I'm reading on coordination structures in relation to Chomsky's proposal of the Labelling Algorithm and stumbled upon Kayne (1994) The Antisymmetry of Syntax. In it, Kayne takes the view that coordination take the form [DP1 [and DP2]], saying that and heads the phrase [and DP2]. This claim was reinforced by with the fact that languages such as French and Japanese allow and to appear before each conjunct, or in the case of Japanese, after each conjunct. The French example given was:
(1) Jean connait et Paul et Michel.
i.e. Jean knows and Paul and Michel.
I've seen other authors use this example in their research. I know that I must be missing something as I don't believe so many people could get this wrong but, I know French and the construction given in (1) is ungrammatical. So how can so many authors get it wrong by saying that it is in fact grammatical and use it to support their claim?
Kayne goes further to say that the first et in [et [Paul [et [Michel]]]] takes as its complement the phrase headed by the second et and that the first et is in fact a distributive operator. But how when this sentences is inherently incorrect?