A return to semiolinguistics, with its great hall, the Textual Investigation Model (TI) which dates from the 90s, seems essential. If in the 90s linguistic studies and literary studies were still in a dichotomous relationship, the birth of sub-fields (textual linguistics and discourse analysis) produced a great upheaval, if not even a conceptual anarchy, the first effect of which remains, no doubt, the bringing together of the oppositional terms linguistics/literature. Today's globalization (Internet, corpus and the digital humanities, revision of the oral/written couple, multilingualism, etc.) reinstates a rather "unifying" vision of the object of study, Language with its media or political corpora above all. The era of “models” has passed even if, in this inter-, multi- and transdisciplinary space that is today's linguistics, a reassessment of concepts and operational values proves to be useful and necessary. From here the need to "divert" linguistics from the laboratory, from its neutral and dull "dress", from its "tiling" for methodological reasons, to take it out into the street, where its creators and beneficiaries live, that is, the speakers, or speaking subjects, or enunciators, or users.