BACK TO THE PAST: RUSSIAN PAST TENSE REVISITED

Russian copular sentences can be either predicative (1a), with Instrumental Case-marking on the
post-copular DP in the past tense, or equative (1b), with Nominative Case-marking on the post-
copular DP in the past tense. As examples (2) show, the Russian demonstrative èto can appear in
identity sentences, but not in predicative ones. This becomes particularly clear when we consider
adjectival and prepositional predicates (3), which are incompatible with èto. It is important to
note that, although the two terms linked by èto may not always appear to be in an identity relation
(especially if one of them is indefinite), Pereltsvaig (2001), Matushansky and Spector (2003,
2004) and others, demonstrate that we are dealing with equative clauses even in these cases.
   The goal of this paper is to explain an unexpected contrast that occurs in copular constructions
when the tense is changed from present to past. As (4) shows, for sentences with quantified (or
generic) subjects this change causes ungrammaticality.
    The explanation we propose is semantic in nature: if copular constructions like the one in (4)
are identity statements, then their truth or falsehood cannot be restricted to a particular moment in
time (either two entities are identical or they are not). The only situation in which an identity
statement can hold in the past is if the entities involved no longer exist or are dead (Pereltsvaig
(2001)) – the well-known lifetime effect of Russian Nominative copulas, also observed with
individual-level predicates (Musan (1997), Jäger (1999), among others). This effect explains the
acceptability of the copular construction with a definite subject given in (2). It also explains the
fact that generic identity statements such as (5a) can be saved when applied to extinct species
    However, there exists another way of rescuing identity copular sentences in the past tense:
shifting the point of view. As (6) shows, this does not have to be done via embedding under an
intensional verb, and thus (5a) is acceptable if understood as a (prevailing) point of view at some
past time. The effect disappears if an explicit past tense adverbial having nothing to do with the
point of view shift is introduced (7).
   Russian is generally assumed to have no Sequence Of Tense, except in relative clauses
(Kondrashova (1998)): present tense indicates simultaneity, either with utterance time or with the
time of the embedding attitude verb, while past tense indicates temporal precedence. This is why
Schlenker (2003) argues that Russian present tense is a shiftable indexical whose point of
evaluation may be the time of either the actual or the reported speech act. We complement this
theory with the proposal that Russian past tense is purely indexical and is always evaluated with
respect to the actual utterance time. This means that the past tense morphology may appear in an
embedded clause that is simultaneous with its embedding clause, and this is why an opinion shift
rescues past tense identity statements.
   The same effects in identity statements can be observed in Dutch, which is an SOT language.
This confirms our conclusion that identity statements cannot be restricted to a particular time, as
well as our primary observation that an opinion shift can rescue a past tense identity statement.
Grammatical effects of opinion shifts have been independently observed by Coppieters (1974,
1975, 1983) for the choice of French pronominal subjects in embedded copular sentences and
Matushansky and Spector (2003, 2004) for post-copular NP-marking in French copular clauses.
Likewise, Cabredo-Hofherr (2004) shows that the weak deictic ce/das is possible with French and
German impersonals only if the speaker has emotional involvement in the situation – embedded
contexts included.
   We demonstrate that both simultaneous and past-shifted readings are available for Russian
past tense clauses embedded under past-tense attitude verbs, but their accessibility varies with the
choice of the matrix verb and the aspect of the embedded verb. We also show grammatical effects
of opinion shifts in other attitude contexts, such as embedding under intensional adjectives and
nominalizations, which further confirms our conclusion that the phenomenon is semantic.
(1)   a.      Ciceron byl velikij    orator/       Tullij.
              Cicero was great       orator-Nom    Tully-Nom
      b.      Ciceron byl velikim oratorom/ * Tullijem.
              Cicero was great        orator-Instr Tully-Instr
              ‘Cicero was a great orator/Tully.’
(2)   a.      Ciceron èto byl velikij      orator/        Tullij.
              Cicero this was great        orator-Nom     Tully-Nom
      b.   * Ciceron èto byl velikim oratorom/            Tullijem
            Cicero this was great          orator-Instr   Tully-Instr
            ‘Cicero was a great orator/Tully.’
(3)   Ciceron (*èto) byl izvesten/ izvestnyj/ v Rime.
      Cicero (*this was famous-SF famous-LF in Rome
      ‘Cicero was famous/in Rome.’
(4)   Každyj učenik        èto (*byl) potencial’nyj student.
      Every student        this (*was potential            (under)graduate
      ‘Every school pupil is a potential college student.’
(5)   a. * Slony      eto byli vodoplavajuščie.
           elephants this were aquatic animals
           ‘Elephants used to be aquatic animals.’
      b.      Dinozavry eto byli presmykajuščiesja.
              dinosaurs this were reptiles
              ‘Dinosaurs were reptiles.’
(6)   S    točki zrenja        pervyx xristian,           rimljane eto byli počti čerti
      with point view-Gen first          Christians-Gen Romans this were almost devils
      ‘From the point of view of early Christians, Romans were nearly devils.’
(7) * V      9om godu          N.È.,    rimljane           eto          byli      počti čerti
      In     9th    year       AD       Romans             this         were      almost devils
      ‘In the 9th year AD, Romans were nearly devils.’
Cabredo-Hofherr, P. (2004). The Alternation of Subjects in Weather Predicates in French and in
       Westgermanic. Paper presented at Séminaire de l’UMR 7023, Paris, December 13, 2004.
Coppieters, R. (1974). Pronouns and Adjectives in French. A Further Confrontation. In:
       Proceedings of NELS 5.
Coppieters, R. (1975). The Opposition Il and Ce and the Place of the Adjective in French.
       Harvard Studies in Syntax and Semantics 1, pp. 221-280.
Coppieters, R. (1983). Descriptions and Attitudes: The Problem of Reference to Individuals.
       Studies in Language 6, pp. 1-22.
Jäger, G. (1999). Stage Levels, States, and the Semantics of the Copula. In: E. Lang and L. Geist,
       eds., ZAS Papers in Linguistics 14. Berlin: ZAS, pp. 65-94.
Kondrashova, N. (1998). Embedded Tenses in English and Russian. Ms., Cornell University.
Matushansky, O. and B. Spector (2003). To Be (a) Human. Paper presented at Journées
       scientifiques Sémantique et Modélisation, Paris.
Matushansky, O. and B. Spector (2004). Tinker, Tailor, Soldier, Spy. Paper presented at Sinn und
       Bedeutung 9, Nijmegen, November 1-3, 2004.
Musan, R. (1997). Tense, Predicates, and Lifetime Effects. Natural Language Semantics 5, pp.
Pereltsvaig, A. (2001). On the Nature of Intra-Clausal Relations: A Study of Copular Sentences in
       Russian and Italian. Ph.D. thesis, McGill.
Schlenker, P. (2003). A Plea for Monsters. Linguistics and Philosophy 26, pp. 29-120.

To top