Slot:       28C-5          Dec. 28, 1:30 p.m. – 3:30 p.m.                                                

Panel:     Syntax

Chair:     Robert Channon, Purdue University


Title:       Reflexive Verbs in Russian: Dative Subject Reflexives

Author:   Olga S. Eremina, Michigan State University

Verb reflexive markers (VRMs) have many uses across languages. In this paper I argue that Lidz 2001 analysis of VRMs and Dative markers in Kannada can account for “classical” reflexives (умывался), unaccusatives (разбился), reciprocals (обнялись), passives (строится) , but not Dative experiencer subject cases (DS) in Russian (1). I propose an alternative analysis (based on Kallulli 2005) providing a unified account of all the VRM cases in Russian.

(1)            a.           Им (хорошо) работалось                             

                 (b)          Они (хорошо) работали


Lidz 2001 assumes that verbal predicates not only have an argument structure representation (that interfaces with the syntax) but also an aspectual representation. VRM or Dative case appears to signal a mismatch/misalignment between the two representations, resulting in an argument unlinked in the syntax:

(2)            Он разбил вазу. MATCH:


  разбить [[1 ACT-ON2] CAUSE [3CHANGE]];

  (x(y)) x = “agent”, y = “theme”


                 (x                         (y))                                                   thematic



                 [1 ACT-ON 2] CAUSE [3 CHANGE]              aspectual decomposition


(3)            Ваза разбилась. MISMATCH:        


                                              (y))                                                   thematic



                 [1 ACT-ON 2] CAUSE [3 CHANGE]              aspectual decomposition


The two strategies are alternative ways to resolve the mismatch and predicted not to co-occur, contrary to fact (1).

We modify Lidz aspectual representations as in Kalluli 2005 and we argue that co-occurrence of VRM and DS (1) is a morphological indication of two mismatches (XX).

Thus, we account for cases of “DS + VRM” (1) and “DS – VRM” (4):

(4)            (a)  Ему дуло (в спину)   

                (b)  Ветер дул ему в спину.


Representation of (1) is given in (5), representation of (4) is given in (6).


(5)            (a)          (x)          [+intend], [+act]  работали                                        


                 [1 action]: [+intend], [+act]



  (b)          (x)              [+intend], [+act]                            работалось        


                                             1               2        

                 [1 state]: [+intend], [+act]


Two mismatches, DS + VRM


(6)            (a)          (x)          [+act]                   дул                                                   


                 [1 state]: [+act]


  (b)          (x)Theme: [+act]                               дуло



                 [1 state]: [+act]


One mismatch, DS


The analysis accounts for the complexity of the VRM occurrence in Russian, and the modality in VRM+DS.



Dalina Kallulli. 2005. "Unaccusatives with dative causers and experiencers: A unified account." To appear in W. Abraham, D. Hole and A. Meinunger (eds.) Datives and Similar Cases. Amsterdam: John Benjamins.

Lidz, Jeffrey. 2001. "The Argument Structure of Verbal Reflexives." Natural Language and Linguistic Theory, 19: 311–353.


Title:       Deriving Discontinuity

Author:   Steven Franks, Indiana University

The talk compares mechanisms for generating discontinuous nominal expressions. Slavic exhibits several such expressions, suggesting various derivational routes. Consider Serbian (1), from Progovac 2006:


(1) Vukina odlazi ćerka.  ‘It is VUKA’s daughter who is leaving.’


There are at least four credible approaches countenanced by recent linguistic theory (with movement expressed using the Copy-and-Delete system):


(2)  a. [AP Vukina] odlazi [NP Vukina ćerka]              b. [AP/DP/KP Vukina ćerka] odlazi [NP ćerka].

      c. [NP Vukina ćerka] odlazi [NP Vukina ćerka].   d. [AP Vukina] odlazi [NP ćerka].                


(2a) is the traditional LBE analysis; (2b) is a remnant movement analysis; (2c) employs distributed deletion (cf. Fanselow&Ćavar 2002); (2d) reflects an approach to scrambling in which the A and N are base generated independently (cf. Bošković in press).


The LBE account distinguishes languages that project DP from those which do not, hence Bulgarian (3):


(3) *Vukinata si otiva dâšterja. (cf. Vukinata dâšterja si otiva.)


Movement targets syntactic constituents, yet in Croatian (4) neither piece is one:


(4) U izuzetno sam veliku sobu ušao.

‘It was an EXCEPTIONALLY large room that I entered.’


My proposal exploits focus features and distributed deletion:


(5)   [u izuzetno veliku sobu [sam [u izuzetno veliku sobu [ušao ...



A phrase containing a element moves to SpecFP/CP. The operative principle is that there can be no focus to the right of the [+Foc] element. All material following it within SpecFP/CP bears “flat” intonation and is subsequently deleted. This results in pronunciation of the next highest copy; crucially, neither side of the [+Foc] element need be a constituent.



Bošković, Željko. in press. "Left branch extraction, structure of NP, and scrambling." In Sabel&Saito, eds. Scrambling.

Fanselow, Gisbert & Damir Ćavar. 2002. "Distributed deletion." In Alexiadou, ed. Theoretical approaches to universals.

Progovac, Ljiljana. 2006. A syntax of Serbian. Bloomington, Ind: Slavica.


Title:       Constituent Order in Subordinate Clauses in Russian

Author:   Sarah Turner, Oxford University

Syntactic studies of complex sentences in Russian generally focus on the conjunction used and on the placement of subordinate clauses relative to main clauses. The order of constituents within subordinate clauses is discussed rarely, and normally only in connection with indirect yes-no questions. The matter is similarly overlooked in studies devoted specifically to word order, which either draw their material exclusively from main clauses, or assume that the same principles of organization operate in main and subordinate clauses.                                                                                  

In this paper the order of major constituents within subordinate clauses is considered with reference to aktual′noe členenie predloženija (for convenience: FSP). The material is classified into three groups: clauses in which the subordinating conjunction arguably forms the theme; clauses in which it forms the rheme; and clauses whose conjunctions cannot be described straightforwardly in pragmatic terms. Though not without its difficulties, this classification provides a starting-point for investigation into an area in which the FSP model lacks descriptive power: since FSP analysis typically divides the clause into no more than two discourse units, the internal organization of these units when they contain more than one syntactic constituent remains problematic. The first two sub-sets of data are examined in this connection.            

The paper is concerned primarily with Contemporary Standard Russian. The main set of data is drawn from samples of academic, creative and journalistic writing composed since 1970. This chronological limit allows for comparison with material on colloquial Russian collected since that time. The organization of subordinate clauses in the colloquial language can differ strikingly from the conventions observed in the standard language. The implications of these differences for the pragmatic analysis of constituent order in the third sub-set of data are discussed.