Statistical Considerations in Criminal Law

We've been covering sentencing recently in criminal law. The chief end of sentencing is to protect the public. We may sentence to punish, to keep offenders away from the public, or to rehabilitate. But what is the best manner to do this?

We have statistical data that tell us who is more likely to reoffend. Should we sentence those who are statistically more likely to reoffend based on race/gender, etc.? If the true goal is public safety, the answer is yes. However, this ignores several important issues.

First, the offender has not actually done anything to indicate that they (as opposed to their statistical class of people) will reoffend. With regard to rehabilitation, there is little correlation between these programs and preventing future crime. But if there is a minor correlation, we as humans must allow criminals to be rehabilitated. If we do not take these factors into account, I believe we deny that people have the power to change, and ascribe to criminals an almost certain fate.

What of criminals' demeanor? Should we base sentencing off of good behavior when many criminals will game the system to get out earlier? And what of apologizing to crime victims as basis for more lenient punishment? Many defense attornies advise against apologies becuase they open the door for civil suits. Aain, I believe the best course of action is to allow convicts to be released for good behavior on the slim possibility they may be genuine. In a sense, we should believe people's actions until they prove otherwise. This is strikingly similar to our innocent until proven guilty standard in criminal actions.

What about apologies? Victims often want apologies more than anything else. Here, I believe we need judicial reform in not opening civil suit liability if a criminal apologizes. This would promote less civil litigation if victims are truly satisfied with the reslt of their criminal trial and would also lower prison terms for those who are contrite.