Advances in the Syntax of DPs - Structure, agreement, and case

130 Małgorzata Krzek

derivation. Differences in the interpretation of null impersonal subject pronouns^1 will

depend on the kind of elements its sub-features will be bound by. In consequence,

the SIĘ particle in the impersonal construction in question is analysed as a functional

element^2 ,^3 heading its own projection, dubbed SIĘP, and taking pro as its complement.

This makes the SIĘP very similar to the DP.^4

1.1 Overview of the paper

This paper is organized as follows: Section  2 presents two types of impersonal con-

structions found in Polish with a concise description of their morpho-syntactic prop-

erties. Section 3 shows the mechanics of how different phi-feature values are assigned

to null subject DPs and where they come from. The point of departure for the analy-

sis presented here are hypotheses put forward by Sigurðsson (2004, 2009 ), Frascarelli

(2007), and Holmberg (2010a,b). It is argued that pronouns are complex variables

that enter the derivation as bundles of unbound but interpretable features (Pesetsky &

Torrego 2004). These features are bound in the course of a narrow syntactic derivation.

It is proposed that in order to capture the possible interpretations of null impersonal

pronouns, the feature geometry of Harley & Ritter (2002) will have to be modified

slightly by the introduction of an additional dependent feature dubbed [indefinite] to

the feature [group]. Section 4 proposes an alternative analysis of the SIĘ particle that

is a consequence of the assumptions made above. Section 5 discusses issues connected

with agreement in the two impersonal constructions. Section 6 focuses on case. The

paper finishes with conclusions.

1. Pronouns are assumed here to be DPs, not D heads. Thus, the term null subject DP and
   null subject pronoun are used interchangeably.


2. As is argued below, the particle SIĘ is a kind of D(eterminer).


3. Following Chomsky (1995), functional categories (e.g. T(ense), C(omplementiser),
   D(determiner), and Agr(eement)), contrary to lexical categories (e.g. noun, verb), are closed
   set items that are assumed to drive the computation and determine the interpretation (Borer
   2005a,b). In other words, an item becomes a noun if it is embedded within D, while it becomes
   a verb once it has a layer of v above it.


4. The functional element D of the DP converts the nominal expression into a referential
   [or non-referential – MK] phrase that can be used as an argument. D, then, can be argued to
   be parallel to the complementizer of sentential complements: each turns its complement (NP
   and TP respectively) into an expression that can appear in an argument position (Alexiadou
   2001 : 29).