The past fifteen years have witnessed a revival of interest in multiple realization and multiply realized kinds, most of it highly critical. After providing an overview of this anti-MR literature, I set out what I consider to be a significant concern regarding the ontology of levels that bodes unfavourably for accounts on both sides of this dispute. I do this first by explicating a particular notion of levels that is tacit within much of the literature of the mind sciences, including the literature of MR, then by following through with the implications of this conception for disputes concerning MR. The problems stem from an understanding of MR as a relation between levels. Here I shall defend the alternative (“flat”) account of MR which circumvents or at least suspends the problem of levels and is implicit in the two papers which inaugurated the recent round of critical commentary. This understanding encourages the view that psychological kinds are not multiply realized, and modestly vindicates the anti-MR line of recent years.