Abstract
The proposed study aimed to investigate the frequency effects of a relatively rarely used grammar structure, namely had better, and its impact on the process of second language learning. 43 university students participated in the study, who were undergraduate engineering students at a public university in Saudi Arabia. Their English proficiency level ranged between lower-intermediate and upper intermediate. The second language acquisition (SLA) literature indicates that the more frequently a language structure is encountered by a learner, the more likely the successful acquisition of that language structure is, whether this is for the learner’s first (L1) or second (L2) language (Ellis, 2002; Gries, 2008). Two modal auxiliary verbs, had better and should, were particularly selected for this study. They were selected because one of them (had better) is rather rarely used (both in spoken and written discourse), whereas the other modal auxiliary (should) is used very frequently. The findings obtained from this study were analyzed in terms of form and meaning relationship. While the findings indicated that only one-third of the participants correctly identified that [‘d] in the [‘d better] pattern corresponded to had, more than half of the students (51%) thought that [’d] corresponded to would. The results for meaning, on the other hand, displayed a very different pattern from the findings for form. 65 percent of the participants correctly identified that “you’d better see a dentist” could alternatively be expressed by saying “your teeth are not in a good state. I advise you to see the dentist”. These findings suggest that learners acquire the meaning of a grammatical pattern significantly more easily than the actual grammatical pattern itself, which means that they had far more trouble with form than meaning. The implication of these findings is multifold, which this article will address.