A reader writes:

I am open to any new and true evidence from science of previously unsuspected factors that help explain the movement of crime rates after the middle 1960s. But, so far, both Levitt's and Nevin's work (blaming lead) look to me like efforts to find something, anything, just the slim sliver of anything, that liberalism might have done, by getting the lead out or allowing abortions, that might, just might, have just slightly offset the obviously catastrophic damage that liberalism did by the Warren Court's dismantling of the state criminal justice systems in the crucial middle 1960s, which dismantlement dramatically reduced the conviction rate for crime and was immediately followed by an explosion in, uh, crime.

I am certainly prepared to believe that other forces, probably mostly social, contributed to the increase in crime after the middle 1960s. Post-World War II society was becoming richer, easier and more tolerant of deviations from traditional norms in sex, child raising, work habits and responsible behavior, and thus was more vulnerable to extreme forms of misbehavior, such as crime. But the important thing is that, in the middle 1960s, we lost control of our most direct means of dealing with the potentially violent consequences of all these changes, the justice system.

And that loss of control is perhaps one reason why social scientists continue to play with alternative explanations for crime rates. We can endlessly fiddle with and adjust laws on pollutants, such as lead, and attempt to nudge sexual behavior this way or that by lectures, if not laws. But the post-Warren Court criminal justice system was put in a pseudo-Constitutional lock-box. We cannot get at the thing, to change or improve it. We can only lengthen sentences, hire more cops and build more prisons. No change in the rules governing how police attempt to arrest criminals or how the courts try them is within the control of elected officials. So we do not even discuss such changes any longer or attempt to estimate their possible effects.