This dissertation starts from one core question: Should we let ourselves be guided by our emotions when we make ethical decisions? I give a positive answer to this question. This is not a new proposal. However, my dissertation lays out a novel argument, one that tries to avoid cognitivism about emotions. That is, I argue that there is a kind of emotion that is not cognitive or belief-like, and that can nevertheless act as a normative guide for us. By showing that such emotions can be normative guides, I aim to show that normativity should not be identified with rationality or cognition. The way in which non-cognitive emotions can be normative guides, I argue, is by disrupting engrained habits and beliefs when necessary. This is the second new suggestion I make in this dissertation: that an important aspect in normative guidance has been neglected so far, namely the importance of being able to reconsider one’s ways in light of new circumstances. Philosophers have put a lot of effort into showing how we can have stable commitments and beliefs over time. But not much has been said about how we can break open such commitments and beliefs again if they are not appropriate anymore. I argue that this is a far-reaching omission. We live in a constantly changing world, and our circumstances demand of us different kinds of habits and beliefs as time goes by. I argue that as a result of these considerations, we need to introduce a virtue that has not been considered so far, the virtue of flexibility. Like the virtue of stability in Aristotle, I argue, the virtue of flexibility is a meta-virtue, a good-maker of all virtues.