Font Size: 

Pointing signs are used for pronominal reference (among many other functions) in sign languages. In terms of their form, many pointing signs do not look very different from pointing gestures in non-signers (Kendon 2004; Kita 2003c). Indeed, for this reason, Evans and Levinson (2009) recently suggested that sign languages lack pronouns entirely, and thus present a test case for linguistic diversity. Most sign language researchers, regardless of their theoretical perspective, assume (and would probably explicitly argue) that there is evidence for considering these pointing signs to be pronouns (i.e., distinct from pointing gestures as used by non-signers). In this paper, we compare canonical properties of a) pronouns in spoken languages, b) pointing gestures used by non-signers, and c) pronominal pointing signs in sign languages. Such a comparison is crucial for supporting (or challenging) assumptions that arise from applying linguistic concepts and terminology from one language/modality to another. Such comparisons lead us to two complementary conclusions. Firstly, the features that make pronominal signs difficult to characterize morphosyntactically (e.g., how person/participant role is expressed) are those features they share with pointing gestures and not with pronouns. Secondly, the features that make pronominal signs difficult to characterize gesturally (e.g., syntactic distribution) are those features they share with pronouns and not with pointing gestures. Therefore, we conclude that pronominal signs cannot be characterized simply either as pronouns, or as pointing gestures. Pronouns, however, have been studied in far more detail than pointing gestures have, and more research on pointing gestures could provide the necessary data to support one or the other analysis. In particular, comparative work on pronominal signs and pointing gestures is very much needed.