Mandana Kolahdouz Mohamadi; Ali Reza Gholi Famian; Ferdows Aghagolzadeh
Abstract
Antonymy is a kind of lexical-semantic relationship between two words that are similar in many aspects out imply opposite meanings. The few previous studies as well as the authors’ ...
Read More
Antonymy is a kind of lexical-semantic relationship between two words that are similar in many aspects out imply opposite meanings. The few previous studies as well as the authors’ initial study on the oral and written corpus reveal that antonyms’ sequence is subject to a hidden regularity. In the study of antonyms’ distribution, their sequence, i.e. A-B or B-A is considered. Earlier studies and observations of writers regarding oral and written corpus indicates a hidden order in the sequence of antonymous pairs. Jones (2002) attempts to reveal this order in light of seven rules, i.e., morphological derivation, positivity, magnitude, chronology, gender, phonology and idiomaticity. In this study, we investigated the sequence of antonymous pairs in a researcher made Persian corpus. The detected and extracted samples include 4000 sentences taken from written and oral genres produced since 1970s to 2010s in Iranian mass media. The texts include journalistic, narrative and scientific texts as well as oral files extracted from movies, TV, and radio programs. To consider radical changes in Persian language, texts exchanged in virtual space such as facebook messages are also taken into account. The diversity of subject and historical order of the texts would help up to work on a natural, well-balanced corpus, and would provide us a platform for a diachronic approach. The samples are collected randomly. As far as the part of speech is concerned, in forming antonymous pairs, the category of nouns tops followed respectively by verbs, adjectives, adverbs and pronouns. The findings indicate that the framework introduced in Jones (2002) could explain almost half of the sequences attested in the corpus. Rules of idiomaticity, phonology, positivity, morphological derivation are more effective and in some cases the cultural and contextual differences may explain the sequence. It was also revealed that the unmarked member mostly precedes the marked one. This finding is consistent with previous studies conducted on markedness.
