LINGUIST List 15.757

Tue Mar 2 2004

Diss: Syntax: Kremers: 'The Arabic...'

Editor for this issue: Takako Matsui <takolinguistlist.org>

Directory

1. joostkremers, The Arabic noun phrase: a minimalist approach

Institution: University of Nijmegen
Program: Department of Arabic and Islam
Dissertation Status: Completed
Degree Date: 2003

Author: Joost Kremers 

Dissertation Title: The Arabic noun phrase: a minimalist approach

Dissertation URL: http://www.lotpublications.nl

Linguistic Field: Linguistic Theories, Syntax 

Subject Language: Arabic, Standard (code: ABV)

Dissertation Director 1: Kees Versteegh
Dissertation Director 2: Eric Reuland
Dissertation Director 3: Ad Foolen

Dissertation Abstract: 

The aim of this study is to develop a model of the Arabic noun phrase
from a minimalist perspective. It provides an analysis of many
phenomena in the Arabic noun phrase, such as the construct state,
adjectival agreement, definiteness inheritance, the formation of
deverbal nouns and participles, etc.

Next to this discussion, the study also focuses on a more theoretical
aspect of syntax: linearisation. The Minimalist Program as developed
by Chomsky (1995) and subsequent work provides the basis for a
syntactic theory that uses as few primitive notions as
possible. However, because the Minimalist Program is not a fully
worked out theory, some aspects of it remain to be developed, and
linearisation is one of them: Chomsky remarks that a tree structure is
in principle not linearly ordered. Given this assumption, it is
necessary to develop a way do derive the linear order of any given
tree structure.

This study argues that linearisation is an important part of syntax,
in that it is responsible for certain word order variation. This idea
challenges the notions developed by Kayne (1994), who argues that
human language has a universal underlying word order, and that word
order variation is the result of movement.

Contrary to Kayne's model, the linearisation procedure developed in
this study is minimalist in nature. It starts out with the basic
observation that any procedure that linearises a hierarchical tree
structure needs to search the tree in order to find the terminal
elements that are to be spelled out. By specifying how this search
takes place, it becomes clear that we do not need to resort to
additional mechanisms to account for word order. Instead, the
linearisation procedure uses two paraeters that guide the order in
which branches of the tree are searched, which determines the order in
which the terminal elements are found and spelled out.

With this model, it becomes possible to account for the word order
difference in a (sub)domain of the English and Arabic noun phrases
without having to posit radically different tree structures for the
two languages. English for example has orders such as adjective-noun
and (adjective)-possessor-(adjective)-noun. In contrast, Arabic has
the orders noun-adjective, and noun-possessor-(adjective)-(adjective).
Although these word orders are very different, it is shown that the
underlying tree structures are in fact highly similar. Arabic and
English merely differ in the settings of two specific parameters,
which have the result that the trees are linearised in a different
order.