Hume's Moral Subjectivism

Class notes for 8 March

Main points

We took up three questions about Hume’s moral philosophy

  1. Does Hume’s theory entail that people only have to do their duty or be virtuous if they want to? And isn’t he committed to saying that there is no way of evaluating passions as right or wrong?
  2. Is Hume’s theory of vice and virtue an objective one? Virtues and vices are character traits that typically produce good or bad results. The typical results produced by character traits are a matter of objective fact that is discovered by probable reasoning.
  3. How does Hume handle cultural differences? Isn’t he committed to an implausible empirical claim about the uniformity of human nature, namely, that human beings always regard the same things as virtues and vices?

I tried to give answers on Hume’s behalf while, at the same time, pointing out how they might be challenged.

The class ended on the theme I announced at the beginning: Hume’s attempt to undermine the view that we are made in the image of God, that reason enables us to understand the world and how to behave like God understands these things, and that our lives have value insofar as we are like God.

Question One

I keep repeating that the problem for Hume isn’t that it’s too hard to evaluate other people but that it’s too easy. I don’t think I have done a very good job of explaining what I mean, though. Here’s my last attempt.

Evaluating people, thinking that they are good or bad, is straightforward on Hume’s view. The evaluator has a feeling. That’s all that is involved in thinking that someone is good or bad. If you genuinely, sincerely have one of these feelings about someone, that’s all the justification you need for your evaluation. And it’s all the justification that is available. No one has any better justification for thinking that someone is good or bad.

What’s unsettling about this is that the people you are evaluating seem to have as much reason to think what they do as you have to think what you do. They feel that what they are doing is right, you feel that what they are doing is wrong, and there is no way of settling the issue that is independent of either of your perspectives.

The last bit is crucial since, of course, each of you will be as justified as anyone can be in thinking what you do. The absence of an independent way of settling your disagreement does not show that either of you lack justification for thinking what you do. So you are perfectly right to say that the other person is wrong: you can evaluate that person to your heart’s content.

Would our confidence in our moral beliefs collapse if we accepted this characterization of them? Would it no longer make sense to care about those beliefs as much as we do if we realized that we couldn’t prove that those who disagree with us are wrong? We discussed a few examples that, in my opinion, suggest that this is not what most of us think: Katie’s question about the possibility of discovering a hideous moral fact, my experience of teaching human rights without proofs of their existence, and the simple observation that we love people for what are admittedly subjective reasons.

Question Two

There is a sense in which virtue and vice are a matter of objective fact on Hume’s view. But there is a more important sense in which they are not: virtues have to be agreeable or useful to us, they have to serve our interests.

As Hume plausibly points out in “A Dialogue”, we would never have begun drawing moral distinctions, much less would we have established social institutions for their enforcement, if they did not serve our interests.

Question Three

Hume may be committed to an implausible thesis about the uniformity of human nature, but he is very clever at defending it. He maintains that there is, in fact, a single list of virtues that are praised in all human cultures.

The differences among cultures that we observe are due to the following factors:

  1. Ignorance plus eloquence: you can make any culture look ridiculous if you give a superficial, ethically loaded description of its practices.
  2. The looseness of our values: our values do not determine our actions. The love of one’s children is compatible with a wide range of actions towards them.
  3. Historical circumstance: different qualities are valuable in different circumstances.
  4. Cultural circumstances: different qualities are valuable in different societies, under different social rules.
  5. Conventional differences: justice is valuable in every society, but different societies could have different conventions assigning property rights or governing promises than others.
  6. Religion: the influence of Christianity is an important reason why ancient and modern values are different. Those under its influence, in Hume’s opinion, struggle against human nature and its virtues.

Is that good enough? Bernard Williams, who shared Hume’s tastes and a fair amount of his theory, thought it was not. The problem with Hume’s theory, he held, is that it is insufficiently modern: it underestimates the extent of disagreement and overstates the uniformity of human nature.

Is Williams right? I leave that for another day.


The biographical information about Pascal and Diogenes I read came from Pierre Bayle’s Historical and Critical Dictionary. These are on reserve for this course. I cannot say enough good things about Bayle: I don’t know why he isn’t on the list of the greats. Hume, like most educated people in his time, had a copy of Bayle’s Dictionary on his shelf.

I’m told that the translation by Des Maizeaux, published 1735-38, is excellent. Plus the typography they used is amazing. Anyway, you can find a copy in the library, reprinted by Routledge. Or you can download and print facsimiles from Eighteenth Century Collections Online (University of Chicago connection required).

The snippet about duels is from a wonderful book: Russell Hardin’s One for All: The Logic of Group Conflict. Princeton: Princeton University Press, 1995. This offers a very persuasive explanation of group identification in terms of the dynamics produced by largely self-interested behavior.

The remark about Galileo is based on Dava Sobel’s book Galileo’s Daughter. I had the pleasure of reading Galileo for the first time this summer: he was an amazing analytical writer. Sobel’s book gives you a feel for life at the time that is especially relevant to our point about how it’s possible to love your children while locking them up in “jails.”

The specific reference to Williams is from “Reply to Simon Blackburn,” Philosophical Books, 27 (1986), p. 206. There are quite a number of points on which Williams found Hume’s moral philosophy congenial. His book Morality: an introduction to ethics includes a sympathetic discussion of moral subjectivism (pp. 3-37) and expresses a distinct preference for Hume’s views of moral motivation over Kant’s (pp. 63-72). I cannot speak highly enough about that book, but I frequently try. There’s nothing “merely introductory” about it.

For more on Williams’s views about the relationship between morality and the emotions (a reasonable synonym for Hume’s “passions”), see “Ethical Consistency” and “Morality and the emotions,” both in Williams’s Problems of the Self. Hume’s favorable comparison of classical ethical thought with modern, Christian-influenced morality is mirrored in Williams’s Shame and Necessity. And, finally, Williams has some interesting and, perhaps, too charitable remarks about Hume’s account of the artificial virtues in Truth and Truthfulness.