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

142 Małgorzata Krzek

bound by various elements located in the CP.^12 ,^13 With respect to features found in

the feature geometry of pronouns, I will argue that their content is established in

the process of binding rather than valuation. It has to do with the fact that these

features, as elaborated on below, are privative, not binary and in this respect dif-

ferent from Φ-features found on other elements that enter various Agree relations.

3.3 The feature geometry of pronouns: Harley & Ritter (2002)

Following Harley & Ritter (2002), it is assumed that the structure of pronouns can be

illustrated by the feature geometry in (24).^14

(24)

PRONOUNS^14

Participant Individuation

Speaker Addressee Group Minimal Class

Animate Inanimate/Neuter

Feminine Masculine

Augmented

The features in the feature geometry of pronouns in (24), as already mentioned above,

are bound (by elements such as ΛA, ΛP , operators, Q-adverbs) in the course of deriva-

tion.^15 Those features that are not bound are eliminated. I assume here, departing from

12. The idea that some features enter the derivation as unbound and then get bound in the
    course of derivation and as a result become interpretable is very similar to that by Pesetsky &
    Torrego (2004) in whose account some features are unvalued but interpretable.


13. The idea that pronouns are ‘constructed’ in the Narrow Syntax appears to be in line with
    a program of ‘syntactic contexualism’ advocated by Borer (2005a, b) who claims that many
    properties of lexical items are not truly properties of those lexical items but are contributed by
    the syntactic environment in which those lexical items occur.


14. The Participant node and its dependents, Speaker and Addressee, are used to represent
    person, specifically, 1st and 2nd person (3rd person being unmarked). The Individuation node
    and its dependents, Group, Minimal and Augmented, are used to represent number systems.
    The Class node encodes gender and other grammatical class information.


15. The gist of the analysis proposed here is concurrent with the one proposed by
    D’Alessandro & Alexiadou (2003). ΛA and ΛP will be located in D’Alessandro & Alexiadou’s
    (2003) Speech Act Projection (SAPP).