To: Alistair L.
From: Geoffrey Klempner
Subject: Hume on reason and morality
Date: 25th October 2010 14:54

Dear Alistair,

Thank you for your email of 14 October, with your essay for the University of London BA Ethics Historical Perspectives module, in response to the question, ''According to Hume, reason alone does not move us to act, though our moral opinions do, and he rightly infers from this that reason cannot be the source of our moral opinions.' Discuss.'

At one point in your essay, you noted that the question moves from 'reason alone does not move us to act' to 'reason cannot be the source of our moral opinions'. This is really the key question here.

You say a lot about Hume's rejection of the view that reason an supply the ends of action. It would be helpful here to avail oneself of Kantian terminology, and talk of the contrast between a 'categorical imperative' and 'hypothetical imperatives'. According to Hume, all imperatives are hypothetical: that is to say, the decision to do X is a consequence of a practical syllogism, 'I want A. In order to obtain A I must do X. Therefore I must do X.' There is no class of actions which are such that, as Kant believed, one must do them, regardless of what one desires.

However, that is only half the story. In your essay you repeatedly refer to the notion of 'moral opinions' but you don't really give an explanation of how these arise in the first place. How is it that we even have a notion of the moral 'ought'? How could there even be such a thing as criticism, or moral criticism, of another agent's actions?

So far as the theory that all actions proceed from desires is concerned, nothing has been said about what makes an action, or a desire, something to be 'morally' approved or disapproved. What does 'moral' mean?

According to Hume, the subject matter of 'moral opinions' are not actions as such, taken in themselves, but rather those actions seen as proceeding from a particular motive of the agent which reveals an aspect of the agent's character, as virtuous or vicious. Moreover, Hume distinguishes two kinds of virtue, the natural virtues such as benevolence and kindness, and artificial virtues such as justice.

As another of my students writing on this topic remarked, for Hume there are in principle three persons involved when a moral judgement is made: the agent, the patient, and the observer who makes a moral judgement about the agent. This is of course consistent with, e.g. the observer and the agent being one and the same person, in a particular case.

Crucial to his account, therefore, is the idea of a certain kind of reflection on the things we do and our motives for doing those things. Various things that you say in your essay are relevant to this point; but you never actually state it clearly, in a way that shows that you understand the logical transition from what moves us to act to what moves us to form moral views of the actions of others.

Is he right? The way I have set up the distinction between categorical and hypothetical imperatives might make it seems as though the only alternative to a Humean view is Kant's theory of the categorical imperative. However, there has been much discussion of this in contemporary literature on moral theory, stimulated by Philippa Foot's seminal article, 'Morality as a System of Hypothetical Imperatives' (1972). Perhaps the most important response to Foot comes from John McDowell 'Are Moral Requirements Hypothetical Imperatives' (1978).

McDowell argues, against Foot, that there are kinds of factual judgement which are such that the very condition for the possibility of making these judgements presupposes certain attitudes or desires. For example, if I criticize you for treating that person cruelly, unless you are capable of the appropriate feelings, you will not which actions do, or do not count as 'cruel'. Moral properties are analogous to secondary properties such as colour, in that they presuppose a capacity for perception which we share with other members of the moral community.

It is interesting to speculate what Hume would say about this. McDowell is not defending the rationalist approach to ethics which Hume seeks to undermine, but rather looking for a third alternative which Hume has overlooked. For Hume, it is a matter of brute, empirical fact that human beings have the desires that they do, and in particular that they characteristically feel 'sympathy' for the situation of others. But it would be perfectly possible, say, for alien visitors to learn our language while remaining totally puzzled about the fuss we kick up about moral rights and wrongs. What we see, or rather what we feel, they don't see or feel.

If McDowell is right, these aliens would simply be unable to fully translate or understand our language; they would be lacking something that was a prerequisite for applying our moral notions (in their case dispassionately).

I'm sceptical about this, and for that reason am more inclined to seek a rationalist response, along the lines of the a priori necessity for recognition of the moral claims of others (as I do in my book 'Naive Metaphysics'.

All the best,

Geoffrey