A novel comparative method for young sign languages: Base comparison
As studies of diverse sign languages (SLs) have grown, scholars have been interested in establishing historical and synchronic relationships between them. This has traditionally been done by identifying lexical cognates between SLs, using versions of the Swadesh (1971) basic vocabulary list, most famously that of Woodward (1991). The most common method of identifying cognates is by using sublexical contrastive parameters of signs. A sign is made up of three major contrastive parameters: handshape, location (in space or on the body), and movement pattern (cf. Brentari, 2011). Figure 1 demonstrates minimal pairs on these three parameters in Israeli SL.
Scholars traditionally establish cognates between SLs depending on how many contrastive parameters vary between two signs. This method has been used to hypothesise relationships between SLs of the Australian North and Central Deserts (Kendon, 1988); between Auslan, New Zealand SL and British SL (Johnston, 2003); and between Spanish SL, Mexican SL, French SL and Japanese SL (Guerra Currie, Meier, & Walters, 2002). In order for this lexicostatistical method to work, however, signs need to have a prototypical, "citation" form, where signers adhere to sublexical contrastive parameters. If signers do not do so - if there is significant intra-signer variation in form - then the sublexical parameter comparison method is not effective in quantifying relationships beteween SLs.
This work is based on fieldwork with 12 deaf signers in the Nebilyer/Kaugel region of rural Western Highlands, Papua New Guinea. The author administered a wordlist to these signers, most of whom did not know one another and who lived in different communities. Signers in this area do have established signs for referents, such as MAN which is articulated on the chin, referring to the beard of an adult man. However, the form of this sign varies even by single signers. Figures 2 and 3 demonstrates two tokens of MAN, articulated by the same signer within 20 seconds. Even by the same user, in very close proximity, two tokens of the same sign differ in terms of the usually contrastive parameter of handshape.
As such, the standard parameter comparison method is not appropriate for these rural SLs. In order to explore sign sharedness between these SLs, I instead consider the baseof each sign; that is, the "object or action that the production of the sign is derived from" (Kendon, 1980, p. 83). All signs in Nebilyer/Kaugel are iconic; that is, they have a motivated form-meaning relationship. Hence, all signs have a base, and thus we can compare these bases. Even iconic signs carry a measure of arbitrariness in that there are a number of bases that could be selected (Planer & Kalkman, 2019). For example, signs for PIG I have recorded across Papua New Guinea reference the pig's ears; its snout; traditional tying methods; its rooting in the earth; and so on.
While similar identification of sign base has been employed by Hou (2016), and Richie, Fanghella and Coppola (2012), these studies were not specifically directed at exploring relationships between SLs. The base comparison method I present is a valuable tool for scholars working with young and/or rural SLs such as those in Papua New Guinea, where signs are best characterised as wholes, lacking stable internal structure. It provides an effective alternative to standard sublexical comparative parameter models as discussed above.
This paper begins by presenting the theoretical evidence for base as a valid comparative element of signs. I discuss how I created my wordlist, collected data, coded, and carried out statistical analysis. I present challenges with the method, such as when signers produced multiple signs with different bases throughout the session for a given referent. Finally, I present the results of base comparision for seven SLs of Papua New Guinea. Levels of base similarity pattern highest for five SLs in the Nebilyer/Kaugel rural area, while base similarity is lowest between those languages and two other SLs from other parts of Papua New Guinea. Base comparision is shown to be a promising and effective tool for addition to the sign linguistic toolkit.
(Full abstract available at: https://als.asn.au/Resources/PageContent/Files/5efcac11-1ab8-4f31-a82f-85741e2fd876.pdf)
