engleski

Semantic preference and semantic prosody of the most common verb-noun collocations in English. A corpus-based study.

Semantic preference and semantic prosody are two notions that were carefully analysed in post-Firthian corpus linguistics and since the 1990s there has been a growing interest in them. As corpora have become larger in size and tools for extracting different lexical items for different purposes have been developed, the two notions were addressed more frequently. Still there are some terminological difficulties and many open and interesting questions that need to be answered. In this paper, semantic preference and semantic prosody are defined and more light is shed on the two notions, and some relevant issues are discussed. Some of the most common V-N collocations (make sense, make decision, make mistake, take place and take advantage) are extracted from the Corpus of Contemporary American English with the help of association measure Mutual Information (MI). All the collocations in question are observed in the span of 10 to both sides of the node while examining their lexical environment thus revealing their semantic preference and semantic prosody. All the word forms of the targeted collocations are taken into consideration and observed in two different registers, i.e. the newspaper and the academic and the results for 1991 examples are compared and analysed. The results indicate the pivotal role of the two notions, semantic preference and semantic prosody, in forming a clearer picture about the meaning of the collocations in question. Moreover, the million-word corpora are considered to be the only true sources that provide accurate and objective information about natural language.

semantic prosody; semantic preference; corpus linguistics

nije evidentirano