We cannot justify feelings; we can only 'have' them. Since they are not justifiable, (I take it) epistemology rejects them.
Can 'feelings'/intuitions count as epistemic justification? Although my initial intuition suggests 'no', science may provide us with reason to reconsider. (This could give rise to a fun Liar-style paradox. Should I trust my gut instinct that tells me I shouldn't?)
It might be the case that our feelings arise due to (reliable and accurate) cognitive processing of subconscious cues. For example, I recall Steven Pinker claiming that spouses' suspicions of infidelity are very often accurate, despite their conscious "reason" for the belief being entirely absurd. So if we accept some sort of externalism about justification, then we might well find that beliefs based on inexplicable 'gut feelings' do qualify as justified after all.
So now I'm half-convinced that they should count, I'm wondering how to reconcile that idea with my own views about justification. Maybe I'm an externalist after all? Or can we count feelings as internal evidence?
Well, feelings certainly are internal, at least. But their reliability is not something that is known to most of us. So, to use them as internal justification, one would first have to acquire the complementary belief that gut instincts are reliable indicators. They could then be used just like any other form of perception.
I think I'm happy enough with that, maybe. If someone's only evidence for a belief was their gut feeling, I wouldn't normally consider the belief justified. But if they could say: "I have a gut feeling, and I also know* that gut feelings are often reliable in this context" then I'd be much happier to consider it justified.
However, I don't know what I'd think if I knew the feelings were reliable but they didn't. And that's really the critical case here. According to my previously outlined views, I should say their belief is not (internally) justified, yet might still count as knowledge. I'm not sure that I do want to say that however. It sounds a bit odd, at least.
* = Given my internalist justification / externalist knowledge split, should I want this sort of second-order evidence to be known, or internally justified? I'm confused.
Lastly, I should warn that I'm really exhausted (had to wake up earlier than I'm used to, due to jury service), so this post might not make any sense. I might add a few more thoughts (or clarify existing ones) at some later date. When I'm awake, say.