Part 3: Retributivism and its Critics
David Lyons: Punishment as Retribution
Retribution is giving people what they deserve, hitting them back with equal force to a blow they have struck, and treating someone as they have treated others. A classic legal statement often put forth by retributionists is: lex talionis , which means "an eye for an eye, a tooth for a tooth, and a life for a life." How is retribution different from revenge since revenge is also the hitting back of the wrongdoer? The difference is that retribution involves hitting back with equal force whereas revenge often involves hitting back harder than we have been struck. Revenge exceeds what a person deserves, often to the satisfaction of the vengeful.
Retributive claims for punishment can arise not only when a harm to others has been committed (the standard case) but also when someone violates the "minimal standards of decency and consideration for others." For example, someone who attempts, but does not succeed in harming another still generates a retributive call for punishment. So also the person who through non-action (negligence) puts others at risk or displays disregard for others generates a retributive call for punishment.
In addition to lex talionis, a more complex statement of retribution is offered by Immanuel Kant. Kant takes into 
account the attitude or motive of the wrongdoer. This makes retribution more flexible than lex talionis as it would allow that unintentional crimes warrant less punishment than intentional ones. Kant views people as rational agents who understand what they do and therefore when we intentionally act, we indicate how we think actions should be performed by everyone. Punishment then serves as a way to give people back what they indicate ought to be done to others. Thus, punishment respects the individual person as we are only doing to you as you have intentionally done to others. Our duty, for Kant, is to only perform actions as we believe all rational people should act. Actions such as lying, stealing, and murder cannot be said to be universalizable. In other words, rational people cannot (and do not) believe that everyone should lie, steal, and murder so these actions are wrong. For example, when people murder they indicate their belief that others should do these things to them. This is where retributive punishment comes in as it serves to give them what they seem to be asking for-execution.
The primary feature that attracts us to retributivism over the utilitarian justification for punishment is that retributivism ties punishment to what the person deserves, thereby avoiding the problems of over-punishing, under-punishing, and punishing the innocent. However, several challenges to a strictly retributivist theory of punishment arise.
- What is the moral purpose of punishment?
The utilitarian can offer some goal or gain for punishment-the greater good achieved through deterrence and rehabilitation. Retributivism seems to uphold punishment for its own sake rather than for any particular benefit. For example, any serious retributivist would hold that even if the world were to end tomorrow, we have a duty to ensure that all those awaiting execution are executed. Retributive punishment therefore serves no future purpose. - How do we fit punishments for offenses?
Okay, we execute the murderer, take an eye for an eye, but do we torture the torturer, rape the rapist, and steal
from the thief? Outside of a few simple cases, what guides us in determining what people deserve? What is the deserved punishment for stealing a candy bar? What about rape? Is multiple murder deserving of more punishment than single murder? In answering each question, remember you can not appeal to rehabilitation or deterrence in your answer as they are not justifiable reasons to punish for the retributionist. Further, your answer must be supportable by reasons other than personal preference as society will punish all like cases based upon your decision.
Thought Question: Punishing the Three Drunks?
Three neighbors get drunk at a bar. Each takes the same route home but 15 minutes after the one before them. The first drives home without incident except that a police officer happens to pull him over right next to his home on a random check and arrests him for drunk driving. Shortly after the first man is arrested, it begins to rain. The second drives home but skids off the wet road during a sharp corner into a telephone pole where police find him and arrest him for drunk driving. The third drives home and skids at the same wet sharp corner but due to bad luck or timing hits a family of four head on, who are killed. The third man is arrested for drunk driving (and four counts of manslaughter). You, a retributionist, are assigned to determine the deserved punishments for each of the men. Each of the three intended to drive home drunk, but due to rain and bad timing the results of each drunk driving attempt varied widely. What is the deserved punishment for each of the men?
- Why should the state have a monopoly on punishment?
If someone walks up and punches me and I punch them back, I have provided a justified retributive punishment. Yet the state will call me a vigilante and punish me for my action because the state claims sole authority for punishment. The utilitarian can justify this state monopoly by holding that general utility is best served by reserving punishment decisions to the state. Yet, there is nothing grounded in a retributionist theory that would uphold the state's right to monopolize punishment. Is it an acceptable outcome for our system of punishment to allow individuals to punish others so long as they do not do so excessively? - Is there any possibility for remediation of harms?
Some might argue that retribution can also be said to restore the moral balance by taking away from the
offender any gains from his offense through punishment and perhaps, where possible, forcing the offender to compensate the victim for his loss. Yet Lyons argues this result is problematic in its application for several reasons. First, to restore the moral balance may not always be possible as many victims cannot be compensated for their loss. Second, restoring what was before the crime would only be morally acceptable if the prior condition was itself justified. Third, retribution itself focuses on punishment, not remedy. If I poke out another man's eye, but surgery can replace the eye, will the retributionist be satisfied if I offer to pay for the surgery (thereby forgoing any punishment to me) or will the retributionist prefer that I be deprived of something equivalent to what I have taken? It seems difficult for retributionists to forgo punishment for crimes in exchange for remediation of harms. (Would the state forgo an assault prosecution if after breaking your legs I offer you a sufficient sum of money in compensation?) - Does retribution go too far?
Retribution, unlike law, seems to justify punishment for any action that is wrong. This means that we appear justified in punishing any wrong action (even if there is nothing illegal). Not only does the law traditionally limit the amount of immoral action it punishes, but it also reserves punishment for actions currently illegal. Retribution seems to expand punishment to be justified in all immoral actions (something Kant, Aristotle, and Devlin might agree upon, though for different reasons) and even allows for ex post facto punishments as if your action was wrong when you did it, it deserves punishment (even if we passed the law after you committed the act). - Does retribution go far enough?
Just as retribution expands punishment beyond traditional legal boundaries, it also provides no reason to punish for many legal infractions we have today. For example, most traffic laws and laws promoting economic competition are not based upon any immorality of action nor can punishments be justified based upon desert. Instead these laws are designed to promote some common good (consequence) rather than prevent any immoral or ill motivated behavior. For instance, there appears to be nothing immoral about jaywalking or changing lanes without using a blinker when no one else is around, but these behaviors are punished by law and retribution provides no justification for this.
Herbert Morris: Punishment and Fairness
Morris offers an alternative retributive theory of punishment to lex talionis and Kant. According to Morris, society is based upon a contract of rules. We understand the rules as those things necessary to ensure fairness. When someone commits a crime they have violated the rules in order to gain some good in society without paying the appropriate cost. In other words, they have cheated to get ahead. Punishment, as for all retributivists, is giving people what they deserve. As such, Morris argues that punishment is simply taking away any good or advantage a criminal has obtained against the rules. They deserve to be put back the way they were before the crime was committed. When someone knowingly violates the rules, they bring punishment on themselves-they incur a dept to society that the institution of punishment repays.
C. L. Ten: Is Punishment Fair?
Ten offers a rebuttal to the theory of punishment put forth by Morris. According to Ten, Morris does not provide a pure retributionist theory of punishment at all. Instead he provides a mixed justification in which:
- Punishment is justified to take away the gains made through crimes (a retributionist reason) and
- Punishment is justified to ensure that those who obey the law rest assured that law breakers are punished (a utilitarian reason).
This theory will fail in practice because most benefits obtained through crime cannot be "paid back" to the person wronged (as one would expect to restore things to how they were). Instead Morris' theory would be "paying back" society (not the victim) by ensuring to them that lawbreakers do not "get away with it." This begins to look less like a theory of retributive punishment and more like a theory of political obligation as "deserving punishments" becomes synonymous with "breaking the rules of the game." This stands apart from punishing for the greater good, and punishing because the person has done something deserving of punishment. With Morris, punishment becomes a matter of holding that everyone is obligated to obey the rules (without regard to what those rules are) and that violators of the rules deserve punishment to ensure that rule breaking does not pay. Since everyone benefits from obeying the rules, then you are obligated to obey them too.
Even if we accept this idea of an obligation to obey the law (and there is substantial literature in political philosophy that rejects this very idea), Morris' theory is also vague in that it does not provide any way of determining what punishment is deserved, only that some type of proportionality is due for violators of the rules. This avoids the real challenge for any non-utilitarian theory; what punishments fit what crime and why? Might the answer be to employ the general utility of deterrence? If so, does Morris' theory collapse into utilitarianism? Not entirely, as the utilitarian takes into account all consequences in determining a course of action, Morris' theory limits this to consequences of fairness to the law-abiding. One significant advantage that Ten finds with Morris' theory is that it provides a defense against punishing the innocent. If punishment is deserved only when people voluntarily act unfairly by breaking the rules, then it can never be possible to punish an innocent person as they do not deserve it.
Further Objections to Morris
- Morris's theory works by punishing acts that are unfair to law-abiding citizens.
This works to justify many punishments but not others. For example, how could Morris justify punishing cruelty to animals (kicking your own dog perhaps)? This theory is therefore incomplete and requires supplemental justifications for other punishments. - More needs to be said about harm to the victim.
Morris would find the murderer deserving of punishment for an act unfair to law-abiders, but speaks nothing about the harm to the victim. -
This theory enables perpetuation of social injustices.
In a society with vast inequalities of wealth and opportunity, it seems likely that the bulk of criminals will come from those at the bottom of the socio-economic spectrum. They may benefit from some from the rules, but the bulk of benefits are reserved for the higher classes. Here Morris' theory of "fairness" continually justifies punishment for those who are not trying to "gain an advantage" as much as they are trying to "catch up." It is not clear how punishing the impoverished protects a fair distribution of benefits and burdens. -
Misplaced motives for obeying the law.
Morris seems to think that the decision to obey the law is motivated by the desire to continue the rules or the system of mutually beneficial social cooperation. Yet, most people's decision to obey is based upon fear of punishment. For instance, a great many law-abiders might reject the rules, but obey them for fear of punishment. Morris seems to take too large a liberty in pointing to law abiders as "supporters of the system" in whose name those who violate the system deserve to be punished when a majority might agree with the criminal's action.
Ten concludes that all retributionist theories fail on their own accord. Yet, retributionist theories of punishment seem to have a necessary component in a theory of punishment that will constrain the worst of utilitarianism (punishment of the innocent) as well as justification for punishment in certain cases.
George Schedler: Can Retributivists Support Legal Punishment?
Schedler offers a further challenge to any retributivist theory. His claim is that retributivists provide a moral justification for punishment which will stand in judgment of systems of punishment (legal punishment) and will reject them. Since, we know for certain that our system of punishment will convict the innocent on occasion (no system is perfect), then retributivists must reject our system. As no system we know of is immune from this criticism, retributivism is incompatible with any system of legal punishment.
One defense of retributivism is that mistakes happen and we recognize that. This defense will be 
insufficient as our system of legal punishment, and its burden of proof are prone to human errors. So long as we know our current system is flawed and we have alternatives to improve protections against the convicting of the innocent, but refuse to implement those protections (due to expense or hassle or fear of letting more guilty go free), then we are knowingly and intentionally endorsing a system that convicts more innocents than a possible alternative. One way of viewing this dilemma is to consider a move from "reasonable doubt" to "no doubt" as a standard of proof for conviction. Juries under the "reasonable doubt" standard will today find innocent men guilty because the evidence seemed convincing. In many trials, the "reasonable doubt" standard is met strictly through circumstantial evidence, which is hardly the surest way to protect the innocent. Under a system of no doubt, the burden of proof would be very high, more guilty would go free, but innocents would be protected. So long as we adopt the system that will convict more innocents, we cannot be true retributivists.