The Milwaukee Domestic Violence Experiment
In 1992, results from the Milwaukee Domestic Violence Experiment were published, revealing that the implementation of mandatory arrest laws had failed to decrease rates of re-assault in domestic violence (DV) cases. The experiment compared the impact of arrest and detention, with the offender eligible for release on $250 bail; arrest and detention, with the offender quickly released on personal recognizance; and finally, no arrest, with the suspect read a standardized warning by responding officers.
After comparing results from 1,200 DV cases, researchers found that the arrest of offenders had variable deterrence effects depending on specific offender characteristics. For example, when the offender was white, employed, and/or married to the victim, arrest was found to have a strong deterrence effect. While for offenders who were Black, unemployed, and/or cohabitating with an unmarried partner, arrest increased both the prevalence and the severity of future violent incidents. The study calculated, “….that 10,000 arrested whites produce 2,504 fewer acts of domestic violence a year than warned whites, while 10,000 arrested blacks produce 1,803 more acts of violence per year than warned blacks…” (Sherman et al., pp. 160, 1992). Looking at employment status as a variable, researchers found that “With 958 fewer acts of violence committed against victims of 10,000 employed suspects who had been arrested than those who had been warned, the price equals 2,274 more acts of violence per 10,000 unemployed suspects who had been arrested than if they had only been warned” (Sherman et al., pp. 160, 1992).
Follow-up Study: Increased Premature Death of Domestic Violence Victims from Arrest
Twenty-three years later researchers Lawrence W. Sherman and Heather M. Harris followed up on the landmark Milwaukee Domestic Violence Experiment and found shocking results. After analyzing state and national death data on 1,125 victims enrolled in the Milwaukee Experiment, researchers found that victims whose partners had been arrested and jailed, rather than warned, were 64% more likely to have died prematurely. Heart disease, cancer, and other “internal illnesses” caused the overwhelming majority of deaths. Such illnesses are associated with chronic stress, leading researchers to postulate that the stress of having their partner arrested contributed greatly to these detrimental health outcomes.
As in the original study, race was found to be a significant variable in predicting premature death among victims. White victims whose partners were arrested rather than warned had a 9% higher death rate. Black victims on the other hand, had a 98% higher death rate when their partners were arrested. Employment was, again, an important variable. For white and Black victims alike, victim employment at the time of their partner’s arrest was correlated with higher victim mortality (Sherman & Harris, 2014). However, victim mortality among employed Black victims was the greatest. Out of the 125 employed Black victims whose partners were arrested following a DV offense, 14 (11%) died prematurely while none (0%) of the 67 employed Black victims whose partners were warned died at the 23-year follow-up (Sherman & Harris, 2014).
The results of both studies call into question the efficacy of mandatory arrest policies for addressing the problem of partner violence. Although certain offenders in the Milwaukee Domestic Violence Experiment were deterred by arrest, others were not. This means that while some victims can in fact benefit from the involvement of law enforcement officials, other victims not only experience little reprieve from partner violence, the violence perpetrated against them increases in frequency and severity following their partners arrest.
Sherman and Harris’s follow up study reveals even more troubling findings. Namely, that in addition to more frequent experiences of partner violence over their lifetime, premature death due to stress-related illnesses increased significantly among Black victims whose partners were arrested.
Findings such as these might reveal a reality that is hard to accept, Sherman concedes. However, Sherman continues, “the moral burden of proof now lies with those who wish to continue this mass arrest policy”.