Overt and zero marking of spatial relations in Cappadocian Greek: synchrony, diachrony, typology

A bulk of research has demonstrated that languages differ with respect to the pattern used to mark the distinction between Goal, Place and Source (Creissels 2006). Among the major findings is that all polysemic patterns in the tripartite distinction Goal—Place—Source are possible, but some permutations are more dominant than others; e.g., Goal—Place polysemy is particularly common (Lestrade 2010, Nikitina 2009). Most studies to date have been concerned with instances whereby the locative role receives an overt marker. Recent work by Stolz et al. (2014), however, has drawn attention to the phenomenon of zero-marking of spatial relations showing that the generalisations formulated for overt markers generally hold for zero marking, as well. In this paper, we investigate overt and zero marking of Goal, Place and Source in the varieties of Cappadocian Greek, which fall into three different categories based on the inventory they use to mark the three spatial relations: (a) conservative varieties, in which Goal and Place are syncretically marked by the inherited preposition EIS whereas Source is marked by the inherited preposition APO; (b) the intermediate varieties of Phloita and Silli, in which Goal and Place are syncretically marked by either EIS or zero, the distribution of which is syntagmatically conditioned (in the sense of Stolz et al. 2014), whereas Source is marked by APO; and, (c) the innovative variety of Ulaghatsh, in which Goal and Place are syncretically marked by zero whereas Source is marked by APO. Both EIS and APO form Prepositional Phrases of the type [EIS/APO + NPACC]PrepP. In the cases where zero is used, we find bare accusative-marked NPs. We first observe that the reorganisation of the system used for the marking of spatial relations in Cappadocian may have had a local effect—in our case, the loss of a member of the prepositional paradigm—but kept the original global picture intact (Goal = Place ≠ Source), thus conforming to crosslinguistically robust tendencies (Lestrade 2010, Nikitina 2009). We subsequently focus on the consequences of the diachronic replacement of EIS by zero for the marking of Goal and Place in Ulaghatsh Cappadocian with the aim of highlighting the variety’s typological rarity. The major consequence in that connection was that the functions originally encoded by EIS (inter alia, GOAL, PLACE, RECIPIENT) were added to the set of functions that were already encoded by bare accusative-marked NPs. This included PATIENT and THEME, which normally occupy the direct object position as complements of (di-)transitive verbs. In this respect, Ulaghatsh Cappadocian belongs to the rare type of language in which the same means of formal marking is used for the encoding of direct object, indirect object, goal and place (see Blansitt 1988). What is more, the fact that accusative and nominative forms of nouns are identical in the variety has the even rarer consequence that the markers used for the direct object (PATIENT, THEME), indirect object (RECIPIENT, ADDRESSEE, BENEFICIARY), GOAL and PLACE functions are always formally identical to those used for the subject functions (AGENT, EXPERIENCER, THEME). References Blansitt, E. 1988. Datives and allatives. In Μ. Hammond, E. Moravcsik, & J. Wirth (Eds.), Studies in Syntactic Typology, 173–191. Amsterdam: Benjamins. Creissels, D. 2006. Encoding the distinction between location and destination: A typological study. In M. Hickmann & S. Robert (Eds.) Space in languages: Linguistic systems and cognitive categories, 19–28. Amsterdam & Philadelphia: John Benjamins. Lestrade, S. 2010. The Space of Case. Nijmegen: Radboud Universities Nijmegen. Nikitina, T. 2009. Subcategorization pattern and lexical meaning of motion verbs: a study of the source/goal ambiguity. Linguistics 47(5), 1113–1141. Stolz, T., Lestrade, S. & C. Stolz. 2014. The Crosslinguistics of Zero-Marking of Spatial Relations. Berlin/Boston: De Gruyter Mouton.