These features show that it makes little sense to blame the police, judges or correctional personnel for being soft on criminals. There is not much else they can do. The police can't find most criminals and those they do find are difficult and costly to convict. Those convicted can't all be sent to prison. The society demands that we do everything we can against crime. The practical reality is that there is very little the police, courts or prisons can do about the crime problem. The criminal justice system must then become as powerless as a parent who has charge of hundreds of teenage children and who is nonetheless expected to answer the TV message: "It's 10 o'clock! Do you know where your children are?"

A few statistics from the Justice Department's recent "Report to the Nation on Crime and Justice" illustrate my point. Of every 100 serious crimes committed in America, only 33 are actually reported to the police. Of the 33 reported, about six lead to arrest. Of the six arrested, only three are prosecuted and convicted. The others are rejected or dismissed due to evidence or witness problems or are sent elsewhere for medical treatment instead of punishment. Of the three convicted, only one is sent to prison. The other two are allowed to live in their community under supervision. Of the select few sent to prison, more than half receive a maximum sentence of five years. The average inmate, however, leaves prison in about two years. Most prisoners gain early release not because parole boards are too easy on crime, but because it is much cheaper to supervise a criminal in the community. And, of course, prison officials must make room for the new prisoners sent almost daily from the courts.

We could, of course, get tough with the people we already have in prison and keep them locked up for longer periods of time. Yet when measured against the lower crime rates this would probably produce, longer prison sentences are not worth the cost to state and local governments. Besides, those states that have tried to gain voters' approval for bonds to build new prisons often discover that the public is unwilling to pay for prison construction.

And if it were willing to pay, long prison sentences may not be effective in reducing crime. In 1981, 124,000 convicts were released from prison. If we had kept them in jail for an additional year, how many crimes would have been prevented? While it is not possible to know the true amount of crime committed by

people released from prison in any given year, we do know the extent to which those under parole are jailed again for major crime convictions. This number is a surprisingly low 6 percent (after three years it rises to only 11 percent). Even if released prisoners commit an average of two crimes each, this would amount to only 15,000 crimes prevented: a drop in the bucket when measured against the 41 million crimes committed each year.

More time spent in prison is also more expensive. The best estimates are that it costs an average of $13,000 to keep a person in prison for one year. If we had a place to keep the 124,000 released prisoners, it would have cost us $1.6 billion to prevent 15,000 crimes. This works out to more than $100,000 per crime prevented. But there is more. With the average cost of prison construction running around $50,000 per bed, it would cost more than $6 billion to build the necessary cells. The first-year operating cost would be $150,000 per crime prevented, worth it if the victim were you or me, but much too expensive to be feasible as a national policy.

Faced with the reality of the numbers, I will not be so foolish as to suggest a solution to the crime problem. My contribution to the public debate begins and ends with this simple observation: getting tough with criminals is not the answer.