Locative inversion in Bantu and predication

Abstract

In generative grammar, locative inversion in Bantu languages is typically analyzed in terms of A-movement of the locative from a VP-internal position to the subject position. I present an alternative analysis, according to which the locative subject-DP is introduced above the νP/VP, in the specifier of a functional category whose head selects the νP/VP as its complement. I suggest that this category is Pr (for "predication"), i.e., the same category that also introduces the subject argument of adjectival or nominal predicates in non-verbal predication constructions (see Bowers 1993, and especially Baker 2003a for Bantu). In locative inversion, Pr establishes a non-canonical predication relation between a νP/ VP that expresses a state or event, and a DP that denotes the location of which this state/event is predicated as a property. My analysis is developed on the basis of a detailed discussion of "semantic" locative inversion in the Bantu language Zulu (Buell 2007), a construction in which the inverted subject-DP is not formally marked as a locative, but receives its interpretation solely by virtue of the locative semantics of its head noun.

A typology of Bantu subject inversion

“…If SLI and FLI differ in this respect, and are structurally distinct, it would be possible for them to co-occur in a single language, but this does not seem to be strongly supported by our data. The largely complementary distribution of SLI and FLI in our data would rather indicate that Zeller's (2013) analysis might apply to both SLI and FLI. However, this proposal and its crosslinguistic validity needs further investigation.…”

“…(57) (Halpert 2012: 231) This, together with the restriction of semantic locative inversion in Zulu, motivated Zeller's (2013) analysis, which claims that the locative in SLI originates in a different position (a predication phrase above the verb phrase). If SLI and FLI differ in this respect, and are structurally distinct, it would be possible for them to co-occur in a single language, but this does not seem to be strongly supported by our data.…”

“…If adverbs are DPs and if inversion is about topicality rather than role reversal, there is no obvious reason why temporal adverbs like 'last year' or 'today' would not be allowed as preverbal DP in an inversion construction. This is especially so if the proposals by Zeller (2012Zeller ( , 2013) are taken seriously: as mentioned above, he suggests that instruments and semantic locatives in II and SLI originate in a different position, where they are semantically interpreted as the 'holder' of the state denoted by the predicate. For example, in an SLI construction like 'the garden grows grass' the predicate would be 'grass-growing', and the DP 'garden' "denotes an entity of which this situation is predicated as a property" (Zeller 2013(Zeller : 1111.…”

“…Bresnan and Kanerva 1989, Demuth and Mmusi 1997, Marten 2006, Zerbian 2006, Khumalo 2010, Salzmann 2011, Creissels 2011, van der Wal 2008, Marten and Gibson 2013, Ngoboka and Zeller 2013, Zeller 2013). …”

What the Giant Tells Us About Agreeing Post-Verbal Subjects in Xhosa

“…This has been extensively studied in Bantu linguistics, both for individual languages and comparatively (e.g. Bresnan and Kanerva 1989;Demuth 1990;Demuth and Mmusi 1997;Marten 2006;Zerbian 2006;Buell 2007;Marten 2011;Creissels 2011;Zeller 2013;Marten and van der Wal 2014). The core inversion constructions are recognised through certain diagnostics, to be presented in section 3.3, and have in common that the post-verbal subject is non-topical.…”