Sunday, February 06, 2005

Cool Linguistics Stuff

Linguists have often used sign languages, and other recently constructed languages (e.g., pigdins), to understand the structure and evolution of language in general. Recently, linguists have studied a very unique sign language that they believe may provide unique insights on both of these issues. Here's part of the news release:
The Al-Sayyid Bedouin Sign Language (ABSL), which serves as an alternative language of a community of about 3,500 deaf and hearing people, has developed a distinct grammatical structure early in its evolution, researchers report, and the structure favors a particular word order: verbs after objects.

The study – the first linguistic analysis of a language arising naturally with no outside influence – is being published online in the Proceedings of the National Academy of Sciences the week of Jan. 31 to Feb. 4 [Though I can't find it there! - Chris].

...

"The grammatical structure of the Bedouin sign language shows no influence from either the dialect of Arabic spoken by hearing members of the community or the predominant sign language in the surrounding area, Israeli Sign Language," said study coauthor Carol Padden, professor of communication at UC San Diego. "Because ABSL developed independently, it may reflect fundamental properties of language in general and provide insight into basic questions about the way in which human language develops from the very beginning."

ABSL arose in the last 70 years and is now in its third generation of use. Remarkably, the fixed word order of ABSL emerged within a generation after the inception of the language.

"Our findings support the idea that word order is one of the first features of a language, and that it appears very early," Padden said.

The research also supports the notion that languages can and do evolve quickly.
Elsewhere, the authors of the article have drawn conclusions about the the development universal features of language from their research on this sign language, writing:

The [verb agreement] system [of sign languages] has regular and productive morphological characteristics that are found across all sign languages that have been well studied: (1) Only a subset of verbs are marked for agreement (Padden, 1988). (2) That subset is predictable on the basis of their semantics; they involve transfer (Meir, 1998). (3) The grammatical roles that control agreement are source and goal. (4) The system is fully productive. (5) The formal instantiation of agreement is simultaneous rather than sequential. Our claim is that the universality of this system in sign languages, and the relatively short time span over which it develops, derive from the interaction of language with the visuo-spatial domain of transmission.

Yet at the same time, as its label suggests, verb agreement in sign languages follows the same syntactic restrictions as in spoken languages: in all languages, verbs may agree only with indexically identifiable properties of their subjects and objects (person, number, and gender in spoken languages; referential indices in sign languages).

This indicates that the mechanism of agreement is universally available to human language (Aronoff, Meir, & Sandler, 2000).

We present new evidence that even iconic, sign language universal morphology does not arise overnight. Current work on a new, isolated sign language used in a Beduin village reveals the kernels of verb agreement that have not yet developed into a full-fledged morphological system. We conclude that: (1) universal morphological properties underlie sign language typical grammar, (2) modality of transmission can have a profound influence on grammatical form, and (3) despite the predictable influence of modality on language form, the normal course of language development and change is detectable in sign language.
Until I can actually find a copy of the PNAS article, these tantalizing tidbits will have to suffice.


1 comment:

Anonymous said...

Signed languages are not 'constructed' languages: they are natural languages that evolve where-ever communities of deaf people exist.
PS Not everyone in the signed language linguistics community is in total agreement with all the claims made by Aronoff et al. No surprises there, eh?