A common objection raised against naturalism is that a naturalized epistemology cannot account for the essential normative character of epistemology. Following an analysis of different ways in which this charge could be understood, it will be argued that either epistemology is not normative in the relevant sense, or if it is, then in a way which a naturalized epistemology can account for with an instrumental and hypothetical model of normativity. Naturalism is here captured by the two doctrines of empiricism and gradualism. Epistemology is a descriptive discipline about what knowledge is and under what conditions a knowledge-claim is justified. However, we can choose to adopt a standard of justification and by doing so be evaluated by it. In this sense our epistemic practices have a normative character, but this is a form of normativity a naturalized epistemology can make room for. The normativity objection thus fails. However, in the course of this discussion, as yet another attempt to clarify the normativity objection, such a naturalistic model will be contrasted with Donald Davidson's theory of interpretation. Even though this comparison will not improve upon the negative verdict upon the original objection, it will be argued that naturalism cannot accept Davidson's theory since it contains at least one constitutive principle – the principle of charity – whose epistemic status is incompatible with the naturalistic doctrine of gradualism. So, if this principle has this role, then epistemology cannot be naturalized.