Do deaf signers of different signed languages do reference the same way? Here we compare how signers of five signed languages coordinate fully conventionalised forms (such as lexical manual signs, fingerspelling and/or spoken language mouthings) with more richly improvised semiotics (such as indicating verbs, pointing signs, depicting signs, visible surrogates and/or invisible surrogates) to identify and talk about referents of varying agency. The five languages are Auslan, Norwegian Sign Language, Finnish Sign Language, Swedish Sign Language and Irish Sign Language. Using 10 retellings of Frog, Where Are You? (Mayer, 1969) from each language, we analyse tokens of referring expressions with respect to: (a) activation status (new vs. maintained vs. re-introduced); (b) semiotic strategy (e.g. pointing sign, fingerspelling, enactment, etc); and (c) animacy (human vs. animals vs. inanimate objects), and assess how they are similar or different with regard to these parameters.

Statistical analysis reveals expected similarities across the five languages. For example, signers of each language typically used conventionalised forms to identify new referents, and less conventional strategies to maintain and reintroduce referents. Signers also preferred to enact animate referents, and manually depict or index inanimate referents. These patterns mirror observations from a larger corpus-based investigation of Auslan using the same method (Hodge, Ferrara & Anible, 2019). However, there are also some differences across languages. While Auslan and ISL signers frequently use fingerspelling to identify referents across all activation contexts, signers of Scandinavian signed languages chose to use other semiotic strategies. We also observed that patterns for specific semiotic strategies are more widespread in some languages than others. For example, Auslan signers prefer using depicting signs in maintained contexts; FinSL signers prefer using depicting signs in reintroduced contexts; while ISL, NTS, and STS signers tend to use depicting signs more equally across activation status. We suggest that doing reference in a signed language involves both cross-linguistic and ecology-specific strategies. The latter may be attributed to the different social and historical trajectories of each language, including possible language contact effects.