The central question of ontology has long been thought to be ‘What is there?’. The central way of answering it has been to consider which entities we must posit as part of a best total explanatory theory. This paper argues against this ‘explanatory’ conception of metaphysics, by showing that it relies on an unarticulated assumption that all the terms at issue in these metaphysical debates serve an explanatory function. Making use of work in systemic functional linguistics enables us to identify (...) the many different functions played by terms of interest in metaphysics. And that makes it clear that ‘contribution of explanatory power’ should be rejected as an across-the-board criterion in ontology. This work in functional linguistics also enables us to see why it is useful to have a language that entitles us to use redundant inferences to introduce terms for properties, numbers, and the like, giving us new reason to accept ‘easy’ inferences that there are such things. As a result, we should give up thinking that ‘what is there?’ provides a deep and interesting question for a discipline called ‘ontology’ to answer, and give up thinking that the task for ontology is to determine which entities to ‘posit’ as part of a best total explanatory theory. (shrink)