Imagine you’re discussing something with someone and then, just after you make a point that they think is foolish, they slap you in the face and tell you to stop being silly. Perhaps they jokingly wag their finger at you. Then they continue the discussion. What do you do?
I can’t speak for everyone, but I think one common reaction would be an outrage out of all proportion to the actual pain suffered, and based instead on a feeling of humiliation and disrespect. The desire this might produce could be expressed principally as ‘desire that things not go on as normal’. Maybe you slap them back, or shout at them, or leave immediately. In each case, the goal is (at least in part) to ‘mark’ the unacceptability of that action – or to put it another way, to ‘refute’ the ‘message’ that the action expressed, namely that it’s ok for this person to do that.
That is, the action taken in response would be not aimed at producing any effect, not at causal power, but would be communicative. Its rationale would be as part of an ongoing ‘discourse’ about how to act. If somebody watching had assumed that actions all ‘aim at ends’ in the sense of some result they produce, or was in another way committed to looking at actions as actions, and not as assertions, then they might well be confused, and find it hard to make sense of your response. What did it acheive?
This, I think, is what is often going on in discussions about ‘justifications for punishment’. There are broadly two sorts of theories about why we punish/why we should punish/whatever – I want to consider the debates without implying endorsement of any of their assumptions, let alone their real applications and history. Some say ‘it’s useful’ – whether by ‘deterrence’, ‘incapacitation’, or ‘rehabilitation’, it aims at some sort of good. The other says ‘it’s deserved’, or ‘it’s proportionate’ or ‘it’s justice’.
And one of the notable things about this latter is how hard it’s often been to explain. The former, ‘acheive some good’ approach makes a fairly obvious sense: you do something so that fewer bad things will happen. Everybody wants fewer bad things to happen. That’s pretty much definitional. There’s no problem combining this with some broader ethical theory.
But why does hurting one person make any sense, apart from this? After all, now a bad thing is happening, that could easily have been avoided: isn’t that the opposite of good? That is, there’s a certain theoretical standpoint from which retribution becomes entirely opaque. And not just a theoretical standpoint: this opacity appears for many people as natural and intuitive. It feels obvious that two wrongs don’t make a right.
And often the ways used to express retributive intuitions just reinforce this bafflement. One locution that appears in some philosphers is that ‘the punishment erases the crime’, or ‘nullifies the guilt’. But this sounds crazy. Killing a murderer doesn’t bring their victim back to life.
But then – contemplating a serious crime (whether illegal or not) can usually produce some retributive impulses, even in the most scrupulous consequentialist. To avoid that would basically mean never feeling anger – but to lump everything under the heading ‘anger = emotion = irrelevant’ is to distort, not clarify, the issue.
So my suspicion is that what’s happening here is a conflict between seeing acts of punishment as actions, with the standards of justification appropriate to actions (such as producing a desirable result), and seeing them as statements.
The latter idea could be put like this: why do we find something more abhorrent about murder than death from illness? Why do actions produced by deliberate human action have an emotional charge absent from things that just happen? It’s because they add more than the tragic result: they make a statement is that ongoing ‘discourse’ I referred to earlier – the murderer says ‘murdering this person is reasonable’.
And while we cannot undo the action’s effects, we can (and must!) deny the statement. We must deny it for if we do not, it will stand unchallenged, and our assent will be implicit, and we must deny it in actions because words are cheap and easy, and cannot convey our force of condemnation.
That’s not necessarily a position confined to any particular political tendency. It includes both a policeman tasing someone to ‘assert’ (that they have) authority, or a country making war to avenge a ‘slight’ to its ‘dignity’, and someone refusing to submit to oppression, despite the personal costs that defiance imposes, because they feel it utterly unacceptable to be silent in the face of injustice. It’s a dimension that most political tendencies participate in to some degree.
This is supported by a few observations about common (though not universal) intuitions on which crimes are more horrendous. For some people, a gay person’s murder by homophobes provokes less outrage than would a straight person’s merely larcenous murder, because the ‘statement made’, that gays should die (under certain conditions), has a certain naturalness and reasonableness about it (they probably did something to provoke it…), while for other people, it provokes a stronger outrage, because the statement made is ‘more powerful’ from the history of precisely such attitudes – it draws on the societal endorsement of this and related attitudes, and because it is a ‘stronger’ statement, it is even more important to deny it.
That is, identities and concepts like ‘gay’ and ‘gay-bashing’ affect the emotional significance that people perceive in the action. It would be absurd to say that they do this because they make the result itself more or less significant – one person’s life is equal to one other person’s. But the statement is different.
So what if this is all correct, then? What if ‘retributive’ punishment can only be understood as communicative?
One implication is that retribution makes certain assumptions.
It assumes that we are engaged, merely by acting, in a sort of ‘discourse’ or ‘conversation’ with other people about ‘how to act’ – it assumes that this not a mere figure of speech or fanciful feeling, but something significant and worth taking seriously.
But moreover, it assumes that such a conversation, with millions of participants, can be made consistent overall. What I mean is – that we can take a million people’s different estimations of what constitutes a ‘crime’, and their different interpretations of what a single act means, and their different estimations of what sort of ‘denial in action’ is necessary, if any, and make out of these a single system that can be applied generally.
Thirdly, it assumes that a certain way of acting in this ‘conversation’ – that of denying the statements made by people doing bad things, through doing bad things back to them, is reasonable and should be continued – it assumes, for instance, that ‘forgiveness’ as a general rule is not suitable.
Are these three assumptions true? I’m not sure.
I think the first is probably true, and while the second is certainly false, the need to try it nevertheless could certainly be argued – after all, it seems to be involved alike in the very question of whether so many people can co-exist in a single society, with shared norms and mores. But the difficulty may still have important implications.
And what of the third? That’s obviously the big question. I won’t try to answer it, but hope that just clarifying the question is useful. Coming posts in the future will look more at this.