A Sociophilological Study of Late Latin (review)

Roger Wright's scholarly work has focused on a single problem: when and how did the Romance languages become distinct from Latin? His answer was articulated at length in his Late Latin and Early Romance (1982). Until roughly the year 800, as Wright puts it, "late Latin was early Romance." The difference between the two is simply the difference between the written and spoken forms of a language. Just as English-speakers write boatswain but say [bo'sun], so early medieval Latin speakers wrote quattuor, but pronounced it [katt'r], [kwattro] or what-have-you. This "complex monolingualism" first changed under Charlemagne with the educational reforms of the English scholar Alcuin. Henceforth the clergy were to pronounce Latin words as they were written (and, not coincidentally, as they had always been pronounced in Alcuin's Britain). Quattuor was no longer to be [katt'r] or [kwattro] but [kwa-too-ore]. As a result, "Latin became . . . a foreign language for everybody," just as contemporary English would if its speakers were suddenly forced to pronounce tough, nation, and Leicester phonetically. Since written quattuor now had a new pronunciation, a new written form had to be devised for the ordinary pronunciation [katt'r]: such a system is first attested in the Strasbourg Oaths of 842—generally regarded as the first example of written Romance.

The "Wright thesis" has not convinced everyone, but it is undeniably provocative. Even most skeptics would agree that it has reframed the debate in interesting ways. The twenty-five pieces collected here defend the basic model, while developing its implications or exploring particular aspects in more depth. With the exception of one 1981 article, all were originally published between 1993 and 2002. Many have been revised or updated, and ten pieces originally published in other languages have been translated by the author.

On the one hand Wright is concerned to restate and bolster the argument against any Latin/Romance differentiation before c. 800. Here he finds welcome support in Michel Banniard's Viva Voce (1992), an extended review of which appears as Chapter 4. But the weight of the collection falls on the post-800 period, with special attention to Ibero-Romance. While Wright still regards the Carolingian reforms as initiating the split, he now sees a "more nuanced process," in which consciousness of the Romance languages as distinct from Latin and (later) from one another may have taken firm hold only in the twelfth or thirteenth century. In the mid-800's, Wright argues, a Spaniard traveling along the "dialect continuum" from Cordoba to Strasbourg would have recognized almost every word in the Strasbourg Oaths (chap. 12). Tenth- and eleventh-century glosses from San Millán and Silos, sometimes regarded as the first written "Spanish," in fact show no conceptual differentiation between Latin and (Ibero-)Romance (chaps. 14-16). It is only around 1200, under figures like Rodrigo Ximénez de Rada, the Paris-educated archbishop of Toledo, that the two are clearly distinguished. The subsequent division of Ibero-Romance into "Castilian," "Galician," "Portuguese," etc., may reflect political more than linguistic developments.

Wright's observations will be of interest to both Latinists and Romance linguists. But they also open the door to more general issues: the arbitrary nature of linguistic periodization (chaps. 3 and 13), the role of textual evidence in historical linguistics (chap. 23) and the perplexing question of "What Actually Changes During a 'Sound-Change'" (chap. 24). The central thesis lends the collection a substantial unity (even if it does not quite justify the singular "study" of the title); some repetition is inevitable, and forgivable. Few university libraries will own all of the original pieces—my own institution scores only eleven out of twenty-five—and it is useful to have them accessible in one place.