Academic journal article Journal of Slavic Linguistics

On the Absence of Long-Distance A-Movement in Russian

Academic journal article Journal of Slavic Linguistics

On the Absence of Long-Distance A-Movement in Russian

Article excerpt

Abstract: Lasnik (1998) observes that Russian lacks long-distance subject-to-object and subject-to-subject raising, where "long-distance" is understood in the sense of crossing the boundary of a clausal domain defined in terms of an independent Infl (Tense/Agreement) system. In Lasnik's terms, this state of affairs arises because Russian infinitival clauses are necessarily Tensed, whereas English infinitivals (which do allow long-distance raising) may appear "tenseless." In this article I discuss examples of raising with aspectual and modal predicates in Russian, whose grammaticality appears to call into question the validity of Lasnik's claim and show that raising in these contexts is in fact limited to a single TP domain. Realizing the monoclausal character of raising removes the apparent challenge to Lasnik's generalization and reaffirms the radically "local" behavior of Russian in the domain of A-movement.

1. Introduction

Generative research identifies two standard varieties of cross-clausal A-dependencies: subject-to-subject raising and subject-to-object raising, the latter also known under the rubric of Exceptional Case Marking (ECM). (1) It is well known that some languages have a cross-clausal ECM construction only in limited contexts (e.g., French has it with wh-traces only), and yet others do not have it at all. Given that both kinds of cross-clausal raising usually receive a similar kind of analysis (e.g., Chomsky 1981, Stowell 1981, among others), it is therefore natural to expect that not all languages necessarily feature long distance subject-to-subject raising either.

Russian has been recognized as a language which has neither cross-clausal ECM constructions, nor subject-to-subject raising in infinitival clauses (Lasnik 1998). In this article I endorse this view by evaluating a number of sentence types that may potentially involve cross-clausal A-movement with infinitivals and show that those constructions are in fact monoclausal and involve only "short-distance" rather than long-distance raising. Russian, then, is a language which lacks cross-clausal effects in the A-domain altogether.

2. Background

Cross-clausal raising occurs with raising predicates, which typically constitute a closed class set in a language. The traditional method for distinguishing a raising predicate from a control predicate is a battery of tests involving selectional properties, agentivity, and controllability, as well as certain scope effects and particular behavior in idiomatic contexts. Perhaps the best indicator of the raising status of a predicate is that, unlike control predicates, it does not impose any selectional restrictions on its subject, as illustrated for seem and is likely below:

(1) a. The sun seems t to have been obscured t by the clouds.

b. John seems t to be smart.

c. This candidate is likely t to win.

(1) shows characteristics of what I consider for the present purposes truly cross-clausal or long-distance raising: both the lower clause(s) and the matrix clause contain their own Tense/Agreement (or Infl) systems. According to the standard Minimalist assumptions, the lower clause(s) each contain "defective" (phi-incomplete) Tense with only partial (namely, person) agreement, and the matrix clause contains a "full" (phi-complete) Tense with full-fledged phi-feature agreement (Chomsky 2001). The subject starts out in the clause with a phi-incomplete Tense and ends up in the Specifier of the phi-complete Tense. In contrast, upon raising within a single clause the subject starts out and ends up within the domain of a single Tense, either phi-complete or phi-incomplete. The example of the former is a classical passive; a familiar example of the latter is a passivization involving quirky Case in Icelandic:

(2) John was arrested.

(3) Stolunum hafoi verio stolio a uppbooinu.

[the-chairs.DAT] [had.subDEF.AGR been stolen at the-auction

'The chairs had been stolen at the auction. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.