Word Order Variation and Acquisition in American Sign Language
, D. Pichler 2001
This study examines two crosslinguistic generalizations generated by previous studies of word order: (1) that the word order parameters (i.e. the spec-head and head-complement parameters) are universally set early, and (2) that word order variation in languages with rich and regular inflection is acquired earlier than in languages with poor or irregular inflection. These generalizations are evaluated using spontaneous production data of four deaf children between the ages of ~20-30 months, acquiring ASL as their first language.
Both noncanonical and canonical word order are attested during this period. I argue that instances of early noncanonical order are not necessarily errors, since many exhibit properties of grammatical adult word order variation from underlying/canonical SVO. The relevant movement operations are subject-pronoun copy, yielding noncanonical VS order, and rightward verb raising licensed by handling, spatial and aspectual inflection, which I group together as reordering morphology, resulting in noncanonical OV order. I argue that the occurrence of grammatical word order variation side by side with canonically ordered sentences is evidence that the word order parameters are set early in ASL, consistent with the first crosslinguistic generalization.
ASL has a relatively rich but irregular verbal inflection system. The fact that children acquire word order variation early despite the irregular inflectional system of ASL supports a modified version of the second crosslinguistic generalization: Early acquisition of order variation depends on early acquisition of the morphological cues associated with noncanonical order. Alternatively, noncanonical orders with no morphological cue are also acquired early, provided there are no syntactic restrictions on their application.
Finally, this thesis challenges previous claims that topicalization is acquired late (not before 3;0) in ASL. Examination of one child's OV combinations not accounted for by reordering morphology reveals that roughly half feature a simple prosodic break between the object and verb. These breaks are reminiscent of those used to mark topics in Israeli Sign Language, and I propose that they serve the same function in early ASL. This analysis puts acquisition of topicalization movement at as early as 24 months, although other aspects of ASL topicalization (i.e. adult nonmanual marking and pragmatic appropriateness) have yet to be mastered.