Some accounts of mental content represent the objects of beliefas structured, using entities that formally resemble the sentencesused to express and report attitudes in natural language; others adopta relatively unstructured approach, typically using sets or functions. Currently popular variants of the latter include classical andneoclassical propositionalism, which represent belief contents as setsof possible worlds and sets of centered possible worlds, respectively;and property self-ascriptionism, which employs sets of possibleindividuals. I argue against their contemporary proponents that allthree views are ineluctably plagued by generation gaps: they eitherovergenerate beliefs, undergenerate them, or both.