Synonymy and language teaching

abstract This paper aims to analyse the phenomenon of synonymy from a corpusbased perspective. Synonymy has been the object of research of many linguists (Cruse 1986; Lyons 1981, 1995; Carter 1998; Pearson 1998; Biber, Conrad, Reppen 1998; Partington 1998; Tognini Bonelli 2001) and many of them agree on the need to consider selection restrictions to identify semantic incompatibilities between apparently synonymous words. In this paper the investigation of synonymy is pursued by studying the behaviour of four adjectives, ‘set’, ‘situated’, ‘located’ and ‘placed’, in the English language of tourism. The data used for analysis are taken from a corpus made up of British Farmhouse Holiday websites. Collocates of the four node words have been identifi ed and grouped into four semantic fi elds that have been labelled General Location, Specific Location, Measured Location and Convenient Location. The semantic fields identifi ed for each collocational profile have been compared and contrasted in order to detect similarities and differences of usage. The results will show that although these four words are apparently synonymous and are reported as such by some monolingual dictionaries, their patterns of association reveal some differences in usage which act as constraints in the substitutability process. A corpus approach to synonymy may have interesting pedagogical implications in that it guides students to understand the importance of considering typical phraseological patterns and contextual relations in the identifi cation of the meaning and usage of words