This paper is based on a number of objects that were in full expansion in the eighteenth century: the practices of formalizing and standardizing scientific language (natural history and chemistry are the most prominent examples), the practices of classifying scientific language (dictionaries, lexicons, systems, lists), the disputes over scientific language (which were not limited to Buffon), and the many places where science and language intersected. Analyses of how these sources converged poinedt to a common goal: the control of language used by scholars to communicate with each other about scientific matters, whether in treatises, memoirs, or correspondence. These diverse linguistic and communicative practices, as well as the disputes over scientific language, revealed a common trend that manifested itself throughout Europe in the second half of the eighteenth century. This veritable linguistic turn, in which scholars turned their attention to language and not just to scientific objects, gradually led to models and solutions for controlling language through which different disciplines would successfully engage in general reforms. Once scientific language became the object of a strong scientific control, as it did in the early 19th century, it now contained a strong element of artificiality. By the end of the eighteenth century, it represented a definitive break with the classical model of the Republic of Letters, for which the written word and the text were still the canon and the guarantee of both knowledge and an author's reputation. Henceforth, the models of language control introduced by the linguistic turn would serve as the basis for the constitution of new scientific disciplines, detached from the scholarly linguistic practices of the Ancien Régime, allowing scholars to free themselves once and for all from both the Republic of Letters and the problem of scientific writing.