Abstract

"Results" are relevant observations that do not appear prone to an early confutation. It is by no means easy to identify criteria for such a characterization. One of these criteria is the agreement with results established in neighbouring scientific fields. I take into consideration here linguistics and genetics and pose two main questions: What genetics can do for linguistics; and what linguistics can do for genetics. The answer to the first question requires some time and is the subject of a series of articles of mine on this journal. In the light of the second question I consider here what linguistics can teach to biologists involved in the interpretation of genomes seen as texts. In particular, I discuss concepts like segmentation of speech, the so-called commutation test and the existence of dialects and idiolects.