Sunday, February 18, 2007

Is Normativity Just Semantics?

I'm over a year late responding to this one, but Computational Truth had an interesting post about whether disagreements over well-being are substantive. Suppose that we all agree on the empirical facts: Alan experiences more pleasure, but Betty has fulfilled more of her heartfelt desires. What are hedonists and desire theorists disagreeing about, then, when they dispute which of the two is "better off"?

The problem: According to analytical reductionism (or "descriptivism"), normative terms like 'wellbeing' simply mean whatever wellbeing reduces to -- happiness, desire fulfilment, or whatever the case may be. This would seem to suggest that the dispute is merely terminological. Either one of the theorists is confused about what their words mean, or else they're speaking subtly different idiolects. In that case, we can translate the apparent dispute:

H. "Alan is better off!"
D. "No, Betty is better off!"


H. "Alan is happier!"
D. "No, Betty has fulfilled more of her desires!"

so that they're not really in disagreement at all. (At most, they disagree about what the term 'better off' means. But words aren't worth arguing over. It's the proposition, or what is said with the words, that really matters. We mean to argue about the world, not just the language used to describe it.)

This seems like a pretty good reason to reject analytical reductionism. Normative disputes, e.g. between theories of wellbeing, are surely more substantive than is allowed for by this account.

A proposed solution: recall my recent claim that philosophical truth just is the idealized limit of a priori inquiry. If we grant this rational normativity as primitive, I've previously suggested that we can use this to define our other normative terms. So, for example, 'wellbeing' means something like "what it is ideally rational to value for a person's own sake".

This yields the happy result that disagreements about wellbeing can be substantive after all. Hedonists and desire theorists disagree about what to value (or what it would be ideally rational to value for a person's own sake). That sounds right to me, at least.


  1. when people debate they tend to each take a position and then try to win the debate.

    Now if one side was to conceed the definition of "better off" to the other in this case they would probably be crippled, and, at least to third parties, I will probably loose. So the FIRST thing I need to do is to trick you into conceeding the terminological debate.

    In theory two people thus might debate such an issue and have no disagreement at all but fight based on the nature of debate and their initial assumptions.

    As to the issue itself I'm not sure if "number of desires filled" makes for a meausre that will have the sort of mathmatical properties we would like to see. So it would be interesting to see if it could be tied down.

    For example is 2 fufilled desires better than 1? (think how many desires are there? what if filling one desire causes another). Is it twice as good? then there is the problem of who's desires? (you now? you later? you 'hypothetical' later? ideally rational you?)


  2. That does sound right. (Although it should perhaps be noted that the 'pretty good reason to reject analytical reductionism' given in the post is actually nothing other than the open question argument; Moore had less sophisticated views in mind, but this is precisely the sort of argument he makes against people of his day who were trying to do similar things.)

  3. Richard, it seems to me that you are right in general but wrong in the specific case of well-being. In general, I would say that there can be real disputes over the "higher level properties or concepts", provided we use these concepts in order to make sense of our world and practical commmitments.
    You may argue that "good" or "right" are such concepts- although I am even skeptical about it. (I like making sense of my world in terms of thick moral concepts like corageous more!!!)

    But in the case of well-being, I go with Scanlon and hold that well-being is a transparent good: the fact that X contributes to my well-being does not give me an additional reason to pursue X (beside the reasons to pursue X that derive from other properties of X).So why should we care in practical life if X adds to our well-being or not? Anyway we pursue it for other reasons.

    If SCanlon is right, and there is no important role that the concept of well-being can play, than queries about well-being's essential nature end up being more or less terminological.

    Also: Sumner distinguishes the "normative adequacy" of a theory of well-being and its "descriptive adequacy". If you judge a theory of well-being (or welfare) by its normative adequacy alone, you end up justifying the point of view of Computational Truth.

  4. Hmm, thanks for the link and response to my post.

    Now I agree one could turn the debate about what is welfare into a substantive question by just defining it to mean 'that which is morally good to maximize.' However, my claim is that we couldn't turn it into a substantive question that is worthwhile to discuss prior to deciding on our moral theory. That is if we define welfare as I just mentioned above debates about what is welfare, while substantial, are just poorly veiled debates about what is moral and serve no purpose.

    So what then of your attempt to define it as "what it is ideally rational to value for a person's own sake." I don't think this actually accomplishes anything. Terms like 'rational to value' or what is 'good for one's own sake' are no more clear or well defined than the concept you are hoping to explicate. Presumably these terms themselves are themselves reducible to terms that only reference the ontologically fundamental moral property and empirical facts (or something of the kind).

    In some sense I feel this is like answering Quine's charge that one can't define what it means to be analytic by saying, "yes I can things are analytic iff they are true by definition." Yes you may have given a synonym for the statement but what is required is at least to gesture at a definition in terms of primitive terms.

    In other words suppose I only understand statements about substantive moral facts (say I believe in a fundamental relation X ought to Y in situation Z) and totally physical claims. At least suggest to me how I can define what welfare means.


  5. Hmm, I think ultimately our disagreement comes down to the nature of rationality.

    I think you can apply the exact same objection to rationality itself. Namely either you believe that there is some external objective property that undergirds rationality or you mean to introduce rationality by construction/definition. If the later it can't hope to bootstrap us up into the realm of substantive disagreement and if the former we ought to make what the argument is about clear.

  6. I guess I'm inclined to take rational normativity as primitive (rather than having anything else 'undergird' it).

    What do you find most mysterious or unclear about it?

  7. Ahh, well that is a reasonable way to make the debate about what is welfare substantive. However, I don't think this is at all what people assume when they naively start debating about welfare. It isn't the case that if you just went up to philosophers engaged in this question they would say:

    Ohh yes we both agree that there is this primitive notion of rational normativity and we are debating in what way it applies to this situation.


    As far as taking this notion of rational normativity as primitive I frankly don't have any grip on the notion of rationality except as relative to a set of desired ends. You know the standard Hume quote.

    If you take the acceptable ends to be merely part and parcel of your primitive concept then all discussions about what is welfare are just disguised discussions about what are the ends that are part and parcel of rational normativity. That may be interesting but this discussion can never serve the purpose that people seem to want it to serve, namely as a way to illuminate or get a handle on what is good since the only way to figure out what is in people's welfare is to first explicate this more fundamental moral notion that would solve the problem anyway.


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.)