British empiricism and Transformational Grammar

Our aim in this paper is to show, that, although Chomsky’s interest in empiricist British linguistics was always very limited, and the controversy of Transformational Grammar against British empiricism was led more by Neo- Firthians against Chomsky than by Chomskyans against Firth and his followers, the debate has gone through different phases since the 1950s, and is not yet over. It was very vivid in the 1960s, when British empiricist linguists, such as Randolph Quirk and M.A.K. Halliday, discussed mostly grammaticalness and competence versus ppperformance acceptability and gradience. Contrary to what happened with the Neo-Bloomfieldians, the debate did not touch on the validity of corpora. After a decline of several decades, the discussion was revived in the 1980s–90s with the rise of large computerized corpora. At this point the debate concerned mostly new topics still pertaining to TG, such as corpora, statistics, and creativity. It was carried out within the area of computational linguistics and is still currently vivid. The issue is to investigate whether this debate with TG is still relevant, in particular for corpus linguistics, or whether these arguments are past their time and are revived only for strategic reasons, in pretending to justify the emergence of a new linguistics.