran was one of them. The televised conversion to Islam was a second. The employment with Individual A was a third. The Snowden-tinged correspondence was a fourth. The visit to the embassy in Kabul was a fifth. Each successive step narrowed the space of exit, until Witt’s final message to Individual A before boarding the plane out of Dubai read: I’m signing off and heading out! Coming home (par. 24 of the indictment).

The word “home” tells you everything about the state of the operation at that moment. It is not a word describing a destination airport. It is a word describing the completion of a psychological process that had begun eighteen months earlier, when someone invited a disenchanted former American counterintelligence officer to a conference under the auspices of the Islamic Revolutionary Guard Corps. Witt was not going to Iran as a country of emigration. She was going home in the sense that developmental psychology gives the word. To the place in which the identity she had arrived at over those eighteen months was congruent with the surroundings. The United States, in that sense, had ceased to be home three months after her first visit to Tehran. The rest was already geography.

What Iran received can be summarized in three categories. The cryptonym and mission of the special-access program to which she herself had been assigned as a desk officer. The real name of a counterintelligence officer and the target against whom that officer was working. Eighteen months of target packages—dossiers built to find, locate, track, and neutralize her former colleagues. The term “target package” is precisely defined in the indictment: a neutralization plan may involve detention, recruitment, cyber exploitation, or a capture-or-kill operation.

But the literature on which this essay rests draws a distinction that, until now, I have set aside, and that now organizes the remainder of the analysis. John Horgan and Max Taylor, reading the classical espionage studies through the lens of developmental criminology, divide the cycle of an agent’s life into three phases: becoming, being, and leaving. Everything I have analyzed up to this point concerns the phase of becoming. It is the phase in which most of the books are written, because it is the phase in which most of the indictments are returned. The second phase, being, is less visible; the third phase, leaving, is nearly invisible. Each, however, requires a different theory and a different set of instruments. More than that: each offers the other side a different operational window.

Witt, in 2026, is in the second phase, in its thirteenth year. Her handlers are in the second phase with her. Wilder writes that a good handler manipulates the recruited asset’s vulnerabilities to maintain the person’s long-term engagement in espionage. Manipulation in this phase is daily labor, not an event. The apartment is granted. The computer is granted. The stipend is granted. Access to family in the States is denied. Travel is denied. Dependence is engineered, materially and emotionally, so that each day of continuation costs less than a day of attempted departure.

This is, strictly speaking, the situation Martha Crenshaw, cited by Borum, describes among members of terrorist organizations. Continued membership is sustained by mutual interdependence, peer pressure, sensitivity to betrayal, and security risks. Hans Joachim Klein of the Baader-Meinhof Gang said of his own efforts to leave that the only way out led through the cemetery. For Witt, the cemetery is not a figure of speech. It is, literally, geography: the Iranian border, in one direction. Iran does not permit the departure of persons who know what Witt knows—not from the harshness of the regime, but from the ordinary logic of the service. A source who leaves becomes a source for someone else.

The second phase, however, has its internal fissures, which counterintelligence learns to recognize. Borum, analyzing the weaknesses of extremist groups, lists four: internal mistrust, boredom and inactivity, internal competition for power, and serious disagreement. The first two, under the conditions of a regime in protracted crisis, escalate in predictable ways. Internal mistrust grows as the regime turns paranoid toward its own apparatus. This is a phenomenon observed historically many times: every security service under pressure begins to consume its own tail, because it is easier to suspect a colleague than to suspect an adversary. Boredom and inactivity afflict those members of the organization whose operational function was defined by a particular campaign—the cyber campaign against Witt’s former colleagues, for example—and who, in the thirteenth year after her arrival, have less and less to do, because the information she supplied is older and older, her former colleagues have changed assignments, been promoted, retired, or, in some cases, died. The operational value of a source in the second phase declines monotonically. These are precisely the conditions under which Borum locates the moment of susceptibility to departure from the organization.

This is where the third phase of Horgan and Taylor begins—the phase that the F.B.I., with its two-hundred-thousand-dollar reward, has just entered.

The Bureau’s announcement of May 14, 2026, leaves no doubt as to its intent. “The F.B.I. believes that, at this critical moment in Iran’s history, someone knows something.” The sentence must be read twice. On one level, it is a standard appeal to potential informants. On another, more significant level, it is a signal that the American intelligence community judges the current internal situation in Iran to be fluid enough to warrant the risk of publicly raising the stakes.

And here the symmetry that is the proper thesis of this essay reveals itself. The operation the Bureau is mounting in 2026 uses precisely the same instruments that the Corps used in 2012. Spotting and assessment: the Bureau is identifying who in Witt’s circle is in a state of dispositional crisis. Insinuating itself into the situation: the reward is a communication that someone is noticing the crisis and is prepared to treat it as legitimate. Provision of resources: money, presumably extraction, presumably a new identity. The gradual construction of dependence: contact begins with a single piece of information and ends in full coöperation. Reframing: the person who is considering betraying the regime receives, from the American side, the very gift that Witt received from the Iranian side—a vocabulary in which the act is not betrayal but fidelity to values higher than the loyalty owed to one’s state.

The mechanics are identical. This is not an analogy; it is a structural identity. What differs is the aim, and, importantly, what Borum calls the “direction of activity” of the ideology. The Iranian apparatus in 2012 operated in destructive mode, directing the recruitment cycle toward the elimination of the American officers whose identities Witt knew. The Bureau, in 2026, operates in recovery mode, directing the cycle toward the apprehension of the perpetrator and the protection of those still in harm’s way. This is not an ethical symmetry. It is a technical symmetry. The ethical asymmetry persists, but it is a different chapter of the same story.

The symmetry of instruments is, in a certain sense, the most important lesson of the Witt case. The mechanisms of recruitment are universal. They belong to no culture and to no service. They can be described in the terms of Wilder, of Borum, of Horgan and Taylor; or in the terms of operational writers from the nineteen-thirties, who used the same mechanisms toward different ends. Every society produces persons in a phase of cognitive opening. Every service may notice them. What decides the outcome is the speed of response and the quality of the offer. In 2012, the Corps noticed Witt faster than the F.B.I. did. In 2026, the question is whether the F.B.I. will notice the persons around Witt faster than the Iranian internal apparatus does.

The Witt case will not end with an arrest in a courtroom. Iran will not surrender her to the United States. It may, however, end in a fashion that will never be published in the form of an indictment: in the office of a counterintelligence officer to whom, one day, someone from Witt’s Tehran circle walks in and says that he has a few hours before his absence is noticed. That is the proper function of the two-hundred-thousand-dollar reward at this phase of the case. It is not about Witt as a person to be apprehended. It is about Witt as a node in a network in which institutional memory is measured not in years but in decades—and in which the instruments of recruitment turn out, in the end, to be held in both hands.

What remains is a question that is, in truth, not a legal question or even a counterintelligence question. It is an anthropological one. How does it happen that a person who, over fifteen years, signs one non-disclosure agreement after another, writes one day to a friend “Coming home,” and means Tehran?

The answer, Wilder suggests, is never singular. It is composed of a constellation of personality, of crisis, and of opportunity, all fanned by a professional recruiter who can present himself as a safe patron. The constellation can form anywhere—in allied services, in corporations, in law firms, in regulated industries. The Witt case is an American parable, but the warning it issues is universal in scope.

Sometimes the largest gap in the system is not a server. It is a human being who, long ago, stopped feeling at home.

Sources

This essay draws on four sets of primary sources: the indictment in United States v. Witt et al., No. 1:19-cr-43, U.S. District Court for the District of Columbia, January, 2019; the Department of Justice press release of February 13, 2019; the F.B.I. notice of May 14, 2026, accompanying Witt’s current wanted listing; and Ursula M. Wilder, “The Psychology of Espionage,” Studies in Intelligence Vol. 61, No. 2, June, 2017. The theoretical layer on radicalization draws on Randy Borum, Psychology of Terrorism, University of South Florida, 2004. Quantitative data on security clearances are drawn from the annual reports of the Office of the Director of National Intelligence. Quotations from Witt’s correspondence, set within quotation marks, are translations from the original English, the full text of which is found in the indictment.

Topical publications

2026-05-03

Internet Frauds

Why a Nineteenth-Century Doctrine of Defective Consent Cannot See the Twenty-First-Century Confidence Game — and Whose Interests That Blindness Quietly…

2026-05-02

The Same Trick Twice

How Poland’s financial system spent twenty years perfecting the art of never learning from its mistakes In the spring of…