Monday, February 02, 2009

Moral Principles, Objective Generalizations

Folk often confuse the question whether morality is objective with the very different question of whether general moral principles, e.g. 'Stealing is wrong', are exceptionless. This is pretty silly, when you think about it. After all, nobody would think that the general biological principle "cats have four legs" is refuted -- or rendered merely 'subjective' -- by the existence of a three-legged cat. General principles may merely claim to describe normal cases, in which case they are not threatened by the odd exception.

So there's nothing to prevent a moral objectivist from affirming that normally, stealing is wrong, but in special cases it may be justified. Indeed, 'particularists' go even further, and claim that there are no (useful, informative) moral principles at all. One can still be a moral objectivist, and hold that each particular ('token') action is either objectively permissible or impermissible, depending on the precise details of the situation in question. They're merely claiming that we can't usefully generalize from the moral status of token actions to general act types. This suggests that morality is complicated, but complexity is obviously compatible with objectivity. As I put it in an old post:
Any adequate theory must be sensitive to the morally relevant features of a situation. It would be morally obtuse to claim that lying (for example) is always wrong, no matter the specific context. But this isn't relativism, so long as we agree that there's an objective fact of the matter in any particular case. Some lies are permissible and others aren't; but there's no one particular (token) act of lying that is at once both right and wrong, "relative" to different observers.

One might object, "If you can't give a straight answer to the question whether lying is wrong -- an answer that's true for everyone, in all times and places, then isn't that practically the definition of relativism?" But no, that's just as stupid as demanding a straight and universal answer to the question whether cats are tabby. The fact is: some are, some aren't. In each particular case, there's a perfectly objective fact of the matter. It may vary from case to case, however, rendering the general question underspecified. (Which cat? Which instance of lying?) Point to a cat, and I'll tell you whether that one is tabby. Likewise, point to a particular instance of lying, and we may determine whether that action was wrong or not. Any more general questions may lack answers. This isn't "relativism", in any interesting sense. It's just the perfectly ordinary point that (to put it in slogan form) variation precludes generalization.

General lesson: to think clearly about such fundamental issues, focus on tokens, not types.

8 comments:

  1. Good point Richard. This raises yet another worry, though.
    The universal principle theorist has a natural story to tell about how moral reasoning goes:
    (1) Actions of type A are wrong.
    (2) This particular action is of type A.
    (C) This particular action is wrong.
    However, how can the token theorist identify certain (token) actions as wrong or right? She might say "this particular act of stealing was wrong" but how did she reach that conclusion?
    One immediate answer that comes to mind is that she merely felt/intuited that that action was wrong. But then, this might again strengthen a relativist interpretation - after all, feelings and intuitions are subjective.
    On the other hand, if the token-theorist can't explain to us how she figures out what's morally right or wrong, this might lead to some sort of skepticism about moral knowledge. Yes, there may be particular acts that are right or wrong or merely permissible but how are we to know them in the absence of true general principles?
    This reminds me vaguely of Davidson's paper "Mental Events" - I'm sure there is a fruitful connection to be made but I'm too lazy and I don't really understand Davidson.

    ReplyDelete
  2. Whilst your argument that objectivism need not collapse to relativism works, it doesn't do anything to show that objectivism is the correct understanding of human ethical behaviour/judgement/activity.

    Remember that the choice isn't between objectivism OR relativism - there is moral subjectivism as well, which properly understood is a very sophisticated position delineated from relativism. David Hume, Bernard Williams, SImon Blackburn - all ethical subjectivists, but none of them relativists.

    However, i'm intruiged to know, suppose that there are some cases of lying which you would deem to be instances of the general statement "lying is wrong" (or perhaps "lying is bad").

    As a moral objectivist, please explain to me how it is that you come to perceive this truth of the world/universe/construct of reason all rational beings are capable of formulating?

    If there is a cat (tabby or otherwise) before us, we can point at it and say "there is a cat". If you can't see it, I can offer a physio-biological explanation: "you are blind, my poor friend".

    Yet if I look on an act of lying and think it is ok but you think it is wrong (or bad), what is the explanation for our diverging opinions on the matter? What is the apparatus via which you perceive the wrongness and I don't? And what is wrong with my apparatus causing me to commit such moral failing?

    Of course, an alternative explanation is that neither of us perceives anything - that rather we project our sentiments out onto the world around us, "guilding and staining" it with sentiment, as Hume put it.

    But of course, whilst that explanation goes a long way to account for why ethical judgements differ, it rules out ethical objectivism pretty firmly.

    toodle pip

    ReplyDelete
  3. Just to stoke the fire a little...

    this:

    "but then, this might again strengthen a relativist interpretation - after all, feelings and intuitions are subjective."

    is very bad reasoning.

    There is nothing which necessitates that feelings or intuitions are subjective. It is quite possible that all human beings could intuit some objective truth. That's what GE Moore thought.

    Of course, there are all sorts of problems with proposing that human beings intuit objective truths (the outstanding one being, "how, exactly?"), but intuition (and by the same argument, feeling) need not *necessarily* by subjective.

    And as I wrote above, something being subjective does not reduce it to relativism

    ReplyDelete
  4. Paul - it would go beyond the scope of this post to assess moral objectivism itself. (Instead, see my old post: Non-spooky Moral Realism.)

    Anlam - to expand on Paul's criticism, you seem to be confusing metaphyics and epistemology. Even if intuitions provide our epistemic access to some class of truths, it doesn't follow that those intuitions are what (metaphysically) makes the true things true, as relativism claims. Compare: we perceive that a cat is tabby, but "perception is subjective" (in some trivial sense). Is that any reason to think that the cat-truths are relative? Of course not.

    More importantly, you ask: "how can the token theorist identify certain (token) actions as wrong or right?"

    I respond: How does the type theorist identify certain act-types as wrong or right? As discussed in comments here, I see no reason to think that our more general intuitions are more reliable. Quite the opposite, in fact.

    Having said that, theorists (at least, non-particularists) may still reason from general principles even if they are admitted to contain exceptions. We just need to add an extra premise to check for any possible 'defeaters' to the general principle. Consider:

    (1) Stealing is normally wrong.
    (2) This particular action is stealing.
    (3) There are no special circumstances that would justifying stealing in this case.
    (C) This particular action is wrong.

    ReplyDelete
  5. Richard - I didn't confuse "metaphysics and epistemology" (if you want to put it that way, fine) but my point was that, once you accept a certain metaethical (metaphysical?) theory, relevant epistemological issues can arise as objections. (Consider problems with Moore's intuitionism.)

    And the epistemological objection to the particularist/tokenist is that, how do you gain knowledge about the moral status of some token? You say that,


    respond: How does the type theorist identify certain act-types as wrong or right?

    I don't know. Read Kant, Bentham, Mill or Aristotle. I'm not that interested in ethics (or philosophy for that matter).

    Even if our intuitions about specific cases are more reliable, you still will have to generalize from these reliable cases to full-fledged principles.

    You also write that,


    (1) Stealing is normally wrong.
    (2) This particular action is stealing.
    (3) There are no special circumstances that would justifying stealing in this case.
    (C) This particular action is wrong.


    As far as I can see, this doesn't answer the original problem. Either premise 3 is itself going to come from general principles that declare what special circumstances justify stealing or it's going to declare "this special circumstance does not justify stealing simpliciter".

    I don't see how the latter option is plausible, therefore I don't see how you can plausibly be an ethical-particularist but it's possible (very much so indeed) - I'll give you that.

    ReplyDelete
  6. The 'M&E' confusion was when you wrote that intuitionism "might again strengthen a relativist interpretation - after all, feelings and intuitions are subjective." Your epistemological point was distinct from this point about relativism (which is a metaphysical thesis).

    But with that cleared up, let's focus on the epistemological objection. My response is basically just to point out that moral justification or knowledge is no more mysterious for tokens than it is for types. (Maybe you think it's all mysterious, but this isn't the place to discuss general moral skepticism.)

    My example argument was just to illustrate how moral reasoning might work with fallible principles. Putting the reasoning aside, one can always question whether the premises are justified, but again that's no less true in your original case, so I don't see any relevant objection here.

    You write: "Even if our intuitions about specific cases are more reliable, you still will have to generalize from these reliable cases to full-fledged principles"

    Only if we want "full-fledged principles", which a particularist will deny.

    (Of course, I'm not a particularist myself. But nor do I think that all tokens of an act type have the same moral status. Some lies have better consequences than others, after all, and that's what I think matters. Indeed, utilitarianism offers a simple counterexample to the folk who think it's somehow "relativism" to hold that lying is sometimes right and sometimes wrong. "Always do whatever maximizes net well-being" is pretty obviously a form of objectivism still!)

    ReplyDelete
  7. I think one metaphysics/epistemology confusion operating here may be something like "if case-by-case reasoning is permitted, any action can be defended through casuistry, therefore there is no true morality." You know, the same kind of "reasoning" that leads people to dismiss logic because "you can prove anything with it."

    ReplyDelete

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.)