Er-nominalisations which take CP complements are rare in English, but possible. A common construction involving one is to be a firm believer that…. I propose that the behaviour of CP-taking er-nominalisations (‘CoPTErs’) results from a tension. On the one hand, they are Argument Structure Nominals in the sense of Grimshaw (1990), and they ‘inherit’ the argument-taking properties of their parent verb. So if the parent verb believe can take a CP argument, the corresponding er-nominalisation believer should be able to take a CP argument too. On the other hand, they are nouns. And since Stowell (1981), a long line of work has argued that a noun simply cannot take a CP as an argument. I argue that this tension is usually fatal, which is why CoPTErs are fairly unacceptable when placed in argument positions. It's only when they are used as predicate nouns that they become acceptable – but even then, the CP does not pattern like a true argument of the noun. I sketch a possible analysis, in which the CP complement to a CoPTEr adjoins to the predication and binds a variable (of category D) in the CoPTEr's argument position.