In this paper I consider a criticism of the current `two aspect' interpretation of Kant's idealism. On the `two aspect' interpretation, talk of appearances and things in themselves is seen as two ways to describe one and the same object, as opposed to older interpretations which held there to be two separate objects. However, there is thought to be a problem here. If Kant wants to claim that an appearance is identical to some thing in itself, then he must be prepared to explain how an appearance description picks out the same object as a thing in itself description. And yet, so the criticism goes, this identification is impossible since we cannot know any true descriptions of things in themselves on Kant's account. I defend Kant against this charge by arguing that a proper understanding of the `two aspect' interpretation does not require Kant to show that two descriptions pick out the same object.