Saturday, March 26, 2005

Facts, Values, Apples

It's commonly thought that there is a fundamental divide between fact and value, 'is' and 'ought', such that no amount of knowledge about the former could entail any conclusions involving the latter. But I wonder if the same could be said of facts and apples, 'is' and 'eat'. No matter how many non-apple facts you list, you're not going to end up with a ripe juicy apple. You can't even make any claims about apples. Does it thereby follow that apples are non-natural and fundamentally separate from the rest of reality?

I'm probably missing something. But let's look at the naturalistic fallacy. The 'naturalistic fallacy' is the fallacy of disagreeing with G.E. Moore about whether some property can be given a reductive definition. One proves that an opponent is guilty of this fallacy by professing one's own ignorance; this is known as the 'Open Question argument'. For example, if a chemist proposes that salt can be reduced to sodium chloride, one can expose their fallacy by asking: "I know this shaker contains salt, but does it contain sodium chloride?" The fact that this is an 'open question', as you do not know the answer, demonstrates that salt and sodium chloride cannot be identical after all. For if they were, it would be equivalent to asking "I know this shaker contains salt, but does it contain salt?" and even Cambridge philosophers know the answer to that one!

[Retraction: this post is badly confused.  See here and especially here.]


  1. Richard, Yes, you're right. Some ink has been spilt on that. The non-reductive naturalists (Railton being one as I recall) would say to Moore that there may by an identity between "good" and some natural property even though the identity isn't by definition (analytical). Similar non-reductive naturalistic tricks are done in philosophy of mind (though Jaegwon Kim has devoted his career to refuting them.)

    Just my two cents: I agree with that reply to Moore, but I think that it veers into your shmethics. If someone proposes an identity of goodness and some natural property that is not by definition, then we can imagine a possible world in which the identity doesn't hold and, say, killing kids for fun is "good" in that world. That's not the ethics we know so well. So, I think only reductive (definitional) naturalism and not the non-reductive sort will work. Most importantly Moore's open question argument is no argument, really, but rather a test that any proposed naturalistic definition of "good" or "right" must pass. (I told you my definition, which I believe passes the test.)

    Stephen Ball's articles on this business are superb explications of the issues:

    "Linguistic Intuitions and Varieties of Ethical Naturalism." Philosophy and Phenomenological Research, 1991.

    "Reductionism in Ethics and Science: A Contemporary Look at G.E. Moore's Open-Question Argument." American Philosophical Quarterly, 1988.

  2. I thought identities held necessarily? There might be a world with salt-like stuff that isn't NaCl, but if so it still isn't salt. Presumably the same could be said of 'good' and our synthetic reduction basis.

    Of course the really tricky thing is showing how to discover the reference of 'good' using empirical methods. So I'm not completely sold on the synthetic approach, though I think I need to re-read Railton.

    Still, analytic reductions strike me as even more problematic. The very notion of synonymy appealed to by the definitional approach seems less than clear. And, as you note, the open question argument does actually pull some weight against analytic reductions. For any proposed definition whatsoever, it does seem that a competent user of moral language could reply, "eh? That's not what I mean when I say 'good'!"

    Must we simply conclude that they weren't competent language users after all? It might not be such a terrible move. People make superficial use of shape-language all the time, without knowing the precise definitions provided by geometers. Maybe morality is like that?

    Also, there are a couple of things puzzling me about the open question paragraph in your paper. Would you mind if I quote and discuss it here on my blog sometime?

  3. Correct me if I'm wrong, but you are saying that analytic philosophy can answer questions of value? Values are construed according to historical, ethological, cultural, linguistic, norms that are construed outside of any form of "analysis of words, ideas, concepts, etc." We must be able to reach beyond simple analysis of 'givens' if we are to do anything at all apart from forming a community of "philosophers" that exists outside of normal society. Richard Rorty is a good example of one who realized the need to explore the pragmatism of the everyday.


Visitors: check my comments policy first.
Non-Blogger users: If the comment form isn't working for you, email me your comment and I can post it on your behalf. (If your comment is too long, first try breaking it into two parts.)