« Hold 'Em Up | Main | Beyond Carter »

Intimate Partner Murder

Alina Stefanescu ponders some evidence that being pregnant is more likely to get you killed. I don't know about that (as in literally don't know), but the specific case she cites strikes me as not necessarily having much of anything to do with the pregnancy as such. What this does bring to mind are some facts I recall from doing some general crime research a while back. Namely, that while people in general -- and women especially -- tend to have a great deal of anxiety about street crime (especially in a high-crime city such as DC), women really have much, much, much more to fear from husbands/boyfriends/exes than from random violence.

I wouldn't want to say that this sort of thing -- "intimate partner murder," as they call it -- is common. Rather, murder in general is rare, and the murder of women is especially rare. Within that context of the murder of women being rare, intimate partner murder is an extremely high proportion of the aggregate. "40% to 70% of female murder victims were killed by their husbands or boyfriends," according to the CDC. The Bureau of Justice Statistics pegs it at a lower, but still quite high, 30 percent. Interestingly, as of the mid-1970s, intimate partner murders had a close to 50-50 gender split. Since that time, women killing their husbands (or...) has gotten much rarer, but men killing their wives (or...) has only dropped a little, transforming a once general phenomenon into something of a ghettoized "women's issue."

The data seems to suggest that there is, in fact, a silver lining to the "decline of marriage" and a dark side to its promotion. The overall decline in intimate partner murder is composed entirely of a steep decline in spousal murder that has been only partially offset by an increase in boyfriend/girlfriend murder while ex murder has stayed pretty flat. At least one seemingly serious researcher I've read hypothesized that marriages are particularly likely to end in murder, because the high level of commitment tends to dissuade people from dissolving the relationship before problems get really out of hand. It's also interesting that intimate partner violence in general, when directed against women, has a very heavy class component, whereas the rarer intimate partner violence directed against men doesn't really.

February 23, 2005 | Permalink

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8345160fd69e200d834226de953ef

Listed below are links to weblogs that reference Intimate Partner Murder:

» Intimate Partner Murder from Figleaf's Real Adult Sex
As a libertine prude I'm particularly fond of the institution of marriage, and nearly as fond of long-term domestic partnership, and nearly as fond of that as I'm fond of short-term domestic partnerships. Plus I like seeing each other regularly, dating... [Read More]

Tracked on Feb 23, 2005 4:13:34 AM

» Puleeze from Pseudo-Polymath
If marriage were indeed just about a relationship between two adults, Mr Yglesias might be onto something. But alas, ... [Read More]

Tracked on Feb 23, 2005 8:58:01 PM

» Shocking Statistics from Nerve Endings Firing Away
Statistics can lie but they hold tremendous power to herald change. While sitting through CNN’s Headline news, I was about to change the channel on a breaking story about yet-another woman and a kid killed by her boyfriend when a [Read More]

Tracked on Feb 24, 2005 3:00:33 AM

» Puleeze from Pseudo-Polymath
Matthew Yglesias (of the eponymous blog) has a short post on "intimate partner murder" in which he decides:The data seems to suggest that there is, in fact, a silver lining to the "decline of marriage" and a dark side to... [Read More]

Tracked on Feb 28, 2005 7:20:01 PM

Comments

In this context, I think it's very important to point out that this does NOT mean that one's friends, lovers, and aquaintances are necessarilly hugely dangerous. Rather, it means that hugely dangerous people happen to have friends, lovers, and aquaintances, and particularly prey upon them out of the simple fact that that's who they're around most of the time.

Posted by: Brett Bellmore | Feb 23, 2005 6:14:34 AM

To amplify, the notion of murderers as ordinary people who suddenly "snap" is a strangely persistant myth, having very little basis in reality. (Oh, sure, such people exist, but they're VERY rare indeed.) The vast majority of murderers are psychologically abherent individuals with long histories of violent behavior.

This is somewhat blurred by the sealing of juvenile crime records, and the FBI's rather inaccurate approach to classifying deaths on the basis of initial police reports rather than final judicial determinations. Which results in a fair number of justified homicides ending up listed as "murder" in the federal databases... But the conclusion is not at all contraversial in criminological circles: Murderers are NOT ordinary people.

To put it simply: Only marriages to violent people have any real chance of ending in murder. Marriage doesn't cause people to kill, it just puts some people in close contact with killers.

Posted by: Brett Bellmore | Feb 23, 2005 6:43:55 AM

The facts here are pretty striking, and for safety's sake I've permanently said goodbye to all of my friends and family.

Every year here a local rag does a story listing every homicide in Portland, Oregon. After deducting justifiable homicides, the deaths of criminals at other criminals' hands, and family-and-friend killings, there are about 10-20 unexplained or stranger-to-stranger homicides a year in a city of ~400,000. (This is my own rough estimate, but I think it's about right).

An exception probably should be made for "friends" who are friends only because they live nearby and can't be avoided. Often someone might befriend someone just to try to keep on their good side, knowing full well that they're trouble.

Posted by: John Emerson | Feb 23, 2005 8:50:10 AM

Did the drop in murders coinside with the no fault divorce?

Posted by: chef | Feb 23, 2005 9:09:24 AM

Is there a link to a data set that disaggregates IPV into boyfriend/girlfriend vs. spousal? I couldn't find one clicking through.

Posted by: Ugh | Feb 23, 2005 9:22:55 AM

Two points:

One of the big reasons for the drop in murders in the mid-to-late 70s was the implementation of the 911 system and improvements in emergency medical treatment. An older cop once told me that they used to get guys dying in knife fights every Saturday night, but now the same fight is just aggravated assault, because the emergency room stiches them up.

Wasn't that study of the astonishingly high murder rate of pregnant women debunked when it was pointed out that it was roughly equal to the percentage of women who are pregnant?

Posted by: Maureen Hay | Feb 23, 2005 10:08:42 AM

Overlooked is the fact that if a woman pays or otherwise convinces someone to off her husband, it's not counted as spousal violence in DOJ stats, but a mulitiple offender incident.

Posted by: Michael Maier | Feb 23, 2005 10:22:37 AM

Matt, we're waiting for you to draw the obvious connection with your insight here and the posting of a few days ago relating to the alleged criminogenic tendency of smart growth land use planning: http://yglesias.typepad.com/matthew/2005/02/crime_and_new_u.html
The bottom line is, unless you're already married to a killer, you are far more likely to get killed on they highway while driving away from the supposedly dangerous city than you are to be killed in the city by someone you don't know.

Posted by: luis | Feb 23, 2005 11:45:13 AM

Back when I used to work in a battered women's shelter, the two most common times for the onset of abuse were (a) the honeymoon and (b) the first pregnancy. There definitely seemed to be something about pregnancy that set abusers off.

About gender equity in spousal murders: mere anecdotal evidence, but I have talked at least three women out of killing their husbands (also back when I was working hotlines; it's not as though I have a homicidal group of friends or anything.) In each case, the reason the woman wanted to kill her husband was that she was being abused, and was convinced that she and/or her kids would be killed if she didn't kill him first.

Posted by: hilzoy | Feb 23, 2005 11:49:48 AM

I am too distracted to try and compare your analyses, Alina's analyses, and Jack Shafer's, but it's not clear to me you've seen Shafer's, so you might want to take a look.

http://slate.msn.com/id/2111390

Posted by: Saheli | Feb 23, 2005 12:05:15 PM

Hmm, actually, I'm just going to repost my comment to Alina's blog, with all the relevant slate and WaPo links.

Three part series by Donna St. George in WaPo about this issue:

http://www.washingtonpost.com/ac2/wp-dyn/A10074-2004Dec18?language=printer%5d
http://www.washingtonpost.com/wp-dyn/articles/A12359-2004Dec19.html
http://www.washingtonpost.com/ac2/wp-dyn/A14920-2004Dec20?language=printer

Jack Shafer's two part analyses (take-down?) of this article in Slate:
http://slate.msn.com/id/2111254/
http://slate.msn.com/id/2111390

a related WaPo article on how this connects with "crimes against the unborn"
http://www.washingtonpost.com/wp-dyn/articles/A12393-2004Dec19.html

a follow up article by St. George in today's WP, based on apparently new data from the CDC:
http://www.washingtonpost.com/wp-dyn/articles/A45626-2005Feb22.html

Yikes, all done.

Posted by: Saheli | Feb 23, 2005 12:37:27 PM

Intimate-partner killings would be considerably more common if the category were expanded to include love-triangle murders, as happens when two men are in love with the same woman (or two women in love with the same man) and one kills his or her rival. There's some evidence that these account for a non-trivial percentage of acquaintance killings.
Also, do murders within gay relationships count as intimate-partner killings?

Posted by: Peter | Feb 23, 2005 12:49:45 PM

I don't see why not; If you murder your cab driver at the end of the ride, instead of the beginning, I'm pretty sure they count it as an "aquaintance" murder. You really have to work at it to get them to officially admit you've killed a stranger.

Posted by: Brett Bellmore | Feb 23, 2005 12:57:24 PM

Relatedly, child murder is also very rare, but when it does happen the perpetrator is almost always a caretaker (parent, stepparent, etc.) Again, the phenomenom here is not that parenthood makes people violent, but that some parents were already violent people, and a helpless child who is in frequent close proximity, and who does the annoying things that all children do, is very vulnerable to a violent parent.

Posted by: Rebecca Allen, PhD | Feb 23, 2005 1:06:55 PM

I think the problem is that men have no way to terminate their parental rights and responsibilities after a woman becomes pregnant. A pregnant woman always has the right to get an abortion or give the child up for adoption, but after conception fathers are simply stuck with the kid. I think there should be a mechanism where potential fathers can disclaim or rights and responsibilities in a fetus. Then the mother would have the right to abort, give the child up for adoption, or raise it independently.

This situation is another argument for such a system. When killing the mother is the only option fathers have for asserting their reproductive rights, pregnant women are in a very dangerous position.

Posted by: Xavier | Feb 23, 2005 1:17:44 PM

Xavier--think how many men would take up that option. How many deadbeat dads are there in the world? Were there, even when women didn't really have the easy option to terminate their pregnancy? The "my father abandoned me before I was even born" tune is an ancient genre of literature and song. It's a biological asymmettry that two people have sex, but one gets stuck with the baby for 9 mos plus breastfeeding (not to mention not insignificant risk of bodily damage or death, all of this violence stuff aside.) So therefore it's a social asymmetry that the other doesn't get much say in the matter. Society can hold him to it, or honor can hold him to it, or whatever, but there's clearly an obligation that goes along with being able to so easily and fearlessly have sex. When Uterine replicators come out we can revisit this argument.

Posted by: Silence Do Good | Feb 23, 2005 1:29:57 PM


"The overall decline in intimate partner murder is composed entirely of a steep decline in spousal murder that has been only partially offset by an increase in boyfriend/girlfriend murder while ex murder has stayed pretty flat."

Since there is no apprent coorresponding "steep decline" in the number of married persons, some factor other than marital status must be involved.

One of the problems with stats in this area, among those pointed out earlier, is that its fairly easy to for the record keepers to identify a person as a spouse, but boyfriend/girlfriend status is much more in the eye of the beholder.

Posted by: Abdul Abulbul Amir | Feb 23, 2005 1:46:07 PM

The mother has the option to abort or give the child up for adoption. You say that this have wouldn't worked back when women didn't have an easy option for to terminate their pregnancy. Maybe that's true, but it's irrelevant. Women have that option now.

I don't see why reproductive rights for women are so much more sacred than reproductive rights for men. Pro-choice advocates are rightly offended when pro-lifers say that a woman commits herself to raise a potential child every time she has sex. Women deserve more control over their future than that and so do men.

If my proposal were enacted, deadbeat dads would be virtually eliminated. Only men who choose to be fathers would responsible for their kids. They would presumably have some resources available to support the child, they would have some interest in raising the child, and even if they did renege, they would be even more unsympathetic than they are now. Women would not mistakenly rely on child support from men who are unwilling or (more likely) unable to provide it.

Posted by: Xavier | Feb 23, 2005 1:49:03 PM


"I think the problem is that men have no way to terminate their parental rights and responsibilities after a woman becomes pregnant. A pregnant woman always has the right to get an abortion or give the child up for adoption, but after conception fathers are simply stuck with the kid. I think there should be a mechanism where potential fathers can disclaim or rights and responsibilities in a fetus."


This system used to be in place in the first half of the last century. It was called marriage. Those that got married were committed legally and socially to raising offspring. Women who chose to find a sperm donor outside of marriage were able to do so, but were responsible for the consequences of that choice.

Posted by: Abdul Abulbul Amir | Feb 23, 2005 1:55:20 PM

At least one seemingly serious researcher I've read hypothesized that marriages are particularly likely to end in murder, because the high level of commitment tends to dissuade people from dissolving the relationship before problems get really out of hand.

Well, is this true? The chart here shows that the number of murders by boyfriends/girlfriends is about equal to the number of murders by spouses. But what is the number of marriages compared to the number of boyfriend/girlfriend pairs? Without knowing those figures, you can't know the percentage of each relationship that ends in murder.

Personally, I suspect that the number of married couples far outnumbers the number of dating couples, if only because most people spend much more of their lives as married than as part of a dating couple. That would mean that the spousal murder rate is far lower than the boyfriend murder rate.

Posted by: Functional | Feb 23, 2005 2:17:09 PM

40% to 70% strikes me as a rather large spread, and it gets broader if you 'tack on' that alternative 30%. It reminds me of those weather reports where the newscaster said that air quality would be 'good to unhealthful' (ruling out, apparently, only fantastic and lethal).

Posted by: Achillea | Feb 23, 2005 5:34:35 PM

chef: Did the drop in murders coinside with the no fault divorce?

Well, Tyler Cowen blogged last March on The benefits of no-fault divorce where he quoted the Washington Post (scroll to the bottom) citing a Stevenson-Wolfers study:

After states adopted no-fault divorce laws, suicides among women dropped by 20 percent, the rate of domestic abuse fell by a third, and the number of women murdered by their partners dropped by about 10 percent, Stevenson and Wolfers found.

Cowen's take: "I'm all for family values, but let's not forget that some families should split up."

Posted by: fling93 | Feb 23, 2005 9:35:33 PM

One of the problems with stats in this area, among those pointed out earlier, is that its fairly easy to for the record keepers to identify a person as a spouse, but boyfriend/girlfriend status is much more in the eye of the beholder.

IIRC, couples are counted as "intimate partners" (at least for purposes of crime statistics) if they live together or have had a child together. Mere dating relationships don't count.

Posted by: Peter | Feb 23, 2005 10:37:46 PM


"IIRC, couples are counted as "intimate partners" (at least for purposes of crime statistics) if they live together or have had a child together. Mere dating relationships don't count."

Live together for an hour? A day? A week? A minute? Can you be certain when the "living together" started? All of that is subject to reasonable people making different judgments. The point remains that marriage is a verifiable status, where "living together" is to a greater extent a judgment.

Posted by: abdul abulbul amir | Feb 23, 2005 11:25:46 PM

Marriage is a verifiable status Abdul. But, forgive me if I am wrong, last time I checked, it was pretty easy to tell if two people are living together.

Dating and living together are different activities. If they were not, I would never date. As for having a child together, I would not consider the father of my baby an "intimate partner" for statistical purposes, although it does seem that, if he murdered me while I was pregnant, he should be included under this stat.

Posted by: alina | Feb 25, 2005 3:19:33 PM

The comments to this entry are closed.