Verb Movement and Phase-Sliding
A central issue being investigated within the Minimalist Program is whether (and how) access to the Lexicon
must be restricted in some manner. Under standard assumptions, this process is supposed to unfold phase by
phase, with `small' subarrays of Lexical Items placed in active memory (cf. [3],[4]). This view immediately raises
the question of what the metrics to establish such sequential accesses are. Chomsky has provided both
conceptual and interface/output motivations supporting the claim that v*P and CP constitute the strong phases,
and extending conclusions from previous work, in Chomsky (2005), he claims that all the relevant features are
generated in the phase heads. In the same breath, Chomsky (2005) argues that A-/A'-movements are now
distinguished by the triggering head: only phase heads launch A'-movement by means of what Chomsky (2005)
calls edge feature (presumably the current version of the "generalized EPP" of Chomsky (2000)); any others are
A-movements, driven by -features, which appear in T and V derivatively, as a process of inheritance.
Interestingly, only phase heads bear edge features, these being an optional device to yield surface-semantics
(i.e., specificity, presupposition, etc.) or cyclicity through phase edges (i.e., the head and SPECs of C and v*; cf.
[4]). The fact that edge features are optional in phase heads is explicitly pointed out in Chomsky (2000) (cf. (1)).
This said, one might wonder whether inheritance can feed parametric variation. Let me elaborate.
Chomsky (2005) actually considers two different scenarios: either inheritance of -features by T and V is a
device of UG (perhaps parametrized) or else it constitutes an optimal solution to capture the A/A' distinction.
Chomsky (2005) follows the second route (the strongest one), suggesting that the SEM component may require
distinguishing Case/Agreement systems from peripheral ones through inheritance. Be that as it may, it is
tempting to explore the first possibility. To be specific, suppose that T can be endowed not only with -features,
but also with edge features in the relevant languages. Consider, therefore, a parameter roughly as indicated in
(2). Here I would like to argue that (2) is marked positively in Null Subject Languages (NSLs), a move that
virtually turns T into a strong phase head in those languages (cf. (1)). However, given that such a possibility is
conceptually problematic (only C and v* are the strong phase heads; cf. [3],[4],[5]), I propose that the phase
effects manifested in T, though pervasive and robust, can be regarded as a side effect of v*-to-T movement:
when internally merged to T, v* re-labels the whole structure, forcing a species of Reprojection (cf. [6],[7]), to
which I will refer here as phase-sliding (metaphorically, it is as if v*'s movement pushed the v*P phase up to the
TP level). The analysis clearly revamps Chomsky's (1986) idea that V-T amalgamation can freed the VP of its
"barrierhood", and, if correct, refutes the empirical phenomena threatening the phase status of v* in NSLs, for all
the operations that appear to take place at the T level, actually occur at an "extended v*" level. The proposal is
interesting, at least in two respects: first, it trivially derives the hybrid A/A' nature that SPEC-T has been said to
have (cf. [10]); and, second, it still allows us to say that v* and C are the strong phase heads universally, for T
acts as a phase just derivatively, due to the hybrid label it forms with v* (in other words, if v* does not merge with
T, T cannot display phase-like properties). Consequently, notice that T seems to exist just as a feature holder (of
both edge and -features) feeding parametric variation.
This phase-sliding proposal makes immediate predictions. Consider first intervention effects in VOS
configurations, taking the object to be in an outer-SPEC-v* (cf. [8]). If we assume that movement of the object to
SPEC-v* and nominative case assignment to the subject by T operate in different phases (say, v*P and CP
Chomsky's system), we predict that a sentence like (3) must show intervention effects. Intervention, however,
does not obtain, and the sentence is fine. The conclusion, then, reinforces the hypothesis defended here: all
operations (both object raising and nominative case assignment) take place at the first strong phase level (i.e.,
v*/TP). A second prediction concerns subextraction. As Chomsky (2005) shows, in English, the Subject
Condition affects SPEC-v* a phase edge-, not SPEC-T (for otherwise the contrast in (4) would not be
explained). If what I have argued for is right, the prediction is clear: in NSLs, subextraction should be fine from
subjects in SPEC-v* (i.e., postverbal ones), since that position no longer qualifies as a phase edge after v*-to-T
movement, SPEC-v*/T does instead (cf. (5)). It is possible to draw a connection between this proposal and more
remarkable facts of NSLs if we endorse Pesetsky & Torrego's (2004) claim that `case' is an uninterpretable tense
feature [uT]. In Pesetsky & Torrego (2004), C is also endowed with [uT], which must be deleted by a pure T head
(that, in (6a)) or the subject DP (cf. (6b)), crucially assuming that the case feature of such DP can participate in
checking processes within the phase it has been "marked for deletion", something which is feasible if `valuation
of features' eliminates the interpretable/uninterpretable distinction for a short period of time (specifically, until a
strong phase level is reached; cf. [4]). Importantly for our concerns, if v*/TP is a strong phase in NSLs, we expect
that the [uT] of subject DPs will never remain `computationally alive' to check C's [uT]. The prediction is borne
out, as these subjects never trigger that-trace effects (cf. (7)), and that-deletion is almost barred (cf. (8)).
Synthesizing, the logic of the proposal outlined here argues that parameters can be associated with
whether or not (functional) heads are endowed with edge or -features (cf.[1],[4],[5]). Viewed that way,
parametric variation boils down to feature assembling into Lexical Items, reconciling a long-standing insight
about linguistic universals and their parameters- with the worthy goal (embodied in the inquiry undertaken by
the Minimalist Program) of accounting for properties of the Faculty of Language in a principled way (i.e., reducing
them to interface conditions or conceptual necessities).
(1) The head H of a [strong] phase Ph may be assigned an EPP-feature.
[from Chomsky (2000:109)]
(2) Parameter: T is endowed with edge features.
(3) [CP [TP Mercouz T[u] [v*P [o coche][i]k [v*P Xoán[i] [VP tz tk ]]]] (Galician)
Bought-3SG the car Xoán
`Xoán bought the car'
(4)
a. [CP Of which carj was [TP [the driver tj]z [vP awarded tz a prize]]]? (English)
b. *[CP Of which carj did [TP [the driver tj]z [v*P tz cause a scandal]]]? (English)
(5)
a. ??[CP De qué universidadj dices que [TP [muchos estudiantes tj]z protestaroni [v*P tz ti ]]]? (Spanish)
Of which university say-2SG that many students protested-3PL
`Of which university do you say that many students protested?'
b. [CP De qué universidadj dices que [TP protestaroni [v*P [muchos estudiantes tj] ti]]]? (Spanish)
Of which university say-2SG that protested-3PL many students
`Of which university do you say that many students protested?'
(6)
a. John said [CP that[iT]k C[uT, EPP] [TP Mary Tk called him ]] (English)
b. John said [CP Mary[uT]j C[uT, EPP] [TP tj called him ]] (English)
(7) [CP Chij credi[iT]k C[uT,EPP] [EF] Tk pro [CP tj chez C[uT,EPP] [TP tj Tz ha parlato]]]]? (Italian)
Who think-2SG that have-3SG talked
`Who do you think has talked?'
(8) *En Joan diu [CP C[uT,EPP] [TP la Maria no ha vingut]] (Catalan)
The Joan say-3SG the Maria not have-3SG come
`Joan says Maria has not come'
References
[1] Borer, H. (1984): Parametric Syntax: Case Studies in Semitic and Romance Languages, Dordrecht:Foris.
[2] Chomsky, N. (1986): Barriers, Cambridge (Mass.):MIT Press.
[3] Chomsky, N. (2000): "Minimalist Inquiries: The Framework", in Martin, Michaels, & Uriagereka, Step by Step.
Essays on Minimalist Syntax in Honour of H. Lasnik, Cambridge (Mass.):MIT Press,89-155.
[4] Chomsky, N. (2001): "Derivation by Phase", in Kenstowicz (ed.), Ken Hale: A Life in Language, Cambridge
(Mass.):MIT Press,1-52.
[5] Chomsky, N. (2005): "On Phases", Ms., MIT.
[6] Donati, C. (2000): La sintassi della comparazione, Padova:Unipress.
[7] Hornstein, N. & J. Uriagereka (2002): "Reprojections", in Epstein & Seely (eds.), Derivation and Explanation
in the Minimalist Program, Malden (Mass.):Blackwell,106-132.
[8] Ordóñez, F. (1998): "Postverbal Asymmetries in Spanish", NLLT,16:313-346.
[9] Pesetsky, D. & E. Torrego (2004): "Tense, Case, and the Nature of Syntactic Categories", in Guéron &
Lecarme (eds.), The Syntax of Time, Cambridge (Mass.):MIT Press,495-537.
[10] Uribe-Etxebarria, M. (1992): "On the Structural Positions of the Subject in Spanish, their Nature and their
Consequences for Quantification", in Lakarra & Ortiz de Urbina (eds.), Syntactic Theory and Basque
Syntax, San Sebastian:ASJU,447-491.