Drinking, so can't be arsed to wall-o-text, sorry. :/

Subconscious activity is still activity in the brain, right? And I don't see a (big) problem with there possibly having to be several definitions of good, as long as they can be modeled. It's all still solvable. None of what we've been discussing have been originally excavated from dino bones, all of these are human constructs. Good and bad, right and wrong, morality, life, emotions, money. Clearly any study into those wouldn't be physics or chemistry, but rather some mathematical models, statistics and psychology.

And I didn't say crimes need to be penalized, but whether they do or not shouldn't be based on a whim, but models and data.