Since the work of Woodward in the 1970s, researchers have proposed genetic groupings of signed languages based on comparisons of basic vocabulary. Their claims have typically relied on methods such as lexicostatistics. To date, no historical study of signed languages has successfully applied the comparative method to identify the types of regular correspondences that conclusively establish cognate vocabulary. I explore theoretical and methodological reasons for the failure of the primary method in historical linguistics when applied to signed languages. I argue that this failure to demonstrate cognacy stems from three problems, which I illustrate with examples from my research on signed language in Tajikistan: 1) the prevalence of iconicity and indexicality in the lexicons of all known signed languages, 2) differences in the transmission of signed languages compared to spoken languages, and 3) the lack of cross-linguistic lexical databases using comparable representations.
The prevalence of motivated signs has implications for whether lexical items in signed languages undergo the type of regular change that produces regular correspondences. Research on spoken languages suggests that sound symbolic forms may change in irregular ways (Malkiel 1994) or may not participate in otherwise regular sound changes (Mithun 1982, Kaufman 1995). Because iconic and indexical form-meaning relations are prominent in the lexicons of signed languages, change may be characterized by irregularity and regular correspondences may not exist. On the other hand, research on American Sign Language (ASL) identified sets of signs defined by sublexical units that have undergone similar (i.e., putatively regular) changes (Frishberg 1975, Battison et al 1975). In addition, studies of phonetic variation in ASL (Lucas et al 2001) and in Australian and New Zealand Sign Languages (Schembri et al 2009) seem to show similar changes in progress (e.g., lowering signs articulated at the forehead). I conclude that lexical items in signed languages undergo regular change, but that the prominence of iconicity and indexicality in the lexicons of signed languages yields irregular change and resistance to change. I discuss examples of putatively regular diachronic changes in signed language called symmetry and displacement (Frishberg 1975), which have numerous, previously-unexplained exceptions. I show that, once an appropriate conditioning environment is described, many—but not all—exceptions can be explained as showing irregular change or resistance to change due to iconicity or indexicality.
This analysis implies both that regular correspondences may exist amongst related signed languages but that their identification will be more complicated compared to the same task for related spoken languages. Thus, change in signed languages differs not in kind compared to change in spoken languages, but in proportion: the greater the presence of iconicity and indexicality in the lexicon, the more frequently irregular change and resistance to change will occur. While sound symbolism is sometimes treated as a marginal phenomenon in comparative historical work on spoken languages, signed languages provide the opportunity to study change in human languages which feature iconicity.