Abstract

My current work compares the ways Czech and Hungarian learners at early and advanced levels of English vs. English native speakers organise temporal information in discourse. Two production experiments were designed to test the claim from previous research on reference to time that L2 discourse largely reflects principles of temporal information organisation typical of the speaker’s mother tongue. This claim has been supported by findings from numerous cross-linguistic studies focusing on a variety of discourse dimensions such as topic time management (cf. Carroll & von Stutterheim 2003), regulation of granularity and condensation levels (cf. Noyau et al. 2005), as well as selecting event perspectivation frames (cf. von Stutterheim & Lambert 2005).

Based on contrasts linked to temporal reference in the three discussed languages, the main question addressed here is how highly advanced L2 speakers from typologically unrelated L1s go about when construing events in English as a second language. Written as well as spoken production data was elicited to detect whether advanced learners who have mastered the formal devices for temporal reference in L2 (a) still rely on principles of temporal information structure typical of their L1s; (b) adopt target-like strategies for organising temporality in discourse alongside learning a new language; or© employ other techniques. Analyses of early L2 data were added to provide a richer developmental picture.

Results show that the acquisition of new temporal forms in L2 does not necessarily imply the acquisition of their TL functions. Although advanced L2 speakers have come a very long way in acquiring TL structures, at the level of organising temporal information they remain fixed in some principles typical of their L1, resulting in inconsistencies in L2. Inconsistencies typical of advanced English discourse produced by Czech and Hungarian learners include overgeneralisations of phasal segmentation and combinations of temporally incompatible expressive means such as linking inchoative structures directly to endpoints. Mismatches like these demonstrate that the reorganisation of discourse principles in moving from L1 to L2 is a great challenge even at very advanced stages of L2 acquisition.