This paper discusses data from Lithuanian and North Russian that exhibit a "Case–in–Tiers" effect (Yip et al. 1987), the strong cross–linguistic correlation between NOM objects and the presence of oblique subjects. The basic idea is that NOM is predictably assigned (checked, valued) "downstairs" only in case it fails to be assigned "upstairs". The empirical domain of the paper concerns non–agreeing Lithuanian participial predicates in word–final /–
(1) a. Lithuanian –
Gal Jonuko tie grybai atne&shachek;ta.
'Maybe Jonukas brought these mushrooms.' [Ambrazas et al., 1997:281]
b. North Russian –
U nas vzjato ot Van&soft;kinyx takoj ploxoj kot.
'We have taken such a bad cat from the Vankin's.' [Kuz'mina & Nemchenko 1971:35]
A plausible typological explanation for these construction types as "embryonic" ergatives is presented in Lavine 1999, though many questions remain regarding the case–assigning mechanisms involved, especially in light of Chomsky's 2001 "Derivation by Phase" [DbP] model, which proposes a strict phase–based cyclic Spell–Out approach to derivational syntax. First, I show that the oblique (or PP) marking on the subject of the Lithuanian and North Russian constructions in (1) is assigned as a lexical property of the /–
Indeed, if T's uninterpretable phi–features normally value NOM case, as is standardly assumed, an Agree relation can be argued to hold between T's probe and the NOM object, provided that v
Further argumentation against T as the source for the object's NOM case follows from the fact that T itself is defective: it is non–agreeing and, by hypothesis, lacks the phi–features that value NOM. In addition, elsewhere in both languages, NOM objects regularly cooccur with non–finite predicates. It will be shown that the NOM NPs in the constructions in (1) are well–behaved structurally case–marked objects in all respects (e.g., they undergo Genitive of Negation and Partitive Genitive formation, and can serve as the target of reflexivization).
The central claim of this paper is that the realization of morphological NOM is a clausal property: case on the object NP is determined not solely by its position in the clause, but by the structure of the rest of the clause (for languages that admit a Case–in–Tiers–like system) (see Marantz 1991 and Harley 1995). This conclusion presents an empirical challenge to the strict cyclic Spell–Out requirement of the DbP system.
References
Ambrazas, V. et al. (1997). Lithuanian grammar. Vilnius: Baltos lankos.
Chomsky, N. (2001). "Derivation by phase." In Ken Hale: A life in language, ed. Michael Kenstowicz, 1–52. Cambridge, Mass.: MIT Press.
Harley, H. (1995). Subjects, events, and licensing. Doctoral dissertation, MIT.
Kuz'mina, I. B. and E. V. Nemchenko. (1971). Sintaksis prichastnyx form v russkix govorax. Moskva> Nauka.
Lavine, J. (1999). "Subject properties and ergativity in North Russian and Lithuanian." In FASL 7: The Seattle Meeting, ed. K. Dziwirek et al., 307–328. Ann Arbor: Michigan Slavic Publications.
Marantz, A. (1991). "Case and licensing." Proceedings of ESCOL '91, 234–253.
Yip, M. et al. (1987). Case in tiers. Language 63:217–250.