Reconstructing Semantic Roles: Proto-Indo- European *-bhi
Traditional approaches to the reconstruction of Proto-Indo-European nominal morphosyntax have operated by first reconstructing the set of morphological cases for every declensional type, and then attempting to establish the meaning of the different cases, regardless of the specific ending that was used for each declensional type. However, more insight can be gained into the reconstruction of the nominal syntax of proto-languages by applying the concepts and methodologies developed in recent
... loped in recent years in functional-typological approaches to language study. Under this approach, the aim of syntactic reconstruction in the nominal domain lies not in determining the meaning of a given case as a whole but rather in elucidating the semantic role(s) that a specific formative could be used for and, to the extent that this is possible, how those semantic roles relate to each other in historical terms. In this article we survey the semantic roles related to *-bhi-endings in the old Indo-European languages. In the traditional reconstruction, *-bhi has been considered the suffix expressing the Instrumental plural of the athematic declension. However, in the various branches of the family in which it is attested, *-bhi-endings express a broad array of semantic roles. When charted on a diachronic semantic map of Instrument and related semantic roles, the *-bhi-endings appear to cover neighbouring areas, and it becomes clear that they have followed well-known paths of semantic change. If we add the information about *-bhi in the pronominal declension and its etymology, a neat grammaticalisation process is revealed. This results in a 'dynamic' reconstruction of the morphosyntax of the protolanguage, which is more in accord with what we know about the actual processes of semantic change in grammatical markers and paradigmatisation of markers more generally. Eugenio R. Luján and Ángel López Chala -9789004392007 Downloaded from Brill.com06/16/2020 04:36:25AM via free access 1 This article is a result of the research project "Studies on nominal morpho-syntax in Palaeohispanic and Ancient Indo-European languages" (FFI2015-63981-C3-2), which has the financial support of the Spanish Ministry of Economy and Competitiveness. We are very grateful to two anonymous reviewers, as well as to Jóhanna Barðdal, for their comments and suggestions, from which we have greatly benefited in the final version of our article.