Pick a Fight or Bite your Tongue: Investigation of Gender Differences in Idiomatic Language Usage

Proceedings of the 28th International Conference on Computational Linguistics, pages 5181–5192 Barcelona, Spain (Online), December 8-13, 2020 5181 Pick a Fight or Bite your Tongue: Investigation of Gender Differences in Idiomatic Language Usage Ella Rabinovich1 Hila Gonen2 Suzanne Stevenson1 1Dept. of Computer Science, University of Toronto, Canada 2Dept. of Computer Science, Bar-Ilan University, Israel {ella,hilagnn}@gmail.com suzanne@cs.toronto.edu Abstract A large body of research on gender-linked language has established foundations regarding cross- gender differences in lexical, emotional, and topical preferences, along with their sociological un- derpinnings. We compile a novel, large and diverse corpus of spontaneous linguistic productions annotated with speakers’ gender, and perform a first large-scale empirical study of distinctions in the usage of figurative language between male and female authors. Our analyses suggest that (1) idiomatic choices reflect gender-specific lexical and semantic preferences in general language, (2) men’s and women’s idiomatic usages express higher emotion than their literal language, with detectable, albeit more subtle, differences between male and female authors along the dimen- sion of dominance compared to similar distinctions in their literal utterances, and (3) contextual analysis of idiomatic expressions reveals considerable differences, reflecting subtle divergences in usage environments, shaped by cross-gender communication styles and semantic biases. 1 Introduction Distinctions in language usage between men and women open a window into the differences in their perception of the world around them. Over the years, these differences have been studied extensively, both in empirical linguistics (Lakoff, 1973; Labov, 1990; Coates and Pichler, 2011), and using data- driven computational approaches (e.g., Koppel et al. (2002), Schler et al. (2006), Mihalcea and Garimella (2016)). Striking cross-gender deviations in lexical and syntactic properties have been identified, igniting debate as to their underlying psycholinguistic and sociological interpretations (Lakoff, 1973; Cameron et al., 1988; Bergvall, 2014). In contrast to the study of literal linguistic utterances, only sparse prior research exists on gender differences in figurative language (for review see Colston and Katz (2004)). Figurative language is a common communicative device in which utterances use words in ways that deviate from their literal interpretation (e.g., Nunberg et al. (1994)), including a range of devices, such as irony, sarcasm, personification, and idiomatic expressions. Various communicative motivations for non- literal language have been proposed, such as distinguishing different intensities of the same emotion, and communicating ideas and subjective experiences more vividly (for a review, see Gibbs (1994)). Indeed, figurative language has been suggested to reflect the way that concepts are structured within human thought (e.g., Lakoff and Johnson (2008)). Consequently, differences in the way men and women perceive experiences and emotions give rise to a plausible hypothesis that male (M) and female (F) authors differ in their preferences in the choice of non-literal language; however, there has been no large-scale computational study conducted on the distinctions in such usage patterns. This gap in the research stems in part from the sparsity of large-scale corpora of linguistic productions employing figurative language, annotated for the gender of a speaker. The increasing popularity of online discussion platforms offers an unprecedented opportunity …

Content truncated to 500 words