How frequently have we heard a judge say to a defendant on the day of sentencing, “I’m sending you to prison because I want to send a message that we won’t tolerate this kind of behavior?” The “message” is supposed to be a deterrent to potential offenders to cause them to refrain from engaging in crime out of concern that they’ll end up in prison. Aside from the fact that only in the unusual case does anyone outside the courtroom even hear the message, there’s little evidence that harsh sentences produce any significant deterrent effects. Criminologists have known this for hundreds of years.

Writing in the 18th century, the Italian criminologist Cesare Beccaria argued that deterrence in the criminal justice system was basically a function of the certainty of punishment, not its severity, a conclusion that still resonates today. Think about how we drive on a highway. Virtually all of us drive over the speed limit to some extent, but if there’s an increased presence of state troopers patrolling on a holiday weekend most of us will slow down. That’s because the certainty of punishment has increased and we’re trying to avoid getting a ticket. But if the legislature has recently increased the severity of punishment, by raising the fine for speeding, that will have little effect. Most of us will be unaware of that change, and we’re not expecting to be caught anyway.

The recent report by the National Research Council on The Growth of Incarceration in the United States: Exploring Causes and Consequences, a comprehensive assessment by the nation’s leading criminologists, confirms these findings on deterrence. The report concludes that “all of the evidence on the deterrent effect of certainty of punishment pertains to the deterrent effect of the certainty of apprehension, not to the certainty of postarrest outcomes (including certainty of imprisonment given conviction).”

The threat of criminal penalties is therefore only significant if an individual believes there is a high risk of being caught for committing a crime. But since most crimes, whether minor property offenses or serious violent crimes, don’t in fact result in arrest and conviction, there is little reason for most potential offenders to consider the severity of penalties.

A further problem with deterrence theory is that it assumes that people are rational actors who calculate the consequences of their actions. But we know that this is frequently not the case, since many crimes are committed impulsively. A national survey of state prisoners found that half the offenders reported that they were under the influence of drugs or alcohol at the time of their offense.

If we care about enhancing public safety we need to compare the effects of severe punishment with alternative approaches. Looking at deterring individual offenders one might suspect that the more severe the sentence, the less likely that the person would reoffend. But a 1999 meta-analysis reviewing 50 studies over several decades found that longer prison sentences were actually associated with a 3|PERCENT| increase in recidivism. Further, controlling for various individual characteristics, the researchers also found that being incarcerated versus receiving a community-based sanction was associated with a 7|PERCENT| increase in recidivism.

None of these findings suggest that prison is not appropriate in certain circumstances, either for its incapacitation effects or its retributive value. But if we want to actually deter crime, there is little to be gained by enhancing sentences that are already too severe in many cases. And, of course, preventing crime in the first place by providing family support and creating opportunity is by far the most effective – and compassionate – means of promoting public safety.

Marc Mauer is the Executive Director of The Sentencing Project and the author of Race to Incarcerate.