Thursday, May 17, 2012

Truth of moral judgements and the limits of ethics

To: Matthew A.
From: Geoffrey Klempner
Subject: Truth of moral judgements and the limits of ethics
Date: 21 September 2007 12:04

Dear Matthew,

Thank you for your email of 10 September with your fourth and fifth essays for the Moral Philosophy program, in response to the questions:

'According to the ethics of dialogue how is it possible for moral judgements to be true? What does your answer to that question show about the nature of truth?'

'Identify one area in which the subject of ethics, or the resources of the moral philosopher, may be said to be 'limited'. What is the practical significance of that observation?'

Truth of moral judgements

This is an excellent essay which shows a good understanding of the issues surrounding the difficult question of realism vs anti-realism in ethics and in science.

You pose a challenge for me, when you state: 'To move from having justified moral beliefs to having true moral beliefs it is still necessary for such a moral realist to establish the explanatory power of moral values. Otherwise justified moral beliefs established by dialogue would exist without connection to the reality of the natural world.'

This is an aspect which I did not sufficiently consider when I addressed the question of realism vs anti-realism, and the nature of 'moral truth'. I said that the 'marks of truth' were convergence, stability over time and the possibility of explaining error (11/224). Is it also a mark of truth that a true proposition may be used as a premiss in a hypothetico-deductive explanation?

Michael Dummett, in his seminal paper, 'Truth' (reprinted in 'Truth and Other Enigmas' Duckworth 1978) emphasizes that a proposition may be regarded as capable of truth or falsity if it can be the antecedent of a conditional statement. In effect, this is all that hypothetico-deductive explanation amounts to. We say that if there is an entity (which we call a 'photon') travelling at such and such a speed, colliding with such and such particles, then a visible vapour trail will be produced in a cloud chamber. The observed vapour trail is taken as confirmation (not proof) of the theory, whereas failure to observe what was predicted is - in principle if not in scientific practice - a disproof.

Similarly, if it is true that a particular action was 'cruel', then we may draw conclusions from a hypothetical statement of the form, 'If action X was cruel then Y'. This is what we do. If cutting off the arm of the civilian suspect was cruel, then the person who did it is either a merciless sadist who deserves to be subjected to the full weight of the law, or a psychopath in urgent need of psychiatric treatment. We then investigate to see if that was indeed the motive for cutting off the suspect's arm. Perhaps the suspect was injured in a bomb blast and it was necessary to remove his arm surgically. In which case we cannot, after all, assert the antecedent of the conditional.

What this doesn't give us are explanations which have independently verifiable empirical consequences. If that was a requirement, then any attempt at attributing truth to moral statements would have to be given up once and for all.

The response would be that empirical, scientific truth is central to understanding our world. But it doesn't account for everything. There is more to the world than the world of science. Human beings, in effect, inhabit two worlds, the world of causes and effects and a 'human world', in which the appropriate form of explanation takes the form of reasons rather than causes. (John McDowell refers to this as the 'logical space of reasons' see 'Mind and World' Harvard 1994.) This claim is fully consistent with materialism. We are not resurrecting Descartes' 'ghost in the machine'. Rather, the point is about irreducibility.

Even within science, it could be argued, we recognize irreducibility: for example, the impossibility of translating an explanation in biology into the language of physics, or indeed the impossibility of explaining on an atomic level why you can't place a 'square peg in a round hole'.

Donald Davidson, in a seminal paper 'Actions, Reasons, and Causes', Journal of Philosophy 1963, argues that viewing actions has having reasons is consistent with also viewing them as having causes. The agent's reason IS the cause of the action. This explains how the 'logical space of reasons' and the 'logical space of causes' can map onto one and the same material world.

Limits of ethics

This essay ranges rather more widely than the title would suggest. However, the general theme relates to the third of the three 'limits' explored in unit 15: supererogation, politics and the idea of theory.

There is, as you advised, a certain amount of overlap with the previous essay.

The core idea of this essay involves the distinction between normative ethics and meta-ethics. I have argued that philosophy has a 'limited' role in normative ethics - against ethical theories which attempt to generate moral judgements, such as utilitarianism.

Although you do not state this in so many words, it seems reasonable to raise the question whether philosophy encounters 'limits' at the level of meta-ethics. No philosopher would claim to have a complete and adequate metaphysics, with so many metaphysical questions (realism vs anti-realism is just one) still being debated today. Perhaps there will never be a complete and final metaphysical theory. But, then, why insist on a complete and adequate meta-ethics?

In this program, we have taken on the relatively modest task of engaging the subjectivist 'error theorist' dialectically, showing how the refutation of solipsism and anti-solipsism leads to a third alternative -- the asymmetry of self and other -- which implies a 'logical basis for moral conduct': the claim that the recognition of the other is a necessary condition for the very idea of 'truth'.

The claim is controversial. Philosophers seeking to establish some form or other of 'objectivism' have adopted a variety of dialectical strategies against the subjectivist, and there is no general consensus that any of these strategies is wholly adequate. It could perhaps be said to be a 'limit' to meta-ethics that this question has not yet been resolved despite many centuries of debate, and perhaps never will be.

All the best,