Friday, September 28, 2007

Prisons and crime

Most of the letters I got in response to my column "Life Sentence" were positive, but reader CW -- perhaps reflecting the CW -- had a different take:

Not once do you even bring up the actions that put these people in jail. Yes, they are disproportionately black and poor, but they did commit these crimes to wind up in jail. Or perhaps you could explain to the mostly black victims of their crimes about how the offenders are the real victims.

Now, it appears you're trying to describe a public policy matter and how society should help ex-cons. Again, you paint one side of the picture -- which is how these people are victims. Isn't there evidence that suggests people with past criminal records are more likely to engage in criminal activity again in the future? Maybe, maybe not. I don't have a hard time imagining such evidence being suppressed by the liberal academic establishment.

Before you chalk me up as some wacko right winger, I'll point out that I've voted for Democrats in the last 2 presidential elections and opposed the war in Iraq from the get-go.

Bottom line, you're article is biased and seems to serve an agenda. It fails to examine the flip-side of policies or alternatives that you've proposed -- one of the most basic tenants of critical thinking.

I probably should have included at least an aside about the arguments over the relationship between imprisonment and crime. Since the early '70s, as the article noted, the number of people behind bars has grown seven-fold. It's been a straight march upward. And the number of prisoners per 100,000 citizens has grown by a comparable amount.

On the other hand, the number of crimes reported per 100,000 citizens rose sharply from 1970 (when the number was roughly 4,000) to 1980 (peaking at 6,000), dipped until 1985, rose again until roughly 1991 (not quite to 6,000)-- and has outright plummeted since. Now we're back down nearly to the early-'70s level of reported incidents.

What's the link between the two phenomenon? The Harvard sociologist Bruce Western, whose work I profiled, thinks there basically isn't one: He argues that only 2 to 5 percent of the huge crime drop in the 1990s was caused by locking up people. (Many of today's prisoners, he points out, are there for nonviolent drug offenses or because they violated technical conditions of their parole; he's also not sure the perpetrators of violent crimes of revenge or passion carry around in their heads the increased likelihood of conviction today, or the higher minimum sentences.)

On the other hand, the economist Steve Levitt, co-author of "Freakonomics," has argued that more than one-third of the drop in violent crime in the 1990s can be chalked up to tougher sentencing. Levitt didn't respond to my request for an interview, but a scholar who essentially agrees with Levitt on this point, Bill Spelman, of the University of Texas, did. He told me that despite his belief that prisons are more effective than Western thinks, he still thinks we've long since passed a point of diminishing returns, incarceration-wise. He'd roll back prison populations to the level they were in 1990 -- half what they are today, but still higher than Western would prefer and more than double the figure for most Western European countries. Western declined to give a firm number for his ideal, saying he had not crunched the numbers and that the issue was hugely complicated.*

Certainly people of different ideologies, and experiences, will disagree on the cost-benefit calculus of any imprisonment rate. But it's puzzling that, since 1970, prison populations have boomed when crime is rising -- and they have boomed when crime is falling precipitously.

*Western, as I did say in the article, also makes the case that what prison does to men, and in turn to communities, itself breeds crime -- and the amount of that crime, he suspects, is larger than the amount decreased by keeping people off the streets. I don't know if that's true -- but it has to be part of the equation.

Posted by Christopher Shea at 04:05 PM

Print | E-mail to a friend | Permalink | Subscribe via rss