Should punishment be about retribution or deterrence?

One of the things I found particularly interesting in ‘Who’s in Charge?’ by Michael Gazzaniga, is experimental evidence suggesting that people who believe they have free will tend to be better behaved than those who believe in determinism. People who don’t believe they have free will are apparently less likely to control their impulses to act selfishly or aggressively. This might be an instance where a little knowledge can be a dangerous thing.

The main point made by Gazzaniga – who is referred to as the father of cognitive neuroscience – is that individual responsibility is a dimension of life that cannot be illuminated by analysing single brains in isolation. In isolation, single brains seem to be governed by unconscious intentions – awareness comes after the event. Individual responsibility comes from social exchange. If we want to understand how individuals are responsible for their actions we need to look at the whole picture of a brain interacting with other brains.

Readers who are interested in a general overview of the book should read the review by Benedict Carey in the NYT.

Gazzaniga discusses the question of whether punishment should be retributive or utilitarian near the end of his book. Retributive justice is concerned primarily with giving criminals the punishment they deserve – the crucial variable is the degree of moral outrage the crime engenders. Utilitarian justice is concerned primarily with the future good of society. The author suggests that means it is concerned with deterrence, incapacitation (e.g. jailing criminals to prevent them from re-offending) and rehabilitation.

The interesting point is that although many people label themselves deterrists rather than retributivists, when it comes to actually handing out punishments self-labelling counts for little – people have a strong tendency to behave as retributivists. Irrespective of what they say, they tend to punish for harm done even when there is little likelihood that the person will re-offend in future.

Gazzaniga suggests that retributive justice has deep moral foundations in human evolution. We can use abstract consequentialist thinking when faced with abstract questions of public policy, but we resort to fairness judgements when faced with an individual who is to be punished.

For example, should harsh sentences be applied to minor offences to increase the deterrence effect? If you think about it in abstract terms, sending a person to prison for a relatively minor first offence (e.g. low range drink driving) might seem likely to increase the sum total of human happiness by deterring others from an anti-social behaviour that endangers human life. But would it be fair to hand out punishments that are disproportionate to the additional risks involved in particular instances?

Consider a more extreme example. Should judges make an example of celebrities by giving them greater punishment for minor offences? Since the punishment of celebrities would receive greater publicity it could be expected to have a greater deterrent effect, but in my view it would still be unfair.

Does utilitarianism necessarily imply that it is OK to impose unfair sacrifices on individuals for the future good of society? No. Some people who subscribe to utilitarianism, as a theory of normative ethics which views human happiness as the fundamental value judgement or ultimate criterion, consider the best test of actions or rules of action to be the extent to which they promote social cooperation. For example, Henry Hazlitt argued that ‘for each of us social cooperation is the great means of attaining nearly all our ends’. He noted that social cooperation ‘has the great advantage that no unanimity with regard to value judgements is required to make it work’. It enables the disparate goals of different individuals to be reconciled and harmonized. (‘The Foundations of Morality’: 35-36).

Leland Yeager has argued:

‘Regardless of just what plausible interpretation we give to happiness, social cooperation is prerequisite to its effective pursuit. Lying, cheating, and stealing subvert happiness because they subvert the prerequisite cooperation. Telling the truth, keeping promises, and respecting other people’s rights and property are conducive to cooperation. We come to believe propositions like these through factual and logical analysis of what conditions help individuals pursue their own diverse goals effectively’ (‘Ethics as Social Science’, 2001: 83).

Yeager also makes the point:

‘Emphasis on voluntary cooperation warns against imposing unfair sacrifices on individuals for the supposed greater good of a greater number’ (p. 82).

This line of reasoning suggests to me that if social cooperation is the objective we should be seeking retribution i.e. giving criminals the punishment that they deserve – even if we are sufficiently civilized not to seek pleasure from their suffering.

Michael Gazzaniga comes to a similar conclusion in looking at the issue from an evolutionary perspective. He suggests that humans have evolved to cooperate on a massive scale with unrelated others partly by punishing noncooperators. He leaves us to ponder: ‘If we don’t incapacitate the offenders, will the noncooperators take over and society fall apart?’

Postscript:
I was aware that I was being provocative in coming out in favour of retribution, when the point I was really trying to make is that justice should be primarily about fairness rather than deterrence. (As I acknowledge in response to a comment from kvd, however, there is an element of deterrence involved in fair punishment.) Retribution may mean giving people fair punishment but it can be mistaken for vengeance. Jim Belshaw had an interesting post yesterday about vengeance - which I largely agree with.

Comments from TBT (below) prompt me to acknowledge that I hope that evolution in public perceptions of fairness might enable able us to move further towards some kind of restorative system of justice.