Riku Erkkilä (University of Helsinki):

The construal of the landmark as the basis of a grammatical distinction

In my paper I will explore the grammatical distinction between the goal-oriented cases in Erzya Mordvin. Erzya has two spatial cases denoting action towards a goal, namely illative and lative. Compared to many other Finno-Ugric languages, this is somewhat unexpected, because otherwise the paradigm of spatial cases is symmetric: There is one case denoting static location, one case denoting source, and one case denoting path. Both goal-oriented cases are productive members of the paradigm, and there are no morphosyntactic restrictions for their occurrence, i. e. both can occur with nouns and postpositions. The study is based on corpus data from the MokshEr-corpus of University of Turku (MokshEr), from where 200 examples of the illative and lative were drawn and analyzed.

            Previous studies have explained the distinction as a lexically ruled feature (i. e. certain words marked as goals display the illative case, and others the lative case, Bartens 1978: 166–174). Furthermore, it has been suggested that case selection depends on semantic properties, especially the dimensionality of the landmark (Alhoniemi 1985: 52). Though these explanations are on the right track, there are certain instances where they don’t apply. I will propose a new explanation for the variation occurring in the goal-marking in Erzya, namely that the choice between the goal-marking cases depends on the construal of the landmark. This can be seen in such pairs as (1) and (2).

(1)       ton       kudo-zo-m                   sa-t      miljit͜sjija          marto

            you      home-ill-1px come-2sg        police  with

            ‘You come to my home with the militia’

(2)       Borja   ava-nzo           marto  jaka-sj              Rjabinjina-v [- -]

            Borya  mother-3px     with     travel-pst.3sg Ryabinina-lat

            ‘Borya traveled with his mom to Ryabinina [- -]’

Arguably both landmarks in the examples (kudo ‘home’ and Rjabinjina ‘(village of) Ryabinina’) are similar because their referents are known, definite, and unique. Still, they display a different goal-marking pattern. This is due to the fact that the landmarks have different profiles in the construal. In (1) the landmark is construed to be more or less fully in the functional domain of the trajectory, and so the whole referent of the landmark is profiled in the construal. In (2), the functional domain of the trajectory is small compared to the size of the landmark entity. In this case only the part of the landmark that is in the functional domain of the trajectory is profiled in the construal. This difference in profiling the landmarks in Erzya leads to the variation in the grammatical marking of the landmark.


Alhoniemi, Alho 1985: Über die wohin-Kasus im Mordwinischen. – Ural-Altaische Jahrbücher 5 s. 45–53.

Bartens, Raija 1978: Synteettiset ja analyyttiset rakenteet lapin paikanilmauksissa. Mémoires de la Société Finno-Ougrienne 166. Helsinki: Société Finno-Ougrienne.

MokshEr = MokshEr V.3 2010. Electronic corpus of Mokša and Erzya. University of Turku.