Part I · Chapter 10
The Perfect Alibi
The psychiatric label is the most perfect alibi ever built: it does not hide the perpetrator, it destroys the witness — recategorizing a true account as a symptom, so the more accurate the testimony, the sicker it sounds.
A perfect crime is not one committed without evidence. It is one where the evidence exists but has already been sorted, in advance, into a category nobody will believe. This chapter is about the most efficient tool ever built for that sorting, and it belongs to no police force or intelligence service. It belongs to medicine.
The deniability structure traced from the opening pages runs in ascending order of perfection: an accident, which only requires that no one look too closely; a mystery illness, which hides the perpetrator behind the appearance of misfortune; and, at the top of the scale, a diagnosis of mental illness in the person reporting the harm. That last one does something neither of the others can. It does not merely hide the perpetrator. It destroys the witness. Once a target’s testimony has been recategorized as a symptom, no quantity of accurate description can incriminate anyone, because the more precisely the testimony matches what actually happened, the more it reads, to an untrained listener, as elaborate delusion. This is the logical endpoint of the reversal examined in the third chapter — she’s not a victim, she’s ill — and it is where the book’s contested zone, the terrain where an accurate account and a paranoid one are described in identical language, finally closes shut. The label is the lock.
The clearest version of this mechanism was not run by an individual against a partner but by states against their own citizens, with a paper trail thorough enough to leave no honest doubt about what was happening or why.
Between the 1960s and the 1980s, the Soviet Union committed at least twenty thousand dissidents to psychiatric hospitals for what amounted to political crimes. Forensic evaluation ran through the Serbsky Institute in Moscow, where the psychiatrist Andrei Snezhnevsky engineered a diagnosis purpose-built for the task: sluggish schizophrenia, a form supposedly so mild the patient appeared entirely sane, with symptoms that included, by name, “reformism,” a “struggle for truth,” and “perseverance.”1 Opposing the state was, within this framework, definitionally a delusion — a “delusion of reformism.” The purpose was never to treat anyone but to discredit the dissident so completely that nothing he said afterward would carry any weight, and the state built the apparatus with that purpose stated plainly enough to survive in its own records.
The mechanism did not end with the Soviet collapse. China has operated roughly twenty police-run psychiatric institutions, known as Ankang hospitals, since 1987, under the authority of the Public Security Bureau rather than any health ministry. Human Rights Watch has documented political dissidents, petitioners, and members of the Falun Gong movement detained inside them, forcibly drugged and, in some cases, electroshocked; by 2002, more than three hundred Falun Gong members had reported being committed this way.2 The mechanism is operational today, not merely on record from a vanished regime.
The profession itself eventually recognized what had been built in its name, though grudgingly. At the Sixth World Psychiatric Association Congress in Honolulu in 1977, it passed the Declaration of Hawaii, condemning psychiatric abuse generally, and a separate resolution condemning the Soviet abuse specifically. That resolution passed ninety to eighty-eight.3 The narrowness of the margin is its own evidence: an international body of psychiatrists, shown twenty thousand documented political commitments, could barely bring itself to say so out loud.
None of this began with the Soviet program, and the deeper history is more damning still, because it shows the mechanism working just as cleanly with no state apparatus behind the accusation — only an accuser with more social standing than the person being labeled.
In 1851, the physician Samuel Cartwright invented a diagnosis called drapetomania, whose defining symptom was an enslaved person’s desire to escape captivity. For the diagnosis to make sense at all, slavery itself had to be so benign that only a diseased mind could want to leave it. Cartwright’s recommended treatment ran from whipping to, in severe cases, amputating toes.4 It is the purest documented example of resistance reclassified as pathology — and Cartwright’s colleagues, at the time, considered him a credible physician making a credible diagnosis.
For roughly two centuries, a parallel diagnosis did the same work against women who deviated from expected feminine behavior: hysteria, a catch-all broad enough to cover almost any assertion of independence, on the premise that such independence was itself, medically speaking, a form of madness. A woman could be committed on nothing more than her husband’s or father’s demand. Elizabeth Packard, committed to an Illinois asylum in 1860 for holding religious and political opinions her husband disapproved of, campaigned for commitment-law reform once she secured her release.5 The pattern is identical to Cartwright’s: assertion, then diagnosis, then removal — of liberty in one case, of legal personhood in the other.
Neither invention stood alone. Both were entries in a broader nineteenth-century pattern: manufacturing categories to medicalize behavior that threatened a power arrangement, then using the diagnosis to strip a person of legal or bodily autonomy. Coverture, the English common-law doctrine carried into American law, was not a psychiatric category but worked toward the identical end. A married woman’s legal identity was absorbed into her husband’s, leaving her generally unable to own property, sign a contract, or testify in her own right — the same erasure, accomplished through a legal instrument rather than a medical one.6 James Cowles Prichard’s 1835 category of “moral insanity” supplied the direct conceptual ancestor of both drapetomania and hysteria: a finding of insanity based on behavior or morals alone, no delusion or impaired reasoning required — a person could be declared insane purely for failing to conform.7 And in 1860s London, women judged to have excessive sexual desire or independence were diagnosed with what was then called nymphomania and, in the documented practice of the surgeon Isaac Baker Brown, treated with clitoridectomy — which he performed and publicly defended as a cure for nymphomania, hysteria, and masturbation alike. Baker Brown was expelled from the Obstetrical Society of London in 1867 over these practices; even his own professional body eventually recognized the diagnosis as pretext for non-consensual bodily control — though only after the harm had been inflicted at scale.8 This is the chapter’s mechanism at its most literal endpoint: a diagnosis authorizing irreversible physical intervention, not merely commitment or the loss of custody.
The history of the DSM itself supplies the cleanest available proof that a diagnostic boundary moves according to political pressure, not solely according to new biological evidence — and it directly undercuts the assumption that “the DSM says so” is equivalent to “this is settled biological fact.”
Homosexuality remained listed as a mental disorder in the DSM until 1973, when the American Psychiatric Association removed it, following years of activist pressure and one pivotal moment: a 1972 APA convention panel at which the psychiatrist John Fryer testified that he was himself gay, under the pseudonym “Dr. Henry Anonymous,” wearing a rubber mask and speaking through a voice-distorting microphone, because being identified would have ended his career.9 The diagnosis was removed the following year — not because any new biological finding had emerged, but because the professional and social cost of holding the position had shifted. A diagnosis the profession was confident in during 1972 was gone in 1973, and a psychiatrist had to disguise his own face and voice to help move it. A diagnostic category, in other words, is a negotiated boundary, not a fixed fact the profession simply detects.
At interpersonal scale, the medical literature described the mechanism plainly decades before the word entered popular use. “Gaslighting” comes from a 1938 stage play in which a husband manipulates his wife into believing she is losing her mind, to control her and seize her inheritance.10 In 1969, the physicians R. Barton and J.A. Whitehead published, in The Lancet, three real case studies of relatives deliberately fabricating or exaggerating evidence of insanity — staging events, hiding objects — to engineer a person’s involuntary psychiatric commitment, with motives that included seizing assets and removing an inconvenient relative.11 This is weaponized commitment in peer-reviewed medical literature more than fifty years before “gaslighting” became a household word: not a folk concept retrofitted with clinical legitimacy, but something observed and published by clinicians first.
The trap this produces is self-sealing. Coercive control operates through patterns of surveillance, isolation, financial restriction, and intermittent kindness — precisely the kind of detailed, cumulative description that, to a clinician untrained in coercive control specifically, resembles ideas of reference, persecutory delusion, or paranoia. The perpetrator, meanwhile, offers a simple account requiring no framework at all: she’s emotionally unstable, she’s irrational. His account is credible because it demands nothing complex from the listener. Hers sounds implausible for the same reason a conspiracy theory does — not because it’s false, but because it’s intricate, and intricacy reads as fabrication to an unprepared ear.
Borderline personality disorder has become, in the documented literature, the diagnosis most often used to formalize this discrediting. A 1997 study by Zanarini and colleagues at McLean Hospital, published in the American Journal of Psychiatry, found that of three hundred fifty-eight patients diagnosed with the disorder, ninety-one percent reported childhood abuse and ninety-two percent reported childhood neglect before the age of eighteen12 — and the symptom set clinicians associate with it (anger, dissociation, mood swings) overlaps substantially with ordinary trauma responses. Zanarini’s research carries a nuance: childhood sexual abuse specifically is neither necessary nor sufficient to produce the diagnosis; neglect is the stronger predictor. That weakens any simple equation of “borderline diagnosis equals sexual abuse survivor,” but not the broader finding that the diagnosis correlates overwhelmingly with childhood trauma of some kind. Because the diagnosis hinges on a clinician’s judgment about whether an emotional response is “appropriate” to its context, it imports the clinician’s own biases wholesale — a woman’s anger, read through this lens, becomes evidence of pathology rather than a proportionate response to what happened to her. The label then follows her into court, custody proceedings, housing, employment — used to invalidate her account of her own life, for the rest of that life, regardless of how the circumstances that produced her symptoms eventually change.
The pattern by which perpetrators install this outcome is consistent enough to call a playbook, not a coincidence. The abuser brings the victim to therapy himself, frames the presenting problem as her dysregulation or paranoia or jealousy, and establishes himself, before she ever speaks alone with the clinician, as the reasonable narrator of the relationship. Some tell their partners directly that the therapist already agrees she’s the problem. Others attend sessions and gently, persistently contradict her account: she exaggerates, she’s very sensitive, it didn’t happen the way she’s describing. By the time she is alone with the clinician, the credibility asymmetry is already installed. If she becomes visibly upset describing the control, that emotion gets read as confirming the very instability the perpetrator has already named. He rarely needs to lie outright. He needs only a clinician unfamiliar with what coercive control looks like, and the rest runs itself. Evan Stark’s documented case studies, in his foundational text on coercive control, trace victims previously diagnosed with depression, anxiety, and post-traumatic stress — diagnoses that vanished once the victim left the relationship that produced them, evidence that the label had been tracking the perpetrator’s narrative rather than any independent, enduring pathology.13 A related contemporary reframe, the “high-conflict personality” construct popular in family-law circles, does the identical work with softer branding: justified vigilance in a parent who suspects or knows about abuse gets redescribed as obsessive, paranoid, or personality-disordered — the same mechanism in a gentler vocabulary.
Institutions carry their own version of this weapon, built from the same logic. US federal data show that whistleblowers facing retaliation were subjected to a retaliatory psychiatric or fitness-for-duty examination at a rate that rose from 1.6 percent in 1992 to 6.1 percent in 2010. One documented case involved a Department of Transportation secretary who exposed overtime fraud and was forced into a psychiatric exam where an in-house psychiatrist diagnosed her as a paranoid schizophrenic without ever interviewing her. A 2018 paper in the Journal of Business Ethics, titled “Mental Health as a Weapon: Whistleblower Retaliation and Normative Violence,” frames this as the institutional sibling of both the state psychiatric programs examined earlier and the DARVO mechanism examined in the chapter before it.14
The stakes are highest of all in family courts, where a psychiatric label can determine custody of a child rather than merely someone’s reputation. A study led by Joan Meier, funded by the National Institute of Justice and analyzing more than two thousand court opinions between 2005 and 2014, found that mothers’ claims of abuse — especially claims involving child physical or sexual abuse — measurably increase their own risk of losing custody, while fathers’ cross-claims of parental alienation roughly double that risk again. In Meier’s earlier pilot study of two hundred forty cases, mothers who alleged child sexual abuse lost primary custody twenty percent more often than mothers who made no such allegation, and even where a court formally validated the father’s history of domestic violence, fathers still won custody more than forty percent of the time.15 Prior psychiatric diagnosis compounds this: once a mother has any history of mental-health treatment on record, courts appear to discount her abuse claims still more readily, using the diagnosis as grounds for doubt before the allegation is even examined on its merits.
Why well-intentioned clinicians fall for this matters, because the mechanism does not depend on malice but on a handful of ordinary cognitive vulnerabilities operating in sequence. Whoever reaches the clinician first tends to anchor everything that follows — if the initial frame comes from the perpetrator, calmly describing her as dysregulated, all subsequent evidence gets filtered through that anchor, including evidence that should contradict it. Confirmation bias does the rest: once a hypothesis like “she has borderline personality disorder” forms, disconfirming evidence gets reinterpreted as further proof instead of weighed on its own terms. Therapists trained only in generic couple or family conflict, rather than in coercive control, default to a “high conflict” or “toxic dynamic” frame that treats both partners as equally responsible — not merely unhelpful but dangerous, since it can increase lethality by encouraging continued proximity between an abuser and a target he should never be near unsupervised. And the “she’s sick” narrative is simply cheaper than learning an entirely new framework: a single diagnosis explains isolation, surveillance, financial control, and intermittent kindness at once. It is easier, and easier tends to win.
None of this leaves the mechanism unanswerable; the forensic path out is the same discipline the final chapters return to. The most reliable anchor is the longitudinal record: if a woman’s diagnosed symptoms — depression, anxiety, post-traumatic stress — largely resolve after she leaves the relationship, with no other major life change to account for the shift, that divergence is itself forensic evidence that the diagnosis had been capturing the trauma of an ongoing relationship rather than an enduring illness. Comparing symptom patterns during the relationship against those afterward, using therapy notes, medical records, and outside observation, can show the diagnosis was context-dependent rather than pathological in the clinical sense. A second thread examines the perpetrator’s own documented role in producing the diagnosis: whether he attended sessions, contradicted her account there, told her — in any message or recording — that the therapist already doubts her, or pushed for the psychiatric evaluation in the first place. And the deepest reframe of all is an inversion of the ordinary presumption: the mere existence of a psychiatric label is not, on its own, evidence of pathology. In a case built on this mechanism, the label is an artifact of the harm, not proof of it — Exhibit A for the abuse, rather than evidence against the person carrying the diagnosis.
Two theoretical frameworks give this argument its spine. Miranda Fricker’s concept of testimonial injustice names the specific wrong of being disbelieved not for any failure in the substance of one’s account but purely because of who one is — a credibility deficit assigned by prejudice rather than earned by unreliability.16 It is the philosophical name for what the psychiatric label industrializes: a permanent, institutionally backed discount applied to a person’s own account of their own life. Thomas Szasz, in his 1961 The Myth of Mental Illness, argued that psychiatry functions, in part, as a state-sanctioned instrument of social control, using the power to label deviance as illness and to commit people without trial.17 Szasz demands discipline: mental illness is real; the diagnostic apparatus built to identify it is also, demonstrably, abusable; both are true at once. Documenting the abuse of a real medical instrument is a different claim than declaring the instrument a myth, and collapsing the two would tip the argument into antipsychiatry — a much weaker, more easily refuted position. He does not speak for this book.
This is the contested zone at its most sealed: an account filed as illness precisely because it is accurate. But the zone has a second door, and the next chapter has to open it before the book goes any further — because sometimes an account sounds like a delusion for the oldest of reasons, which is that it is one, and a book about the machinery of compliance has to be able to tell the true report from its counterfeit.
Notes
Soviet punitive psychiatry, 1960s–80s: at least 20,000 dissidents committed to psychiatric hospitals for political “crimes”; forensic evaluation through the Serbsky Institute, Moscow; psychiatrist Andrei Snezhnevsky’s diagnosis of “sluggish schizophrenia.” Sidney Bloch and Peter Reddaway, Russia’s Political Hospitals: The Abuse of Psychiatry in the Soviet Union (Gollancz, 1977), the definitive account.↑
China’s Ankang hospitals (police-run psychiatric institutions under the Public Security Bureau, since 1987); Human Rights Watch documentation of detained dissidents, petitioners, and Falun Gong members; 300+ Falun Gong members reporting commitment by 2002. https://www.chinafile.com/library/nyrb-china-archive/chinas-psychiatric-terror↑
Sixth World Psychiatric Association Congress, Honolulu, 1977: Declaration of Hawaii and a separate resolution condemning Soviet psychiatric abuse, passed 90–88. Sidney Bloch and Peter Reddaway, Soviet Psychiatric Abuse: The Shadow over World Psychiatry (Gollancz, 1984); Robert van Voren, “Comparing Soviet and Chinese Political Psychiatry,” Journal of the American Academy of Psychiatry and the Law 30 (2002): 131–135.↑
Samuel Cartwright, “drapetomania,” in “Report on the Diseases and Physical Peculiarities of the Negro Race,” DeBow’s Review (1851).↑
Elizabeth Packard, committed to an Illinois asylum in 1860; later campaigned for commitment-law reform. https://time.com/6074783/psychiatry-history-women-mental-health/↑
Coverture, the English common-law doctrine absorbing a married woman’s legal identity into her husband’s; see William Blackstone, Commentaries on the Laws of England (1765–69).↑
James Cowles Prichard, “moral insanity,” A Treatise on Insanity and Other Disorders Affecting the Mind (1835).↑
Isaac Baker Brown, London, 1860s: clitoridectomy performed and publicly defended as treatment for “nymphomania,” hysteria, and masturbation; expelled from the Obstetrical Society of London in 1867. Isaac Baker Brown, On the Curability of Certain Forms of Insanity, Epilepsy, Catalepsy, and Hysteria in Females (Robert Hardwicke, 1866); his 1867 expulsion is recorded in the proceedings of the Obstetrical Society of London.↑
Homosexuality removed from the DSM in 1973 by the American Psychiatric Association, following a 1972 APA convention panel at which psychiatrist John Fryer testified under the pseudonym “Dr. Henry Anonymous,” disguised with a rubber mask and voice-distorting microphone. Ronald Bayer, Homosexuality and American Psychiatry: The Politics of Diagnosis (Basic Books, 1981).↑
The word “gaslighting” derives from the 1938 stage play Gas Light. Patrick Hamilton, Gas Light: A Victorian Thriller (1938).↑
Barton, R. & Whitehead, J.A., “The Gas-Light Phenomenon,” The Lancet 293:7608 (21 June 1969): 1258–1260. https://www.thelancet.com/journals/lancet/article/PIIS0140-6736(69)92133-3/fulltext; https://pubmed.ncbi.nlm.nih.gov/4182427/↑
Zanarini, M.C. et al., McLean Hospital, American Journal of Psychiatry 154:8 (1997): 1101–1106 — of 358 patients diagnosed with borderline personality disorder, 91% reported childhood abuse and 92% reported childhood neglect before age 18. https://psychiatryonline.org/doi/pdf/10.1176/ajp.154.8.1101↑
Evan Stark, Coercive Control (2007), documented case studies of victims previously diagnosed with depression, anxiety, and PTSD, with diagnoses resolving after leaving the relationship.↑
The rise in retaliatory psychiatric / fitness-for-duty exams for US federal whistleblowers, from 1.6% (1992) to 6.1% (2010), is reported by the US Merit Systems Protection Board, Blowing the Whistle: Barriers to Federal Employees Making Disclosures (2011), comparing its 1992 and 2010 employee surveys. On psychiatric evaluation used as an instrument of retaliation, see also “Mental Health as a Weapon: Whistleblower Retaliation and Normative Violence,” Journal of Business Ethics (2018), https://link.springer.com/article/10.1007/s10551-018-3868-4↑
Meier, J.S., Dickson, S., O’Sullivan, C., Rosen, L. & Hayes, J., National Institute of Justice-funded study analyzing 2,000+ court opinions (2005–2014), published Journal of Social Welfare & Family Law 42:1 (2020); earlier 240-case pilot study. https://scholarship.law.gwu.edu/faculty_publications/1456/↑
Miranda Fricker, testimonial injustice / epistemic injustice (2007). https://iep.utm.edu/epistemic-injustice/↑
Thomas Szasz, The Myth of Mental Illness (1961).↑
From The Machinery of Compliance by Willow Whitman · edition 1.0.2, · free under CC BY-NC-ND 4.0 · corrections