Sub-extraction from Subjects
Since Huang's (1982) seminal work, it is a well-known fact that subjects and adjuncts constitute
opaque domains to sub-extraction (cf. (1)). Different proposals have tried to account for this fact,
either capitalizing on PF-based constraints like Kayne's (1994) LCA (cf. Uriagereka (1999)), or
assuming a representational constraint along the lines of Chain Uniformity (cf. Boeckx (2003),
Stepanov (2001), Takahashi (1994), i.a.). Each type of approach makes different predictions for
the subject case: in particular, while for the first type the problem arises the minute a complex
specifier is created (SPEC-v* being the first candidate), for the second one only sub-extraction
from derived specifiers matters, since only those pose chain-uniformity concerns.
Chomsky (2005) has reassessed this issue, arguing that sub-extraction from subjects is
blocked in the base position (i.e., SPEC-v*, a strong phase edge), and not the surface position
(i.e., SPEC-T), for otherwise the contrast in (2) would remain unexplained. As Chomsky himself
points out, an immediate prediction is that sub-extraction from SPEC-C (another strong phase
edge) should be as degraded as sub-extraction from SPEC-v*; that prediction is borne out, at
least according to data of the sort in (3), from Lasnik & Saito (1992). Although Chomsky (2005)
does not provide a straightforward explication of this state of affairs (he merely speculates that
sub-extraction from strong phase edges raises locality problems), the parallelism is interesting in
itself. That said, the minimal pair in (4), originally noted by Torrego in 1985, casts some doubt
on the whole approach: in Spanish, it would appear, sub-extraction from SPEC-T is worse than
sub-extraction from SPEC-C. To complicate things further, Uriagereka (1988) noted that sub-
extraction from a postverbal subject (which we take to be in SPEC-v*) is much better than from
a preverbal one arguably in SPEC-T, as in Torrego's (1985) examples- (cf. (5)).
Here we want to argue that, despite initial appearances, Spanish data can be
accommodated into Chomsky's "phase theory". Consider first the subject case; what we must
find out is why sub-extraction from postverbal subjects is fine even if they constitute a phase
edge. Our solution relies on v*-to-T movement: we propose that verb movement in languages
like Spanish is a bona fide instance of syntactic movement (contra Chomsky (2001)) that forces
a reprojection at the TP level, with both labels coexisting (cf. Chomsky (2005) and Hornstein &
Uriagereka (2002)). Technically, we claim that the verb `pied-pipes its phasehood' to TP so that
v*P is not a phase domain any longer, a process we refer to as phase-sliding (the device is
clearly equivalent to Chomsky's (1986) idea that V-to-I movement, where it obtains, frees VP
from its `barrierhood'). The analysis is reinforced by the simple possibility of having VOS
sequences in Spanish, which, following Ordóñez (1998), we analyze as involving scrambling of
the object to an outer-SPEC-v*: if v*P were a phase in Spanish, nominative case assignment
should be blocked by the object (cf. (6)); if, however, v*/TP is, both operations (object raising
and nominative case assignment) can take place in parallel within that phase boundary. The logic
of the proposal predicts that extracting from the true phase edge in Spanish (under our account,
SPEC-v*/T) should be deviant and it is, as shown in (4a). Consider next sub-extraction from
SPEC-C; although we agree with Torrego's (1985) judgments, we argue that trying to sub-
extract here is actually as bad in Spanish as it is in English: the key for sub-extraction from
SPEC-C to be possible we argue is the presence of negation, which `rescues' sub-extraction by
forcing a presuppositional (de re) reading of the wh-phrase undergoing movement. Notice that,
when negation is dropped, sub-extraction is out (7a), as expected.
In sum, the sketched analysis argues that sub-extraction from what Chomsky (2005) calls
phase edge runs into locality problems in all languages, with apparent counterexamples emerging
from independent language-specific factors (i.e., parameters). One such parameter (the crucial
one, for us) rests on v*-to-T movement, which, if our account is on track, should not be cornered
to the PF component: it has non-trivial computational consequences. Recall, to conclude, that
neither of the theories proposed to predict sub-extractions works for (2); while this is radically so
for the Chain Uniformity approach, the PF-based theory could be made compatible with the data
if it is assumed that linearization (hence freezing) becomes definitive only at phase edges.
(1) a. ?*[CP Whoi did [TP [a friend of ti ]j [v*P tj admire a picture of Aristotle]]]? Subject Condition
b. ?*[CP Whoi did [TP Maryj [v*P tj cry [afterP after Peter hit ti ]]]]? Adjunct Condition
[from Stepanov (2001)]
(2) a. ?*[CP Of which cari did [TP [the driver ti ]j [v*P tj cause a scandal]]]?
b. [CP Of which cari was [TP [the driver ti ]j [vP awarded tj a prize]]]?
[from Chomsky (2005)]
(3) a. ?*[CP Whoi do you wonder [CP [which picture of ti ]j [TP Maryz [v*P tz bought tj]]]]?
b. ?*[CP Whoi do you wonder [CP [which picture of ti ]j [TP tj [vP is tj on sale]]]]?
[from Lasnik & Saito (1992)]
(4) Spanish
a.??Esta es la autora [CP[de la que]i [TP[varias traducciones ti]j han ganado [v*P tj premios inter.]]]
This is-3SG the author of the that several translations have-3PL won awards international
`This is the author whom several translations by have won international awards'
b. [CP De qué autorai no sabes [CP [qué traducciones ti ]j [TP tj están [vP tj a la venta]]]?
Of what author not know-2SG what translations are-3PL to the sale
`Which author don't you know which translations by are on sale?' [from Torrego (1985)]
(5) Spanish
a. [CPDe qué conferenciantesi te parece que [TP me van a impresionar [v*P [las propuestas ti] ]]]?
Of what speakers CL-to-you seems-3SG that CL-me go-3PL to to-impress the proposals
`Which speakers does it seem to you that the proposals by will impress me?'
b.* [CPDe qué conferenciantesi te parece que [TP[las propuestas ti]j me van a impresionar [v*P tj]]]?
Of what speakers CL-to-you seems-3SG that the proposals CL-me go-3PL to to-impress
`Which speakers does it seem to you that the proposals by will impress me?'
[from Uriagereka (1988)]
(6) [CP C [TP compróvT[u] [v*P el cochej [i] [v*P Isabel[i] [VP tv tj ]]]] (Spanish)
_________________ ____________
(7) Spanish
a. [CP De qué directori no ha dicho María [CP [qué películas ti ] j tj han ganado un Óscar]]?
Of what director not has-3SG said María what films have-3PL won an Oscar
`Which director hasn't María said which films by have won an Oscar?'
b. ??/* [CP De qué directori ha dicho María [CP [qué películas ti ]j tj han ganado un Óscar]]?
Of what director has-3SG said María what films have-3PL won an Oscar
`Which director has María said which films by have won an Oscar?'
References:
Boeckx, C. (2003): Islands and Chains, Amsterdam: John Benjamins.
Chomsky, N. (1986): Barriers, Cambridge (Mass.): MIT Press.
Chomsky, N. (2001): "Derivation by Phase", in Kenstowicz (ed.), Ken Hale: A Life in Language, Cambridge
(Mass.): MIT Press, 1-52.
Chomsky, N. (2005): "On Phases", Ms. MIT.
Hornstein, N. and J. Uriagereka (2002): "Reprojections", in Epstein & Seely (eds.), Derivation and
Explanation in the Minimalist Program, London: Blackwell, 106-132.
Huang, J. (1982): Logical Relations in Chinese and the Theory of Grammar, Ph.D. dissertation, MIT.
Kayne, R. (1994): The Antisymmetry of Syntax, Cambridge (Mass.): MIT Press.
Lasnik, H. & M. Saito (1992): Move , Cambridge (Mass.): MIT Press.
Ordóñez, F. (1998): "Post-verbal asymmetries in Spanish", NLLT, 16, 313-346.
Stepanov, A. (2001): Cyclic Domains in Syntactic Theory, Ph.D. dissertation, U.Conn.
Takahashi, D. (1994): Minimality of Movement, Ph.D. dissertation, U.Conn.
Torrego, E. (1985): "On Empty Categories in Nominals", Ms. U.Mass Boston.
Uriagereka, J. (1988): On Government, Ph.D. dissertation, U.Conn.
Uriagereka, J. (1999): "Multiple Spell-out", in Epstein & Hornstein (eds.), Working Minimalism, Cambridge
(Mass.): MIT Press.