Historians of Linguistics have found it difficult to characterize Italian linguistics of the period between the two world wars. However, the analysis of the works of Benvenuto Terracini (1886–1968) offers us a clear insight into what actually happened. First of all, Terracini was very much familiar with structuralist trends in linguistics beginning with Saussure. Indeed, he sketched a very interesting historical interpretation of these currents. Furthermore, he made use of portions of both Saussure’s and Charles Bally’s theoretical positions. If his orientation remained nevertheless quite different (also from those of Devoto and others), it is because he remained faithful to the programs of historical linguistics and to a socio-cultural interpretation of language development. On the other hand, the alleged relationship between Terracini’s linguistic thinking and Benedetto Groce’s philosophy was in effect minimal: Terracini clearly saw that Croce’s philosophy of language meant an end to the autonomy of linguistic science; besides, his acknowledged masters were Humboldt and Hugo Schuchardt in questions of general linguistic theory, Ernst Cassirer in matters concerning the symbolic nature of language, and Ascoli, Jakob Jud, and Karl Jaberg with respect to a dynamic conception of dialectology.

Terracini’s reputation in the area of stylistic analysis should not make us forget that the basic concepts of his method had been developed in the fields of the history of Vulgar Latin and of Italian dialect geography. Terracini has deepened our appreciation of the results of bilingualism in the study of the relationships between substrata (in particular Celtic and Mediterranean) and Latin, and sketched a careful phenomenology of linguistic transformations. What is perhaps surprising is the fact that Terracini, in his dialectological research into ‘languages in contact’, had already put into practice, understandably in a somewhat dated terminology, concepts such as ‘diasystem’, ‘interference’, ‘diglossia’, and ‘register’. In fact, a comparison between studies by Uriel Weinreich and those by Terracini reveals an impressive number of similarities. Indeed, it is from these parallels that it would be possible to develop the conditions and themes of a linguistic historiography which, while using the principal concepts of structuralism, proposes to approach the fundamental problems of historical linguistics, which continue to remain the historical bases of culture through time. It may be noted in this connection that sociolinguistics addresses a considerable portion of problems in historical linguistics; yet the frame has remained largely empty following the rise of the new methods. To conclude, Terracini appears to us faithful to the programs of an historical linguistics whose achievements have been admirable, while at the same time gravitating toward an experimentation with analytical procedures which has permitted him to anticipate certain attainments of contemporary linguistics.