Distinguishing languages from dialects: A litmus test using the picture-word interference task

Linguists have been working to develop objective criteria for distinguishing languages from dialects for well over half a century. The prevailing view amongst sociolinguists is that no objective criteria can be formulated. The aim of this study is to examine whether language processing can provide insights into this problem by comparing bidialectal behavioural effects to bilingual effects reported in the literature. Previous research has demonstrated that when bilinguals name an object in Lx while simultaneously processing a translation equivalent distractor word in Ly, naming times are sped up relative to an unrelated condition (Costa, Miozzo, & Caramazza, 1999). Using the same methodology, we evaluated whether a comparable facilitation effect arises when the distractor word is a dialectal or register variant of the picture name. Across 5 experiments we found no trace of translation equivalent facilitation. Instead, we repeatedly observed between-dialect and between-register interference, in contrast to the between-language facilitation effect. This behavioural divergence between bilingual vs. bidialectal processing suggests that this paradigm could provide an objective litmus tests for identifying the boundary between dialects and languages.