Co-occurrence Tendencies of Loanwords in Corpora

This paper investigates some major approaches to the analysis of foreign material in text, commonly known as loanwords. While the nature of data may differ in various fields of linguistic research (e.g., bilingual vs. monolingual corpora), perspectives on the analysis of such material have not been different, and they have traditionally been analysed as singly-occurring items out of context. However, corpus research has shown that words rarely occur in isolation. On the basis of a number of English loans in a corpus of Turkish compiled in a multilingual setting, and a number of Italian loans in a corpus of English compiled in a monolingual setting, we conclude that collocational patterns growing around loanwords are significant and should be included in the treatment of loanwords.