On the Verbal Origin of the Verbal Suffixes of Bantu Verb Suffixes

What follows below is a hypothesis. It is by no means fully proven. I will attempt to bring a certain amount of data to show that it has definite empirical contents, then argue from several more universal considerations. The more general view which underlies the hypothesis presented here may be found in Givon [1971a]. Briefly, it is there suggested that in order to understand the current morphology and morphotactics of a language, one must construct highly specific hypotheses concerning the syntax of that language at some earlier historical stage of its development. Conversely, it is also suggested there that synchronic morphologies and morphotactics are a most powerful tool for reconstructing earlier diachronic stages in the syntax of a language.