One of the more dangerous misconceptions in family law, and in popular understanding, is that separation ends intimate partner abuse. The peer-reviewed evidence is unequivocal on this. Separation is a risk factor, not a solution. For survivors of coercive control, the months and years after a separation are often the most dangerous period of the relationship, and the tactics shift to fit the new context.
This article is about what coercive control is, how it continues and escalates after separation, the specific tactics used in the post-separation phase, and why family courts so often fail to recognize it. It draws on peer-reviewed research, CDC data, and the work of researchers who have spent decades studying both the dynamics and the institutional responses.
What coercive control actually is
The foundational text on coercive control is Evan Stark's 2007 book of the same name (Oxford University Press), now in an updated edition. Stark, a sociologist and forensic social worker, reframed intimate partner violence away from the older incident-based model (count the bruises, count the police calls) toward a pattern-based model (look at the whole architecture of control).
Stark defines coercive control as a pattern of behavior, not a collection of isolated incidents, that "seeks to take away a person's liberty or freedom and strip away their sense of self, including bodily integrity and human rights." The pattern is "designed to make an individual dependent by isolating them from support, exploiting them, depriving them of independence, and regulating their everyday behavior."
The reframing matters. Stark argues that coercive control is best understood as a liberty crime, not a violence crime. Physical violence may be present, but the most damaging effects are the systematic deprivation of a victim's capacity to act freely, make decisions, seek help, and maintain a sense of self. This is what distinguishes coercive control from what researcher Michael Johnson calls "situational couple violence": conflict-based, bidirectional incidents without a pattern of domination.
Stark's framework directly shaped the criminalization of coercive control in the United Kingdom (Serious Crime Act 2015, section 76), Scotland (Domestic Abuse (Scotland) Act 2018), Ireland (Domestic Violence Act 2018, section 39), and Australian states including New South Wales and Queensland. In the United States, California, Hawaii, Connecticut and a growing list of other states have added coercive control to the conduct that can support a domestic violence protective order. Hawaii's statutory definition (Haw. Rev. Stat. § 586-1) tracks Stark's language almost word for word.
Separation is not safety
The 2023 literature review by Spearman, Vaughan-Eden, Hardesty and Campbell in the Journal of Family Trauma, Child Custody and Child Development states the position cleanly: separation is commonly assumed to be the solution to ending intimate partner violence, but a robust body of research has identified that separation is a risk factor for lethality, for continued or worsened abuse, and occasionally for the first onset of physical violence.
The lethality data are blunt. In the Campbell et al. (2003) landmark femicide case-control study, 44 percent of women murdered by an intimate partner had separated or were in the process of leaving. The combination of a highly controlling partner and separation produced a roughly ninefold increase in the risk of femicide (adjusted odds ratio 8.98). Across 42 reporting states in the CDC's National Violent Death Reporting System, 50.8 percent of women murdered are killed by a current or former intimate partner, compared with 7.2 percent of men. Roughly 46 percent of family mass shootings involve separation, divorce, or child custody disputes as a precursor (Fridel, 2021).
The structural reason for the elevated risk, in Lundy Bancroft's framing, is that an abuser's drive for control intensifies as he senses the relationship slipping. The departure is experienced as the ultimate loss of control, and the tactics escalate in response. Bancroft has spent more than a decade working directly with over 1,500 abusive men in intervention programs, and his observation is consistent with what survivors report and what the homicide data confirm.
The tactics shift to fit the new context
What ends with separation is often the cohabitation, not the control. The Spearman 2023 review identified five major categories of post-separation abuse tactics, each well documented in the peer-reviewed literature.
Litigation abuse
Spearman and colleagues describe legal abuse as a form of abuse that arises specifically in the post-separation context. It provides another avenue for forcing contact and inflicting harm through repeated court proceedings. The specific tactics documented in the literature include filing frivolous motions and emergency hearings to drain finances and psychological resources, filing counter-restraining orders against the victim, making false allegations of child abuse or "parental alienation," and using custody litigation to force ongoing contact. Bancroft notes that batterers may continue harassment for years through legal channels, causing periodic retraumatization and destroying the family's financial position.
A validated psychometric instrument now exists to measure this. Gutowski, Lawrence and Goodman developed the Legal Abuse Scale (LAS) and published their work in 2022. The scale captures how perpetrators use the legal system itself as a tool of ongoing coercive control. A 2026 study in the Journal of Family Violence found higher rates of court filings in separating couples with documented intimate partner violence, consistent with litigation as a tactical extension of control.
Financial and economic control
Spearman, citing Lin et al. (2022), reports that 94 to 99 percent of survivors in one study experienced economic abuse. Post-separation tactics include hiding assets, refusing financial disclosure, failing to pay court-ordered child support, withholding medical expenses, sabotaging the survivor's employment by calling repeatedly or showing up at the workplace, and deliberately exhausting the survivor's legal resources through endless motions. The interaction between legal abuse and economic abuse is direct. Litigation can be a tool of economic control. Survivors often find themselves unable to obtain adequate representation because their resources have been deliberately depleted.
Weaponizing children
The 2023 review describes a category that survivors recognize immediately: using children to coerce and control, and using tactics that harm survivors' identities as mothers. The specific behaviors include neglecting children's needs in order to cause the other parent distress, using children to keep track of the other parent's whereabouts or to force contact, threatening to harm or kidnap children, manipulating children's statements to custody evaluators, denying children access to medications or healthcare by using joint legal custody to veto medical decisions, and what Clements and colleagues (2021) call "custody stalking": using the court to obtain parenting time not for meaningful involvement with the children but to retaliate against the other parent. Bancroft documents the same pattern, including unfounded accusations against new partners and abrupt custody filings as retaliation.
Surveillance and technological abuse
Established risk factors for stalking include shared children and recent separation. Technology-facilitated tactics include spyware and tracking apps installed on shared or children's devices, monitoring via family location apps, constant contact across multiple channels, weaponizing shared streaming and smart-home accounts, image-based abuse, and enlisting strangers in coordinated online harassment campaigns. The geographic boundary that physical separation used to provide has been substantially eroded by technology.
Isolation and "mesosystem" abuse
Spearman uses the term mesosystem abuse for tactics aimed at the survivor's wider community: spreading rumors to family, friends, employers and faith communities, filing reports with child protective services as harassment, contacting the survivor's employer with false allegations, intimidating attorneys and advocates, and running defamation campaigns on social media. Bancroft notes that in the most extreme cases, custody evaluators have themselves been afraid to release their recommendations out of fear of the abuser's retaliation.
What the prevalence data show
The CDC's National Intimate Partner and Sexual Violence Survey, in its 2023/2024 data brief published in early 2026, found that 34 percent of women in the United States experience contact sexual violence, physical violence, or stalking from an intimate partner in their lifetime, and that 30.2 percent experience psychological aggression. Coercive control and entrapment specifically were reported by 27.2 percent of women and 19.5 percent of men over a lifetime. The most commonly reported specific tactics affecting women included a partner demanding to know where they were at all times (18.6 percent), trying to keep them from seeing family or friends (16.0 percent), making decisions that should have been theirs to make (15.2 percent), and threatening to harm themselves or threaten suicide (14.0 percent). These are baseline national figures, not worst-case estimates.
Peterson and colleagues estimated the lifetime cost of intimate partner violence at approximately $103,767 per female victim, with a national total approaching $3.6 trillion. The economic argument for getting the institutional response right is, on its own, substantial.
Why family courts miss it
The failure of family courts to identify and respond appropriately to coercive control is one of the most extensively documented failures in the civil justice system. The empirical work of Joan Meier at George Washington Law School is the most thorough. The Family Court Outcomes Study, a ten-year national analysis of 4,338 cases (Meier, 2020, Journal of Social Welfare and Family Law), found that fewer than half (41 percent) of mothers' abuse claims were credited by courts. Child sexual abuse claims were credited only 15 percent of the time in non-alienation cases. When a father cross-claimed "alienation," courts were more than twice as likely to disbelieve mothers' abuse claims of any type, and almost four times more likely to disbelieve claims of child abuse. In Meier's data, child sexual abuse became "virtually impossible to prove" when an alienation defense was raised: 1 case out of 51 (2 percent) was believed. When fathers' alienation claims were credited, mothers' custody loss rates rose to 73 percent. Meier summarized: "alienation trumps abuse."
The structural causes are several. Family courts were built around an incident-based model of intimate partner violence, looking for physical acts, bruises, police reports. Coercive control is a pattern, and much of it is non-physical. The "high-conflict" label is routinely applied to cases involving genuine coercive control, prompting interventions (co-parenting therapy, parenting coordination, shared custody) that require continued contact between the survivor and the perpetrator. Strong presumptions in favor of shared parenting can frame a survivor's protective behavior as interference. And Bancroft has documented an uncomfortable asymmetry in court presentations: a victim of long-term coercive control often presents as hostile, agitated and disjointed (because they are traumatized), while the perpetrator presents as articulate, calm and friendly (because for a brief, strategically important period, they can be).
The widely held belief that custody disputes are full of intentionally false abuse allegations is not supported by the data. The US Department of Health and Human Services (2022) reported that of 3.3 million child abuse reports, only 1,223 (0.04 percent) were intentionally false. The Canadian Trocmé and Bala study (2005) found that of 135,574 child maltreatment reports, only 4 percent were intentionally false, and that custodial mothers (14 percent) and children (2 percent) made the fewest. Most intentionally false allegations in custody disputes come from non-custodial parents, typically fathers (43 percent).
What survivors actually need to know
Three points worth being direct about.
The first is that the elevated lethality risk in the months around separation is real and worth taking seriously. If you have left or are planning to leave a coercively controlling partner, a domestic violence advocate can help you build a safety plan that is calibrated to your actual risk. The National Domestic Violence Hotline (1-800-799-7233) is available 24/7 in the United States. Local DV programs typically offer safety planning, legal advocacy, and access to attorneys with specific domestic violence training.
The second is that documentation matters and the form of documentation matters. Pattern, not incident, is what shows the architecture of control. A timeline of communications, financial records, contact attempts, restraining-order violations, and third-party witness statements is more useful than any single dramatic incident. Records that are tamper-evident and verifiable carry more weight than screenshots. This is one of the practical reasons Kinduit was built the way it was: a permanent, court-admissible record of every communication is, on its own, a meaningful counterweight to gaslighting and to the "he said, she said" problem.
The third is that the institutional barriers are real and survivors are not imagining them. Meier's data, and decades of consistent advocacy literature, confirm that family courts often respond poorly to coercive control. That does not mean the legal system has nothing to offer, but it does mean that finding an attorney with specific training in coercive control and DV-informed practice is one of the highest-leverage decisions a survivor can make. National organizations like DV LEAP and state coalitions against domestic violence maintain referral networks.
Coercive control is, as Stark argued, a crime against liberty. The recognition of that fact is moving forward in statute and in the research literature. The recognition inside any given courtroom is more uneven. Survivors deserve to know the landscape they are operating in, and they deserve to have it described to them straight.
Read next
- What actually makes a co-parenting message court-admissible — Screenshots get challenged. Timestamps get questioned. Here is what family courts actually look for in a digital communication record, and how to make sure the one you bring holds up.
- Children's rights in co-parenting: what they are, and what they are not — Children have a legal and moral right to two functioning parents and a calm environment. Here is what that looks like in practice, and what it does not.
- Parents' rights in family court: what you keep, and what you trade — Filing in family court does not erase your rights as a parent. Here is a plain-language guide to what those rights are, where they bend, and how to protect them.