Tuesday, April 10, 2012


John Derbyshire, who writes for conservative, especially alt-conservative magazines, has a firestorm article over at Taki's Magazine. It was written in response to the writings of black columnists in the wake of the Trayvon Martin shooting. (Several links to same in the Derb article.) It's an important bit to keep in mind in any defense of Mr. Derbyshire, as those articles have some pretty racist and certainly intellectually soft bits themselves. But a mainstream-acceptable racism, you see.

Derb defends himself as well as he might over at Gawker, so I don't need to give thumbs up or thumbs down here.

Steve Sailer had an interesting take, relating The Talk, black version to a conversation parents of any race should have with their child. He's a bit harsher on cops than I would be - perhaps attributable to my living in NH - but this information went out to my various sons in direct or indirect form. (The Romanian sons have mixed American and Eastern European attitudes.
My sons aren't black, but I have given them the exactly same Talk about cops, although mine was probably less polite toward cops. I made ensure to emphasize that "Cops have guns, with which, if you piss them off or make them feel uneasy, they might kill you. The kind of people who become cops have personalities where they want to dominate others around them. (If they just wanted to save people, they could have become firemen.) So, let them. Remember, cops have guns."
Jonah Goldberg's "Goldberg File" from NRO is by free subscription, not on the site, but his take was of the Straw, camel's back variety to Rich Lowry "firing" Derbyshire over the article. More importantly he notes what I think would be important to readers here, as we have covered it before: Derb keeps saying he's using science - specifically, statistical data - but a good deal of what he puts forward as evidence is in fact, anecdote. I have mentioned before, including quite recently, how much I think that is essentially deceptive. For example, Derb asks us to suspend our ordinary rules for whether we would help a stranger in distress by linking to an incident in which a Good Samaritan was killed by a black couple. Try as he might, I don't think Derb can maintain he's got solid Good-Samaritan-ends-up-dead numbers broken down by race. The sample size is too small.

As my son Jonathan points out, there is also a tone problem with the essay. It is one thing to report the IQ numbers (and I could state those same numbers a different way in full honesty), quite another to lean almost gleefully, or at least pompously, on the topic. He's a professional writer. He knows tone.

I do think John Derbyshire made some good points in the column. In the aggregate, however, it is weak and unpersuasive.


Texan99 said...

The article struck me as an irritable "two can play at this game" response, and the links to the articles made the argument take the form of stating a universal rule and "proving" it by citing to a single, selected incident. Weak, really just a loss of temper. Of course he was provoked by some serious nonsense, but if you're going to do that kind of thing you should at least make it funny.

james said...

I agree with Grim that his description of the utility of black friends is an insult to any friends he has. If serious, very bad--if not, very poor work.

Hanging around an all-black church wouldn't expose him to anything more dangerous than an invitation to a pot-luck. And the gospel, of course, which has dangers of its own.

karrde said...

I hadn't look carefully to separate statistics from anecdote.

Maybe I should do that...

Or, in reverse, maybe I should dust off some queries that I did at the CDC website about six months ago, and the FBI's Uniform Crime Reports about a year ago.

The CDC's WISQARS tool gives victims of homicide by race, with no data on perpetrator. The racial category "Black" contains slightly less than half of all homicide victims. (I just looked at 2009, but I recall the racial-ratio being fairly constant over most of the years reported by the CDC.)

--Point against Derbyshire: he says Blacks commit half of all murders, but fails to mentions how many such murders are murders of Black, and how many are murders of Whites.
--Complicating point: any death by the action of another human, even deaths that the police would not charge as murder/manslaughter, are reported as homicide by the CDC.

The UCR for 2009 has a lower number of murders reported than CDC-reported homicides.

The UCR has a slightly-higher rate of Black homicide victims. (Tables 1,2 on linked page.) Still, it's in the realm of 50%. The same report gives 37% of murder offenders as Black, 33% as White. Most of the remainder are of race Unknown, about 1.5% are race Other. (Table 3)

--Point against Derbyshire: He implies that 50% of all murderers are Black. It appears that 50% of murderers of known race are Black, and 37% of all murderers are known to be Black. A disturbing ~28% of murderers are of unknown race.

Significantly, the UCR also shows that ~90% of all murders are White-victim-White-offender OR Black-victim-Black-offender. (Table 6. Values are for category single-victim/single-offender, see on Table 4.)

--Point against Derbyshire: the number of cross-racial murders is small, both as a raw number and as a percentage of murders.

It doesn't look too good for Derbyshire, though he is right that the level of criminal behavior among Blacks is uncomfortably high.