Variation in American Sign Language
Research on African American Signing
African American signing has been the object of several recent investigations. Studies include Aramburo (1989), Guggenheim (1993), Lewis, Palmer, and Williams (1995), and Lewis (1996). Aramburo and Guggenheim observed lexical variation during the course of structured but informal interviews. Lewis et al. (1995) studied the existence of and attitudes toward African American varieties. Specifically, they described the differences in body movement, mouth movement, and the use of space in the signing of one African American signer who codeswitched during the course of a monologue. In addition, they explored how interpreters handled the codeswitching in spoken language from Standard English to African American Vernacular English (AAVE). Lewis (1996) continued the examination of African American signing styles in his paper on the parallels between communication styles of hearing and deaf African Americans. His investigation took its departure from two observations: first, ASL users recognize the existence of what is often referred to as “black signing” but have difficulty in explaining what makes it black; second, uniquely black or “ebonic” (Asante 1990) kinesic and nonverbal features exist, and these features occur in the communication of both hearing and Deaf African Americans. His investigation described some of these kinesic and nonverbal features—specifically, body postures and rhythmic patterns accompanying the production of signs—in the language that a deaf adult African American woman used. The frequently articulated perspective that African American signing differs markedly from Caucasian signing in all areas of structure—and not just lexically—is thus beginning to be explored.
Perspectives on the Structure of Sign Languages
To understand variation in ASL, we also need to consider the changing perspectives on the basic structure of ASL and of sign languages in general. Current thinking about the linguistic structure of sign languages sheds new light on some of the earlier studies. At the same time, advances in our understanding of sign languages raise important issues for the analysis of the data we describe in this book. For example, the perspective on the fundamental structure of signs has changed dramatically since the earliest studies of variation. Stokoe’s perspective, which shaped sign language studies from 1960 until fairly recently, held that signs are composed of three basic parts or parameters (the location at which the sign is produced, the handshape, and the movement of the sign) and that, unlike the sequentially produced segments of spoken languages, these parts are produced simultaneously. In a more recent perspective developed by Liddell (1984) and Liddell and Johnson (1989), signs are viewed as composed of movement and hold segments, sequentially produced, somewhat analogous to the consonants and vowels of spoken languages. In this model, the handshape, location, orientation, and nonmanual signals constitute a bundle of articulatory features that are a part of each movement or hold segment. Currently there is considerable debate about the nature of the segments in question (see e.g., Coulter 1992, Sandler 1992, and Perlmutter 1992). We have chosen to work within the Liddell-Johnson framework because, as Liddell (1992) amply demonstrates, it allows not only for an accurate account of the description of any individual sign but also for an accurate account of phonological processes such as assimilation, metathesis, epenthesis, and segment deletion, processes that play central roles in variation.
Naturally, a central concern of any variation study is to define clearly the linguistic variables under examination and to ensure that they are indeed variable. Current thinking about the linguistic structure of sign languages and about data collection methodology (we address the latter in chapter 2) has implications for the identification of variables in the earlier studies of sign language variation. It is not clear, for example, that the rules of negative incorporation, agent-beneficiary directionality, and verb reduplication in Woodward’s (1973b, 1973c, 1974) studies are actually variable in native signer ASL. The apparent variability of these rules merits reexamination because the variability observed may simply have been an artifact of combining data from native and non-native signers. For example, in terms of the semantic continuum proposed for agent-beneficiary directionality (from signs that have “beneficial” connotations to those that have “harmful” connotations), directionality may be obligatory in most of the verbs in question and unrelated to semantic considerations. It is basically the way in which agreement is shown with the subject and the object of the verb and is not optional. Failure to use space properly in these verbs would seem to result not in a variable form but in an ungrammatical one. Although the semantic categorization does seem to work for some verbs (e.g., “beneficial” for give and “harmful” for hate), it does not work at all for others. For example, it is not clear at all why fingerspell would be labeled as “harmful.” It may be that at the time of the study, fingerspell as an agreement verb was an innovation and hence not widely attested, placing it at the “less frequent use of directionality” end of the continuum. But fingerspell cannot therefore be said to have a semantic characteristic of “harmful,” the researchers’ account of this end of the continuum that they set up.