The present paper examines the semantic prosodies in the specialized texts of JDEST Corpus, by observing behaviors of such
lexical items as cause, incur, utterly, probability and career. By semantic prosody is meant the consistent aura of meaning with which a lexical item is imbued by its collocates. Studies show that some words have a strong tendency to attract, or even almost exclusively attract, a category of words which share roughly the same or similar semantic characteristics. As a result, a heavy semantic aura has been formed, which pervades through the entire span, a colligation and even an extended context. This study adopts two conventional methods to conduct the research: (1) observe the prosodic features of a node word on the basis of its significant collocates which are extracted within a defined span and established with a criterion of statistical significance and (2) extract, on a random basis, a fairly large number of concordances of a node word, and then examine the prosodic profiles of collocates with reference to colligations. The study has found, among other things, the word "cause" has a stronger negative prosody than the case in general English texts whereas the word "career" has a weaker or less obvious positive prosody than the case in general English texts. The study shows that lexical meanings entangle in the shared environment; one just cannot disentangle them. The present study also discusses the phenomenon of
prosodic exploitation, in which an established semantic prosody is purposefully violated or flouted by language users for a special communicative effect by use of
unusual collocations which present a prosodic clash.