Authors

Discussion

The paper provides Czech data to demonstrate the characteristics of a consistent null subject (NS) language (full pro-drop) in terms of Holmberg (2005). Czech finite predicates, although located in a low position, carry a rich agreement morphology allowing NS with a non-contrastive definite (specific) interpretation in all Tense-Aspect contexts for all three Persons both singular and plural. On the other hand, a standard Czech generic (non-specific) subjects require an overt proform jeden (one) or a standardized NS.

The paper concentrates on Czech structures with (i) obligatory NS and (ii) those which do not allow NS. The former include agentless predicates, generic 2sg subjects, or anaphoric subjects in non-root clauses with a hierarchically symmetrical interpretation, e.g. identifying (pointing) copular structures with pronominal predicates: *Ø/To jsem já. (It is me.).

The paper also refers to the NS with analytic predicates as described in Veselovská 2020. It proposes a complex agree process including several independent checking domains for (i) the nominal φ feature set, (ii) the [D] (Person, Number?) feature set and (iii) a Topic/Focus feature.

The theoretical framework used for the discussion and analyses is based on the diagnostics and claims made in studies by Alexiadou & Anagnostopoulou (1998), Biberauer et al. (2010), Kučerová (2014).