If you are not alone right now, this page will be in your browser history. You can read it in a private window, or clear it when you are done.
Coercive control is a pattern of behavior — isolating, monitoring, regulating and demeaning — that one person uses over time to remove another person’s freedom, rather than a set of separate incidents. Most women living inside coercive control describe it as a series of small things. That description is not a failure of description. It is the accurate one, and the smallness is how it works.
Coercive control is not the same thing as an unhappy marriage, and not every controlling act is coercive control. A husband who is rigid about money, or who sulks for two days, or who argues badly, may be genuinely difficult without controlling anyone. What distinguishes coercive control is a course of conduct that narrows one person’s freedom while leaving the other’s intact. Whether that describes any particular marriage is a question for assessment, not for a page.
What does coercive control actually mean?
Coercive control means a strategic, ongoing pattern in which one partner uses some combination of intimidation, isolation, and regulation of daily life to dominate the other and strip her of basic liberties. The term was developed by Evan Stark, a US forensic social worker and researcher, in Coercive Control: How Men Entrap Women in Personal Life (Oxford University Press, 2007).
Stark’s argument is that the important harm is not the harm to the body. In testimony to the Vermont legislature in February 2015, he defined it as “a strategic course of oppressive behavior in which some combination of physical and sexual violence, intimidation, isolation and control are used to dominate, exploit and/or subjugate a partner and deprive them of basic rights and liberties.” And: coercive control “not only violates women’s rights to physical safety/integrity, but also their basic rights to autonomy, dignity and liberty.”
The clinical construct underneath that is a liberty harm. Ordinary crime frameworks measure injury. Stark’s measure is how much room a person has left to move — how many opinions she no longer says out loud, how many friendships have gone quiet, how much of her week is organized around forecasting one man’s mood. That quantity can be enormous in a marriage where nobody has ever been hit.
You may have gone looking for the word abuse and found it did not fit. Coercive control is the more precise word, and it needs no injury to apply.
Why does it feel like it’s all little things?
Coercive control feels like a series of little things because it is built out of little things. The acts are small on purpose — a small act is deniable, unremarkable to anyone outside the house, and impossible to describe afterwards without sounding as though you are making a fuss about nothing. The pattern is the mechanism, and the pieces were never meant to survive being looked at one at a time.
The most useful clinical model of how this works comes from Mary Ann Dutton and Lisa Goodman, in Sex Roles in 2005. They describe coercion as a process with four moving parts: a demand, a credible threat of some consequence if the demand is not met, surveillance to establish whether it was met, and, occasionally, delivery of the consequence — which keeps the threat credible without needing to be repeated. They also describe setting the stage: the earlier work of demonstrating what he is willing to do, and of creating or exploiting the dependencies that make the threat land.
Dutton and Goodman’s model explains the thing women find hardest to explain. Monitoring is not a quirk and it is not jealousy. Surveillance is a structural requirement of the system. A contingent threat is worthless unless somebody is checking compliance, which is why the monitoring is not separate from the control. It is the part of it that has to keep running.
In a qualitative study of fifteen Australian women who had experienced coercive control, one participant described it in almost exactly the words that recur in clinical work: “these little things all add up, and slowly, over time, you completely lose yourself, your independence, and your desire to make choices” (Kassing & Collins, 2026). The smallness is not evidence that it was not much. The smallness is the design.
What does it look like on an ordinary Tuesday?
Coercive control looks, on an ordinary Tuesday, like a week quietly rearranged around one person’s requirements — and like nothing at all to anyone watching from outside. Four areas come up most often, and one honest note comes first.
The literature does not agree on a tidy set of domains, and anyone who tells you it does is simplifying. Stark’s own grouping is violence, intimidation, isolation and control. Dutton and Goodman sort not tactics but demands, into eight domains including appearance, social life, money, health and the children. Colorado’s statute lists eleven examples. The CDC’s survey measures seven behaviors. The four below are a teaching frame, used because it maps onto how women actually report it. It organizes recognition. It is not a finding.
Isolation
Isolation on a Tuesday is not a locked door. It is that your sister has stopped ringing, because the last four times there was an atmosphere afterwards, and it was easier not to have her ring. It is declining the thing at work. It is the friend he decided against in 2019, and the fact that you cannot now remember whether you agreed with him. Nationally, 16.0% of US women — about 20.4 million — report a partner keeping them from seeing family or friends (CDC NISVS 2023/2024).
Surveillance and monitoring
Monitoring on a Tuesday is a text at 2:40 asking where you are, when he knows where you are. It is the location sharing set up for safety and never turned off. It is composing a truthful account of your afternoon on the drive home, because a truthful account still has to be ready. Tracking or monitoring a partner’s whereabouts is the most-reported form in the CDC data — 18.6% of US women, about 23.8 million.
Micro-regulation of daily life
Micro-regulation is Stark’s term, and it lands hardest because nobody else has a name for it. It is the correct way to load the dishwasher and the fact that there is a correct way. It is what you wear to your mother’s. It is asking before you spend forty dollars in a household where you earn. It is the tone in which a subject may be raised, and the twenty minutes spent selecting that tone.
Degradation
Degradation on a Tuesday is rarely shouting. It is being told, pleasantly, in front of the children, that you cannot be trusted with directions. It is a joke about your body that you are then unreasonable for minding. It is being told you said something you know you did not say, often enough that you have started keeping a private tally to check yourself against.
In clinical work, almost nobody opens with an incident. They open with the forecasting. They describe reading his face in the hallway before the coat is off, and adjusting the whole evening in about a second and a half, and they apologize for how small it sounds. That skill is not a personality trait. It is a trained response to an unpredictable contingency, and it is one of the clearest things a clinician can recognize from across a room.
If you are in immediate danger, call 911.
National Domestic Violence Hotline — 1-800-799-7233, 24 hours · text START to 88788 · thehotline.org
988 Suicide and Crisis Lifeline — call or text 988
Deaf, DeafBlind and hard-of-hearing callers: National Deaf Domestic Violence Hotline videophone 855-812-1001, 24 hours (a partnership between the National Domestic Violence Hotline and Abused Deaf Women’s Advocacy Services).
Calling leaves the number in your call log and texting leaves the thread in your messages. Both can be deleted afterwards. If that is not something you can do safely, a friend’s phone or a payphone is the safer route.
How do I tell this from a difficult marriage?
Telling coercive control from a difficult marriage is not done by counting incidents, and it is not done by deciding whether any single incident was bad enough. It is done on structure: what the conflict is about, what it leaves behind, what repair looks like afterwards, and whether the behavior shows up anywhere other than with you. Marriages contain a great deal of unresolved, recurring, genuinely unpleasant conflict without any of it being control.
| An ordinary difficult marriage | Coercive control | |
|---|---|---|
| What the conflict is about | A thing. Money, the in-laws, the dishes, sex. The subject is real, and it keeps coming back because it is unsolved. | Position. The stated subject changes constantly. What stays constant is who ends up conceding. |
| What it does to your freedom | Nothing durable. You are as free on Wednesday as you were on Monday. | It narrows. After each round there is one more thing you do not say and one more person you do not ring. |
| What happens afterwards | Repair, eventually, and it goes both ways. Each of you gives something. It can take days and it can be graceless. | Restoration of access, not repair. The temperature comes back up when you concede, apologize, or stop asking. |
| Whether it generalizes | It happens with other people too. He is short with his brother, his boss, the man at the counter. | It is selective. It happens with you, and largely in private. Other people describe a different man. |
| What you do to prevent the next one | Nothing in particular. You argue, it ends, you get on with the evening. | Forecasting. You read his face at the door, adjust the plan, and edit the sentence before it leaves your mouth. |
The table above draws a distinction; it is not a scoring instrument. No row on it settles a marriage, and a row that fits does not convict anyone. The clinical value of these axes is that they are about the system rather than about the worst night — and the worst night is the thing you have been made to argue about.
The field’s own attempt to formalize this distinction is Michael P. Johnson’s typology, which separates intimate terrorism — violence embedded in a general pattern of control — from situational couple violence, which arises from specific arguments and is not part of a control pattern (Johnson, Leone & Xu, 2014). The typology is genuinely contested, and you should know that. Joan Meier of George Washington University Law School argues that the underlying data were never designed to measure control, that the cut-off used to sort couples is arbitrary, and that family courts have used the “situational” label to wave away substantiated abuse. Sylvia Walby and Jude Towers go further and argue for a different frame altogether — treating all domestic physical violence as coercive and controlling, rather than carving off a subset and calling that the controlling kind (Walby & Towers, 2018). The distinction is real and useful in a clinical room. It is not settled science, and it is not a verdict.
Why did it take so long for this to have a name?
Coercive control took so long to have a name because the instruments built to detect abuse were built to count incidents, and coercive control is not an incident. It is a course of conduct. An instrument that counts assaults will register a marriage in which nothing countable ever happened as a marriage in which nothing happened. That is a fact about the instrument. It has never been a fact about your marriage.
Stark’s charge against the incident model is blunt. Interventions focused on discrete assaults rather than course of conduct, he told the Vermont legislature, “miss the significance of abuse in the vast majority of cases, fragment and trivialize the reality of partner violence for women and children.” The word fragment is the operative one. Every time you have tried to explain this to somebody and watched it break into pieces in your hands — that is what he is describing, and it is a property of the framework, not of your account.
The clinical lineage is older than the term. Albert Biderman, studying prisoners of war in 1957, produced a chart of coercive methods — isolation, monopolization of perception, induced exhaustion, threats, occasional indulgences, demonstrating omnipotence, degradation, and the enforcing of trivial demands. Domestic violence practitioners recognized their own caseload in it decades before anyone had a phrase. Judith Herman made the same connection in 1992, arguing that survivors of prolonged, repeated trauma under captivity — political prisoners, and women in abusive marriages — develop a syndrome the diagnostic categories did not cover (Herman, 1992). That work became complex PTSD, recognized in the ICD-11 in 2019.
Why this matters to you, specifically: if you have been told you are exaggerating, the absence of a word was doing work in that conversation. The absence of a word was a fact about the vocabulary, not a fact about your life. Almost two decades after Stark named it, US federal data measures it. In the CDC’s National Intimate Partner and Sexual Violence Survey — a nationally representative survey of 15,609 US adults, including 8,842 women, collected September 2023 to September 2024 and published February 2026 — 27.2% of US women, about 34.8 million, reported experiencing one or more forms of coercive control and entrapment by an intimate partner in their lifetime (Zhang Kudon et al., 2026, Table 7). That figure is calculated across all US women, not only across women who reported abuse.
Is coercive control against the law where I live?
Coercive control appears in the statute books of a substantial minority of US states, and in most of them it is a civil concept rather than a criminal one — it makes someone eligible for a protection order, or it is something a court weighs when deciding parenting arrangements. Very few states have made it a crime in itself. The Battered Women’s Justice Project keeps the running national matrix, and it changes. Three states illustrate the range: one where coercive control is defined in the family code, one where it is defined inside the restraining-order statute, and one where a bill defining it has not become law.
Colorado defines coercive control in its family code, at C.R.S. § 14-10-124, as “a pattern of threatening, humiliating, or intimidating actions, including assaults or other abuse, that is used to harm, punish, or frighten an individual” — followed by a list of eleven examples covering isolation, monitoring of money and communications, degrading language, threats to people and animals, threats to publish private material, property damage and immigration threats. In May 2026, HB26-1309 changed how domestic violence, including coercive control, is handled in decisions about parental responsibilities.
California puts coercive control inside its restraining orders instead. Family Code § 6320 lets a court enjoin “disturbing the peace of the other party,” and defines that phrase as conduct destroying the other party’s mental or emotional calm — expressly including “coercive control, which is a pattern of behavior that in purpose or effect unreasonably interferes with a person’s free will and personal liberty,” with five examples covering much the same ground as Colorado’s eleven (Cal. Fam. Code § 6320). South Carolina’s pending bill uses that same sentence below, which is why it reads twice.
South Carolina has no coercive control statute. A criminal coercive control bill, S.702, was introduced in January 2026 and remains in the Senate Judiciary Committee. It defines coercive control as “a pattern of behavior that in purpose or effect unreasonably interferes with a person’s free will and personal liberty.” It has not become law, and nothing in this paragraph should be read as saying otherwise.
This page explains a clinical concept that some legislatures have borrowed. It is not legal advice, and no disclaimer would make a legal statement into a non-legal one. What a statute means for your situation, what evidence a court considers, what you might file and when — those go to a family lawyer in your state and to WomensLaw.org, which is free and organized state by state.
What is not coercive control?
A great many difficult marriages are not coercive control, and a woman who reads this page carefully and finds that it does not fit her has been given a real answer, not a failing grade. Recognition works in both directions or it is not recognition. Four things in particular are not.
A single controlling act is not a pattern. A man who is unreasonable about one thing — money, or your friend from university, or how the kitchen is left — has one unreasonable position. Coercive control is a course of conduct, characterized by generalization across domains and by the narrowing of your freedom over time. One rigid rule in an otherwise open life is a rigid rule.
Unresolved recurring conflict is normal. The observational research on marriage has consistently found that most of what long-term couples argue about is never solved; the same disagreements recur for decades in stable, affectionate marriages. A marriage in which the same fight keeps coming round is not an abusive one for that reason. What matters is whether both people are still free between rounds.
A real condition can be present, and it does not settle the question either way. Depression, ADHD, autism, addiction and brain injury are real, and they can make somebody hard to live with. They are also non-selective — an impairment does not check who is in the room first. Whether it is abuse or whether it is his condition is a legitimate and answerable question, and the honest answer is that both can be true at once.
Coercive control is not a diagnosis, and this page diagnoses nobody. It is not in the DSM-5-TR as a mental disorder; the nearest entry is spouse or partner abuse, psychological, which sits among the conditions that may be a focus of clinical attention rather than among the disorders. Nobody can be diagnosed from a description.
The clinical concern about over-application is also real, and the psychologist Nick Haslam’s work on concept creep — the tendency of harm concepts to widen until they lose their edge — deserves to be taken seriously rather than dismissed.
The women who struggle most with this section are usually the ones it does not fit. They arrive having decided the page will convict him or convict them, and when it does neither they feel they have failed a test. There was no test. A hard marriage with a difficult man in it is a real problem, it deserves real attention, and it is not made smaller by not having this name on it.
What this page can tell you, and what it cannot
This page can tell you what coercive control is, how the research carves it up, why it took so long to name, and what distinguishes it from ordinary marital conflict. This page cannot tell you whether it is happening in your marriage. That is an assessment, it requires a person, and nothing on the internet — this included — is in a position to make it.
A next step, if you want one, does not have to be a decision. It can be understanding why you cannot leave even though you know it is bad, which is about the mechanism rather than about your character. It can be working out what else there is if you are not leaving. If you tried couples work and came out of it feeling like the identified problem, there is a mechanism behind that too, and it is not your failure.
If you would rather talk to somebody: the National Domestic Violence Hotline is 1-800-799-7233, 24 hours, or text START to 88788. Advocates there do safety planning, which is skilled work, done free, with information a website cannot see.
Finding somebody who already knows what coercive control is saves you the work of explaining it from the beginning, so it is worth settling what that specialist training actually consists of and what to ask before a first session.
You do not have to have decided anything to start. Not leaving. Not staying. Nothing at all.
SOURCES
- Stark, E. (2007). Coercive Control: How Men Entrap Women in Personal Life. New York: Oxford University Press.
- Stark, E. (2015). Testimony on Coercive Control, Vermont House Judiciary Committee, H.103, 25 February 2015. PDF
- Zhang Kudon, H., Zhu, S., Chen, B., Breiding, M.J., Leemis, R.W., Zhang, X., Schwank, A., & Basile, K.C. (2026). The National Intimate Partner and Sexual Violence Survey (NISVS): 2023/2024 Intimate Partner Violence Data Brief. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control. PDF
- Dutton, M.A., & Goodman, L.A. (2005). Coercion in intimate partner violence: Toward a new conceptualization. Sex Roles, 52(11–12), 743–756. doi:10.1007/s11199-005-4196-6
- Herman, J.L. (1992). Complex PTSD: A syndrome in survivors of prolonged and repeated trauma. Journal of Traumatic Stress, 5(3), 377–391. doi:10.1002/jts.2490050305
- Johnson, M.P., Leone, J.M., & Xu, Y. (2014). Intimate terrorism and situational couple violence in general surveys: Ex-spouses required. Violence Against Women, 20(2), 186–207. doi:10.1177/1077801214521324
- Meier, J.S. Johnson’s Differentiation Theory — Is It Really Empirically Supported? George Washington University Law School. PDF
- Walby, S., & Towers, J. (2018). Untangling the concept of coercive control: Theorizing domestic violent crime. Criminology & Criminal Justice, 18(1), 7–28. doi:10.1177/1748895817743541
- Kassing, K., & Collins, A. (2026). “Slowly, over time, you completely lose yourself”: Conceptualizing coercive control trauma in intimate partner relationships. Journal of Interpersonal Violence, 41(3–4), 662– 684. doi:10.1177/08862605251320998
- Lohmann, S., Felmingham, K., O’Donnell, M., & Cowlishaw, S. (2024). “It’s like you’re a living hostage, and it never ends”: A qualitative examination of the trauma and mental health impacts of coercive control. Psychology of Women Quarterly. doi:10.1177/03616843241269941
- Tolmie, J., Smith, R., & Wilson, D. (2023). Understanding intimate partner violence: Why coercive control requires a social and systemic entrapment framework. Violence Against Women, 30(1), 54–74. doi:10.1177/10778012231205585
- Biderman, A.D. (1957). Communist attempts to elicit false confessions from Air Force prisoners of war. Bulletin of the New York Academy of Medicine, 33(9), 616–625.
- Colorado Revised Statutes § 14-10-124 (2025), Best interests of the child. Justia
- Colorado HB26-1309, Abuse in Cases of Separation, signed and effective 27 May 2026. Colorado General Assembly
- California Family Code § 6320, Ex parte protective orders — disturbing the peace of the other party — coercive control. Justia, 2025 codification · FindLaw, current text
- South Carolina S.702, Criminal Coercive Control, 126th General Assembly. South Carolina Legislature Online
Educational content. Not therapy, legal, or financial advice. Nothing on this page is an assessment of any person or any relationship.