> According to your definition, pure untested intuition would then be "scientific". As it happens of course, intuition isn't very effective. The point is that all kinds of crazy methods might turn out to be effective, but we wouldn't therefore want to call them scientific.
Maybe I'm just using circular logic here, but the only way we know whether a method is effective is by testing it. Nevertheless, I don't think there's a working scientist out there who wouldn't want to trust a colleague's intuition (that we've already established as reliable) to try to do things we don't quite rationally understand yet. I think the key thing to remember is that ultimately, it has to be an effective means, and the way scientists figure out what is or isn't effective is by experiment.
I wasn't trying to make a point about the use of intuition, this was just a hypothetical example. I could just as well have constructed a hypothetical scenario in which, say, asking a five year old kid turns out to be the most effective method of investigating the world. Surely, even if this turned out to be an effective method, it would not be scientific.
You cannot name a single effective technique that's been shunned because it smells bad to a philosopher. Coincidence? Frankly if anyone tries that in any competitive field they'll lose market share, won't be able to publish, etc.
>You cannot name a single effective technique that's been shunned because it smells bad to a philosopher.
I'm not sure why I should need to do that, since I'm discussing a hypothetical scenario where a crazy method turns out to be effective. Are you saying that it /would/ be scientific to ask a five year old kid if it turned out to be effective? Or what?
Yes. You only say it's crazy because it doesn't work. Like all effective techniques, I don't see why it wouldn't be quite commonplace & eventually well understood.
>Yes. You only say it's crazy because it doesn't work.
How could I be saying that it's crazy because it doesn't work when I'm considering a hypothetical scenario in which it does work? That doesn't make any sense.
It's crazy because we can't construct even the hint of an explanation for why it should be an effective method. Of course, you might in turn imagine a scenario in which we succeed in constructing some kind of explanation -- in which case the method might be rational in addition to being effective -- but there is no guarantee that any such explanation could be constructed, and I am considering the scenario in which it can't.
In any case, in suggesting that scientists should not even be rational (or alternatively, in defining rationality purely in terms of outcomes), you are departing from the mainstream to such an extent that the burden of explanation is clearly on you here. You are certainly not stating a "basic fact" about the nature of science. I'm all for methodological liberalism, but you can't assess a method purely based on its outcomes.
Maybe I'm just using circular logic here, but the only way we know whether a method is effective is by testing it. Nevertheless, I don't think there's a working scientist out there who wouldn't want to trust a colleague's intuition (that we've already established as reliable) to try to do things we don't quite rationally understand yet. I think the key thing to remember is that ultimately, it has to be an effective means, and the way scientists figure out what is or isn't effective is by experiment.