Home Linguistics & Semiotics The null subject parameter and correlating properties: The case of Creole languages
Chapter
Licensed
Unlicensed Requires Authentication

The null subject parameter and correlating properties: The case of Creole languages

  • Marco Nicolis
View more publications by John Benjamins Publishing Company
The Limits of Syntactic Variation
This chapter is in the book The Limits of Syntactic Variation

Abstract

The existence of correlations between the positive setting of the Null Subject

Parameter and other grammatical properties has recently been called into

question (see Newmeyer (2005)). In this paper, I show that the proref pro-ref

correlation appears to be exceptionless and that it follows directly from the theory

of pro–drop in Rizzi (1986a). Rizzi (1982) proposes that the availability of Free

Inversion is what allows for the lack of that–t effects in pro–drop languages;

this, I further show, is empirically untenable. I present data from several Creole

languages that suggest that the lack of that–t effects in these languages is related

to the availability of null expletive pro. I take the latter to be a sufficient, but not

necessary, condition for the lack of that–t effects. I finally discuss the implication

of this finding for the parametric model.

Abstract

The existence of correlations between the positive setting of the Null Subject

Parameter and other grammatical properties has recently been called into

question (see Newmeyer (2005)). In this paper, I show that the proref pro-ref

correlation appears to be exceptionless and that it follows directly from the theory

of pro–drop in Rizzi (1986a). Rizzi (1982) proposes that the availability of Free

Inversion is what allows for the lack of that–t effects in pro–drop languages;

this, I further show, is empirically untenable. I present data from several Creole

languages that suggest that the lack of that–t effects in these languages is related

to the availability of null expletive pro. I take the latter to be a sufficient, but not

necessary, condition for the lack of that–t effects. I finally discuss the implication

of this finding for the parametric model.

Downloaded on 18.9.2025 from https://www.degruyterbrill.com/document/doi/10.1075/la.132.11nic/html
Scroll to top button