A central feature of ethical thought is that it appears to involve not only descriptive belief, belief about what is the case, but also normative belief about what should be done. Suppose we take this at face value and understand normative thought in ethics to consist of attitudes that, at the most basic explanatory level, are genuine beliefs. What then should we say about the basic nature of the normative properties that such beliefs are about? I argue that normative properties (...) are complex naturalistic properties of psychology. In the first chapter, I consider the non-naturalistic realist position, according to which our world contains the instantiation of irreducibly normative, metaphysically sui generis properties. I argue that proponents of non-naturalistic realism have not successfully shown that this view is compatible with confidence in the claims and methodologies of the natural sciences. This gives us powerful (if ultimately defeasible) reason to reject this view. In the second chapter, I consider metaethical ideal attitude theory, exemplified in the work of Michael Smith, according to which normative properties about what an agent A should do concern what an ideal version of A would desire that non-ideal A do. In order a) to maintain a naturalistic account of normative properties, b) to avoid radical skepticism about ethical knowledge, and c) to explain the motivational force of normative judgment, I argue that ideal attitude theorists should hold that what it is for an agent A to be ideal is derived from A’s own evaluative attitudes. I call this a fully agent-attitude-dependent version of ideal attitude theory. In the third chapter, I consider Sharon Street’s recent arguments in favor of metaethical constructivism, according to which normative properties concern what is entailed by an agent’s practical standpoint. I argue that Street’s metaethical constructivism is best developed as a version of agent-attitude-dependent ideal attitude theory. (shrink)