This dataset contains tabular files with information about the comparative and superlative alternation for 139 disyllabic adjectives in written British English. Data are from two corpora, which were compiled in the late 1970s by Albert Braun for his University of Zurich dissertation. The data were originally published in Braun (1982), the first empirical, corpus-based study on syntactic factors underlying English adjective gradation. The texts in the corpora cover four genres ("drama", "fiction", "newspaper", "nonfiction") and were published between 1957 and 1974. Since the focus of Braun (1982) is on syntactic factors underlying the choice between the two gradation forms (-er/-est vs. more/most), only adjectives allowing both strategies were included in the study. Tokens (817 in total) are coded for a range of syntactic variables including position (attributive/predicative/postnominal) and complementation by a than-phrase.