A CORPUS-BASED STUDY OF MODAL VERBS IN NATIVE AND EGYPTIAN LEARNER ENGLISH

Abstract

The present corpus-based study investigates the use of modal verbs in native and Egyptian learner English. Two written corpora - Egyptian learner English Corpus (ELEC) and Michigan Corpus of Undergraduate Native Speakers (MCUNS) - are compiled and processed by the Antconc software (Anthony, 2014). Results are further compared to modal tokens in COCA Academic Sub-section (COCAAS) as a reference corpora. Results show a discrepancy in the frequency and order of modal verbs between native and Egyptian learner English corpora. The overall number of modal tokens is significantly greater in ELEC, but confined to three main overused verbs. Modals in the native English corpora are more diverse. The most frequent modal verbs in ELEC are will, can and should. In MCUNS, would, will, can, could and should occur frequently, while in COCAAS, can comes first followed by may, would, and will. The order in the latter corpus indicates a significant change in the use and frequency of modal verbs in English. The comparison of the findings obtained on the modal verbs in the native English corpora with previous findings confirms that many modal verbs continue to decline such as must and shall, while others retain their positions in the lead such as would and will. Except for epistemic prediction, results show lower proportions of modal tokens marking epistemic functions in ELEC than in MCUNS. Conversely, the proportions of modal tokens expressing deontic functions are relatively higher in ELEC than in the native corpus. Thus, the composition of the Egyptian learners would sound assertive, authorial, direct, and crude. The study ends with a discussion of the pedagogical implications drawn from the analysis, and provides suggestions for further research