Intuitions: When are they Philosophical/Reliable?

Rationalist philosophers enjoy relying upon their intuitions apropos the truth and falsity of various propositions.  By "intuition", I don't mean a warm fuzzy feeling that a proposition is true.  Something like this might count as an "intuition" in common parlance, but philosophical concern is restricted to philosophical intuitions, otherwise known as "rational seemings".  A central tenet of the rationalist project is that such intuitions are justification conferring, and a plausible explanation for the source of the justification is that philosophical intuitions are accompanied with insight into why the proposition is either true or false.  If you understand the proposition in question, you will see that it must be true or else that it could not be true.  This suggests that the best candidates for propositions susceptible of intuitive justification are propositions concerning modal notions.  Oftentimes, those of a rationalist bent restrict this to alethic modality, but this need not be the case.

Those of a more empiricist bent point to cases in which our modal intuitions go awry as evidence that philosophical intuition cannot be justification conferring.  It is a standard rationalist response is to point out that this position is self-defeating.  Whether or not this response is succesfull turns in part on one's theory of justification.  In particular, externalists may be able to escape the charge of self-defeat.  But externalists are commonly sympathetic to reliability, and why think that our modal intuitions are not, in general, reliable?  This is an interesting philosophical question that has seen insufficient study.  Is there a good reason to think that our modal intuitions are reliable?

Classic cases of modal intuitions run awry can be generated from probability theory.  As it turns out, most people cannot probabilistically reason their way out of a paper bag.  But cases might arise in the domain of natural law as well.  For example, suppose we pull two glasses of water out of the tap at the same temperature.  We then heat one glass of water up and place them both in the freezer.  Is it possible that the glass containing the warmer water will freeze faster?  It would seem not.  Indeed, here is an argument that warmer water will not freeze faster than the cooler.  Whatever temperature t the cooler water is at, it will take m minutes to freeze.  Once the warmer water drops to temperature t, it will likewise take it m minutes to freeze.  Therefore, the warmer water will take m minutes to freeze plus the amount of time it takes to cool it down to t.  But this is longer than m minutes, so the cooler water will freeze faster.

As it turns out, that's a bad argument.  It is possible for warmer water to freeze faster than the cooler.  For more on this phenomenon, known as the Mpemba Effect, see here.  Aristotle noted it in Meteorologica I:

The fact that water has previously been warmed contributes to its freezing quickly; for so it cools sooner.  Hence many people, when they want to cool hot water quickly, begin by putting it in the sun. . .

Bacon also mentions it, and Descartes (somewhat carelessly) stated:

Experience shows that water that has been kept for a long time on the fire freezes sooner than other water. - Les Meteores

Go philosophers!  (Cecil, of the Straight Dope, bombs.)  Ok, but what's the point?  Well, before doing the empirical research, a rationalist philosopher might well take himself to have a philosophical intuition about this case.  Several questions present themselves.  Is this a rational seeming?  If not, why not?  One response might be that the proposition has to do with physical possibility.  Perhaps justification conferring philosophical intuitions concern only broad logical possibility?  It's clear that there can be no "insight into the truth" of the proposition "it is not physically possible for warmer water to freeze faster than cooler" since the proposition is false, but the argument I provided is tempting and could easily, I think, be mistakenly taken as the kind of insight into the truth of the proposition responsible for the justification conferring status of philosophical intuitions.  If this is so, then the rationalist is in a bad position.  But maybe only the foolish rationalist who failed to notice that the so-called "insight into the truth" required some (false) assumptions about the physical effects of temperature on bodies of water would be tripped up?  If so, do philosophical intuitions that confer justification never depend upon any assumptions?  It's clear that friends of intuitions (myself included) have a lot of work to do. 

Comments

[Tom]

Indeed, we have work to do.

I think the right thing for the rationalist to say here is either that intuitions are only really reliable when they concern metaphysical (or "broadly logical") possibility, or (towards which you gesture) that it's probable that our physical intuition in the water freezing case is based on false assumptions.

I tend to prefer the former route. I don't think it's special pleading, especially given the way I defend intuitionism. But I could also speculate that it's more plausible intuitively that we would have insight into metaphysically necessary truths than it is that we would have such insight into physically necessary truths. The latter are, er, "more contingent," as it were, so there's less a priori chance that they'll match up with what we intuit. (Obviously this is very sketchy and inchoate.)

One more thing I want to say about intuitions is this. Intuitions about metaphysical impossibility are phenomenologically different from intuitions about physical impossibility, at least for me. When I think about the traditional cases of metaphysical necessities, e.g. the color exclusion thesis or Bealer's "You can't eat yourself all the way up," I do intuit that they could not have failed to be true. But when I think of "Bodies of disparate mass accelerate equally in a vacuum," I get nothing like the same "Of course! It couldn't be any other way!" that I get in the case of metaphysical necessities.

Now, about rationalism per se: I think that one of the final nails in the coffin of empiricism will be when (or if) the rationalist can explain, pace Casullo, why she doesn't need a circular justifying metareason to accept rational insight, but the empiricist does.

Another nail would be a systematic explanation of why rule circularity is stupid. You finished your paper, right? May I have a draft?