The role of pragmatics in focus intervention effects: An experimental study
Focus intervention effects (FIEs) refer to the phenomenon that an interrogative wh-expression cannot be preceded by a focus expression (e.g. a focused phrase associated with a word like "only", as exemplified by the Mandarin example in (1).
(1) *Zhiyou XIAOQI mai-le na-ben zazhi ne?
only Xiaoqi buy-ASP which-CL magazine SFP
'Which magazine is the one that only Xiaoqi buys?'
The majority of the existing analyses of FIEs attributes their ill-formedness to purely structural reasons in syntax or semantics (Beck 1996, 2006; Beck & Kim 1997; Kim 2006; Mayr 2014; Li & Law 2016; Erlewine & Kotek 2017; Kotek 2019; Demirok 2021; a.o.). A (perhaps simplified) prediction of these accounts is that FIEs are categorically ungrammatical. However, as pointed out by Tomioka (2007, 2020), the judgments of FIEs are far from uniform and the variability among native speakers is vast. More recent pragmatic accounts even suggest that FIEs disappear if a context is provided that turns a focus expression into background information (Eiliam 2011, Glasbergen 2022). The variability of the judgments has been commonly acknowledged in the past, but it has rarely been addressed.
The goal of this talk is to report experimental evidence that adjudicates between the grammatical view and the pragmatic view of FIEs. Based on four experiments probing two pragmatic factors, namely, informativity and contextually given focus, we argue that while pragmatics reduce the strength of FIEs, such effects do not reduce entirely to pragmatics. Finally, this talk will also compare FIEs with island effects of wh-movement, concluding that both effects reside at least partially within natural language grammar.
Haoze Li is an assistant professor at Nanyang Technological University. His research interest lies primarily in semantics and its interactions with syntax. He received his Ph.D from New York University. He has worked on topics including form-meaning mismatches, focus—interrogative interactions, measurement, and sentence-final particles. His research methods are not restricted to formal semantics, but also include experimental linguistics.