Language control and lexical access in diglossic speech production: Evidence from variety switching in speakers of Swiss German

Abstract

Diglossic speakers of two varieties of a language switch between their language varieties as bilinguals do between their languages, instead of having the otherwise typical probabilistic distributions of variants across varieties. To investigate the mechanisms involved in diglossic language control and lexical access, we conducted two switching experiments, in which participants who were highly proficient in the two varieties, or languages, named pictures in Swiss German or Standard German (Experiment 1), or in Swiss German or Tamil (Experiment 2), with pictures having cognate or non-cognate names in Swiss German and Standard German. The overall magnitude of switch costs in diglossic switching was comparable to switch costs in bilingual switching. Switch costs were asymmetric, favouring switching into the sociolinguistically preferred variety (Swiss German) in Experiment 1 and the slightly less dominant language (Tamil) in Experiment 2. Cognate gains were higher for Swiss German than Standard German and occurred even for Tamil responses. These results are compatible with a model that assumes that diglossic switching is influenced by similar mechanisms of language control as bilingual switching, that inhibitory processes are involved, that the amount of lexical inhibition depends on the relative strengths of lexical representations and on top-down control, and that diglossic cognates have integrated lexical representations.