Documentation

Linglib.Studies.Istratkova2004

Istratkova (2004): On Multiple Prefixation in Bulgarian #

[Ist04] documents the Bulgarian-distinctive feature of multiple prefixation: up to seven prefixes can stack on a single verbal root, with superlexical prefixes systematically outside lexical ones and outer prefixes taking scope over inner ones. Prefixes attach to both perfective and imperfective stems, and the large class of simplex homogeneous verbs (her ex. (2)) has no perfective counterparts and "remains aspectless", behaving as imperfective only by default — her analytical reclassification of the dictionary-imperfective simplexes, encoded as the study-level homogeneousSimplexes class (the fragment records the consensus imperfective value).

Her po- taxonomy distinguishes three superlexical po-'s: delimitative ('for a while', which does not stack), distributive (occurring only after iz-), and attenuative ('to a low degree', the one that stacks over other superlexicals). Where [Sve04]'s (3a) glosses stacked po- as DLMT, the analyses here follow her own labels (attenuative in a23c, distributive after iz- in a22b).

Main definitions #

Main results #

Her ex. (2) class: the simplex homogeneous verbs, which she analyses as aspectless — with no perfective counterparts, imperfective only by default. An analytical reclassification of fragment stems, so it lives here rather than as a fragment field.

Equations
Instances For

    Single-prefix analyses (her (1)/(3) tables) #

    (1a) za-piša 'put down in writing' — lexical za- on piša.

    Equations
    Instances For

      (3a) iz-misl'a 'make up (a story)' — lexical iz- (idiosyncratic meaning shift) on misl'a.

      Equations
      Instances For

        (3b) za-običam 'start to love' — superlexical inceptive za-.

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

          (3c) po-znam 'guess' — lexical po-: the fully idiosyncratic meaning is [Sve04]'s own lexicality diagnostic (56e); her table assigns no superlexical label.

          Equations
          Instances For

            (3d) za-blest'a 'start to glitter' — superlexical inceptive za-.

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

              (3i) pro-četa 'read completely' — lexical (quantizing) pro-: not in her superlexical inventory; the default perfectivization of četa.

              Equations
              Instances For

                Multi-prefix analyses (her §4 stacking data) #

                (22b) iz-po-na-pro-dam — completive iz-, distributive po- (her distributive po- occurs only after iz-), cumulative na-, all outside lexical pro- on dam.

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

                  (23c) po-na-pro-dam 'sell a few things' — attenuative po- over cumulative na- over lexical pro-: the stacking po- is her attenuative one (contrast [Sve04]'s DLMT gloss of (3a)).

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

                    Results #

                    Every analysis is well-stacked: superlexicals outside lexicals.

                    The (1)/(3)-table derivations attach to her homogeneous class: prefixation quantizes these verbs, after which they come in perfective-imperfective pairs.

                    The homogeneous class is dictionary-imperfective — the consensus value her aspectless analysis reinterprets as default behaviour.

                    Her hyphen segmentation equals each analysis' decomposition (all rows are 1sg-present citation forms).