A corpus-based study on the weak obligation modals

The study investigates the differences in meaning and usage of the three weak obligation modals, should, ought to and had better, through analyzing the 100-million word British National Corpus data. These three modals are thought to have similar meaning and usages, which can easily confuse L2 learners. Grammar references provide some information on the differences between these modals, which are sometimes contradictory as well as incomprehensive. The analysis of authentic data suggests that these three weak obligation modals do have differences in use, which can be best understood when considered in the context in which they are used. Specifically, the quantitative. analysis suggests had better is used mostly with personal pronoun subjects, while should is used more frequently with general noun subjects. The subsequent qualitative analysis sheds more light on the usage patterns of these modals. For example, the deontic force of should differs depending on the subject types it is used with. In addition, the use of had better entails extended discourse with contexts and explanations hedging its directness. As such, the subject of the sentence as well as the context in which these modals are used is found to play a significant role in their choice. The study provides pedagogical data, which can be useful in class when teaching these modals. Furthermore, it shows the importance of context in which a certain lexical and grammatical item is used.