CSLI Publications logo
new books
contact us
for authors
CSLI Publications

Erschler, David: Sluicing-like Phenomena and the Location of CP in Ossetic

In this talk, I show that Digor and Iron Ossetic, closely related Iranian languages spoken in the Central Caucasus, show sluicing in the proper sense and argue on this basis for the low (preverbal) position of complementizers in the Ossetic clause structure. The classical sluicing occurs in wh-fronting languages, and it is only the wh-phrase(s) that is retained:

(1) Anne invited someone, but I don’t know who. (Merchant 2001: 40)

Sluicing-like constructions, however, are also attested in languages that either lack wh-movement or show non-canonical patterns thereof. The issue how to analyze such structures is controversial, see, e.g. the polemics between Ince (2012) and Hankamer (2011) on the proper analysis of Turkish sentences of this type.

In Ossetic, wh-phrases and certain complementizers are obligatorily preverbal, no matter where the verb is situated in the clause, and only certain items may intervene between the wh-phrases/complementizer and the verb, Erschler (2012). It is an obvious challenge to theories of “left periphery” to analyze such a structure: even if the preverbal placement of wh-phrases might be explained away as focusing, it can be hardly done so for complementizers. Despite the unusual patterns of wh-placement and complementation, Ossetic shows a construction that is at least superficially similar to sluicing. I show that, analogously to the “classical” sluicing, this type of ellipsis can go backwards, is unbounded, and is not sensitive to islands.

That Ossetic (pseudo)-sluicing is indeed an independent type of ellipsis, and not a subvariety of stripping [as argued for Turkish pseudo-sluicing in Hankammer (2011)], follows from the fact that, unlike sluicing, Ossetic stripping is ungrammatical in dependent clauses and cannot go backwards.

Another potential analysis of pseudo-sluicing, reduced cleft, is ruled out because the copula cannot be inserted in sluices, cf a similar argument in Toosarvandani (2008: 682-684) for Persian. Thus it is natural to analyze Ossetic (pseudo)-sluicing as the classical sluicing, i.e. as deletion of the material below C. If my proposal is on the right track, it provides additional evidence that Ossetic indeed has C in the preverbal position.


Maintained by Stefan Müller

Last modified: January 09, 2019

pubs @ csli.stanford.edu