Advances in Biolinguistics - The Human Language Faculty and Its Biological Basis

description, head-movement of X^0 effectively replaces Y^0 in an SO with the so-
called “Y0max” category that consists of X^0 and a segment of Y^0 , as shown in
(12), but still behaves as Y^0 as a whole.

(12) a. YP

Y^0 XP

X^0 ZP

b. YP

Y0max

X^0 Y^0

XP

tX 0 ZP

However, if Chomsky (2007, 2008) is right in claiming that every application
of Merge satisfi es the No-tampering Condition (NTC) (13), then X^0 -Y^0 -merger
should not be able to modify the internal composition of Y^0.

(13) No-tampering Condition (NTC) (Chomsky 2008:138; see also Narita
2014):
Merge of X and Y leaves the two SOs unchanged.

Under the NTC, instances of X^0 -Y^0 -merger should not be able to replace Y^0
in (12a) with “Y0max” = {X^0 , Y^0 }. Rather, it is predicted that such an applica-
tion of Merge just generates another set {X^0 , Y^0 }, without tampering with
the SO in (14a)/(14a’). If this operation really exists as an instance of Merge,
and if we keep to the defi nition of Merge as a simple set-formation operation,
then what it yields is two SOs, (i) {Y^0 , {X^0 , ZP}} and (ii) {X^0 , Y^0 } as shown
in (14b’).

(14) a.
Y^0
X^0 ZP

b.

a’. {Y^0 , {X^0 , ZP}} b’. (i) {Y^0 , {X^0 , ZP}}
(ii) {X^0 , Y^0 }

Y^0

X^0 ZP

14 Hiroki Narita and Naoki Fukui