The canonical picture of how creoles emerge through language contact involves the source languages – superstrates, substrates and adstrates – providing the necessary grammatical and lexical input to shape emerging structures (Winford 2009: 22ff.). However, the lack of a certain grammatical feature in an emerging contact language does not automatically lead to its adoption from a source language, even though it might be available (Wurm and Mühlhäusler 1985: 114f.). Tok Pisin, the creole serving as Papua New Guinea’s lingua franca, provides two examples.
I argue that there were language–internal and language–external influences impeding the adoption of these grammatical features in the forms present in the source languages. Showing how these impediments differed at various stages and how language–external and language–internal factors either worked in concert or against each other will help illustrate the complex origin of contact languages.