Corpus-based Analysis on Near-synonym Talk and Speak across Different Text Genres

Near synonyms are words in a certain language that have the same concept of meaning’ but cannot be used interchangeably in all contexts of use. Corpus data could be used to help understand the differences in the use of near synonymous words. This research aims to understand the distribution patterns of the near synonymous verbs ‘talk’ and ‘speak’ in different types of discourse as well as understand the collocation patterns of the verbs ‘talk’ and ‘speak’ in each discourse to understand the differences in the use of these two synonymous words. This research is a quantitative qualitative research and uses the British National Corpus (XML Edition) as the data source and CQPWeb as the analytical instrument. The analysis of the distribution and frequency of use of these verbs show that the verb ‘talk’ is more widely used than ‘speak’, but each verb has different grammatical form preferences in each type of discourse. In addition, the collocation pattern shows that the verb ‘talk’ plays more of a role as the main predicator, while the verb ‘speak’ is more often used as an adverbial phrase. These results prove that these two words cannot completely substitute for each other in every context. These results can also be used as a reference in teaching the use of synonymous words in learning English as a second language.