The grammar of space in two new sign languages∗
Carol Padden
University of California, San Diego
Irit Meir
University of Haifa
Mark Aronoff
Stony Brook University
Wendy Sandler
University of Haifa
Sign languages use space because they can.1 In previous work on verb agreement in sign languages, we have discussed “the ability of a language produced in space to represent certain spatial and visual concepts iconically” (Aronoff, Meir, & Sandler, 2005). We resolved in that work what we called “the paradox of sign language morphology.” Although all sign languages that had been well studied up to that point showed a particular form of complex simultaneous non-affixal verb agreement that has no simple parallel in the morphology of spoken languages, they did not show much ‘run of the mill’ sequential affixal morphology. Why should a language acquire complex morphology before it acquires simple morphology, why sign languages, and why this particular sort of morphology? We argued that the agreement morphology of sign languages is based on an iconic use of space, which sign languages accommodate readily, and that this iconicity is what leads to the quick development of the system. Linear affixal morphology, by contrast, is much slower to emerge and much more varied, precisely because it is not iconic.
In this paper, we will focus a much finer lens on the iconically based grammatical use of space in sign languages. Specifically, we will look at the actual production of verb forms where we expect space to be used. We will compare forms produced across two or three generations of signers of two young sign languages, Al-Sayyid Bedouin Sign Language (ABSL) and Israeli Sign Language (ISL). ABSL is a village sign language that has emerged in a socially insular community in the last seventy years and which we have been documenting for several years (Aronoff, Meir, Padden, & Sandler, 2008; Sandler, Meir, Padden, & Aronoff, 2005) ISL has a similarly short history but it is widely used by many Deaf people in Israel of different language backgrounds. To our own surprise, the use of space in the utterances we elicited diverges quite dramatically from what we had always taken to be the norm among sign languages, in ways that we will discuss in detail below. We also found interesting differences between the two languages and also generational differences within each language. We learn from this study that not even a highly motivated grammatical system like agreement in sign languages emerges overnight, but rather that it may unfold gradually over generations and may take different courses of development. This is not to say that the use of space in ABSL and ISL is unsystematic. To the contrary, we find systematicity in the use of space at all stages, but not in the same way that we had previously found in certain other well-studied sign languages. We propose that early systematicity in a new sign language is powerfully influenced by the signer’s own body. Specifically, the signer exploits the iconicity of his or her own body in the structure of verb forms in the new sign language. Both ISL and ABSL use this iconicity, but they differ from one another in how they balance the iconicity of the signer’s body against other “competing iconicities,” notably spatial directional movement, also emerging in the sign language. These differences may help us to understand how the emergence of a language in a village differs from the way a language develops in a less homogeneous community.
∗ Our work is supported by grants from the National Institute on Deafness and other Communication
Disorders (R01 DC 6473) and the Israel Science Foundation (#553/04). Thanks to Sara Lanseman for
her help in obtaining the ISL data, and to Adi Lifshitz for her help in coding and organizing the ISL
data.
1 We thank Yoav Moriah for this seminal thought.
Background
Broadly, verb forms in many established sign languages divide between those that move in space in front of the signer’s body (agreement and spatial verbs) and those that do not, but instead are anchored to the body (plain verbs). Plain verbs lack the complex morpho-syntactic marking that characterizes verbs involving movement in space, though they do inflect for aspect. Semantically, plain verbs are typically cognitive, emotional or stative in nature. Verbs that exploit space are further divided between those that mark for person and number of the subject and object (agreement verbs), and those that do not (spatial verbs) (Liddell, 1977; Padden, 1988). The distinction between the two verb classes is grounded in their semantics: agreement verbs denote transfer events, whereas spatial verbs denote the motion of an entity in space (Meir, 2002). In a recent study, Thompson, Emmorey & Kluender (2006) used an eye-tracking device to locate signers’ eye gaze during the production of verbs. They identified distinctive eye gaze behavior for each of the three classes of verbs, supporting the view that no two classes can be collapsed into one archetype. Padden (1988) and Liddell (2003) treat agreement and spatial verbs as distinct subtypes because of different grammatical behavior. For example, person marking on agreement verbs in ASL is not specifically indexical but “in the general direction of”, resulting in more gross indexical differences between first person and second and third in agreement verbs. In comparison, spatial verbs have more fine or “gradient” locative distinctions in space. The two subtypes also differ with respect to number inflection. Multiple plural inflection adds a sweeping movement at the end point of agreement verbs but spatial verbs cannot employ this form. Because spatial verbs do not mark person and number inflections, they are ostensibly more free to exploit space in front of the signer’s body, having more distinctive locations and movements available to them. The two types also differ in that there are apparently no “backwards” spatial verbs. In backwards agreement verbs like TAKE, COPY, CATCH, RECEIVE and STEAL, the direction of the verb is reversed: the subject is goal rather than source as is the case with regular agreement verbs, and consequently the verb's path moves from the object’s referential location in space, or the R-locus, to that associated with the subject, the recipient of the transfer event (Meir, 2002; Padden, 1988). Because spatial verbs do not involve “transfer,” they also do not have recipients. The spatial verb MOVE can have directional movement inward toward the body, but it does not change the thematic role of the subject from agent to recipient, only the referential interpretation of the source and goal. MOVE with an inward movement toward the signer’s body can mean ‘to move near my location (as opposed to another more distant location).’
Morphology aside, agreement and spatial verbs can look very much alike. Both agreement and spatial verbs have directional movement from source to goal.2 In ASL as well as other sign languages, there are pairs of verbs such as GIVE (agreement) and CARRY-BY-HAND (spatial) that are formationally identical and become distinct only when inflections are added. In a psycholinguistic experiment, Cormier (1998) performed detailed measurements of signers’ movements and hand placements when producing agreement and spatial verbs and found less distinctiveness in the referential interpretation of their initial and final points than their grammatical analysis would suggest. Another grammatical system of sign languages that exploits space and iconicity is the system of classifier constructions. While these constructions share many characteristics with spatial verbs (Padden 1988, Sandler and Lillo-Martin 2006), they have subcategories and properties of their own and we will not be dealing with them in any detail here.
2 Indeed, Fischer and Gough (1978) refer to the two types of verbs as “directional verbs.”
This paper is actually much more longer and detailed, but here is the conclusion:
Conclusion
To a great extent, language is determined by our physiology: phonological features are determined by our articulatory system (features become grammaticalized within a phonetic space) and the way we conceptualize events is determined by our body. The
body is an important resource that new languages can rely on. But they may use it in different ways (for example, ABSL doesn’t seem to have role shift, but ISL does). Leaving the body and developing grammatical categories in the verb system takes time. The fact that, even after a language has developed such a system (ISL), the signer can still fall back on the body point of view, highlights the centrality of the body in sign language linguistic systems, and maybe in other cognitive systems as
well. Studying new sign languages provides novel evidence for a richer view of grammar than can be obtained from the investigation of spoken languages alone, as the latter are all old or descended from old languages. The gradual emergence of a verb agreement system, one that is a robust phenomenon among the youngest group of ISL signers studied here, but yet still not fully consistent among them, requires us to understand language fundamentally as a social system, characterized by the establishment and spread of a linguistic convention through interaction, both interand intra-generational, within a community.