Johnson's Differentiation Theory: Is It Really Empirically Supported?

Abstract

Michael Johnson's differentiation of “types” of domestic violence has had a significant impact on courts and providers, in part because of its claim to an empirical basis. Unfortunately, the label “situational couple violence” has already been used by judges and evaluators to minimize abuse claims in custody cases. Johnson's repeated assertion that SCV is the most common form of domestic violence reinforces the marginalizing of domestic violence. But what do his data actually show? Here Professor Meier takes a close look at the research Johnson relies on and finds that it fails to prove his thesis. Rather, the data suggest that Johnson has it exactly backward: Control-based abuse is probably far more common than “situational” domestic violence.

KEYWORDS:

• children
• abuse
• custody
• family court
• domestic violence
• quantitative research

ACKNOWLEDGMENTS

The author is grateful for the in-depth work of research assistants Sonia Shaikh, Christy Zlatkus, and numerous other GW law students, as well as helpful feedback from participants at the Feminist Legal Theory conference at University of Baltimore Law School in October 2012, and attendees at a George Washington University Law School work-in-progress presentation in 2011.

Notes

See Meier (2015); Meier Presentation, American Association of Law Schools Workshop on Family Law, Vancouver, Canada (June 2007); Meier Presentation, Jewish Women International's 4th International Conference on Domestic Abuse, Crystal City, VA (April 2009); Meier Keynote, Seminar, University of Akron, Akron Bar Association and Cleveland Bar Association (September 2011), http://www.dvleap.org/Resources/TrainingMaterials.aspx (and on file with author).

Johnson's last two types are not discussed further herein. “Violent Resistance” describes women's violence in reaction to control-based (i.e., IT) violence: It is sometimes, but not always, self-defense, but may also be a violent reaction to a man's dominating violence. “Mutual Violent Control” describes couples in which both parties vie for control with violence. Johnson is, however, uncertain whether this category exists in sufficient numbers to really examine (Johnson, 2008, 23–24; Johnson & Ferraro, 2000).

The court and evaluators ignored the child's own report that she no longer wanted to see him after he violently shook her in an altercation over a toothbrush, in her first visit to his new home.

Leone found 51% SCV and 17% IT, along with 32% “control, no threat.” If the last category is combined with IT, the rates of SCV and IT become virtually identical, and given that control is the defining characteristic of IT, it is not clear why the authors chose not to characterize the last 32% as IT. The label “no threat” is difficult to square with the fact that all of the couples studied had experienced violence. Id. at 486.

Johnson also states that the “IT” and “SCV” populations he identified in the Pittsburgh data can be understood as those who rate high on all seven control behaviors versus those who rate low on all seven. (Johnson, 2008, 97). However, it seems illogical to call couples with “low” amounts of control on all seven control behaviors, “SCV.” It seems at least as appropriate, if not more so, to consider them “low IT.”.

Johnson identifies the cut-point for the Pittsburgh data as 2.74, the cut-point for NVAWS as 3 or more control behaviors, and the cut-point in the Chicago Help-Seeking Study as 5. “Less controlling” couples are called SCV. In his study with Leone, they used a cut point of two or less (Johnson & Leone, 2005).

In his last publication on the subject (his book), Johnson makes a point of noting that “there is considerable variability in the violence involved in intimate terrorism” (Johnson, 2008, 30).

Stark writes “[i]t is unclear whether the violence in coercive control is typically severe as Johnson contends or consists of routine, low-level threats and assaults, as I believe.” (Stark, 2006).

The brief describes Ms. L having had to get permission and to justify to her husband the spending of money, grocery shopping, children's clothes shopping, and even taking a bath.

In the JML case the history of ongoing minor violence and extreme control was ignored. The court then chose not to believe the one very serious incident of choking and sexual assault (JML v CAL, 2012 Opinion,).