Le traitement de qui, qui(s), quod dans la tradition grammaticale latine : quelques jalons pour l'étude du relatif, de Donat à Port-Royal

Bernard Colombat, « The treatment of Qui, qui(s), quod in the Latin grammatical tradition: some stages in the study of relative, from Donatus to Port-Royal ». The aim of this article is to study stages in the description of what we today call the 'relative pronoun' in the Latin tradition over a very long period (from Antiquity to the 17th century). For ancient grammarians, the first difficulty was to show the specificity of the relative among others qu- forms (interrogative, indefinite), a specificity that is represented only imperfectly in the morphology. But they also had to decide whether qui is a 'pronoun' (as it for Donatus) or a 'noun' (as it is for Priscian). Priscian already analyses quite clearly the syntactic function of qui as relative. But it took a very long time to recognize its double function, anaphoric and subordinating. Unexpectedly for us, the relative had to lose its anaphoric function, being only an anaphoric indicator followed by the repetition of the antecedent (for Linacre and Sanctius); by Port-Royal, the relative 'pronoun' fully recovered the anaphoric function, but the propositional analysis also made it possible to insert a proposition incidente into another proposition. As for quod, its analysis as being only a relative and the difficulty in recategorizing it prevented grammarians from recognizing it as a conjunction.Colombat Bernard. Le traitement de qui, qui(s), quod dans la tradition grammaticale latine : quelques jalons pour l'étude du relatif, de Donat à Port-Royal. In: Langue française, n°139, 2003. La grammatisation du français : qui que quoi vs qui(s) quod entre XVIe et XVIIIe siècles, sous la direction de Bernard Colombat. pp. 10-27