Posted by Jay Livingston
Something I read in another blog sent me digging into the statistics on homicide between husbands and wives or other “intimates.” I remembered from my days in the crim biz that the US was unique in that wives here killed their husbands almost as frequently as husbands killed wives. This statistic, the “spousal rate of killing” (SROK), was introduced in a now-classic1992 article by Margo Wilson and Martin Daly. In most countries, that rate is 25-30%. In the US, Wilson and Daly pointed out, it was about 75%.
But something has happened, over the last thirty years or so (data here). And as far as I can tell from a quick search on the Internet, nobody seems to have noticed.
(Click on the graph for a larger view.)
Between 1976 and 2005, the number of women killed by their male partners decreased by about 25%, less than the decrease in all homicides nationwide. But the number of men killed by women dropped dramatically, from 1300 to 330, a 75% decrease (since the population increased in those three decades, the change in rates is probably even greater. The SROK fell from 82% to 28%.
My Internet search for explanations was cursory at best, but it turned up nothing. I have only two ideas:
1. Men Behaving Better. Men have stopped doing those things that made women want to kill them.
I offered this explanation to two women in the Justice Studies department here. They rejected it out of hand and without comment. (Maybe they didn’t like the blaming-the-victim assumption: if women kill men, it’s because of what men do. Or maybe they were using a convenience sample of anecdotal data on men’s behavior.) One of these women, Lisa Anne Zilney, offered a counter-explanation . . .
2. Women Having Options. Women’s shelters and other facilities have given women an alternative. Without these, the only way to escape an intolerable situation at home was to get rid of the cause. Providing abused and desperate women a safe place to go saves lives – and apparently not just the lives of women.
I’m not wild about either of these explanations for the steep decline in the SROK (and as I recall, Wilson and Daly weren’t wild about any of their explanations of why it was so high).
Any ideas?
I reject the first possibility out of hand and without comment (but could provide ample evidence against it if needed)...
ReplyDeleteThis is interesting and I'm amazed no one has noticed this. It would be good to know class/ses info on the intimate partners as well. I would suspect it isn't just greater availability of shelters but also more middle-class women working and declining stigma associated with divorce.
I have a relative who had an abusive husband -- they (actually, she) had a lot of money but her leaving him was not an option in small-town iowa in the 1950s. Luckily, he died young of natural causes so she didn't kill him but she never would have left him either.
I second what newsocprof said. Answer thief!
ReplyDeleteDivorce! The drastic drop in SROK coincides with increasing divorce rates in the 70's and 80's. Notice the pattern levels of as divorce rates level in the 1990's. So I guess I also agree with newsocprof.
ReplyDeleteEmily. There are different ways to measure divorce, but the simple figure often used -- divorces per 1,000 population -- shows divorce increasing till about 1981, then delining steadily over the next 20 years. During that 20 year period, fewer women were killing their men, but divorce was also decreasing. Or to put it another way, more women were staying married, but fewer of them were klling their spouses.
ReplyDeleteI am a family law attorney in California representing women in divorce and other family law matters. Two comments:
ReplyDelete1. people in the domestic violence movement have been aware of the increased safety of men since the movement achievements in shelters and protection for women and children.
2. why do you assume that the statistics you rely on for past numbers of violent deaths of spouses is accurate? Research data from before 1976 is apt to be gender biased and unrealiable.
Anon, I didn't present data from before 1976. As for the accuracy of the data I used, no data set is going to be perfect. But if I had to choose a crime where the stats are the most accurate, I'd choose murder any day. So would every criminologist I know of. And if I could narrow it, I'd specify murders between intimates.
ReplyDelete