source

Sturgeon, “Moral Explanations”

Harman: Many of our beliefs are justified by their providing or helping to provide
reasonable explanations of our observing what we do. Scientific beliefs are often
justified in this way. But moral beliefs cannot be justified in this way, because moral
facts appear unnecessary and indeed “irrelevant” to explaining our moral observations
and beliefs.
Sturgeon: Harman’s thesis is in fact broader than this in a couple of ways:
(i) Harman is not, in fact, just concerned with explaining our moral observations. If
we can show that moral facts are required to explain any of our moral beliefs,
whether or not these are observational, and any of our observations, whether
or not these are moral, this will be enough to support the reality of moral facts.
(Cf. mathematics)
(ii) When Harman says he is asking about whether reference to moral facts is needed
to explain moral beliefs, he isn’t wondering whether there is any way of
explaining our moral beliefs that doesn’t make reference to moral facts; if that
supported skepticism about moral facts, it would equally well support
skepticism about physical facts, and Harman is searching for an argument that
specifically targets moral facts. Harman is arguing, rather, that no explanation
of our observations/beliefs that appeals to moral facts is any better than some
competing explanation that does not. So:
Harman’s broader thesis:
Moral facts aren’t part of the best explanations of any non-moral facts we have reason to
believe in.
Sturgeon’s strategy:
To evaluate Harman’s thesis, we need to ask whether the assumption that there are moral
facts makes available to us any explanations of any non-moral facts that is better that the
explanations available when we don’t assume there are such facts.
(Although Sturgeon sets out the problem this way, he sometimes seems to be arguing for
a somewhat weaker claim, which he also seems to think is enough to undermine
Harman’s skeptical worry: that there are explanations of non-moral facts which we
reasonably appeal to.)
Sturgeon’s counterfactual test for explanatory relevance:
“[I]t is natural to think that if a particular assumption is completely irrelevant to the
explanation of a certain fact, then that fact would have obtained, and we could have
explained it just as well, even if the assumption had been false.” (p. 223)
2
To apply this counterfactual test, Sturgeon assumes, for the sake of argument, that there
are moral facts, and then asks whether they figure in any good explanations of any nonmoral
facts.
Some of Sturgeon’s examples:
(i) Hitler: If Hitler had not been morally depraved, he would not have ordered the
extermination of the Jews. And we would not have judged him to be morally
depraved.
(ii) Slavery: The growth of the anti-slavery movement in the US would have been
slower had slavery been and remained less bad an institution.
(iii)Harman’s cat: If the children had not been treating the cat wrongly, we would
not have judged that they were acting wrongly.
As Sturgeon points out, there are two ways of interpreting the counterfactual test.
According to the first, and, Sturgeon says, most natural interpretation:
(i) we should ask whether Hitler would have ordered the extermination of the Jews
had he not been depraved, and whether we would have judged him depraved
in this case;
(ii) we should ask whether the anti-slavery movement in the US would have grown as
quickly as it did had slavery not been morally wrong;
(iii)we should ask whether we would have judged the children’s acts wrong if they
had not been acting wrongly.
In each case, Sturgeon says, in order to imagine that the acts in question weren’t wrong,
we must imagine that the agents acted differently (because there can be no change in the
moral facts without a corresponding change in the non-moral facts upon which the moral
ones supervene):
(i) If Hitler had not been depraved, he would not have had the racist, paranoid, and
self-aggrandizing personality traits that led to his horrific orders, and he would
not have issued them, and we would probably not have judged him depraved.
(ii) If slavery hadn’t been as bad as it was, it would have involved less suffering for
the enslaved, and the anti-slavery movement would probably have grown less
quickly.
(iii)If the children had not acted wrongly, then they would not have been engaged in
pointless cruelty, and we would probably not have judged their acts wrong.
Sturgeon notes that whether our moral judgments would have changed if the moral facts
had been different is usually an empirical question – if I would have judged the
children’s act wrong no matter what they were doing, because I hate children, and think
any children having fun must be up to no good, then the moral fact (of the wrongness)
indeed will indeed play no role in explaining my judgment. But in general, this will not
be case – in general, our judgments do change when there’s a change in the moral facts.
3
But Sturgeon acknowledges that Harman may have wanted us to consider a different
counterfactual: would we still have formed the judgment about the children, or about
Hitler, and would the anti-slavery movement have formed just as it did, if all the nonmoral
facts were held constant – if the children, for example, had acted just as they did,
and if Hitler had had exactly the psychological profile he in fact had, and if slavery had
had caused exactly as much suffering as it did cause, but these acts and characters and
policies had not been wrong or bad?
In reaching the conclusion that Hitler would not have acted as he did had he not been
morally depraved, Sturgeon acknowledges that he was relying on a (not very
controversial) moral background assumption – that no one would order the extermination
of millions of people, and initiate a world war, unless he was morally depraved.
To apply the counterfactual test in this new way is to allow that such background
assumptions may be systematically and wildly mistaken.
Sturgeon grants that if our moral background assumptions were systematically and wildly
mistaken, then we might still have believed that Hitler was depraved or that the children
acted badly, even if their actions were just as they were but weren’t actually wrong. But,
he points out, the same applies to the physicist observing the proton. If her background
physical theories were wildly and systematically mistaken, and a vapor trail in a cloud
chamber did not in fact indicate the presence of a proton, but she nonetheless thought it
did, seeing the vapor trail would have led her to “observe” a proton going through the
chamber even though no proton was present.
Sturgeon’s conclusion:
On the first reading of the counterfactual test, moral facts pass the test; on the second
reading, moral facts fail the test, but so do physical facts. So Harman’s argument does
not raise a special problem for ethics.
Other worries underlying Harman’s argument:
(1) Moral facts play no role in a good explanation of non-moral facts in general, and
of our moral judgments in particular, because we can give no good account of
how it is that moral facts could lead us to form moral judgments (at least not in
the absence of a reductive account of moral facts).
(2) In every case where a moral explanation of a non-moral fact seems plausible,
there’s a perfectly good non-moral explanation of the non-moral fact in question
that appeals instead to the natural facts on which the moral facts supervene.
Moral facts, if there are such facts, are epiphenomenal. So moral explanations
are never necessary.