Debate over what to do about crime is sometimes like reading by Roman candle. The sparks fly without casting much light on the subject. Arguments heat up over capital punishment, banning handguns, or clearing television of violence. Liberals want to rehabilitate criminals. Conservatives want to impeach lenient judges. And there is little consensus on just what prisons are for.
But the temper of the times is changing, says James Q. Wilson, a Harvard professor and one of the nation's leading authorities on crime. He is also editor of an extensive new book of crime-fighting research.
''There is a consensus on criminal justice that probably didn't exist to the same degree 10 years ago,'' Professor Wilson explains.
As researchers Jan M. Chaiken and Marcia R. Chaiken point out in Wilson's book, ''(It) turns out that a very few criminals actually are responsible for prodigious amounts of crime.''
So the crux of the new consensus is that these few criminals - serious, repeat offenders - should be locked up and taken off the streets as quickly and for as long as fairness allows. Corollary to this, says Wilson, is the view that while society should keep trying, it is not likely to succeed at rehabilitating these criminals.
This is a view, says Wilson, ''around which a relatively nonpartisan, nonideological alliance exists in most states and in Washington.''
Beyond this basic point, he adds, the consensus falls apart. Criminal justice has no clear direction now. By contrast, in the late '60s and early '70s the shift was clearly toward rehabilitation and moving criminals out of prison into alternative programs.
''And I don't think there ever will be again that single, dominant way of thinking,'' Wilson says. People are too aware of crime's complexity. They are more interested in simply what works, and where it works, to cut down crime.
The book Wilson has edited, ''Crime and Public Policy,'' just released by the Institute of Contemporary Studies in San Francisco puts forth several practical steps.
Most prominently, it suggests that serious, repeat offenders be identified and given long sentences, while other criminals be given much shorter sentences, or perhaps no imprisonment at all.
Ideally, this ''selective incapacitation'' could both make the streets safer and the prisons less crowded.
''There's a lot of evidence to suggest that may be true,'' notes Wilson, ''but, like most things, you don't know if it's true unless you try it.
''And judges are in my judgment quite reluctant to try, in an organized and consistent way, any new approach. . . . They are skeptical of anything that reduces their capacity to administer individualized justice.''
To just require a minimum sentence for each prior felony on a convict's record is too clumsy, according to Wilson. ''It's a meat-axe approach.'' To identify the high-rate offender, judges need to look at factors like both adult and juvenile records, time in prison, drug use, and recent employment.
''Crime and Public Policy'' also suggests that police could be more effective if they used their authority and expertise in organizing neighborhood volunteers , like watch groups, to help control crime. It is the police that are often the most receptive to new ideas and scholarly research, Wilson says.