Your complimentary articles

You’ve read one of your four complimentary articles for this month.

You can read four articles free per month. To have complete access to the thousands of philosophy articles on this site, please

SUBSCRIBE NOW

The Politics of Freedom

On Retributive Punishment

Oliver Waters asks, is retributive justice justified in a modern society?

“When I woke up, I went on what the movie advertisements refer to as a ‘Roaring Rampage of Revenge’. I roared, and I rampaged, and I got bloody satisfaction.”
The Bride, Kill Bill: Volume 1 (2003)

We’ve all experienced the undeniable pleasure of sweet vengeance – if not through a full-blown Tarantino-esque massacre, then perhaps while delivering a stinging comeback to an insult. Once our blood comes off the boil, we usually hasten to clarify that our actions weren’t merely enjoyable, but were in fact what justice demanded. After all, the wrongdoers got what they truly deserved.

This justification is enshrined in the criminal justice system as one of four legitimate rationales for punishing criminals, going by the respectable title of ‘retributive justice’. It has always niggled philosophers, though, that retribution fits somewhat uncomfortably with the other three moral justifications for punishment, since the others can each be cashed out in terms of bringing about positive future consequences, while retribution cannot. The first of these ‘forward looking’ justifications concerns the protection of others: a criminal can be incapacitated or imprisoned so that the community is kept safe. The second is that enacting punishment will deter that criminal, and others like them, from committing such crimes in the future. The third is that punishment will reform or rehabilitate the criminal, so that they do not seek to reoffend. In stark contrast, retributive justice is ‘backward-looking’, in that it seeks to inflict suffering upon the perpetrator simply because they deserve to suffer. Even if the universe were going to end tomorrow, retributive justice still demands that a criminal is punished today. It doesn’t care about any future beneficial consequences of doing so: justice simply demands that it be done.

At this point the family members of, say, a murder victim, might protest that the desire for retribution does have positive future consequences, namely their own future feelings of satisfaction at knowing the criminal got what they deserved. This is indeed a benefit of a kind. But we should interrogate this instinct for retributive justice, which likely evolved a crude but necessary mechanism for social control. During our hunter-gatherer origins, with none of our more sophisticated modern ways to contain or rehabilitate criminals, the only available method was either to kill criminals or maim them so badly they couldn’t pose a future threat to anyone, or perhaps to simply enact revenge in kind. This meant that a strong emotional thirst developed in us for retributive justice, in terms of ‘an eye for an eye’. However, like many of our ancient impulses, such as our drive to eat as much sugar as we can get, or for males to have sex as frequently as possible, we can recognise it as decidedly sub-optimal in a modern context. Just as we need to constrain the impact of our sugar cravings with sensible diets, we must also constrain our lust for retribution with a calm and rational approach to criminal justice.

Art © Cecilia Mou 2024 Instagram @moucecilia

Responsibility & Determinism

So far, so good: retribution is an urge we ought to restrain. But should we seek to eliminate it altogether?

Take the comparable dangers of cravings for food or sex. While it makes sense to temper these desires, our species literally could not survive without them. The philosopher Gregg Caruso thinks that when it comes to retribution, however, it’s a different story. In his 2021 book Rejecting Retributivism, he argues that rather than being a legitimate desire that can in some cases go too far, the retributive impulse is incoherent to its core, because it depends upon the concept of free will. We are supposedly morally responsible for our crime because we could have chosen otherwise. In Caruso’s view, however, free will does not exist, and so neither does any philosophical basis for retributive punishment..

As he puts it in a discussion with Daniel Dennett in Aeon online magazine:

“What we do and the way we are is ultimately the result of factors beyond our control, whether that be determinism, chance, or luck, and that because of this, agents are never morally responsible in the sense needed to justify certain kinds of desert-based judgments, attitudes, or treatments – such as resentment, indignation, moral anger, backward-looking blame, and retributive punishment.”

Dennett took issue with this sort of absolute denial of free will and its consequences for moral responsibility. He spent much of his academic career defending a compatibilist form of free will, partly because he saw it as necessary for any coherent view of morality. That is to say, Dennett fully accepts the notion of determinism, defined as ‘at any given time, the universe has only one physically possible future’. His key argument here however is that determinism does not imply fatalism, the idea that we cannot act to avoid bad things happening to us, or indeed, to avoid doing bad things. This is because (as the title of one of his books on the subject suggests) freedom evolves. Through natural selection we have developed many degrees of freedom of movement and thought. It follows, Dennett argues, that just because the future is determined by physics, it’s not the case that our bad decisions and actions are inevitable. Indeed, we avoid nasty things, and doing nasty things, all the time.

Caruso doesn’t buy this compatibilist defence of responsibility. He holds that if our actions really are predetermined by the laws of physics, then we cannot claim any responsibility for the fact that we are such splendid avoiders. After all, mindless robots can avoid obstacles, and it would be silly to attribute free will, and so, moral responsibility, to them.

This point seems perfectly valid to me, and it addresses a key problem with Dennett’s approach. It seems that if the future really is fixed by universal, timeless laws of nature, our consciously-felt choices cannot possibly make any difference to how that future unfolds. It’s rather that when confronted with this apparent contradiction between free will and determinism, Caruso throws out free will, while Dennett redefines it beyond recognition. But there’s another option: discarding determinism. Someone might argue that the idea that the future is fully fixed by the laws of physics is a mistaken extrapolation, made by applying physical principles to simplified thought experiments. Newton’s laws, for example, can help us predict what will occur given certain conditions, but they lose their predictive power once those conditions get more complex. The same is true for quantum theory and general relativity. They are profound, universal explanations of the nature of space, time, matter, and energy, but they cannot be used to predict what you’re going to have for dinner. Indeed, they contain no representations of you or me, our brains, thoughts and decisions, or even of much simpler biological systems.

The physicists David Deutsch and Chiara Marletto at Oxford have offered a powerful critique of the common framing of fundamental physics in terms of ‘initial conditions plus laws of motion’. Their alternative conception they call ‘Constructor Theory’. The right way to think about the laws of physics, they say, is as constraints on what kinds of physical transformations are possible; and this is perfectly compatible with evolved organisms creating explanatory knowledge and using this to transform the world as they desire without being trapped on a single fixed trajectory – that is, with them exercising free will. On this view, there is no slice of the universe made up of particles moving like billiard balls along pre-determined paths. How these particles actually end up moving through the world can depend upon ‘higher-level’ causes. Deutsch uses the example of a copper atom at the tip of the nose of the statue of Sir Winston Churchill outside the Houses of Parliament, pointing out that ideas about leadership, warfare, and sculpting must all feature in a causal account of how that atom got there. A purely physical story simply won’t do.

Certain philosophers, such as Jerry Coyne and Sam Harris, are worried about ditching determinism because they think the alternative is ‘mysticism’, which would allow for miracles and supernatural intervention. But such concerns are not warranted according to Constructor Theory, because we still live in a universe governed by timeless, fixed laws: it’s just that these laws do not dictate how exactly the future will unfold. However, the physical laws that make it possible for us to be conscious, creative human beings, making real choices, are still naturalistic laws that would rule out Jesus spontaneously converting water into wine or rising from the dead.

On this alternative way of thinking about fundamental physics, we don’t need to accept the notion that the universe evolves according to a predetermined plan, set from the beginning of time. Our best theories of physics don’t require it, and our best ethical, psychological, and political theories must reject it.

Free Will & Moral Knowledge

Freed from the shackles of determinism, then, we can start to make some sense out of what free will is. Every sane person rightfully rejects the idea that we are morally omniscient – that is, that we possess perfect moral knowledge. This kind of ‘ultimate’ responsibility for our moral behaviours is untenable given that we are finite, fallible creatures.

One implication of moral infallibility would be that whenever we act, we do so in the full knowledge that what we’re doing is right or wrong, good or bad. But a criminal rarely commits a crime fully knowing it’s morally wrong. Either they’re not thinking and their behaviour’s driven by animalistic urges, or they figure mistakenly that the gain outweighs the harm caused to others. Indeed, I find it incoherent to posit anybody arriving, via reason, at the belief that what they’re about to do is morally wrong, yet choosing to do it anyway.

Another way of putting this is that those who do wrong are either neurologically flawed or conceptually flawed. However, if it’s the former, then it is wrong to morally condemn them for their actions, since they had no conscious control over them. And if the latter, then they acted according to moral beliefs that were flawed or outright false. So what we mean when we say conceptually flawed criminals ‘could have chosen otherwise’, is that they could have chosen the correct course of action if they knew better. This class of criminal has the potential for moral reform, by learning why what they did was wrong, and why they ought to avoid doing similar wrongs in the future. They are however responsible for what they did in a way that a dog is not, because at the time of the crime they possessed the potential to understand the wrongness of their action. Of course, if you wound back the clock to the microsecond before they committed the crime, then they would have done exactly the same thing, but that’s irrelevant. What is relevant is that the criminal is the kind of being who can think about and explain to himself what he ought to do or not do. This makes him responsible for the decisions he makes. He therefore deserves praise for good decisions and blame for bad ones. Taking responsibility for your past action involves recognising that your current self is the self that committed the action, and that only through your deliberate self-reflection can that self be reformed.

Retribution & Repentance

Art © Susan Auletta 2024. Please visit instagram.com/sm _ auletta/

Given this refined concept of free will, does retributive justice still make sense?

Here’s one way to think about it. Suffering is a necessary part of any genuine process of moral reformation; but to apprehend, in detail, that what you did caused terrible, perhaps irreparable, harm, is to suffer a great deal (for non-psychopaths). One key part of punishing (usually by imprisoning) those we find guilty of crimes, is to create the conditions under which they’re likely to experience such shame or guilt. The ideal scenario is that, by the time they’re released, they feel genuine remorse for what they did, and have revised their moral beliefs and habits accordingly, making them far less likely to reoffend in the future.

The negative psychological experience that attends a criminal’s feeling of shame is a sufficient reason to punish that criminal. It doesn’t matter whether he goes on to massively benefit society later as a fully changed person. At the moment of painful self-reflection, there is no net increase in positive affect in the world; and yet, some genuine justice has been done. So the common desire to bring about this kind of suffering is the morally acceptable core of our crude drive for vengeance.

Take the exclamation ‘I hope he suffers for what he did!’ This can be interpreted in two very different ways:

a) ‘I hope he has pain inflicted upon him regardless of any change in his understanding of what he did’; or

b) ‘I hope he feels intense shame and guilt for what he did.’

The first desire is illegitimate, as it is merely a call for further suffering. But the second desire is legitimate. Yet it’s not utilitarian, because it doesn’t necessarily entail an increase in future pleasure or happiness.

Take the following thought experiment. You manage to capture Adolf Hitler before he commits suicide in his bunker in April 1945. Let’s say there are two possibilities open to you. The first is that you get to torture him to death, to get back at him for all the evils he has committed. The second is that you somehow explain to him, in a way that is actually persuasive, why what he has done is morally abhorrent. And imagine (as difficult as it may be) that upon hearing this, he feels an appropriate degree of shame and guilt for his wrongdoing. Which outcome would you find more satisfying: the physical torture imposed upon him from without, or the psychological torture imposed upon him by himself?

Of course, the latter possibility assumes that Hitler was not irredeemably psychopathic, which he may well have been. Supposing he was cognitively limited in this way: why would you feel any more justified torturing him than you would torturing a rabid dog? Neither has any capacity for moral reform. This is what makes it wrong to cause them pointless suffering.

This distinction came out in the debate between Dennett and Caruso when they discussed the idea of a magic pill that could instantly cure a criminal of any untoward inclinations. Caruso held that if such a pill existed, it would be wrong to still insist on punishing criminals, as from a consequentialist perspective, there would be no point. Dennett pushed back, arguing that punishment would still be necessary, even in such a brave new world. But what neither philosopher makes explicit is that the very fact that a criminal can be cured via biochemical intervention implies that their behaviour was neurologically determined. However, if a criminal’s behaviour is the result of bad or false beliefs, no pill can be effective for them, since no pill can selectively destroy some beliefs and implant others. So on this point Dennett is right – a world with a magical pill that cures neurologically-caused criminality would still contain criminals, and they would be precisely those criminals who have the potential to be morally reformed, if the conditions are made right for that. What are these conditions? Certainly not a dark, lonely cell, with no access to other people or information. But also certainly not the other extreme: a pleasurable matrix-like simulator which allows for endless distractions from critical self-examination.

Conclusion

A non-retributivist like Caruso can in principle accept a world in which a prisoner is morally rehabilitated without enduring any suffering. This is a world in which no one, including the prisoner, has a genuine choice as to who they become. But in the real world, a crucial part of coming to terms with what you did is acknowledging that you did in fact have a choice in the matter; it’s just that you lacked the moral knowledge or character to make the right choice. This reflection hopefully leads to the meta-knowledge that in the future you will face more choices like this, so you ought to seek better moral knowledge in order to prepare for them.

Prisoners should indeed suffer for the wrongs they have done, not only because locking them up keeps society safe, and not only because their suffering may deter future offenders, but also because it is better that they come to truly understand the error of their ways, even if they never step back into society afterwards. In bringing about the feelings of shame that attend moral insight, punishment can add moral knowledge to the world. And more knowledge is better than less.

Oliver Waters writes about philosophy and science, and is based in Melbourne.