This paper reports on a corpus-based study of English for Academic Purposes (EAP). It investigates linguistic characteristics of English abstracts written by Korean scholars in the fields of English Literature and of English Linguistics in order to find lexico-grammatical patterns used by native and non-native English scholars specializing in English Literature and English Linguistics. The paper hopes to offer suggestions on suitable academic English pedagogical techniques, especially for English writing. In our study we compiled two English abstracts corpora from journals of English language and literature and journals of English linguistics from English-speaking countries and from Korea for the years 2000 to 2011: Anglophonic Corpus and Korean Corpus. Each corpus further yielded two subcorpora: native and non-native English literature corpora and native and non-native English linguistics corpora. Unlike previous studies focusing more or less on abstracts of native scholars, in this study abstracts of non-native scholars are compared to those of native scholars in the areas of English literature and English linguistics. This study contrastively analyzes abstracts in academic research articles in terms of two aspects: (i) native/non-native speakers’ distinctive use of language and (ii) linguistic features of abstracts as a genre. In particular, four types of methodologies (statistics, wordlists, keywordlists, and key clusters) often adopted in corpus-based analyses are used to explore characteristics of abstracts written by Koreans and by native speakers.