| Summary: |
A typology of grammatical features offered in Corbett (2012) and Kibort & Corbett (2008, 2010) makes a crucial distinction between two types of interface features reflected in morphology: (i) morphosemantic features, which affect semantics but do not participate in syntax, (ii) morphosyntactic features, which are both semantically charged and relevant to syntax. In neutral terms, for a feature to be relevant to syntax means that at least some of its values must be determined through a syntactic relation with another word.
Although focus was listed as a possible candidate for a grammatical feature, its status within this typology remained unspecified. If it is a feature, it is an interface feature since it tends to affect syntax and carries an instruction to phonology and semantics, but for most languages the focus feature is purely abstract and irrelevant for morphology. If focus is expressed by a dedicated morphological marker, there is typically no evidence that it is relevant for agreement or government, so at best we can view focus as a morphosemantic feature. This paper contributes to the typology of grammatical features by analysing how focus works in Tundra Nenets (Uralic). I argue that this language has a dedicated marker of exclusive focus which is fully integrated into the morphology of the word of which it is a part. It appears to be the exponent of two different features which do not necessarily overlap: a morphosemantic focus and a morphosyntactic focus. The latter participates in ‘focus spreading’, i.e. some kind of feature transmission partly similar to the phenomenon of ‘definiteness spreading’. Focus spreading shows at least some canonical properties of grammatical agreement. Based on this, I will conclude that Tundra Nenets comes as close as possible to a language in which postulating a marginal morphosyntactic feature ‘focus’ may be justifiable. |
