We have two chapters: one on “the amoralist” and the other on “subjectivism”.
What unites them is that both figures are thought to pose challenges to morality.
The amoralist rejects it altogether. Here the challenge is simple: can we show the amoralist is wrong?
The subjectivist does not think there is any way of comparing different moral views. There are two challenges here:
Most of us think that our moral beliefs are not merely subjective. We cannot, for instance, pick and choose which moral rules apply to us and which ones do not. By analogy, I cannot choose whether to believe that 2 + 2 = 4 or that I’m sitting in California. I have to acknowledge that is the way things are no matter what I think. For many of us, morality seems to be the same. Lying is not OK for you even if you think it is.
We take the project of comparing different moral views seriously. Say you believe in moral progress or you think that you have to reconsider your own opinions when someone you respect disagrees with you. It is hard to make sense of that while regarding morality as a merely subjective matter.
I think that Williams has two different aims for this chapter.
He wants to address the question of whether it is important to be able to show that the amoralist is irrational, inconsistent, or mistaken. Williams says no.
He wants to use it to define our topic: morality. By looking at what is involved in rejecting morality, we can gain some insight into what is involved in accepting it.
What is moral territory? That is, what thoughts must the amoralist avoid in order to remain an amoralist?
First, thinking in terms of what is permitted and forbidden; resenting or disapproving of the ways others behave (Williams 1972, 5).
Second, thinking about oneself as being an especially excellent or good person.
Williams thinks that rests ultimately on thoughts about the general interests and needs of other people and those thoughts are in moral territory (Williams 1972, 6–7).
Williams also thinks that the formulation “I’m great because I do the things that everyone else would do if they could” is false. “If there is such a thing as what men are really like, it may be that … it is not so different from what they are actually like; that is, creatures in whose lives moral considerations play an important, formative, but often insecure role” (Williams 1972, 9).
Third, caring about others, doing things for others, because they need them (Williams 1972, 11). The contrast is with a psychopath, who cares about no one.
The amoralist has the idea if, say, he cares about his mother. He will take care of her, for instance, because “she’s sick” or “she needs me.”
The amoralist is not separated from a moral person by a huge psychological gulf. Rather, the amoralist is capricious in how he thinks of others, giving vastly more weight to the people he cares about than to others.
There is a kind of inconsistency in the amoralist: “my mother needs care” does not apply to anyone else who needs care. But, Williams thinks, bringing the amoralist fully into the moral realm requires an extension of his imagination and sympathies rather than an argument. The idea is that he has to care about the interests of others to whom he has no personal connection and to have a sense of fairness. So that is what we learn about morality: it involves sympathy for others and a sense of fairness (Williams 1972, 12).
The conclusion of the chapter on the amoralist is that morality involves feelings, especially sympathy. That naturally leads to the suggestion that it is subjective, since it depends on your having the appropriate kinds of feelings.
Hence we get this question: what might it mean to think that moral opinions, judgments, or outlooks are “merely subjective?” Williams gives us three possibilities.
linguistic: “A man’s moral judgements merely state (or express) his own attitudes.”
epistemological: “Moral judgements can’t be proved, established, shown to be true as scientific statements can; they are matters of individual opinion.”1
metaphysical: “There are no moral facts; there are only the sorts of facts that science or common observation can discover, and the values that men place on those facts.”2
One of Williams’s goals in introducing these ideas is to identify a project that he associates with subjectivist moral philosophies: “defusing” subjectivism by showing that it is “not alarming” but rather “essential to the nature of morality” (Williams 1972, 15).
Here is how the rest of the chapter goes.
The linguistic statement is either false or harmless (Williams 1972, 15–16).
If it means that moral judgments are only expressions of attitudes or feelings, then it is false.
If it were true, there would be no disagreements about values; it would be me saying my stomach hurts and you saying yours does not. Since there are disagreements about values, it is false.
If it means that moral judgements do express attitudes or feelings, then it is harmless. Factual judgments do that too: they express the attitude of belief.
The subjectivist’s reply is that we classify moral judgments as right or wrong because we are trying to get others to agree with us (or because we are trying to conform with social opinion, he might add) (Williams 1972, 17).
The idea is that it is not the same as evaluating factual judgments; it is more about trying to get a group of people to agree than it is about figuring out the truth. What we think of as moral reasoning and argument is mostly rationalization, not what it appears to be.
Williams thinks this gets the phenomena wrong. Even if moral argument is mostly (or even completely) rationalization, the analysis of moral language has to leave room for rationalizations, something that feels like reasoning. That would not make sense on the linguistic interpretation of subjectivism.
The subjectivist is then described as saying that moral reasoning happens only within frameworks of assumptions. People can disagree and argue only because they share those assumptions. But there is nothing holding those assumptions up other than the social fact that the members of a society share them. There is no way of showing that they are right or wrong (Williams 1972, 18).
This gets us to the epistemological formulation of subjectivism, that there is no way of showing that moral statements are right or wrong. Only we are swapping out “individual opinion” and putting “social opinion” in its place.
It is a natural thing to do because there are limits to how much moral disagreement there can be within a society, as societies fall apart if there is too much disagreement, but there are no limits to how much disagreement there can be across societies. This introduces the next chapter on relativism (Williams 1972, 19).