Connector placement in EFL learner writing: Focus on 'however'

Dupont, Maïté [UCL] Granger, Sylviane [UCL]

The use of adverbial connectors is a very popular topic in learner corpus research. This is perfectly justified in view of the important role they play in building clear and convincing argumentation and the difficulty even advanced learners experience in using them appropriately in academic writing. Most studies so far have compared the frequency of connectors in learner versus native reference corpora (e.g. Lenko-Szymanska 2008). These comparisons have yielded contradictory results, some pointing to a general overuse (Lei 2012, Güneş 2017), some to overall underuse (Altenberg & Tapper 1998), while others do not reveal any quantitative difference (Granger & Tyson 1996). Native/non-native comparisons have also highlighted differences in learners’ preferred semantic categories of connectors (adversative, causal, etc.). One aspect that has rarely been investigated is placement. A key characteristic of adverbial connectors is that they are mobile. According to Quirk et al. (1985: 643), the default position in English is the initial position; medial position is “quite normal” for conjuncts that cannot be misinterpreted in this position, while final position is restricted to a handful of connectors. The few studies that have studied connector position in learner corpora point to an overuse of initial position (e.g. Narita & Sugiura 2006, Van Vuuren & Berns 2018). However, they tend to be based on very small datasets and cover a limited number of L1 populations. In addition, while the positioning of connectors induces a range of rhetorical effects (e.g., in the case of however, emphasising a contrast between two ideas or laying focus on some parts of the message), this aspect has hardly been investigated in learner corpus research. The aim of our study is to investigate connector placement in a large learner corpus covering a variety L1 populations through the lens of the highly frequent connector however. Learners’ placement patterns will be compared to those observed in the argumentative writing of both (i) expert native writers; and (ii) novice native writers of English. Thus, the influence of both L1 and degree of expertise (expert vs novice) will be assessed. More specifically, the study attempts to answer the following research questions: 1) Does the placement of however by EFL learners differ from the placement preferences of (i) expert native writers and (ii) novice native writers of English? 2) Are there differences between the different L1 learner populations, and do some L1 populations better approximate the placement preferences of expert or novice native writers? 3) Are the different connector positions associated with similar rhetorical effects in learner, novice and expert writing? The study is based on a 5-million-word corpus of argumentative texts written by EFL learners from 24 L1 learner populations extracted from the third version of the International Corpus of Learner English (ICLEv3) (Granger et al. 2020). The placement preferences of expert writers are identified on the basis of a 2-million-word corpus of newspaper editorials, a genre which is arguably close to the argumentative essays of the ICLE in terms of both length and overall communicative purpose (cf. Neff et al. 2004). The LOCNESS (Louvain Corpus of Native English Essays) was used to analyse the patterns of novice native speakers of English. All the occurrences of however were extracted and disambiguated so as to weed out premodifying uses. The disambiguated data set, amounting to 8,000+ occurrences of however, was categorized on the basis of Dupont’s (2021) study of the placement of connectives of contrast. The adverb classification is inspired from Halliday’s Systemic Functional description of thematic structure (see Halliday & Matthiessen 2014) and distinguishes between five positions (thematic 1 and 2 and rhematic 1, 2 and 3), exemplified in (1) to (5). (1) However, members of the royal family tend to act without thinking. (2) Worryingly, however, our survey also found that […]. (3) Russia, however, has failed to rise to the challenge of creating a real democracy. (4) Mr Adams has, however, stopped short of recommending […]. (5) The cute comparisons don't always apply, however. Preliminary results strongly confirm that learners of English markedly overuse however in thematic 1 (i.e. initial) position as compared to expert writers. By contrast, learners tend to largely underuse however in rhematic 1 position, i.e. after the subject or a fronted adjunct (as in example (3)). These two tendencies are observed across all 24 learner populations, thereby pointing to a developmental pattern rather than an L1-induced feature. Once all the data has been analysed, the random forest technique (Levshina 2020) will be applied to assess the relative importance of degree of expertise and L1 on the placement of however. The rhetorical effects achieved through connector placement by the three groups of writers will also be compared. Although our study is focused on only one connector, we will highlight some of its wider implications for the teaching of connectors in general. References: Altenberg, B. & Tapper, M. (1998). The use of adverbial connectors in advanced Swedish learner’s written English. In S. Granger (Ed.) Learner English on Computer. London: Longman, 80-93. Dupont, M. (2021). Conjunctive Markers of Contrast in English and French: From Syntax to Lexis and Discourse. Amsterdam & Philadelphia: Benjamins. Granger, S., Dupont, M., Meunier, F., Naets, H., & Paquot, M. (2020). The International Corpus of Learner English. Version 3. Louvain-la-Neuve: Presses universitaires de Louvain. Granger, S. & Tyson, S. (1996). Connector Usage in the English essay writing of native and non-native EFL speakers of English. World Englishes, 15(1), 17-27. Güneş, H. (2017). A corpus-based study of linking adverbials through contrastive analysis of L1/L2 PhD dissertations. International Journal of Curriculum and Instruction, 9(2), 21-38. Halliday, M.A.K. & C. Matthiessen (2014). Introduction to Functional Grammar. London: Routledge. Lei, L. (2012). Linking adverbials in academic writing on applied linguistics by Chinese doctoral students. Journal of English for Academic Purposes, 11, 267-275. Lenko-Szymanska, A. (2008). Non-native or non-expert? The use of connectors in native and foreign language learners’ texts. Acquisition et interaction en langue étrangère, 27: 91-108. Levshina, N. 2020. Conditional inference trees and random forests. In S. T. Gries & M. Paquot (Eds.). A Practical Handbook of Corpus Linguistics. Springer: New York, 611-643. Narita, M. & Sugiura, M. (2006). The use of adverbial connectors in argumentative essays by Japanese EFL college students. English Corpus Studies, 13, 23-42. Neff, J., Ballesteros, F., Dafouz, E., Martínez, F., Rica, J.P., Díez, M., & Prieto, R. (2004). Formulating writer stance: A contrastive study of EFL learner corpora. In U. Connor & T. Upton (Eds.). Applied Corpus Linguistics: A Multidimensional Perspective. Amsterdam: Rodopi Brill, 73-89. Quirk, R., Greenbaum, S., Leech, G., & Svartvik, J. (1985). A Comprehensive Grammar of the English Language. London: Longman. Van Vuuren, S. & Berns, J. (2018). Same difference? L1 influence in the use of initial adverbials in English novice writing. IRAL, 56(4), 427-461.