Because of their extreme marginalization in dominant colonial society, in more cases than not the significant role of people of African descent in shaping the history, politics, economics, cultures, and languages of the Caribbean and the rest of the Atlantic World has been consistently and systematically ignored, trivialized, underestimated and otherwise erased in academic teaching, research and writing. This erasure is usually perpetuated and replicated unconsciously by scholars who for a variety of reasons find it difficult to critically question some of the many erroneous assumptions that simultaneously underpin and undermine our efforts to account for the facts. This paper deconstructs four of these assumptions which have had a particularly negative impact on the work of creolists: (1) that West African and Caribbean societies were not pluri-lingual, pluri-cultural, and pluri-identified throughout the colonial period; (2) that the languages spoken along the West African coast are less genetically and typologically related than they actually are; (3) that African influences on the Atlantic Creoles can/must be traced to a single African language; and (4) that the acknowledgement of African influences must be restricted to cases where no other influences (superstrate, universals, etc.) can be plausibly invoked.