The impact of register on the use of the passive by L2 and L1 writers
(2025) Register and task variation in Learner Corpus Research (VAR4LCR) conference — Location: Université catholique de Louvain (7.July.2025)
Files
No attached file found for this publication.
Details
- Authors
Granger, Sylviane 
- Abstract
- The passive is a highly register-sensitive linguistic feature. Previous studies on register variation in the use of the passive have primarily focused on L1 English (Svartvik, 1966; Granger, 1983; Biber,1988; Biber et al., 1999), leaving L2 English underexplored. Within learner corpus research (LCR), attention has largely been directed toward other aspects of passive use, in particular frequency of use, which has generally been found to be markedly lower in L2 corpora compared to L1 corpora (Hinkel, 2004; Xiao, 2007; Granger, 2013). In addition to analyses of its error patterns (Oshita, 2000; Cowan et al., 2003), the passive has also been studied within the framework of lexical bundles research (Ädel & Erman, 2012; Chen & Baker, 2010; Lu & Deng, 2019). This study seeks to address the gap in register variation research in LCR by examining passive use in corpora of argumentative essays and discipline-specific research papers written by L2 English learners and comparing them with L1 English texts from corresponding registers. The following corpus data were used: - Argumentative essays from the International Corpus of Learner English (ICLE) (L2 English from four populations: Dutch, French, Norwegian, Swedish) compared to the Multilingual Editorials corpus (Mult-Ed) (expert L1 English) - Discipline-specific research papers (linguistics) from the Varieties of English for Specific Purposes dAtabase (VESPA) (L2 English from the same L1 populations as ICLE) compared to the Louvain Corpus of Research Articles (LOCRA) (expert L1 English) Be + past participle structures were automatically extracted from the POS-tagged versions of the corpora and subsequently subjected to manual screening to eliminate tagging errors and constructions that did not meet our definitional criteria. The following aspects were investigated: frequency of use of the passive, agentful vs agentless passives, and impersonal passive structures. Our overarching hypothesis was that, given the typically less personal style of research papers than argumentative essays (cf. Larsson et al., 2021), passives, agentless passives and passive impersonal structures should be more frequent in research papers than in argumentative essays. Preliminary results reveal an overall overuse of the passive among learners in both argumentative essays and research papers, which goes against most studies on the passive but confirms results from some lexical bundle studies (e.g. Pan et al., 2016; Yakut et al., 2021). Following the trend of the L1 corpora, the passive is more frequent in L2 research papers than in L2 argumentative essays. This confirms our hypothesis and suggests that the students in ICLE and VESPA, who have reached a relatively advanced proficiency level, are able to adapt their use of the passive to this register difference. Further supporting our hypothesis, the learner corpus analysis shows that, despite an overwhelming predominance of agentless passives in both registers, there are more agentful passives in L2 argumentative essays than in L2 research papers. There are also more impersonal passive structures in L2 research papers than in L2 argumentative essays. For these two features, the L1 reference corpora do not display any significant difference. Impersonal passive structures in L2 include some unidiomatic and/or superfluous instances, and these are proportionally more frequent in argumentative essays than in research papers. We conclude by mentioning some avenues for research, such as distinguishing between different types of passives and agents or considering potential L1 transfer in the use of the passive.
- Affiliations
Citations
Gilquin, G., & Granger, S. (2025). The impact of register on the use of the passive by L2 and L1 writers. Register and task variation in Learner Corpus Research (VAR4LCR) conference, Université catholique de Louvain. https://hdl.handle.net/2078.5/274952