Editor for this issue: <>
This is a somewhat delayed reaction to Frederick Newmeyer's posting on basic word order. I was prepared to let someone else comment on the underlying logic of Newmeyer's message. This did not happen, however, so here comes. Newmeyer observes that data from a given language seldom exhibits an unambiguous 'basic word order', and there seem to be no generally agreed-upon criteria to resolve this ambiguity. He further notes that this fact reveals a weakness in functionalist linguistics (or, in more polite terms, 'presents a challenge to it'). He adds, on two occasions, that generative linguistics is increasingly coming to share the same weakness. By `increasingly' he must mean the fact that some applications of the principles-and-parameters approach are taking both language-particular and cross-linguistic data into account more seriously than used to be the case. Before these developments, generative linguistics decreed that every (configurational) language has some sort of basic word order, as specified by the phrase structure (or X-bar) component of its grammar. Newmeyer seems to suggest that this type of basic word order did not share the weakness that is 'increasingly' becoming a characteristic of generative linguistics. The difference between earlier generative linguistics and current generative linguistics, and between their respective conceptions of basic word order, resides in the fact that the former did not, whereas the latter does, pay systematic attention to (cross-)linguistic data. Newmeyer's formulation then seems to suggest that not paying attention to (cross-)linguistic data is a STRENGTH of the theory. Esa ItkonenMail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue