[Author Login]
Title:Anti-*That-Trace Effects in Norwegian
Authors:Edward Keer
Abstract:Anti-*That-Trace Effects in Norwegian

Edward W. Keer

Rutgers University

In Norwegian a complementizer is obligatory when a subject is

wh-extracted locally from an embedded question as in (1).

1. a. Jeg vet hvem som t vant.

I know who Comp t won

b. *Jeg vet hvem t vant.

I know who t won

'I know who won.'

This phenomenon has been labled the Anti-*that-trace effect by

Taraldsen (1986). In contrast, when a subject is wh-extracted

long distance from an embedded clause, the complementizer is

obligatorily absent - *that-trace effects - as in (2).

2. a. *Jeg vet hvem du tror som t vant.

I know who you think Comp t won

b. Jeg vet hvem du tror t vant.

I know who you think t won

'I know who you think won.'

In this paper I argue that the distribution of the complementizer

in Norwegian follows from the OT syntax system proposed in Grimshaw

(to appear) with the assumption that wh-phrases cannot remain in situ

in Norwegian. Specifically, the complementizer is obligatory when it

is the best available governor for a subject trace and absent when it

blocks government of a subject trace from a better governor.

In Grimshaw's system, traces are subject to two head government

constraints. The general constraint T-Gov requires traces to be head

governed. The specific constraint, T-Lex-Gov requires traces to be

lexically governed. In configurations which exhibit *that-trace

effects, the complementizer is obligatorily absent because having it

violates T-Lex-Gov. In constructions showing Anti-*that-trace effects

however, T-Lex-Gov cannot be satisfied. Therefore, Norwegian must

settle for satisfaction of the less strict T-Gov.

Other issues raised in the paper include: a definition of head

government which differs from that in Grimshaw (to appear) based on

multiple wh-extractions in Norwegian, the hypothesis that the forced

movement of wh-phrases is a result of verb second in Norwegian, and

an exploration of the consequences of my analysis for empty operator

constructions (relative clauses, clefts and comparatives).
Type:Paper/tech report
Article:Version 1