Tracking and explaining variation and change in the grammar of American English
Both <i>to</i> infinitive and <i>to</i> -<i>ing</i> complement clauses selected by adjectival heads may involve subject control in English, but there are syntactic differences between them. In spite of the differences, some matrix predicates have shown variation and change between the two types of sentential complement in recent times. The article examines the adjective <i>accustomed</i> from this point of view, with evidence from the <i>TIME</i> Corpus. It is observed that in the first decade covered by the corpus, from 1923 through 1932, <i>to</i> infinitives were slightly more frequent than <i>to</i> -<i>ing</i> complements, but that in the 1960s the <i>to</i> -<i>ing</i> pattern had become predominant in comparison with the <i>to</i> infinitive pattern. The article points to such variation and change affecting the sentential complements of <i>accustomed</i> in one particular text type of American English, and identifies two explanatory principles to account for the variation observed. The first is the Extraction Principle, which is defined more broadly than in some earlier work, and the second is the postulation of a semantic contrast between the two complementation patterns. It is argued that both the Extraction Principle and the semantic distinction are statistically significant in explaining variation at one stage in the overall process of change that has been called the Great Complement Shift.