Matt asked:

There's things I know rationally, like if I drop a ball it will fall to the ground. And there's things I know
otherwise — not rationally — such as my favorite color, or who I love. As the latter category of
knowledge isn't subject to demonstrable and repeatable experiment, how may I have confidence in it?

============

Some people are confident about their favourite colour and others not. I suppose I fall into the latter
category, because if someone asked me my favourite colour, I wouldn't know whether to say blue, or
green. I might even on occasion say red. But if we take a person who confidently asserts, 'Red is my
favourite colour', and we notice that they never choose clothes, or home colour schemes, or cars
which are red or have red in them, then that is pretty good prima facie evidence that they don't know
their own mind. All the same, there is room for doubt. Perhaps the person whose favourite colour is
red is too self-conscious to wear a red shirt, or paint their front door red, or drive a red car, opting
instead for what they see as a 'safer' colour.

Notice, however, that what is going on here is has a parallel with regularities in the physical world. If I
drop a football, I can be confident that it will fall to the ground. But that's provided someone standing
close by doesn't get the boot in first, or provided that it isn't attached to the ceiling by invisible elastic
string, and so on.

Or consider the question, Who is it that I love? Love is not just a preference that we feel strongly.
Love is put to the test. I may feelever so strongly that I love a particular person, yet when put to the
test, when required to do an action that someone who genuinely loved that person would do, my
resolve fails. Yet, following the example of colour preferences, there is always the possibility that my
resolve failed because the genuine love that was there was thwarted by cowardice, or because at the
last moment I doubted my decision concerning which action I ought to do for my loved one's sake.

There is, however, an important difference between knowledge of the physical world, and such cases
of self-knowledge. I don't think it is correct, as I hope I have made clear, to explain the difference by
saying that the former type of knowledge is rational, while the latter is irrational. We know our own
minds without first having to make observations of our behaviour, that is the crucial difference. It
doesn't follow, however, that our confidence in our self-knowledge is immune to experimental
evidence.

Geoffrey Klempner