EARLY ACQUISITION OF WORD ORDER:
          EVIDENCE FROM HINDI-URDU AND
                     MALAYALAM
                      MAYA LEELA
             Adoctoral dissertation submitted for the degree of
             Philosophiæ Doctor in Cognitive Science and Language
                  Supervised by Dr. Anna Gavarró
                   Centre de Lingüística Teòrica
                 Departament de Filologia Catalana
                 Universitat Autònoma de Barcelona
          Maya Leela: Early Acquisition of Word Order: Evidence from Hindi-Urdu
          and Malayalam © July 2016
                    Dedicated to Unniyamma.
            ABSTRACT
            Theacquisitionofwordorderintwounder-researchedlanguages,Hindi-
            Urdu and Malayalam, is explored under the assumptions of Univer-
            sal Grammar and Very Early Parameter Setting. The acquisition of the
            OV order was tested through an experiment that was conducted on
            19-month old native Hindi-Urdu infants using a combination of the
            preferential-lookingmode,theweird-word-orderparadigm,andpseudo-
            verbs. The results showed that the infants parsed the grammatical SOV
            order and did not parse the ungrammatical VSO order, indicating that
            the parameter responsible for the OV order was set at 19 months of age.
            The acquisition of scrambling was investigated by means of three ex-
            periments in Malayalam speaking children. The first experiment tested
            the acquisition of the canonical word order SOV, and the scrambled
            non-canonical word orders OSV, OVS, and SVO, with no discourse con-
            text, through a picture matching task in children aged 3 to 5-years-old.
            Theexperimental results showed above chance performance in the com-
            prehension of the canonical and most scrambled word orders by all
            age groups. The second experiment tested the acquisition of the canon-
            ical and the non-canonical word orders with a discourse context, in
            an act-out task in 2-year-old children. The results revealed at ceiling
            performance by all children. The third experiment, run with 2-year-old
            children, tested the acquisition of canonical and scrambled sentences
            with an anaphor. All children performed above chance indicating that
            anaphor resolution is adult like. Usage-based theories and the impli-
            cations of the theory of Relativized Minimality in child grammar are
            critically analyzed using the results of the Hindi-Urdu and Malayalam
            experiments. The evidence gathered is in support of the fact that adult-
            like competence of abstract syntax is present within child grammar at
            early ages and that movement appears to be adult-like.
                                iv