Three philosophies of punishment and whether or not they work

Retributive justice

Perhaps the most straightforward idea about punishment there is; if someone does something very wrong, they cause themselves to be worthy of punishment. This punishment is good by itself, even if there are no side effects. Most theorists in favor of this system also posit that the punishment should be proportional to the crime and that it should only affect those duly convicted.

Most people have a strong intuition about this. A famous thought experiment with many variations asks people to imagine that murderers enjoy long tropical island getaways where they can't hurt anyone after conviction but appear to be suffering in jail for TV cameras every now and again, to deter other potential murderers. Even if the deterrent works, you might feel that something is off here. Something that can only be corrected by inflicting some kind of punishment on the murderer.

University of Chicago Professor Albert W. Alschuler argues that retributive justice can have positive consequences in addition to any inherent justice it offers.

He puts forward the idea of a neighborhood where no one parks correctly, with drivers frequently boxing in others and parking too close to stop signs and fire hydrants. The laws against this are unenforced in that neighborhood. Because there is no consequence for inconsiderate parking, there is no reason to be considerate yourself; your neighbors will continue to act like this in any case. It ends up being the case that everyone acts this way to avoid being a sucker. He points out that this situation could be resolved by punishing the lawbreakers, as it would drive people back to a state of fair play. He summarizes the concept by saying, "Withholding punishment is inappropriate when doing so would encourage people to conclude, 'Everyone else is looking out for themselves, and I'll be a fool unless I become a little bit like them.'"

Arguments against retributive justice often focus on the difficulties of justifying harsh treatments (rather than just punitive damages or restitution) against the convicted in a way that aligns with broader principals of justice. Many theories that attempt to do so have been deemed unsatisfactory by other philosophers. Others point out that retributive systems only look backward on what has been committed and not forwards, to what situation we'd like to be in after matters are settled.

Deterrence

Deterrence theory is the idea that punishments for crime should exist primarily to discourage others from committing a similar crime or to assure the punished individual won't do it again. For example, making the potential costs of committing a crime too high to justify doing it in the first place.

Jeremy Bentham, the eccentric founder of Utilitarianism, took a consequentialist approach to punishment. Seeing punishments as "always evil," he nevertheless hoped that the use of them could deter crime by others, increasing the total happiness of society overall and reducing the number of criminals in the future. He combined this support for deterrence with elements from other theories.

Ceasre Baccaria, a jurist in Milan during the Enlightenment, argued that crimes strained the social contract and that punishments should be used to assure that people continued to stick to it. Rather than a retributive scheme, this called for a deterrence system to ensure that neither those punished nor those aware of the punishments would desire to commit such crimes in the future.

Of course, there are objections to this idea as well. The most common revolves around the theory's assumption that most people who break the law weigh costs and benefits before doing so. A point many would contest. The previously mentioned thought experiment (with the murderer on the island) also points to another objection to pure deterrence theories. Deterrence can be produced without actually punishing the convicted, a situation that strikes many as unsatisfactory.

Rehabilitation

Rehabilitative theories of punishment are diverse in their foundations. In general, they look at what causes a person to turn to crime and try to remedy the situation.

Many proponents of rehabilitative theories argue that the decision to engage in criminal behavior is not as clear cut as other theories suggest. Factors of economic opportunity, addiction, mental illness, social issues, and circumstance can make it more or less likely that a person will be driven to crime. With that in mind, they suggest that the penal system should focus on resolving or mitigating those issues.

Others are more utilitarian in perspective. They argue that a person who went into jail with a criminal tendency is likely to come out the same way unless some action is taken. What that looks like, be it job training, education, counseling, or something else, depends on the situation. Making it less likely someone will return to crime by providing these services, they argue, benefits society as a whole.

This comparatively holistic and often humane approach doesn't mean there isn't a potential dark side to rehabilitation. The theory is very dependent on our understandings of psychology, sociology, and criminology being accurate. Mistakes can have horrible results. The modern practice of solitary confinement, a practice now deemed torture by the United Nations when used for more than two weeks, goes back to recommendations by the Quakers that leaving criminals alone and slightly sensory deprived would allow for repentance and reformation. They were extremely wrong.

Even when it works, there are concerns about its implications. In his pro-retribution essay on criminal justice, Professor Alschuler cites Francis A. Allen's argument that a dedication to rehabilitation can make it challenging to limit the scope of state involvement, as "one immediate consequence of a rehabilitative regime is a drastic enlargement of state concerns. The state's interests now embrace not only the offender's conduct, but ... his motives, his history, his social environment."

The concerns of libertarians and others interested in a limited state are easy to comprehend.

Enough of this abstract philosophy, what does the data say?

Empirical data does exist in a wide variety of areas related to the criminal justice system. Here, we can use it to see if the above conceptions of justice can do what they set out to do.

Retributive justice benefits from only seeking to deal punishment out to those convicted of crimes, which it often manages to do. It isn't easy to empirically measure such a thing, but its various side effects can be measured.

Studies show that those close to a convicted individual can share the effects of punishment despite them not having committed a crime themselves. Similarly, a criminal record's impact can follow people long after they have "paid their debt to society," suggesting that it is more difficult to assure "proportionality" in sentencing than might be supposed. In the United States, unequal sentencing is a known and well-documented problem, suggesting more difficulties in reaching the ideals of retributive justice in reality.

Deterrence theory has a fair amount of empirical evidence against it. Studies suggest that many crimes are committed under the rationality reducing influence of alcohol, that few people can tell you what the punishments for a given crime are, and that many people don't consider the possibility of being caught when planning a crime.

Longer sentences are associated with slightly higher recidivism rates, the opposite of what a proponent of deterrence theory would expect from people with first-hand knowledge of the prison system. Likewise, programs like "scared straight" don't seem to do much.

However, Professor Daniel Nagin has argued for the existence of a general deterrent effect while also suggesting it is difficult to use this to make any new policy. Dr. Valerie Wright suggests that a deterrent effect does exist, but adds that it is tied to how certain a person is that they will be caught and given a specific punishment rather than how terrible their punishment might be.

Rehabilitation has shown promise in achieving its goals. Efforts at providing education and addiction treatment in the American prison system lead to reductions in recidivism. The Norwegian prison system, based on rehabilitation and renowned for its humanity, boasts one of the lowest recidivism rates in the world.

Despite this, American efforts have yet to match the Norwegian system's effectiveness levels, and some studies also suggest that modern treatment programs have little effect on individuals with psychopathy, who are disproportionately incarcerated and have a high rate of recidivism to begin with.

What this data means is going to be influenced by which of the above arguments appeal to you. Do the side effects of retributive policies or the problems we have in assuring equal punishments for similar crimes outweigh the moral intuition towards punishing criminals? Do failures in rehabilitative practices make the concept worthless? Can deterrence be of use even if we know a disproportionate number of criminals aren't acting along the lines of its assumptions?

The raw numbers can't answer these questions by themselves. Philosophy has to step in and provide the tools for value judgments, answer questions of justice, and help determine where the line between theory and practice has to be drawn.

We'll probably never be rid of the need to do something with people who harm or violate the rights of others. What we do with them is another question. No definitive answer exists for what models of justice and punishment are best. Still, by considering the philosophy and raw data around each model, we might find something that works for our society. While many people would support a system that uses elements of all three of these considered philosophies, alongside others, how much of each to use remains the subject of continual debate.