An example of the limits of the Latin calque in the description of the name: the case theory in the 18th century

Abstract

When describing the different components of French speech, the grammarians of the sixteenth and seventeenth centuries resorted, like their predecessors, to Latin, viewed then as a model of excellence. More specifically, as far as the noun is concerned, their emulation led them to declare that its occurrence varies in gender, number, and case, just as with its ancient counterpart. Far from constituting an obstacle, the difficulty surrounding the lack of case morphemes in the French substantive is circumvented: in French, the declension of the noun is more properly that of the article, an anteposed occurrence that both clarifies the reference implied by the noun element and varies according to the grammatical function occupied by the noun group. Thus the occurrence la femme (subject or direct object) corresponds to the nominative and the accusative, while the occurrence à la femme (indirect object) would equate to a dative. This theory, already endangered in the sixteenth and seventeenth centuries, was very widely challenged in the eighteenth century. The controversy did nothing to discourage Régnier-Desmarais (1706) and Vallart (1744), who defend the theory of case variation and establish declension paradigms that they clarify and justify at length (first part). Their well-supported arguments clash with reports by Girard (1744) and Du Marsais (1769), who perceive the Latin calque as a syntactic misinterpretation and a misunderstanding of the characteristics of the article, at the time a poorly-regarded class of words (second part). Beauzée (1767) reconciles the two trends, claiming that even if the noun does not undergo declension, French remains a case language as far as the pronoun is concerned. Thereby restoring the legitimacy of the case while limiting it to a precise type of word, he more generally questions the role of this grammatical category (third part).