Presentation Open Access
Nini, Andrea; Cameron, Michael; Murphy, Colin
Usage-based approaches predict that each person possesses a unique linguistic system because each person’s experience and interpretation of the linguistic input is also arguably unique. This hypothesis implies that, all other factors being equal, each person has a unique set of preferred variants of lexicogrammatical alternations. Corpus research conducted so far on idiolect or individuality in language is indeed largely consistent with this prediction (Barlow 2013; Anthonissen & Petré 2019; Schmid & Mantlik 2015), and so is the real experience of forensic linguists working on cases of disputed authorship (Coulthard, Johnson & Wright 2017) and computational authorship attribution research (Stamatatos 2009; Grieve 2007). However, despite some relevant work (Dąbrowska 2018; Dąbrowska 2012), not enough experimental research has been conducted to test this hypothesis.
The present paper reports on new experimental evidence on the individuality of sets of lexicogrammatical variants. A sample of eight humanities academics who are native speakers of British English was recruited to perform a test in a psycholinguistic laboratory. These participants were asked to select the most appropriate variant of an alternation to fill in a gap in a sentence for twelve categories of lexicogrammatical alternations (e.g. particle placement, relative pronouns, genitive alternation, help vs. help to, negation not vs. un-). The test sentences were designed so to isolate other well-known factors that influence the choice of variant and to be consistent with an academic written register by using real sentences from the British Academic Written English corpus as model. Participants took the test twice, with the second repetition after at least two weeks from the first. The participants were not told that they would have taken the same test twice and for the second repetition the order of the test sentences was changed randomly and some of the non-influencing lexical items were altered.
The results of the experiment constitute evidence in favour of the hypothesis of unique and personalised lexicogrammatical systems. Despite the highly similar background of the participants, no two sets of alternations were identical and the percentage of similarity between same speaker’s tests was significantly higher than between different speaker’s tests. This result suggests that the uniqueness found is not just arising from chance but is likely to be a reflection of the participant’s idiolect. The paper will deal with the consequences of this finding for both theoretical and forensic linguistics.
Anthonissen, Lynn & Peter Petré. 2019. Grammaticalization and the linguistic individual: new avenues in lifespan research. Linguistics Vanguard. doi:10.1515/lingvan-2018-0037.
Barlow, Michael. 2013. Individual differences and usage-based grammar. International Journal of Corpus Linguistics 18(4). 443–478. doi:10.1075/ijcl.18.4.01bar.
Coulthard, Malcolm, A Johnson & David Wright. 2017. An Introduction to Forensic Linguistics. London: Routledge.
Dąbrowska, Ewa. 2012. Different speakers, different grammars. Linguistic Approaches to Bilingualism 2(3). 219–253. doi:10.1075/lab.2.3.01dab.
Dąbrowska, Ewa. 2018. Experience, aptitude and individual differences in native language ultimate attainment. Cognition 178(May). 222–235. doi:10.1016/j.cognition.2018.05.018.
Grieve, Jack. 2007. Quantitative authorship attribution: An evaluation of techniques. Literary and Linguistic Computing 22(3). 251–270.
Schmid, Hans-Jörg & Annette Mantlik. 2015. Entrenchment in Historical Corpora? Reconstructing Dead Authors' Minds from their Usage Profiles. Anglia 133(4). 583–623. doi:10.1515/ang-2015-0056.
Stamatatos, Efstathios. 2009. A survey of modern authorship attribution methods. Journal of the American Society for Information Science and Technology 60(3). 538–556. doi:10.1002/asi.21001.