Section 4

Chapter 1

Linguistics

Linguistics

428 RLE Progress Report Number 139

Chapter 1. Linguistics

Chapter 1. Linguistics

Academic and Research Staff

Professor Morris Halle, Professor Noam A. Chomsky

1.1 Introduction

The work of the Linguistics group is directedtowards a better understanding of the mentalcapacities of human beings through the study of thenature, acquisition, and use of language. Lan-guage is a uniquely human faculty: only humansappear to be capable of learning and using a lan-guage, and every normal human acquires know-ledge of one or more languages.

We are trying to understand how this linguisticknowledge is represented in the speaker's mind.The central issues of linguistics research are:

1. What is the nature of linguistic knowledge?What do speakers of a particular languagesuch as Latvian, Spanish or Walpiri know, andhow does knowledge of one languageresemble or differ from that of another lan-guage?

2. How do speakers acquire this knowledge?

3. How do speakers put this knowledge to use inproducing and understanding utterances?

4. What are the physiological mechanisms thatprovide the material basis for storage, acquisi-tion and utilization of linguistic knowledge?

Our ability to answer these questions differs consid-erably, and our research reflects these differences.At present, we have progressed further with regardto answering the questions posed by item one andhave made less progress with item four. Currently,our research is heavily concentrated on issues con-cerned with the nature of the knowledge that char-acterizes fluent speakers of various languages.However, we are making a significant effort to solvethe other questions also.

We are studying these topics along a number ofparallel lines. Linguists have investigated the prin-ciples by which words are concatenated to formmeaningful sentences. These principles have beenthe primary domain of inquiry into the disciplines ofsyntax and semantics. Phonology studies thesound structure of words while morphology exam-ines the manner in which different languagescombine different meaning-bearing units (specif-ically, stems, prefixes, suffixes and infixes) to form

words. The latter topic has attracted increasinginterest in recent years and will probably becomemore prominent in our research efforts in the future.

1.2 Abstracts of Doctoral Dissertations

The following are abstracts of dissertations sub-mitted in 1996 to the Department of Linguistics andPhilosophy in partial fulfillment of the requirementsfor the degree of Doctor of Philosophy in Linguis-tics.

1.2.1 Order and Structure

Colin Phillips

Abstract

This thesis argues that (1) grammars of natural lan-guage construct sentences in a strictly left-to-rightfashion, i.e. starting at the beginning of the sen-tence and ending at the end and (2) there is no dis-tinction between the grammar and the parser.

In the area of phrase structure, I show that the left-to-right derivations forced by the principle MergeRight accounts for the apparent contradictions thatdifferent tests of consituency show. Phenomenadiscussed include coordination, movement, ellipsis,binding, right node raising and scope.

I present a preliminary account of the interface ofphonology and morphology with syntax based onleft-to-right derivations. I show that this approach tomorphosyntax allows for a uniform account oflocality in head movement and clitic placement,explains certain directional asymmetries inphonology-syntax mismatches and head movement,and allows for a tighter connection between syn-tactic and phonological phrases than commonlyassumed.

In parsing I argue that a wide range of structuralbiases in ambiguity resolution can be accounted forby the single principle Branch Right, which favorsbuilding right-branching structures wherever pos-sible. Evidence from novel and existing exper-imental work is presented which shows that Branch

429

Chapter 1. Linguistics

Right has broader empirical coverage than otherproposed structural parsing principles. Moreover,Branch Right is not a parsing-specific principle: it isindependently motivated as an economy principle ofsyntax in the chapters on syntax.

The combination of these results from syntax andparsing makes it possible to claim that the parserand the grammar are identical. The possibility thatthe parser and the grammar are identical orextremely similar was explored in the early 1960's,but it is widely considered to have been discreditedby the end of that decade. I show that argumentsagainst this model which were once valid no longerapply given left-to- right syntax and the view of theparser proposed here.

1.2.2 Multiple Feature-Checking: A Theoryof Grammatical Function Splitting

Hiroyuki Ura

Abstract

This thesis investigates some consequences of thetheory of formal feature-checking in the minimalistprogram for linguistic theory (Chomsky 1992,1994a, 1995b). More specifically, I will explore thesignificance and implications of the theory ofmulitple feature-checking. The main purpose of thisthesis is to demonstrate that the theory of multiplefeature-checking enables us to give a naturalandconsistent explanation of some less-familiar phe-nomena in the literature under the generative tradi-tion, phenomena in which some ofthe grammaticalfunctions alleged to be associated with a certaingrammatical relation are split up from them intosome others (Grammatical Function Splitting).

Part I offers a brief sketch of the previous studieson "grammatical relation" and "grammatical func-tion" (chapter 1) and a concise introduction of thefundamentals of the theory of multiple feature-checking together with the other minimalistnotations/techniques particularly prerequisite to thediscussions that follow in this thesis (chapter 2). Inpart II (chapters 3 and 4), I show that some raisingconstructions, which have been scarcely studied inthe literature under the Principal-and-Parametersapproach, can be offered a consistent account bythe theory of multiple feature-checking. Part III(chapters 5 and 6) argues that the theory of mul-tiple feature- checking provides a clue for eluci-dating the optionality of movement, which issomethimes alleged to be seriously problematicunder the theory of economy and movement in theminimalist program of Chomsky (1992, 1994a).

Part IV (chapters 7, 8, 9, and 10) deals directly withphenomena involving grammatical function splitting.There it is demonstrated that these varieties ofgrammatical function splitting can be given a naturaland consistent explanation with the aid of thetheory of multiple feature-checking. Concludingremarks together with a comment on the furtherapplications of the theory of multiple feature-checking are in chapter 11.

1.2.3 On the Distribution of DutchReflexives

Fleur Veraart

Abstract

This thesis discusses the distribution of the Dutchreflexives zich and zichzelf. I show that Reinhart &Reuland's (1993) analysis of this topic posesseveral problems. In the present study, a widerange of data is collected, which illustrate that thedistribution of zich and zichzelf needs to beaccounted for in terms of a variety of factors, ratherthan with a simple set of binding conditions.

It is argued that the distribution of zich and zichzelfis partly determined by the possibility of replacingthe reflexive with a non-reflexive element. If it isnot possible to have a non-reflexive instead of thereflexive, zich should be used. Otherwise, zichzelfis usually preferred. Also, contrastive focus isshown to play a role. On the one hand, if thereflexive is focused, it has to be zichzelf. On theother hand, if a constituent other than the reflexiveis focused, zich may sometimes be used instead ofzichzelf. Locality conditions play a role as well; zichcan occur at a greater distance from its antecedentthan zichzelf. Another factor that plays a role in thedistribution of zich and zichzelf, is the morphologicalnature of the predicate with which the reflexiveoccurs. In PPs, the factors mentioned here interactwith the phonological zelf-deletion rule. This rulestates that zelf is deleted from anaphora that occurin prepositional phrases whose preposition is non-contrastively stressed. More generally, the factorsthat determine whether zich or zichzelf should beused, interact with the fact that focus is realized byadding zelf or by stressing the element that is to befocused, and with the fact that zich cannot bestressed, but zichzelf can be. Further, it is arguedthat one needs to distinguish between opaquezichzelf and transparent zichzelf. The latter ismade up of zich and the intensifier zelf which isalso found as a modifier of full DPs.

430 RLE Progress Report Number 139