Problems with these hypotheses:

1. All French Creoles are mutually intelligible, no matter where they are found (Caribbean, AFrica, Indian Ocean)
2. All early reports indicated the masters learned from the subalterns, not vice versa.

3. All Creoles are typologically similar, so could be based on lots of different kinds of baby-talk.
4. Similarities were originally explained as due to the African origin of the first subalterns, such that the universally-seen features were supposedly to be found in the African donor languages.
5. But these similarities surface in areas even where no African influence can be imputed: Pitcairnese, Chinook Jargon, Nagamese, Bazaar Hindi, etc.