Ba…, Michel
▶ Watch on YouTubeCoverage note: full (about thirty minutes; YouTube auto-captions labelled English but clearly a machine translation of a talk delivered in French: many mistranslations, and proper names and Latin/Occitan words are garbled, so the summary follows the overall argument rather than every detail)
Core ideas
Against the diglossia model. The speaker attacks a commonplace among medievalists: a "diglossic" relationship between Latin, a "wise language" in a dominant position, and Romance, a "mad language" in a subordinate one. This functional distinction, he argues, makes sense only in the imagination of contemporary researchers, who find in it a convenient, apparently scientific parameter. He traces it to a nineteenth- and twentieth-century picture in which the birth of the Romance languages is a "pathogenesis", a fall from the lost paradise of the first language, and to a vocabulary inherited from medieval clerics ("vulgar Latin", "vulgar tongue", littérata versus rusticitas). Romance authors themselves never used these terms: troubadours and trouvères name their own language in neutral or positive terms and plead for it before the new feudal elites.
Several Latins, several Romances. After thirty or forty years of diachronic linguistics and sociolinguistics, he says, there is not one Latin but several, some of them in fact "Romance Latin" (latinitas minor), and not one Romance but several, some of them highly learned (Romanitas maior, "acrolects"). A table of language levels in Carolingian Latin separates them; the famous first Romance monuments (the Strasbourg Oaths, the Sequence of Saint Eulalia) have only a partial primacy. The "Ciceronian" pronunciation we learn today exaggerates the distance between medieval Latin and Occitan; read with medieval pronunciation, a line like nos virgines… sounds much closer to Occitan.
Applied to the Sponsus. He analyses the drama of the wise and foolish virgins (in Avalle's edition), where Latin and Occitan alternate. The Latin passages given to the foolish virgins are of an "advanced" level close to Romance, while the Occitan passages belong to the literary acrolect (non-prepositional indirect case, learned vocabulary such as cosseil from concilium, parallels with troubadour poetry). The text points to the Limousin (Saint-Martial, Nontron, Chalus, Confolens), contemporary with William IX. Christ speaks highly learned Latin, then highly learned Occitan: there is no polarisation of roles, gender or knowledge. In particular he rejects a thesis that makes Romance the distant language of childhood, the mother, the woman and the layperson.
Conclusion: a conquest. The arrival of Romance inside Latin was not experienced as a dissonance but as the establishment of a new normality in which Latin and Romance become "relatives"; a literary Romance acrolect emerges as an effect and sign of the rise of the feudal elites, who wanted their own mirror. Using the word farciture ("stuffing") without quotation marks, he adds, reproduces the clerical viewpoint. The discipline, in his view, lacks a neutral metalanguage.
Excerpts
“« There is not one Latin but several Latins. »”
“« The diagnostic model opposing Latin and Romance is simply false, even at the level of the analysis of literary monuments »”
“« this lack of metalanguage goes all the more unnoticed because the clerical vocabulary was, in fact, perfectly in line with the mental models that presided over the birth of Romance philology. »”