Formal Approaches to South Asian Languages (E-Journal)
Not a member yet
116 research outputs found
Sort by
Coda/Onset Asymmetries in Dhivehi
Full text linkThe asymmetry between codas and onsets in neutralizations and assimilations is a challenge for classic OT , which operates only on output constraints and does not distinguish between VC1C1V and VC2C2V as the output of /VC1C2V/. Serial forms of OT capture the asymmetry by making coda neutralization a prerequisite for assimilation. Dhivehi offers evidence that neutralization does in fact precede assimilation, as neutralization of coronal /t/ leaves a coronal glide ‘trace’ that persists after assimilation of a coda to a following onset. However, onsets assimilate to preceding codas in certain morphological environments when the coda is retroflex and the onset is dental. This kind of assimilation cannot be captured by serially ordered neutralization and assimilation, and the analysis requires the use of either morphologically targeted constraints or the reranking of constraints between morphological levels.
Deriving subject and antisubject orientation
Full text linkThis paper investigates subject and antisubject orientation in Hindi-Urdu. We argue that the locus of these two binding constraints is Voice0, the functional head responsible for binding the anaphoric possessor apnaa, wherein the binder of apnaa must raise to [Spec, VoiceP]. Subject orientation reduces to the locality of A-movement. Antisubject orientation is the result of a preference to use the anaphor apnaa whenever possible. We show that this proposal extends to dative–nominative structures, where the complementarity of subject and antisubject orientation for anaphors and pronouns breaks down. Finally, we examine speaker variation of quantifier binding with uskaa in dative–nominative structures.
The syntax of split: The case of Hindi and Magahi
Full text linkThis paper provides a theoretical explanation of split construction in Hindi and Magahi, both SOV, modern Indo-Aryan languages. In both languages, all the prenominal elements can be fronted inde- pendent of the object NP. However, these elements cannot be split off from overtly Case marked object in Hindi. Magahi, on the other hand, allows split even from the Case marked objects. Furthermore, Magahi allows split from all kind of subject whereas Hindi only allows split from Dative and Instru- mental subject and the subject of Unaccusative and Unergative, but not from the subject of transitive clauses. This paper provides a unified account to derive this asymmetry by adopting the theory of “freezing” effects: Once a constituent moves it becomes opaque for extraction (Wexler and Culicover 1980, Takahashi 1994, Stepanov 2007).
Production, Perception, and Distribution of Breathy Sonorants in Marathi
Full text linkBreathy sonorants are crosslinguistically rare, and while a small amount of existing work has focused on their acoustic properties much remains to be learned about their perception and their language-internal distribution. Herein, breathy sonorants in Marathi are investigated via instrumental acoustic analysis, a perception experiment, and corpus analysis. Results reveal that breathy sonorants are under-represented language-internally in addition to being typologically rare. The acoustic differences associated with sonorant phonation contrasts are less robust than those in obstruents. They are also prone to more perception errors than obstruents, and breathy sonorants are more heavily restricted phonotactically than breathy obstruents. These data contribute to a more nuanced understanding of breathy sonorants, and lend potential insight into their typology.