Documentation

Linglib.Fragments.SwissGerman.Case

Swiss German Case and Verb Subcategorization @cite{shieber-1985} #

Swiss German uses the same four-case inventory as Standard German (NOM, ACC, GEN, DAT). The critical fact for @cite{shieber-1985}'s argument is that different verbs in cross-serial constructions subcategorize for different cases on their NP objects:

This case-verb pairing is what makes Swiss German cross-serial dependencies produce the pattern a^m b^n c^m d^n (DAT-NPs, ACC-NPs, DAT-Vs, ACC-Vs), which is not context-free.

Verbs that participate in cross-serial subordinate clause constructions.

These are the verbs from @cite{shieber-1985}'s Swiss German data. Each subcategorizes for a specific case on its NP object.

Instances For
    Equations
    • One or more equations did not get rendered due to their size.
    Instances For

      Case required by each verb on its NP object.

      This is the empirical fact that drives @cite{shieber-1985}'s proof: verbs sort into dative-subcategorizing and accusative-subcategorizing classes, and in the cross-serial construction the case on each NP must match the requirement of its corresponding verb.

      Equations
      Instances For

        The two case classes are genuinely distinct — hälfe requires dative while lönd requires accusative. This non-trivial case distinction is what produces the crossed agreement pattern a^m b^n c^m d^n.