Bentham gets pwned by Kant – or is it the other way around?
posted by Donald Braman
Another post on punishment and what we (don’t) know about it.
A few days ago, I suggested that there are three ways of thinking about the common practice of expensive punishment. If we assume a neoclassical deterrence perspective, there is a puzzle: Neoclassical deterrence theory depends on rationality – the cost of punishment is, after all, what stops criminals from garnering the benefits of crime. But if that is so, then why would individuals impose punishments that are costly to themselves?
One answer that neoclassicists abhor is irrationality – humans, if they have any aims at all, have aims that are not rational. A second answer modifies neoclassicism by suggesting that humans try to behave like neoclassicists suggest, but fail because they are, if not stupid, a little dumb – call this the “bounded rationality” option.
A third option further modifies – indeed you might say subsumes – neoclassicism by proposing that what people maximize includes intangible things like congenial or satisfying meanings. This perspective is both common-sensical, in tune with what little empirical evidence we have regarding punishment-related cognition and behavior, and fairly brutal to most of the theoretical work that has been done to date.
How brutal?
The implications for deterrence are pretty stark: the assumption that punishment has a fixed and measurable function for the recipient, the giver, or society at large depends on the meaning of punishment, and the meaning can vary. There is, on this third approach, no any way to calculate efficient solutions to complex collective action problems without taking social meaning into account. This, in turn, makes economic analysis very, very hard. Most of neoclassical economics, after all, depends on assuming away private and social meaning.
Now one might imagine that retributivists would be inclined to rejoice at this. Humans, it turns out, care about more than costs and benefits – they care about things like justice! The problem is that once you go down the social meaning path, classical (Kantian) conceptions of justice seem more than a little naïve. A punishment that might appear just to those in one social group can appear utterly unjust to those from another social group. There simply is no way, outside of inquiry into social meaning, that one can discern what is just and what is not.
Moreover, the social meaning turn blurs the line between utility and just deserts. If one is just importing a conception of justice into a more complicated utility function, then retribution is really just a part of utility, and retributivism just a part of utilitarianism. As one can imagine, retributivists understandably don’t like the notion of Kant being pwned by Bentham (even – or perhaps especially – if Bentham has turned into an anthropologist).
So where are we? What do we know? We know that people care about social meaning, and that most of criminal legal theory doesn’t take this into account. I’m certainly not the first to point this out, but I do think that the point hasn’t been pressed nearly far enough in revamping theory or practice. More on that in the future.
February 8, 2007 at 11:41 am
Posted in: Criminal Law
Print This Post
Responses (5)
M&M - February 8, 2007 at 12:17 pm
“Most of criminal legal theory doesn’t take this into account”
???
Have you read Joel Feinberg? Jean Hampton? Both are canonical – and have theories centered on “expressive dimension”. I don’t understand where you are coming up with your statements. The difference between you and them is that they are not interested in maximizing anything, whereas you seem to assume that that’s the only way to think about any of this. But that assumption hasn’t been defended.
Donald Braman - February 8, 2007 at 12:31 pm
I am pretty much an expressivist at heart, so yes I have read Hampton and Feinberg, and I pretty much agree with them. That will be the core of my next post along these lines.
On the other point – about maximizing – I’ll both agree and disagree. If you think in terms of expressivism or social meaning, it doesn’t matter whether you talk about maximizing or not. There is still some objective that people seek in the law. Economists would endogenize that objective into a utility function. But once you’ve done that, many of the benefits of economics – that you can just assume away variance of satisfactions – become problematic.
Here’s a question: do you think that Jean Hampton and Joel Feinberg’s approach is incompatible with the notion that people are maximizers who include expressive satisfaction in their utility functions? Or is it that you think (and here I would agree) that it is just an odd way to talk about what people do?
Matt - February 8, 2007 at 1:51 pm
I’m a bit unclear about what you say about Kantian conceptions of justice. Do you just mean that much of what Kant himself says about punishment seems wrong to us now? That’s certainly true, but of course Kant, as I’m sure he’d agree, isn’t the final word on what actually follows from his view, even if the view itself is right.
Paul Gowder - February 9, 2007 at 12:55 am
“There simply is no way, outside of inquiry into social meaning, that one can discern what is just and what is not.” Uh, that premise isn’t established. There are universalistic conceptions of justice… and a Kantian can say that people really want to express those. And the fact (if true) that people want to express those doesn’t mean that Bentham pwns (as it were) Kant: it’s perfectly possible for a deontological moral truth to *also* provide utility for some actor. (Indeed, on some persuasive readings of Kant’s ethical theory, it’s necessary that actors be in fact motivated by the moral law, i.e. that it brings them pleasure in some sense [arguably, insert disclaimers here] from complying with it.)
M&M - February 9, 2007 at 9:23 am
Your post and comment (response to my questions) move very quickly through a number of complex issues, and it won’t make sense to try to untangle all of the various problems with your statements.
But let’s start with an easy one – my first question: where do you get your statements like:
“fairly brutal to most of the theoretical work that has been done to date.”
“Most of criminal legal theory doesn’t take this into account”
I just don’t understand what you are talking about when you say things like this.
Leave a Reply