# Minneapolis Domestic Violence Experiment *A 1981–82 randomized police study of responses to misdemeanor domestic assault calls in Minneapolis, Minnesota.* The **Minneapolis Domestic Violence Experiment** (MDVE) tested police responses to simple (misdemeanor) domestic assaults in [[Minneapolis]] from early 1981 to mid-1982. It covered only calls where both the suspect and the victim were present when officers arrived.[^1] The [[Minneapolis Police Department]] and researcher Lawrence Sherman carried it out.[^2] The [[National Institute of Justice]] (NIJ) sponsored the work under grant 80-IJ-CX-0042. The Police Foundation issued the 1984 report, by L. W. Sherman and R. A. Berk, as number 1 of its *Police Foundation Reports*.[^3] The report found that arrest reduced recidivism more than police counseling of the parties or barring the assailant from the home for eight hours. It ended with a discussion of factors that "suggest a cautious interpretation of the findings."[^4] The Minnesota Legislative Reference Library's account states that the study "concluded that automatic arrest was the primary deterrent to repeat offenses of domestic violence" and that it drew front-page headlines across the nation.[^5] According to the Wikipedia summary, the share of police departments using arrest in domestic violence cases went from 10 percent to 31 percent within a year, and to 46 percent by 1986.[^6] NIJ sponsored six replications beginning in 1986. Five were completed, and none produced an arrest effect as large as the Minneapolis result.[^7] In [[Minnesota]], the experiment followed Laws 1977, chapter 425, under which Minnesota became the first state to fund battered women's shelters.[^8] It began in the same year that Duluth police issued an experimental mandatory arrest policy.[^9] ## Who The Legislative Reference Library names the Minneapolis Police Department and Lawrence Sherman as the study's conductors.[^2] The NCJRS catalogue lists Sherman and R. A. Berk as authors, the Police Foundation as corporate author and NIJ as sponsoring agency.[^3] Wikipedia describes Sherman as the Police Foundation's Director of Research at the time. It reports that 51 Minneapolis patrol officers took part.[^10] ## What Where officers had probable cause for an assault, each qualifying call was assigned one of three responses: arrest, advice and mediation, or ordering the suspect out for eight hours.[^11] The 1984 report found arrest the most effective of the three in reducing repeat violence.[^4] The finding was turned into police directives. Wikipedia reports that New York City Police Commissioner Benjamin Ward ordered officers to make arrests after reading the results in a Police Foundation report. The Houston and Dallas departments also shifted toward arrest.[^12] The same summary reports that the U.S. Attorney General's Task Force on Family Violence leaned on the study in its 1984 report urging criminal-justice handling of domestic violence. It also reports that 15 states and the District of Columbia enacted arrest-requiring domestic violence laws within eight years.[^13] The five completed replications first reported their findings independently beginning in 1990. Their results varied by site and measure: some favored arrest, some showed no difference, and some associated arrest with more repeat offending.[^7] The record thus shows national adoption of arrest policies preceding the replication results.[^6] It does not establish that the Minneapolis effect held elsewhere.[^7] ## When The fieldwork ran from early 1981 to mid-1982.[^1] The Police Foundation report appeared in 1984,[^3] the same year as the Attorney General's task-force report.[^13] NIJ's replication program began in 1986, and initial findings were reported from 1990.[^7] Wikipedia reports that by 2005, 23 states and the District of Columbia had mandatory-arrest laws for domestic assault.[^13] Congress passed the Violence Against Women Act in 1994. The Legislative Reference Library records that the federal response to domestic violence has focused primarily on criminal justice.[^14] ## Where The experiment was confined to Minneapolis.[^1] The replications were run by police departments in Omaha, Nebraska; Charlotte, North Carolina; Milwaukee, Wisconsin; Miami-Dade County, Florida; and Colorado Springs, Colorado. A study begun in Atlanta was not completed.[^7] Critics noted that Minneapolis held arrestees in jail overnight, while other jurisdictions might release them sooner.[^15] ## Why Wikipedia records that before the early 1970s, U.S. police generally avoided intervening in domestic calls and treated arrest as a last resort. During the 1970s, many jurisdictions did not authorize arrest for a misdemeanor assault an officer had not witnessed. A 1976 Police Foundation study found that police had intervened at least once in the prior two years in 85 percent of spouse homicides. Battered women's advocates took legal action against police departments in Oakland and New York City to compel arrests.[^16] The same summary lists the 1978 National Academy of Sciences report *Deterrence and Incapacitation* as part of the research background.[^16] The study rested on deterrence theory, which assumes offenders make rational choices. Critics noted that it did not measure whether arrest raised offenders' fear of future sanctions.[^17] ## How Each officer carried a pad of report forms color-coded for the three responses. For every call meeting the criteria, the officer was to take the action on the top form, and the reports went to the research staff for follow-up.[^18] Repeat violence was measured two ways over the six months after the first police visit: through police records and through victim interviews.[^18] Wikipedia adds that offenders were also interviewed and that the study lasted about 17 months.[^11] Critics questioned whether six months was long enough to capture the cyclical course of domestic violence.[^15] A randomized design estimates an average effect for the group. Some individuals may therefore fare worse after arrest, including through retaliatory violence.[^17] ## How much NCJRS records the NIJ grant number, 80-IJ-CX-0042, but the record does not state the grant's dollar amount.[^3] The Violence Against Women Act of 1994 appropriated $1.6 billion to enhance criminal prosecution and protection efforts.[^14] Minnesota's own appropriations of the period are listed below.[^8] ## How many - Cases: NCJRS reports 330 victims;[^18] Wikipedia reports 330 cases.[^11] - Officers: 51 patrol officers.[^10] - Re-offending: 19 percent after arrest, 37 percent after advice, and 34 percent after being sent away, per Wikipedia.[^19] - Departments using arrest: 10 percent, then 31 percent within a year, then 46 percent by 1986.[^6] - States with mandatory arrest: 15 plus D.C. within eight years; 23 plus D.C. by 2005.[^13] - Replication samples: 907 cases in Metro-Dade, 1,200 in Milwaukee, and more than 1,600 in Colorado Springs.[^7] ## In the Minnesota record ### What Laws 1977, chapter 425, was authored by Representative [[Phyllis Kahn]] and Senator B. Robert Lewis. It provided $500,000 for four pilot shelter and support programs and required a study to guide later sessions. The Library's entry does not record the authors' party affiliations.[^8] A 1979 public safety omnibus act, Laws 1979, chapter 241, expanded the programs with $2.9 million. Domestic violence programs are now housed in the Department of Public Safety under the [[Domestic Abuse Act]], Minn. Stat. § 518B.01.[^20] Arrest policy fed a court-referral system in which the Minnesota-originated [[Duluth model]] became the most common program. NIJ's 2003 special report states: "With the establishment of proarrest policies in the 1980s, increasing numbers of batterers were seen in criminal courts across the country." It adds that batterer intervention programs "were introduced as a way to hold batterers accountable without incarcerating them."[^21] The report calls the Duluth model the most common form of program in the nation and notes that many states mandate conformity to it. It reports the results of two Duluth-model experiments. In Broward County, Florida, there was no change in behavior or attitudes. In Brooklyn, New York, only men who completed a 26-week program had significantly fewer official complaints than the control group.[^22] The report concludes that the methodological limitations of virtually all such evaluations "make it impossible to say how effective BIPs are."[^23] NIJ's [[CrimeSolutions]] profile, posted September 13, 2013, rates the Duluth model Effective for reducing violent-offense recidivism and Promising for reducing victimization. The rating rests on a 2004 meta-analysis of 22 studies with 5,536 participants.[^24] The page now carries the notice "This page is no longer being updated."[^25] Edward W. Gondolf's 2002 book *Batterer Intervention Systems*, sponsored in part by NIJ, reports a multi-site evaluation of batterer programs.[^26] ### Who The Library's entry records three developments. [[Women's Advocates]], a coalition affiliated with Legal Aid, opened the first Twin Cities shelter in St. Paul in the early 1970s and bought a house in 1974. The Women's Consortium helped draft the 1977 law. In 1981, the Duluth project developed a model for a coordinated community response.[^27] Domestic Abuse Intervention Programs states that it was incorporated in 1980 as Minnesota Program Development, Inc. Through its [[Domestic Abuse Intervention Project]], nine city, county and private agencies adopted coordinated policies. The agency describes them this way: "These policies focus on protecting victims from continued acts of violence by combining support services for victims and legal sanctions and rehabilitation programs for offenders."[^28] CrimeSolutions describes the Duluth model as originating in 1981 from the Duluth project. It states that the model's premise is that the principal cause of domestic violence is a patriarchal ideology that has allowed men to control women.[^29] ### When The agency's own chronology records the following outputs. In 1981, Duluth police issued an experimental mandatory arrest policy.[^9] In 1982, nine agencies made the experimental policies permanent after an initial evaluation, and the court raised mandated abuser sessions from eight to twelve weeks. In 1983, neighborhood women's groups developed the [[Power and Control Wheel]].[^30] In 1985, the court raised mandated sessions to twenty-six weeks. In the same year the prosecutor's "no drop" policy was eliminated, and the agency records that "conviction rates gradually fall from 80% to below 47%." In 1987, the prosecutor's office returned to subpoenaing victims. In 1989, police adopted a policy eliminating arrests in self-defense cases.[^31] Separately from the domestic-violence record, the State is the defendant in a pending federal action. On January 14, 2026, the [[United States Department of Justice]] announced the suit in release No. 26-21, titled *United States Department of Justice Files Lawsuit Against Minnesota's 'Affirmative Action' Regime*. According to the release, Attorney General Pamela Bondi certified the case as a matter of general public importance.[^32] The complaint in [[United States v. State of Minnesota (0:26-cv-00273)]] (D. Minn.) is brought under [[Title VII]] of the [[Civil Rights Act of 1964]]. It cites [[Minn. Stat. § 43A.191]] eight times, including subdivision 3(c)'s requirement that an agency explain the reason for hiring a non-affirmative candidate.[^33] The U.S. government uses "regime" for governments it opposes. The [[Iraq Liberation Act]] of 1998 (Pub. L. 105-338) states a policy of supporting efforts to remove "the regime headed by Saddam Hussein." The project's catalogue of U.S.-government "regime" usages ([[How Minnesota Got Here]], note 40) finds Minnesota the only U.S. state so described.[^34] In [[Ames v. Ohio Department of Youth Services]], 605 U.S. 303 (2025), the Supreme Court held that Title VII's disparate-treatment provision "draws no distinctions between majority-group plaintiffs and minority-group plaintiffs."[^35] ## Notes [^1]: Office of Justice Programs, NCJRS Virtual Library, *Minneapolis Domestic Violence Experiment* (NCJ 98905, 1984) — study conducted from early 1981 to mid-1982; limited to misdemeanor domestic assaults with suspect and victim both present. https://www.ojp.gov/library/abstracts/minneapolis-domestic-violence-experiment [^2]: Minnesota Legislative Reference Library, *Minnesota Women's Legislative Timeline — Battered Women/Domestic Violence* (1977 entry) — the Minneapolis Police Department and researcher Lawrence Sherman conducted the 1981–82 study. https://www.lrl.mn.gov/womenstimeline/details?recid=26 [^3]: Office of Justice Programs, NCJRS Virtual Library, *Minneapolis Domestic Violence Experiment* (NCJ 98905, 1984) — authors L. W. Sherman and R. A. Berk; Police Foundation as corporate author; NIJ as sponsoring agency; grant 80-IJ-CX-0042; *Police Foundation Reports*, number 1. https://www.ojp.gov/library/abstracts/minneapolis-domestic-violence-experiment [^4]: Office of Justice Programs, NCJRS Virtual Library, *Minneapolis Domestic Violence Experiment* (NCJ 98905, 1984) — arrest was more successful in reducing recidivism than counseling or an eight-hour bar from the home; the report discusses factors that suggest a cautious interpretation. https://www.ojp.gov/library/abstracts/minneapolis-domestic-violence-experiment [^5]: Minnesota Legislative Reference Library, *Minnesota Women's Legislative Timeline — Battered Women/Domestic Violence* (1977 entry) — the Library's statement of the study's conclusion and its national front-page coverage. https://www.lrl.mn.gov/womenstimeline/details?recid=26 [^6]: Wikipedia contributors, *Minneapolis Domestic Violence Experiment* (revision 1292250260, retrieved 2026-10-28), "Policy response" — secondary summary: departments using arrest rose from 10 to 31 percent within a year and to 46 percent by 1986. https://en.wikipedia.org/wiki/Minneapolis_Domestic_Violence_Experiment [^7]: Wikipedia contributors, *Minneapolis Domestic Violence Experiment* (revision 1292250260, retrieved 2026-10-28), "Replication" — secondary summary: six NIJ replications from 1986; the sites; Atlanta not completed; case counts; findings reported from 1990, mixed, none as strong as Minneapolis. https://en.wikipedia.org/wiki/Minneapolis_Domestic_Violence_Experiment [^8]: Minnesota Legislative Reference Library, *Minnesota Women's Legislative Timeline — Battered Women/Domestic Violence* (1977 entry) — Laws 1977, ch. 425: first state funding of battered women's shelters; $500,000 for four pilot programs; authored by Rep. Phyllis Kahn and Sen. B. Robert Lewis; study requirement. https://www.lrl.mn.gov/womenstimeline/details?recid=26 [^9]: Domestic Abuse Intervention Programs, *History and Recognition* (2017) — 1981: Duluth police issue the experimental "mandatory arrest policy." https://www.theduluthmodel.org/wp-content/uploads/2017/11/DAIP-History-and-Recognition-2017.pdf [^10]: Wikipedia contributors, *Minneapolis Domestic Violence Experiment* (revision 1292250260, retrieved 2026-10-28), lead and "Methodology" — secondary summary: Sherman as Police Foundation Director of Research; 51 participating patrol officers. https://en.wikipedia.org/wiki/Minneapolis_Domestic_Violence_Experiment [^11]: Wikipedia contributors, *Minneapolis Domestic Violence Experiment* (revision 1292250260, retrieved 2026-10-28), "Methodology" — secondary summary: three responses where probable cause existed; victim and offender interviews; about 17 months and 330 cases. https://en.wikipedia.org/wiki/Minneapolis_Domestic_Violence_Experiment [^12]: Wikipedia contributors, *Minneapolis Domestic Violence Experiment* (revision 1292250260, retrieved 2026-10-28), "Policy response" — secondary summary: Commissioner Benjamin Ward's arrest mandate; Houston and Dallas departments' changes. https://en.wikipedia.org/wiki/Minneapolis_Domestic_Violence_Experiment [^13]: Wikipedia contributors, *Minneapolis Domestic Violence Experiment* (revision 1292250260, retrieved 2026-10-28), "Policy response" — secondary summary: 1984 Attorney General's Task Force on Family Violence; 15 states and D.C. within eight years; 23 states and D.C. by 2005. https://en.wikipedia.org/wiki/Minneapolis_Domestic_Violence_Experiment [^14]: Minnesota Legislative Reference Library, *Minnesota Women's Legislative Timeline — Battered Women/Domestic Violence* (1977 entry) — federal response focused primarily on criminal justice; VAWA (1994) appropriated $1.6 billion for prosecution and protection efforts. https://www.lrl.mn.gov/womenstimeline/details?recid=26 [^15]: Wikipedia contributors, *Minneapolis Domestic Violence Experiment* (revision 1292250260, retrieved 2026-10-28), "Criticism" — secondary summary: six-month follow-up possibly too short; Minneapolis held arrestees overnight. https://en.wikipedia.org/wiki/Minneapolis_Domestic_Violence_Experiment [^16]: Wikipedia contributors, *Minneapolis Domestic Violence Experiment* (revision 1292250260, retrieved 2026-10-28), "Background" — secondary summary: pre-1970s police practice; in-presence arrest rule; 1976 Police Foundation 85 percent finding; Oakland and New York City lawsuits; 1978 National Academy of Sciences report. https://en.wikipedia.org/wiki/Minneapolis_Domestic_Violence_Experiment [^17]: Wikipedia contributors, *Minneapolis Domestic Violence Experiment* (revision 1292250260, retrieved 2026-10-28), "Criticism" — secondary summary: average group effects; possible retaliation; deterrence-theory premise; fear of sanctions not measured. https://en.wikipedia.org/wiki/Minneapolis_Domestic_Violence_Experiment [^18]: Office of Justice Programs, NCJRS Virtual Library, *Minneapolis Domestic Violence Experiment* (NCJ 98905, 1984) — color-coded report pads; reports to research staff; 330 victims; police-record and victim-interview measures over six months. https://www.ojp.gov/library/abstracts/minneapolis-domestic-violence-experiment [^19]: Wikipedia contributors, *Minneapolis Domestic Violence Experiment* (revision 1292250260, retrieved 2026-10-28), "Findings" — secondary summary: re-offending of 19 percent (arrest), 37 percent (advice) and 34 percent (sent away). https://en.wikipedia.org/wiki/Minneapolis_Domestic_Violence_Experiment [^20]: Minnesota Legislative Reference Library, *Minnesota Women's Legislative Timeline — Battered Women/Domestic Violence* (1977 entry) — Laws 1979, ch. 241: $2.9 million expansion; programs housed in the Department of Public Safety under Minn. Stat. § 518B.01. https://www.lrl.mn.gov/womenstimeline/details?recid=26 [^21]: National Institute of Justice, *Batterer Intervention Programs: Where Do We Go From Here?* (NCJ 195079, June 2003) — proarrest policies of the 1980s brought more batterers into criminal courts; programs introduced to hold batterers accountable without incarceration. https://www.ojp.gov/txtfiles1/nij/195079.txt [^22]: National Institute of Justice, *Batterer Intervention Programs: Where Do We Go From Here?* (NCJ 195079, June 2003) — Duluth model as the most common program and state mandates; Broward County and Brooklyn findings. https://www.ojp.gov/txtfiles1/nij/195079.txt [^23]: National Institute of Justice, *Batterer Intervention Programs: Where Do We Go From Here?* (NCJ 195079, June 2003) — methodological limitations of virtually all evaluations; low response rates and high attrition. https://www.ojp.gov/txtfiles1/nij/195079.txt [^24]: National Institute of Justice, CrimeSolutions, *Practice Profile: Interventions for Persons Who Committed Intimate-Partner Violence: Duluth Model* (posted September 13, 2013) — rated Effective (violent-offense recidivism) and Promising (victimization); Babcock et al. (2004) meta-analysis of 22 studies and 5,536 participants. https://crimesolutions.ojp.gov/ratedpractices/17 [^25]: National Institute of Justice, CrimeSolutions, *Practice Profile: Interventions for Persons Who Committed Intimate-Partner Violence: Duluth Model* (posted September 13, 2013) — the page states that it is no longer being updated. https://crimesolutions.ojp.gov/ratedpractices/17 [^26]: Office of Justice Programs, NCJRS Virtual Library, *Batterer Intervention Systems: Issues, Outcomes, and Recommendations* (NCJ 247403, 2002) — Edward W. Gondolf's multi-site evaluation of batterer programs; NIJ among sponsoring agencies. https://www.ojp.gov/ncjrs/virtual-library/abstracts/batterer-intervention-systems-issues-outcomes-and-recommendations [^27]: Minnesota Legislative Reference Library, *Minnesota Women's Legislative Timeline — Battered Women/Domestic Violence* (1977 entry) — Women's Advocates shelter (early 1970s; house purchased 1974); Women's Consortium role in the 1977 law; 1981 Duluth coordinated-response model. https://www.lrl.mn.gov/womenstimeline/details?recid=26 [^28]: Domestic Abuse Intervention Programs, *History and Recognition* (2017) — incorporation in 1980 as Minnesota Program Development, Inc.; nine agencies' coordinated policies; the agency's statement of the policies' focus. https://www.theduluthmodel.org/wp-content/uploads/2017/11/DAIP-History-and-Recognition-2017.pdf [^29]: National Institute of Justice, CrimeSolutions, *Practice Profile: Interventions for Persons Who Committed Intimate-Partner Violence: Duluth Model* (posted September 13, 2013) — model originated in 1981 from the Duluth Domestic Abuse Intervention Project; its stated theory of cause. https://crimesolutions.ojp.gov/ratedpractices/17 [^30]: Domestic Abuse Intervention Programs, *History and Recognition* (2017) — 1982: policies made permanent and sessions raised from eight to twelve weeks; 1983: Power and Control Wheel developed. https://www.theduluthmodel.org/wp-content/uploads/2017/11/DAIP-History-and-Recognition-2017.pdf [^31]: Domestic Abuse Intervention Programs, *History and Recognition* (2017) — 1985: twenty-six-week sessions, "no drop" policy eliminated, conviction rates from 80% to below 47%; 1987: subpoenaing resumed; 1989: self-defense arrest policy. https://www.theduluthmodel.org/wp-content/uploads/2017/11/DAIP-History-and-Recognition-2017.pdf [^32]: U.S. Department of Justice, Office of Public Affairs, *United States Department of Justice Files Lawsuit Against Minnesota's 'Affirmative Action' Regime* (Release No. 26-21, January 14, 2026) — release title; Attorney General's certification of general public importance. https://www.justice.gov/opa/pr/united-states-department-justice-files-lawsuit-against-minnesotas-affirmative-action-regime [^33]: U.S. Department of Justice, Civil Rights Division, *Complaint, United States v. State of Minnesota*, No. 0:26-cv-00273 (D. Minn. filed January 14, 2026) — Title VII action; cites Minn. Stat. § 43A.191 eight times, including subd. 3(c). https://www.justice.gov/crt/media/1423361/dl [^34]: U.S. Congress, *Iraq Liberation Act of 1998*, Pub. L. 105-338 (1998) — policy of supporting efforts to remove "the regime headed by Saddam Hussein"; the Minnesota-only finding is the project's catalogue finding (How Minnesota Got Here, note 40). https://www.govinfo.gov/content/pkg/PLAW-105publ338/pdf/PLAW-105publ338.pdf [^35]: Supreme Court of the United States, *Ames v. Ohio Department of Youth Services*, 605 U.S. 303 (2025) — Title VII's disparate-treatment provision draws no distinction between majority-group and minority-group plaintiffs. https://www.supremecourt.gov/opinions/24pdf/23-1039_c0n2.pdf