76 Anna Bondaruk
The lack of PCC-effect in equatives seems to indicate that the two phrases flank-
ing the copula are not affected by Multiple Agree, but must obtain their nomina-
tive case from two different sources.^15 How exactly this happens is addressed in
Section 5.
To sum up, the fact that equatives differ from predicational and specificational
clauses with respect to agreement as well as their sensitivity to the PCC-effect allows
us to draw the conclusion that Polish equatives are structurally different from the other
two types of copular clauses and therefore cannot be associated with either of the
structures provided in (11) or (13) above. Consequently, they seem to call for a struc-
tural analysis different from the one offered for predicational clauses. In this way they
appear to differ from English equatives that are often analysed as being structurally
analogous to predicational clauses, and differing from them solely in their semantics
(cf., for instance, Partee 1987; Mikkelsen 2005; Geist 2008). Two different propos-
als concerning the structure of Polish equatives are examined in the two subsequent
sections.- Asymmetrical structure for Polish equatives
In this section an attempt is made to test whether the structure of Polish equatives can
be reconciled with the asymmetrical representation. One possible route to take is to
follow Reeve’s (2010) proposal. He postulates the existence of a functional head called
Eq, whose semantic effect is that of applying Partee’s (1987) IDENT type-shifting oper-
ator to the second XP, and thus making it an identity predicate which is predicated of
the first XP. He further observes that Eq appears in the extended verbal projection (cf.
Grimshaw 1991, 2000) in specificational and equative sentences, but not in predica-
tional ones. Reeve proposes that Eq is a functional category with no intrinsic semantic- Specificational clauses do not exhibit the PCC-effect, either, as confirmed by (i) below.
(i) Geniusz to ja.
genius.nom cop I.nom
‘#A genius am I.’
The immunity of specificational clauses to the PCC-effect is accounted for by Bondaruk (2012,
2013b) in terms of parallel probing by T and C, where the former targets the subject and the
latter the predicate, and as a result no multiple Agree ever takes place in specificational clauses
(cf. Section 3.1 for an outline of an analysis of Polish specificational clauses).