Statistics From the Ministry of Truth


With much fanfare, the Department of Justice recently announced that domestic violence has been reduced by half over ten years. This is a little like the old Soviet announcements of yet another increase in the wheat harvest.

The news carries several convenient implications for the federal government and its feminist clients: First, it appears to vindicate more than a decade of billion-dollar government programs. But second – in the spirit of having it both ways – it implies that "more needs to be done." As the report's author herself suggests with impeccable bureaucratic logic, figures indicating that domestic violence has decreased may indicate that it has not decreased. "The apparent decline could mean that women are choosing to suffer in silence rather than seek help."

Finally, the announcement preempts opposition. What are skeptics going to do, reply that "No it has not declined"? It is the bureaucratic equivalent of the question, "Have you stopped beating your wife?"

Men's advocates reply that the figures do not include violence against male victims. But larger issues are involved, like the manipulation of information (and us) by the government.

In fact, the figures mean almost nothing, because the figures on domestic violence generally mean almost nothing. And we might begin to wonder about government figures on everything else.

Critics of the domestic violence industry have already shown that any decline is not likely the result of federal programs. Overall crime appears to have been declining since the 1970s – years before domestic violence legislation. At best, family violence went along for the ride.

But the larger question is what precisely are we talking about? The very designation of a special category of "domestic" violence, separate from other forms of assault – a category defined by the relationship between the parties rather than the nature of the deed – raises questions as fundamental as what is the subject under discussion. Statistics purporting to quantify domestic violence are not based on convictions through jury trials or even formal charges; they are based on "reports" that are seldom substantiated. Strong incentives exist not only for women in divorce cases but also for federally funded feminist groups and government officials to manufacture false accusations and exaggerate incidents, and there are few penalties for doing so.

Possibly some justification exists for scholars to examine their subjects in this loose fashion (though scholars, especially when government-funded, have their own incentives to exaggerate the importance of their subject). Yet when this spills over into categories recognized by the penal system, it blurs the distinctions between crime and ordinary personal conflict.

Compounding this tendency is that virtually anything can be considered domestic "violence" if a self-proclaimed "victim" says it is. DOJ defines domestic "violence" to include such offenses as "undermining an individual's self-esteem." "Domestic violence can be physical, sexual, emotional, economic, or psychological actions or threats of actions that influence another person," says DOJ. What "action" does not "influence another person"? Breaking off a romance? Discussing politics? Making a bank deposit? Among the "crimes" that are now included in DOJ's definition of "violent" assault are "constant criticism, diminishing one's abilities, [and] name-calling." So DOJ's latest crime figures mean that, thanks to a $4 billion federal program, name-calling is at last under control?

"Approximately one-third of these offenses were serious violent crimes," DOJ earnestly insists. Wait a minute. If we are talking about "violence," I would have thought that all the offenses are serious violent crimes. But it turns out that two-thirds are not serious. So in what sense are they "crimes" at all? And the one-third that are "serious," are these judicially confirmed? No. So who determines how "serious" they are? DOJ bureaucrats of course.

Similarly nebulous is today's other great hysteria justifying plainclothes federal gendarmes: child abuse. (Child abuse, however violent, is for some reason not "domestic violence," unlike diminishing someone's self-esteem.) Like domestic violence, child abuse is seldom adjudicated as violent assault, so no objective measure of its incidence is available. So here too, what precisely are we talking about? None of the federal studies purporting to quantify child abuse bases its figures on convictions in jury trials or even court actions, even when the cases are classified as "substantiated." (The exceptions prove the rule: When the accused do receive jury trials, "A verdict of not guilty in a criminal court will not affect the [substantiated] finding in Juvenile Court because that finding is based on a different and lower evidentiary standard," according to a San Diego grand jury investigation.) At best, such studies are likewise based on "reports" by social workers and even include alleged incidents that are not reported at all. An official at the National Clearinghouse on Child Abuse and Neglect Information explained to me that the federal studies they conduct are not limited to reports from state agencies (themselves not necessarily judicially substantiated) but also include cases based on proactive interviews with non-specialist "sentinels" about alleged abuse they chose not to report but which officials nevertheless somehow know is happening.

So here too the absence of evidence is held to confirm the problem. In his book Father and Child Reunion, Warren Farrell notes that "Therapists have concluded that a child was abused because the child adamantly denied anything happened." This reasoning is reminiscent of the Korematsu case, justifying internment of Japanese-Americans during the Second World War for crimes not that they had committed but that they were expected to commit: "The very fact that no sabotage has taken place to date is a disturbing and confirming indication that such action will be taken" (emphasis added).

The fact is that it is virtually impossible to say that domestic violence or child abuse have increased, decreased, remained constant, or even taken place at all. These terms are matters of definition, and no definition exists. To "define" means to limit, and there is no limit on what constitutes these subjective infractions. Like "hate crimes," they are creations of ideology, not science.

The effects are not harmless: loose definitions ostensibly limited to purposes of data collection become loose definitions used to justify federal funding, which in turn become loose definitions used to secure convictions or incarcerations without any conviction or even trial. This is precisely what is happening in domestic violence and child abuse cases. What is particularly insidious is that very little opposition exists to question the government line. Potential critics of the left and right become mute whenever phrases like "violence against women" or "child abuse" are invoked.

The effect is spreading throughout the criminal justice system. Dorothy Rabinowitz and William Anderson both argue the similarity between the child abuse hysteria of the 1980s and 1990s and the Duke "rape" case. As both point out, the Duke defendants are lucky compared to victims of child abuse injustice, some of whom have remained imprisoned for years. And these at least had a trial. Unsubstantiated domestic violence accusations are sending citizens in and out of jail on a daily basis and on a much larger scale, and most never see a jury or an attorney, let alone an exposé in the media.

February 12, 2007