Thursday, November 13, 2008

Rational Objectivity

It's possible to be irrational without realizing it. Such a person can then be expected to misjudge what they rationally ought to do. So it's unsurprising that in some situations (e.g. if one is sufficiently muddle-headed), one has little hope of acting rationally. On the contrary, we should object to any theory that implied the opposite. If a theory implies that people can always work out what they (rationally) should do, no matter how muddle-headed they might be, something has gone wrong. It can't be a true theory of rationality, because rationality is sufficiently objective to allow the possibility of a 'no-hoper' -- a person so incompetent as to be beyond the reach of rational guidance.

I bring this up because people are sometimes tempted to treat rationality in an excessively subjective fashion. (See, e.g., here and here.) People say things like, "How could Bob be rationally required to believe that P, if he has misinterpreted the evidence in such a way that he's led to think it supports not-P instead?" (Imagine Bob is a counter-inductivist. He notes that the sun has risen every day in the past, and misinterprets this evidence as supporting the proposition that it will not rise tomorrow.) The answer, naturally, is that Bob's mangling of the evidence has led him astray. It's true that Bob doesn't realize this. There are no alarm bells ringing in his head. But that doesn't mean Bob is somehow rational after all. It merely means he's unaware of his own incompetence. He really should believe that P - that's what the evidence supports - no matter his failure to recognize this epistemic requirement. Ignorance is no excuse.

All this is to say that we should expect the possibility of a situation in which we're beyond the reach of rational guidance. Again, I hear people object to some proposed rational rule R, "but in some situations I will lack the competence to reliably follow rule R; I confuse it with P and Q; my best attempts will not be good enough." This is true, but it is not an objection. Trying hard is not enough to make one rational. This is entirely to be expected.

Now, there may be multiple standards of rationality, corresponding to different degrees of objectivity or idealization. In addition to the ideal norms followed by a perfectly rational agent, we may also need non-ideal norms to guide imperfect agents like ourselves. But my point is that even these must have some minimal degree of objectivity to them. Any norm for which effort entails success is not any kind of rational norm at all. Any rule that can infallibly be followed (no matter how muddle-headed the agent) is too empty to qualify as a genuine rule of rationality. If there's any real substance there, then we must be prepared to admit the possibility of no-hopers: people who violate the norms without realizing it, and so predictably fail to understand rationality's recommendations - or follow its guidance - despite their best efforts.


  1. Aw, man... The pessimism of this post just hurts my heart...

  2. And that doesn't mean that I disagree.

  3. This post just seems to suggest that most part of people are just plain dumb, and aren't even all of them rational, or sometimes even rational at all, which reminds me Ancient Greece where some were born slaves, that is, people not actually human and people who could only exert the servile arts, contrarily to those who were born free and then should perform the liberal arts.

    And, yes, I do know this is quite beyond the inner logic of your text. Or even worse: it may just be extending conclusions that weren't outspoken.

  4. Yes, we need realizable norms for imperfect creatures, even more than we need ideal norms for ideal creatures. I'd prefer to keep using the word "rational" for such norms, but if you object, what word would you suggest we use instead?

  5. Adriano - no, I am not implicitly advocating slavery, if that's your suggestion. I'm not even implying that any actual people at all are in the unfortunate position of the possible 'no-hopers'. My point is purely theoretical. Please stick to what I actually said. (My actual motivation for the post should be clear enough from the start of the second paragraph.)

    Robin - I agree in my final paragraph that there can be non-ideal rational norms, but they still require some degree of objectivity. So when you say, "we need realizable norms for imperfect creatures", do you mean something more than this? In particular, do you mean that we need norms that could infallibly be followed by any agent whatsoever (thus ruling out the possibility of 'no-hopers')? If so, my dispute with you is not over the terminology. I deny that such norms are any use at all. They would have to be completely empty. (I might expand on this claim in a future post.)

  6. We need norms that can be followed with a non-zero reliability, which would constitute "hope", and so exclude "no-hopers." Infallibility is not an option for pretty much any real creature.

  7. Well, there's always some (non-zero) chance the no-hoper will happen to get things right, even on the most objective views. So I take it you're instead demanding norms that anyone can intentionally follow with above-chance reliability? But again, for a norm to be (above-chance) reliably followable by anyone whatsoever - however incompetent or muddle-headed - it must be effectively empty. For any substantive rational norm, we can imagine someone with no hope of competently following the norm (or conforming to its demands at above-chance rates), right?


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.)

Note: only a member of this blog may post a comment.