We defend the view that belief is a psychological category against a recent attempt to recast it as a normative one. Tamar Gendler has argued that to properly understand how beliefs function in the regulation and production of action, we need to contrast beliefs with a class of psychological states and processes she calls “aliefs.” We agree with Gendler that affective states as well as habits and instincts deserve more attention than they receive in the contemporary philosophical psychology literature. But we argue that it is a serious error to align beliefs with the norm of rationality, while building a contrasting category whose members are characterized primarily by their failure to measure up to that normative standard, since these latter ones cannot constitute a distinct psychological category. First, we demonstrate that Gendler gets unwarranted conclusions about the existence of aliefs from belief-discordant cases. Next, we argue that the concept of alief is insufficiently clear. Aliefs cannot be distinguished from other types of states, such as beliefs. Also, when grouping many states under the category of aliefs, Gendler overlooks important differences between phenomena that are clearly distinct, such as habits and instincts. Aliefs simply do not constitute a legitimate psychological category.