CRG Timetable - Term 2: January-March 2010

Monday 11 January 2010

Kenneth Fordyce (Hiroshima University; Lancaster University)

The Use of Learner Corpora in Measuring the Effects of Pedagogical Interventions on Second Language Acquisition

This study used small learner corpora to compare the effects of explicit and implicit pedagogical interventions on the interlanguage pragmatic development of Japanese EFL learners. The intervention focused on epistemic stance, which previous studies have shown to be a difficult aspect of L2 acquisition. Free production data (both spoken and written) was collected from the participants shortly before (pretest), soon after (posttest), and approximately five months after (delayed posttest), the intervention. For both the spoken and written data, participants were asked to describe a picture and express an opinion on a specific topic. The spoken (transcribed) and written texts for the explicit (n=37) and implicit (n=44) groups were organized into learner corpora. Keyword analysis was used to look for evidence of significant differences in the frequency of use of the taught forms as a result of instruction. This analysis showed that the explicit intervention had a powerful short-term effect although attrition was clearly evident by the time of the delayed posttest, particularly in the case of spoken language. By contrast, the implicit intervention could be seen to have only a minimal impact on the learners’ use of the targeted forms.

Considering that the use of learner corpora to investigate the effects of classroom instruction of specific grammatical and lexical forms remains quite rare, it is hoped that this presentation will demonstrate that learner corpora enable more effective analysis of learners’ free response data. However, it is also necessary to consider possible limitations related to the small size of such learner corpora as well as certain transcription difficulties which arise when preparing learners’ speech for corpus analysis.

Downloads: abstract (pdf file); slides (pdf file)

↑top

Wednesday 27 January (4pm, Bowland North, SR 26)

Michael Toolan (Birmingham University)

How do literary texts generate reader emotions? A stylistic approach

A major attraction and effect of literary works, noted since antiquity, is that they often cause readers to be moved, to feel empathy or 'immersed' or involved, to a degree beyond (or different from) how much real and everyday reported events affect them. Are there particular elements or strands in literary texts (e.g., in short stories), isolatable with the aid of linguistic assumptions and corpus stylistic methods, which are especially instrumental in the creation of reader emotional involvement? And can proposed 'hotspots' of emotional engagement (such as the textual projections of individuals' consciousness via desire modality and evaluating mental process verbs, and such individual predicates as feel) be falsifiably demonstrated, by probing reader responses? The stories to be discussed will include John McGahern's "All Sorts of Impossible Things" and Alice Munro's "Passion".

↑top

Pls note changes in day, time and venue:

Fri 05 Feb, 1pm, Bowland North, SR 27

Jane Demmen (Lancaster University)

Investigating the construction of female characters in Shakespeare’s plays through key word-clusters

Shakespeare’s plays have a longstanding literary and cultural presence in the English language, and they also continue to generate academic interest and discussion. Much has been written about the portrayal of women in the plays, particularly in the literary critical tradition, but little empirically-based research exists into the differences between the dialogue of female and male characters. This study aims to help address the gap, using corpus linguistics methodology. Through an investigation of “key” word-clusters (recurrent word combinations of statistical significance; see Scott, 1999; Scott and Tribble 2006), I show that women and men in the plays are represented as using language differently in some respects, through variation in the kinds of formulaic language which they tend to use relatively frequently.
 
Unlike many previous studies of Shakespeare’s plays, this one is rooted in quantitative results derived from all the dialogue in the plays (not from selected extracts or selected characters). These results are categorised according to their functions in the data using a classification system adapted from Culpeper and Kytö (forthcoming), based on Halliday’s (e.g. 1994) interpersonal, textual and ideational metafunctions of language. I demonstrate selected results with particular functions that contribute to the stylistic construction of women and men in different ways, both as social groups and as individuals. Taking in a wide range of research from other corpus linguists and stylisticians, historical sociolinguists and literary critics, I argue that the results cast light on Shakespeare’s contribution to debates about the role and behaviour of women at the time the plays were written. I also show that some types of dramatic device which operationalise the communication of the play to the audience are more prevalent in female dialogue.
 
References
 
Culpeper, J. and Kytö, M. (forthcoming, 2010) Speech in Writing: Explorations in Early Modern English Dialogues. Cambridge: Cambridge University Press.
Halliday, M.A.K. (1994) An Introduction to Functional Grammar. (Second Edition). London and New York: Edward Arnold.
Scott, M.R. (1999) WordSmith Tools Version 3.0. Oxford: Oxford University Press.
Scott, M. and Tribble, C. (2006) Textual Patterns. Key words and corpus analysis in language education. Amsterdam and Philadelphia: John Benjamins Publishing Company.

↑top

wk 15. (08 Feb)

Neil Millar (Lancaster University)

Eye-movements in native speaker reading of learner collocation errors

The prevalence of formulaicity in natural language use points to an important role in the way language is acquired, processed and used. It is widely recommended that second language instruction should ensure that learners develop a rich repertoire of formulaic sequences. If this is justified, it follows that learner failure to use formulaic sequences should present some barrier to communication. However, few researchers have sought to objectively evaluate how learner deviations from the
target-language (formulaic or otherwise) impact on online processing. Operationalising formulaic sequence through collocation, this presentation reports two experiments which used the eye-movement paradigm to investigate how collocation errors produced by learners of English impact on processing in reading by native speakers.

In experiment 1, a word-by-word self-paced reading procedure was used to compare reading times for sentences containing a learner error (e.g. best partner) to sentences containing a formulaic native speaker equivalent (e.g. ideal partner). Using eye-tracking methodology, experiment 2 explored native speaker processing of comparable stimuli under more naturalistic reading conditions. Results show that, in comparison to a formulaic native speaker equivalent, (1) learner collocation errors are
associated with an increased and sustained processing burden, (2) that the size and duration of the burden is substantially greater for morphological errors (e.g. responsibility person vs. responsible person) than for lexical errors (e.g. heavy crime vs. serious crime), and (3) that morphological errors are detected earlier.

The findings support the widely asserted claim that formulaicity in language use is associated with processing advantages and provide empirical support for the importance of formulaicity in language
learning. Based on the results it is suggested that morphological errors are more overtly obvious and may hold greater potential to disrupt communication. Usage-based models form the basis for some hypotheses. In particular, it is proposed that differences in the size and time-course of participants' response to lexical and morphological errors may be indicative of the use of probabilistic knowledge to predict upcoming input. Theoretical and methodological implications for second language learning research are discussed.

↑top

Monday 01 March 2010

↑top

Monday 08 March 2010

↑top

Monday 15 March 2010

↑top