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

138 Małgorzata Krzek

[+human] interpretation appears to be the function of the feature geometry of pro-

nouns, the way in which the features on pronouns are valued and elements that these

values come from; see sections below for details.

3. How are the values of phi-features on predicative complements
   established?

3.1 Sigurðsson (2004, 2009 ), Frascarelli (2007), and Holmberg (2010a,b)

Having identified the possible values of the phi-features on subjects in both construc-

tions, we are now in a position to discuss the mechanics of how these different feature

values are assigned and where they come from. The point of departure for the analysis

to be suggested here will be two hypotheses put forward by Sigurðsson (2004, 2009 ),

Frascarelli (2007), and Holmberg (2010a,b), which is a refinement of Frascarelli (2007).

Sigurðsson (2004) assumes that clause structure has three basic layers, given in (21),

(21) [CP...Speech features [TP Grammatical features [VP Event features...]]]

and extends to Person the idea that grammatical Tense interprets event time in relation

to speech time, in that Person interprets event participants in relation to speech par-

ticipants. More precisely, theta-features in the vP-domain are interpreted in relation

to phi-features in the TP-domain, which in turn are interpreted in relation to active/

passive speech participant-features in the CP-domain. This means that TP-features

take a mediating position between vP-, and CP-features, entering into matching rela-

tionships in both directions. The features contained in the syntactic speech event, as

argued by Sigurðsson (2004), are the time and location of speech, ST, SL respectively,

and the inherent speech participants; that is, the logophoric agent (ΛA) and patient

(ΛP).^7 Before it is explained how the Person feature is established and valued, it is

necessary to mention that this particular approach, if adopted, appears to have a bear-

ing on the way pronouns are perceived. First of all, they are different from bare roots,

7. The inherent speech participants are not simply the speaker and the addressee. They
   should rather be understood as the active vs. passive participants of speech. The following
   example illustrating this comes from Sigurðsson (2004: 9):
   (i) a. I love you.

(^1) SG = the speaker = LOGOPHORIC AGENT (and also the ‘loving one’)
(^2) SG = the addressee = LOGOPHORIC PATIENT (and also the ‘loved one’)
b. John said to me: ‘I love you’.
(^1) SG =John = LOGOPHORIC AGENT (and also the ‘loving one’)
(^2) SG = the speaker = LOGOPHORIC PATIENT (and also the ‘loved one’)