In Interlanguage Pragmatics (ILP) research, the Discourse Completion Task (DCT) is the main source of insight into speakers’ productions of pragmatic phenomena. Its omnipresence as a means of data collection can be explained by the need of comparability of data sets and (sociolinguistic) variable control. However, some studies suggest a discrepancy between surface realisations observed in naturally occurring data and experimental data like DCTs (cf. e.g. Beebe and Cummings 1996; Golato 2003).Unfortunately, the results of these studies are inconclusive and do not offer any information about quantitative differences in realisation patterns and about the impact of different methodological approaches on interlanguage data.
It is therefore the aim of the present study to compare the influence of two methods of data collection, DCTs and task-based elicited conversations, on the realisation of the head act strategies in requests produced by advanced learners of English. Overall, our results show a significant difference in the distributions of request head act strategies across the two methodological conditions. The conversational head acts are substantially more direct than the requests elicited by DCTs. The patterns observed in learner data strongly resemble the ones found in native speaker requests in the same methodological scenarios. This implies that despite earlier claims, advanced learners can display target-like language use. The resemblances furthermore indicate that semi-naturalistic methods of data collection are a more valid means to obtain learner data that is representative of naturally occurring conversations.