Addressee Identity and Morphosyntactic Processing in Basque Allocutive Agreement

Abstract

Information about interlocutor identity is pragmatic in nature and has traditionally been distinguished from explicitly coded linguistic information, including mophosyntax. Study of speaker identity in language processing has questioned this distinction, but addressee identity has been less considered. We used Basque to explore how addressee identity is processed during morphosyntactic analysis. In the familiar register hika, Basque has obligatory allocutive agreement, where verbal morphology represents the gender of a non-argument addressee. We manipulated the gender of the allocutive verb and the congruence of addressee gender in conversations between two interlocutors. Items with person agreement manipulations were included as a control comparison. Basque speakers familiar with hika completed speeded acceptability judgments and unspeeded, offline naturalness ratings for each conversation. Results showed a main effect of addressee identity congruence for naturalness ratings, but there was no main effect for addressee identity congruence for reaction times or accuracy in the acceptability judgment. Interactions and correlations with biographical data showed that the effect of congruence was modulated by the gender of the allocutive verb and that hika proficiency was related to participants' performance for the acceptability judgment. These results show an interaction between morphosyntactic and pragmatic information and are the first experimental data of allocutive processing. In comparison, clear effects were seen for the person agreement condition, indicating that person disagreement is more disruptive to processing than addressee identity incongruence. This study has implications for investigation of the role of extralinguistic information in morphosyntactic processing, and suggests that not all such information plays an equal role.

Who are you talking to? The role of addressee identity in utterance comprehension

“…On one hand, participants seemed to have a different frequency of exposure to feminine and masculine allocutive in their daily life. For instance, our participants reported masculine allocutive to be more used as compared to feminine allocutive (in line with previous literature, Echeverria, ; Wolpert et al, ). On the other hand, there was a difference in frequency of exposure within the experimental section as well, with masculine agreeing allocutive forms being heard more often than feminine agreeing allocutive forms (as person B was always male).…”

“…Auditory conversations were adapted from Wolpert et al () using Basque batua . All stimuli were recorded at 44.1 kHz and normalized to the average root mean squared amplitude.…”

“…They were instructed to listen carefully to the Basque conversations and rate their naturalness from one (not natural at all) to seven (very natural). This task has been reported to be more sensitive to the allocutive manipulation as compared to a grammaticality judgment task (Wolpert et al, ).…”

“…Finally, as an exploratory comparison it is interesting to look at ERP differences between masculine and feminine allocutive forms. Previous behavioral findings showed a difference in the way allocutive forms are processed, with more disruptive effects for the incorrect use of feminine as compared to masculine forms (Wolpert, Mancini, & Caffarra, ), which might be related to frequency differences (masculine forms of allocutive are used more often than feminine forms, including sometimes for female addressees; Echeverria, ). These disruptive effects will be possibly reflected by the ERP waveforms with a consequent effect of allocutive gender.…”

Addressee’s Factor in the Limerick Discourse

“…The recent scientific works analyse the role of an addressee factor in speech communication (Holler, 2011;Wilkin, 2011;Arnold, 2008), pay attention to issues of an addressee's communicative typology (Vorobyova, 1997;Karpchuk, 2006;Kaminskaya, 2009), analyse the categorial and functional essence of addressing (Wolpert, 2017;Mancini, 2017;Caffarra, 2017).…”