Some time ago, I don't know how long, I wrote in this blog about the book MORAL SKEPTICISMS by Walter Sinnott-Armstrong. After a long time away from that book I returned to it recently. Here is a summary.
Although the title, and the programmatic parts of the book, sound as if the plan was the emphasize the diversity of forms that moral skepticism might take, most of the book is devoted to a single argument, which may simply be resisted by moral cognitivists for different reasons. So the real debate is between moral skepticism singular and moral cognitivisms plural.
Before I get to the argument, one critical semantic point. There is a distinction, among meta-ethical philosophers, between a normative proposition and a moral proposition. Some thinkers use "normative" as a broader term than "moral" and "normative but not moral" is a sort of intermediary thing between "is" statements and "ought" statements. So if we can't get from nature to morality directly, we may be able to do so indirectly by passing through amoral normativity. So runs a theory.
I'll leave aside how that is supposed to work for now. Here is an abbreviation of the key argument in the Sinnott-Armstrong book as one finds it at pages 74 to 77.
1) If any person is ever justified in believing any moral claim, it must be by inference.
2) An inference must have either (a) no normative premises, or (b) some normative but no moral premises, or (c) some moral premises.
3) No person is ever justified in believing any moral claim by an inference with no normative premises.
4) No person is ever justified in believing any moral claim by an inference with some normative premises but no moral premises.
5) No person is ever justified in believing an inference to a moral claim from moral premises without a justification for believing that premise, itself a moral claim.
6) Given both (1) and (5), any justification would have to consist in a chain of inferences with moral premises that would be either finite and circular, or infinite.
7) No person is ever justified in believing anything on the basis of a circular argument.
8) No person is ever justified in believing anything on the basis of an infinite chain of premises.
9) Thus, no person is ever justified in believing any moral claim. QED.
The argument is formally tight. If justification requires one of a limited number of possibilities, and you have good reasons for eliminating each of those possibilities, then you have good reasons for believing that justification is not available. If you disagree with (9), you must disagree with something else along the way. And which one you baulk at tells us what sort of cognitivist you are.
An intuitionist believes in non-inferential claims which serve as premises, and so rejects (1).
A moral naturalist will deny premise (3) -- perhaps the fact that certain things give humans pleasure is both a fact of nature and an anchor for moral reasoning. A normativist -- as I mentioned above -- will deny (4). A moral contextualist will deny (5). A moral coherentist will deny (7). And there are even -- odd as it sounds -- infinitists who deny (8).
Accordingly, as Sinnott-Armstrong writes, even if we reject moral skepticism, its central argument serves the valuable purpose of classifying the possible non-skeptical views.
My regular readers will by now surely recognize that I fit happily into the cognitivist-because-intuitionist camp.
Comments
Post a Comment