Linguistics: Semantics

The study of sign language semantics aims to describe how meaning is communicated through sign languages. Clearly, many individual signs carry their own meaning; the study of what individual words and signs mean and how the mind classifies them is called lexical semantics. One major categorization of words is into “open-class” (content) words and “closed-class” (function) words. Open-class words include most nouns (e.g., HOUSE, SHOES), verbs (DANCE, READ), and adjectives (red, friendly), which can, and are, constantly being created anew in a language (e.g., EMAIL) and comprise most of the content of an utterance. Because the meanings of open-class words are tied closely to the cognitive science of concepts, the study of the classification and acquisition of content words is frequently the focus of studies in psychology. In contrast, closed-class words (e.g., NOT, FINISH, NO ONE, IX, IF, SELF) provide the structure for the sentence and are typically added as new items to the language at a much slower pace. These have been the focus of much of natural language semantics research in linguistics for both spoken and sign languages and have contributed greatly to our understanding of compositional semantics—the study of how the meanings of individual words combine to provide the meaning of a sentence.

Locations in Space

One important topic of inquiry in the compositional semantics of sign languages is the representation and contribution of space to sentence meaning. Unique to the visual/spatial language modality, sign languages are able to associate areas of space, known as loci, with things and people that signers want to continue to talk about (“discourse referents”), so that each time a signer returns to the same area in signing space, he or she can refer back to the same referent. This is accomplished by either directly pointing to the area of space uniquely associated with such referent or by having this area incorporated into another sign.

The areas of space glossed in example 1 as a and b are uniquely associated with Mary and Peter, respectively. There are three main types of analyses of these loci in sign languages. The first, based primarily on work by Scott Liddell, takes space to be essentially gestural and so not amenable to analogy to [Page 638]existing linguistic phenomena. The other two analyses draw a comparison to existing phenomena in spoken languages, either as semantic indices or as semantic features, each of which is briefly described in the following.

Semantic Indices

Some researchers—for example, Diane Lillo-Martin and Edward Klima—have argued that loci are pronounced phonological or morphological realizations of semantic indices that have previously been proposed for spoken languages, but which in spoken languages are never pronounced. They suggest that loci simply signal what spoken languages cannot: overt manifestation of the system of keeping track of discourse referents (i.e., the “assignment function”) often associated with anaphoric expressions like pronouns. Implementing this view in example 1, IX (articulated using a closed-fist handshape with the index finger extended, i.e., pointing) is an anaphoric expression (like a pronoun): In location a, IX is necessarily interpreted as Mary, while in location b, IX is interpreted as Peter. One argument in favor of this view is that the assignment of loci is arbitrary: Any particular (nonpresent) discourse referent could be assigned to any location in signing space. A second similarity between indices and loci is the ability to refer to different entities while maintaining the same phonological characteristics of the lexical item.

...