The primary aim of this project is to study the contextualized use of a group of nouns and adjectives as found in a large, computerized corpus in order to develop new types of language description that can be presented in a dictionary format. ″Contextualized″ use of lexis is important, because without information concerning context, most words in the general vocabulary of a language have an indeterminate meaning; i.e., they have a meaning potential that becomes specified once the context of use is established. Under this view of word meaning, the words that co-occur with a certain lexical item are essential in order to determine the meaning of the lexical item because without the co-occurring words in context, the lexical item could have several meanings. This approach, which is much in line with Firth’s well known observation that ″you shall know a word by the company it keeps″, poses a serious challenge to traditional lexicographical description found in general language dictionaries, which list word senses as discreet items under a headword, with little-to-no consideration of contextual information or collocation. To overcome that challenge, the project proposes to study in detail the corpus data for different semantic groups of nouns and adjectives¾with a view towards identifying ways to include contextual information about word senses and collocational networks in new lexicographic products. Data will be taken from Spanish, Catalan, and English.