Restraining orders and what they do and don't do

The civil protection order in its modern form is a creation of the battered women's movement of the 1970s and 1980s. Before then, the legal tools available to survivors were criminal complaint (slow, dependent on prosecutorial discretion, often dismissed) or divorce proceedings (which took months or years and weren't available to unmarried partners). Advocates pushed for a faster, lower-burden, survivor-controlled tool — something that could be obtained in a day or two on a civil standard of proof, that the survivor could choose to invoke without depending on a prosecutor's willingness, and that would have real legal teeth. Pennsylvania's Protection from Abuse Act of 1976 was one of the first; by the mid-1990s every U.S. state had analogous legislation, and similar frameworks existed across the U.K., Canada, Australia, and elsewhere.

Most jurisdictions allow an emergency or ex parte order, granted on the survivor's testimony alone, without notice to the respondent, lasting a short period (typically two to four weeks) until a full hearing can be held. This is the most controversial feature of the system: it removes one party's procedural rights (notice and opportunity to be heard) in order to protect the other party's safety. Courts have repeatedly held this is constitutional given the short duration, the safety stakes, and the prompt opportunity for a contested hearing. In practice, the ex parte order is what gets the survivor through the most dangerous window — the immediate aftermath of leaving or of an attack — until the longer-term order can be sought.

At the contested hearing, both parties present evidence. Survivors often appear without counsel; respondents sometimes appear with counsel. The survivor must establish, on the civil standard, that the statutory predicate for an order is met — typically some combination of abuse, threat, fear, or harassment as defined in the statute. Definitions vary: some statutes require physical violence, some include threats, some include coercive control, some include stalking and harassment. The breadth of the definition matters enormously: a narrow definition excludes many real abuse patterns; a broad one risks being used in disputes that aren't actually about safety. Most jurisdictions have settled on a middle ground that includes physical violence, threats of physical violence, sexual abuse, and stalking; the inclusion of coercive control as such is the current frontier.

Standard provisions: no contact direct or indirect, no approach within a specified distance of the survivor's home, work, school, children's school; no third-party communication; no possession of firearms; exclusion from a shared residence (sometimes called a "kick-out" order); temporary custody of children and a temporary parenting schedule; possession of pets and shared property; financial support pending divorce. The scope is wide. The order is essentially a temporary rewrite of the practical terms of the parties' lives, pending longer-term resolution through divorce, custody proceedings, or criminal cases.

Under federal law (18 U.S.C. § 922(g)(8)) and many state statutes, a person subject to a qualifying domestic violence restraining order is prohibited from possessing firearms. This is the single most important collateral consequence: empirical research consistently shows that the presence of a firearm in an abusive household dramatically elevates lethality risk (Campbell et al., five-fold or more in some studies). The legal provision is strong; the enforcement is uneven. Some jurisdictions actively verify surrender; many do not. Implementation gaps in firearms surrender are one of the most consequential weaknesses of the protection order system, and one of the most addressable through political will.

A specific category that protection orders address poorly is stalking by an ex-partner. The order says don't contact; the stalker contacts anyway, often through means that are hard to document or that look ambiguous (showing up at a coffee shop, posting on social media, sending mutual friends). Each individual contact may seem trivial; the pattern is terrifying. Courts and police are getting better at recognising the pattern, but evidentiary requirements for proving violations are still calibrated to discrete incidents. Some jurisdictions are experimenting with stalking-specific orders with broader prohibitions and lower thresholds for violation.

A persistent problem is the mutual or cross-order, where each party files against the other and the court issues orders against both. Advocates argue, with strong evidence, that mutual orders typically reflect the abuser's strategic counter-filing rather than mutual mistreatment, and that they confuse police response and weaken the survivor's protection. Most jurisdictions have moved away from automatic mutual orders, requiring separate findings against each party, but the problem persists. The reform direction is rigorous primary-aggressor analysis at the order-issuance stage.

The research literature on protection order effectiveness is mixed but instructive. Holt et al. (2003) found a substantial reduction in physical violence among women who obtained permanent orders compared to those who did not, with the protection peaking in the first year. Logan and Walker's research has shown that orders are associated with reduced violence in many cases and continued violence in a minority — and that the predictors of continued violence include severity of prior abuse, history of stalking, prior order violations, and the respondent's substance use and criminal history. The honest reading: orders help on average and don't help everyone, and the cases where they don't help are predictable and identifiable.

Building on this evidence, several jurisdictions now use formal lethality assessment instruments (Campbell's Danger Assessment is the most validated) to identify high-risk cases at the time of order issuance, triggering additional services — high-priority police response, advocate contact, sometimes GPS monitoring of the respondent. The Maryland Lethality Assessment Program and similar initiatives have shown reductions in repeat severe violence. The principle: not every order needs the same supports, and the cases most likely to end in homicide are partly identifiable in advance.

For high-risk respondents, some jurisdictions now order GPS ankle monitoring as a condition of release, with automatic alerts to the survivor if the respondent approaches her location. This is among the most directly protective interventions available, and it has shown meaningful effects in pilot programs. The cost and the civil-liberties profile keep it from being universal; the strongest case for it is in cases identified through lethality assessment as highest-risk.

Orders do not stop bullets. They do not stop a determined attacker who has accepted the criminal consequences in advance. They do not make a survivor visible to police she has not called. They do not protect against attacks at unknown locations, attacks by third parties acting on the respondent's instructions, or attacks following the order's expiration. They do not, in many jurisdictions, attach effective firearms surrender. They do not work in places where police treat violations as paperwork. They do not work for survivors who cannot get to court, cannot get childcare for the hearing, cannot afford counsel, cannot speak the court's language, or cannot risk being seen filing. The list of failure modes is long.

Genuinely false restraining order claims exist and are damaging when they occur. The empirical incidence is contested but generally found by researchers to be low — under 10% in most studies, with the great majority of orders reflecting real concerns. The much larger problem in the other direction is unmet need: survivors who should seek orders and do not, for reasons of fear, lack of access, or distrust of the system. A reform debate that focuses only on false claims is engaging the smaller half of the problem.

The most important thing to know about restraining orders is that they are one element of safety planning, not a substitute for it. Effective safety planning includes: a safe place to go, a packed bag, copies of key documents, financial access, a code word with trusted people, awareness of high-risk moments (separation, court dates, holidays), and connection to a domestic violence advocate. The order works best when it is one piece of this larger plan. Survivors who treat the order as their plan are working with an incomplete safety system. Advocates' single most consistent message is the order helps; it is not enough on its own.

1. Holt, Victoria L., Mary A. Kernic, Marsha E. Wolf, and Frederick P. Rivara. "Civil Protection Orders and Risk of Subsequent Police-Reported Violence." JAMA 288, no. 5 (2002): 589–594. 2. Logan, TK, and Robert Walker. "Civil Protective Order Effectiveness: Justice or Just a Piece of Paper?" Violence and Victims 25, no. 3 (2010): 332–348. 3. Stark, Evan. Coercive Control: How Men Entrap Women in Personal Life. New York: Oxford University Press, 2007. 4. Bancroft, Lundy. Why Does He Do That? Inside the Minds of Angry and Controlling Men. New York: Berkley Books, 2002. 5. Schechter, Susan. Women and Male Violence: The Visions and Struggles of the Battered Women's Movement. Boston: South End Press, 1982. 6. Balos, Beverly. "Domestic Violence Matters: The Case for Appointed Counsel in Protective Order Proceedings." Temple Political and Civil Rights Law Review 15 (2006): 557–589. 7. Campbell, Jacquelyn C., et al. "Risk Factors for Femicide in Abusive Relationships: Results From a Multisite Case Control Study." American Journal of Public Health 93, no. 7 (2003): 1089–1097. 8. Russell, Diana E. H. Rape in Marriage. Rev. ed. Bloomington: Indiana University Press, 1990. 9. Hasday, Jill Elaine. Family Law Reimagined. Cambridge, MA: Harvard University Press, 2014. 10. Johnston, Janet R., and Linda E. G. Campbell. Impasses of Divorce: The Dynamics and Resolution of Family Conflict. New York: Free Press, 1988. 11. Bancroft, Lundy, Jay G. Silverman, and Daniel Ritchie. The Batterer as Parent: Addressing the Impact of Domestic Violence on Family Dynamics. 2nd ed. Thousand Oaks, CA: Sage, 2012. 12. Schepard, Andrew. Children, Courts, and Custody: Interdisciplinary Models for Divorcing Families. Cambridge: Cambridge University Press, 2004.