Studies in Natural Language and Linguistic Theory

The Choctaw language, indigenous to the southeastern United States, now with its greatest concentrations ofspeakers in Missis­ sippi, Oklahoma, and Los Angeles, has in the main escaped the scrutinyoftheoreticallinguistics.ItisnotthatChoctaw isanintrin­ sicallyuninterestinglanguage- aquickglanceataclausewithfive agreement controllers and a mismatch between the caseofafree­ standingnominaland its agreement affix should dispelthat notion. Rather it is, I think, the question of what we can learn from a languageinwhichNPsdon'tmovearound,"WHs"don'tfront, and gaps simply arise from pronominalization. My hope is that the presentvolume,takentogetherwithagrowingliteraturespurredon by the workofPamMunro and her students atUCLA, will bring Choctawintothelightofdayand into the circleoflanguagescon­ sidered when constructing theories that define "possible human language". Thepresentstudy,arevisionofmy 1981dissertation(University ofCalifornia,SanDiego),focusesfirstandforemostontheChoctaw agreementsystem,takingthisasthekeytothestructureofChoctaw syntax. The immediate goal, then, is to provide a unified account ofthestructures and rules underlyingtheagreement system.Along the way a rangeofgrammatical phenomena is examined, taken as evidence for particular structural configurations, and incorporated into awell-integratedaccountofmorphologicaland syntacticfacts. The resultsbearon anumber ofcurrent issues, includingthe Un­ accusative Hypothesis, the existence of demotions, the nature of antipassive, disjunctive rule application, universals of causative constructions, and others.For these reasons Choctawdeserves the scrutinyoftheoreticians. The data forming the corpus for analysis represent a variety of Oklahoma Choctaw.They were collected from a nativespeaker in San Diego between 1978 and 1982 and from various speakers in Oklahoma during two extended visits to Broken Bow in 1980. I lX PREFACE x wishtothankthespeakerswhohelpedmebysharingtheirlanguage andencouragingmystudies.MyworkonChoctawwassupportedin partbyfundsfrom theNationalScienceFoundation(throughgrant numberBNS78-17498totheUniversityofCalifornia,SanDiego), theAmericanPhilosophicalSociety(throughaPhillipsFundgrant), andtheDepartmentofLinguisticsatUCSD.