Synonym Discrimination Based On Co-selection Theory

In English vocabulary,synonyms accounts for a large proportion of the total number.To learn English well,learners must have a good knowledge of synonyms.However,many synonyms are difficult to discriminate if merely referring to their explanations.This phenomenon gives learners the confusion that these synonyms can be exchanged at any circumstances.As a result,learners make grammatically right but pragmatically improper sentences.Therefore,a scientific and practical framework of synonym discrimination is needed to provide guidance for both English teachers and learners.The present study provides a framework of synonym discrimination based on Co-selection Theory.As a case study,four synonyms(absolutely,completely,totally,utterly)are analyzed under the model of Extended Units of Meaning(EUM)to examine their similarities and differences.In addition,this study follows a corpus-based approach and extracts data from COCA,meantime,the concordance tool AntConc 3.4.4w is also employed to process data.Analysis and discussion of the study are carried out following overall distribution,collocation,colligation,semantic preference,and semantic prosody.Results show that in terms of overall distribution,absolutely,completely and totally are most frequently used in spoken,while utterly is most frequently used in fiction.For collocation,the relationship between frequency and MI score is not proportional,collocates of absolutely and completely have a relatively higher frequency,but the number of collocates with both high frequency and MI score are less than that of utterly.In terms of colligation patterns,completely and totally are exactly the same,with five types,while absolutely and utterly have four types and they each has one type that is not shared by the other.However,“ADV+ADJ”,“ADV+V” and“V+ADV” are three colligation patterns that all the four synonyms share and most frequently used.In terms of semantic preference,the four synonyms show great differences.Compared with absolutely,completely,totally and utterly seem to have a relatively wider semantic preference.For semantic prosody,absolutely,totally and utterlyhave strong negative semantic prosody,only completely shows strong neutral semantic prosody.However,differences still exist among absolutely,totally and utterly.For the node word absolutely,neutral semantic prosody is the least.For the node words totally and utterly,positive semantic prosody is the least.All these findings provide implications for English teaching and learning as well as dictionary compiling.