Creolists have long classified creole languages as ranging along a continuum from basilectal to mesolectal to acrolectal, with an implication that the basilectal range is the ‘purest’ or even most prototypical representation of creolization. However, in light of John Holm‘s focus on the category of semi-creoles, in this paper I argue that most creole languages worldwide classify as ‘mesolectal’, much more like semi-creoles than ‘basilectal’ creoles like Sranan and Tok Pisin, and that this even includes the ‘basilectal’ ranges of English Caribbean creoles. I propose that ‘mesolectal’ creolization is in fact the usual result, and that the term ‘basilect’ be restricted to the relatively few creoles that developed under the unusual circumstance of almost complete dissociation from their lexifiers. This includes the Surinam creoles, the Gulf of Guinea creoles, Tok Pisin and its relatives, and very few others.