Monday, December 31, 2012

Kant's categorical imperative as a test for maxims

To: Mark S.
From: Geoffrey Klempner
Subject: Kant's categorical imperative as a test for maxims
Date: 27th January 2009 12:44

Dear Mark,

Thank you for your email of 17 January, with your essay for the University of London paper, Ethics: Historical Perspectives, in response to the question, 'Does Kant's categorical imperative succeed as a test of the acceptability of practical maxims?'

It is important to keep in mind that the question is about Kant's categorical imperative as a test of maxims, not the question of 'categorical' vs 'hypothetical' imperatives generally (Foot vs McDowell). McDowell rejects the view that moral imperatives are hypothetical but this is not in any way intended as a defence of Kant's use of the categorical imperative as a means of sorting necessary/ acceptable/ unacceptable actions.

You say, 'Although Kant argues carefully for the logic of his CI formulations, they are still only based on his intuition. The CI is based on Kant's intuition that there exists something absolutely and necessarily valuable, not because it is desired by humans, but because it is an end-in-itself and so objectively valuable. Since it is present in all rational beings it must be universalised.'

This is something Kant believed, but he would reject this gloss on the rationale of the original formulation. His argument is that if there is ANY constraint on action which is not hypothetical, then the ONLY law can be the one he states. In the light of McDowell's theory this claim might well be questioned.

In response to the charge that he is relying on 'intuition', Kant might concede that it may well be 'intuition' which guides the application of this law, especially when reformulated in terms of increasingly teleological notions. However, this is just an indication of the depth of the categorical imperative -- that human beings strain to grasp its essential meaning, and human reason being fallible we don't always come to the correct conclusions.

You might reply that this is a very convenient let-out. Whenever someone comes up with an apparent counterexample, Kant replies that they haven't thought it through. In fact, in your essay you give some good examples of how carelessness in formulating a maxim leads to the wrong result.

For example, you give the correct response to the objection, 'What if everyone wanted to be a farmer?' However, the reformulation, 'For personal satisfaction, I shall become a farmer, if there is a need for more farmers,' still leaves the question unresolved. What if there is a need for only a few more farmers? Unfortunately, farming has become a very popular occupation and everyone wants to be one. Pursuing my satisfaction would be OK in other circumstances but not here. So we need something like, 'For personal satisfaction I shall pursue my exceptional talent for farming, if there is a need for a few more farmers.'

Of course, my belief in my exceptional talent can be factually false, but that is irrelevant so far as assessing the application of the categorical imperative is concerned.

This suggests a powerful defence of the categorical imperative against apparent counterexamples: where the principle does not give a clear result, and where the initial attempt to formulate our maxim 'precisely and honestly' doesn't solve the problem, we use the categorical imperative together with our knowledge of the relevant facts (e.g. the demand and supply of farmers) to formulate a maxim which is consistent with it.

The problem now is where the categorical imperative apparently gives a 'clear result' but it is the wrong result. How do we know? E.g. the example of the Nazis. This is harder than you make it look, as was demonstrated (notoriously) by R.M. Hare when in defence of his preference utilitarianism (based, as he argues, on the only logically acceptable formulation of the criterion of universalizability -- the principle of 'non-fanaticism') Hare claims that in a society of Nazis sufficiently 'heroic' in their hatred of Jews, the former might under certain circumstances be morally justified in exterminating the latter ('Ethical Theory and Utilitarianism', in 'Contemporary British Philosophy' H.D Lewis Ed. Unwin 1976, cf. pp. 121-2).

This leads into a major question of whether, or to what extent, Kant's categorical imperative can be formulated in terms of a criterion of 'universalizability': the problem is that in pursuing my moral principles, I am ignoring the fact that other people have different goals and ideals (this seems to bear on the point that Millgram is making.) To insist that moral questions should be judged by my lights is 'fanaticism'. I want to do X, and I believe everyone should want to do X. Moreover, a society were everyone wanted to do X would get along just fine. The problem is, in the actual world, a lot of people have the same view about Y which is inconsistent with X.

It seems clear that Kant would never accept Hare's consequentialist reformulation. The only solution in the case of the Nazis seems to me to reject the idea that desires are a 'given' (I've made this point before). Love and trust, for example, are things that can be 'commanded' in Kant's sense. 'I can't help how I feel' is not a defence. We are morally responsible for the way we feel. In other words, not only actions but feelings should be put to the test of the categorical imperative. This would rule out many practical maxims from even being tested, because they refer to feelings which are themselves unacceptable according to the categorical imperative.

Stated baldly, there are all sorts of difficulties and dangers with that view. I am playing devil's advocate to some extent (although I agree with McDowell -- for different reasons -- that moral judgements are not hypothetical imperatives). But I will leave the discussion at that point.

All the best,


P.S. I missed not having a bibliography with this essay. In an exam, it is not inconceivable that you might refer to an article which the examiner is not familiar with. The assertion, 'As Black says...', will have more authority if you can cite a reference ('As Black says (Phil Rev 07)...'), so it's worth memorising these. Whereas if you are referring to someone like Foot or McDowell, a reference is less important.