Haudry Jean. Préhistoire de la flexion nominale indo-européenne

Lyon: Institut d'études indo-européennes de l'Université Jean Moulin (Lyon III), 1982.Starting from the so-called ' syncopated finals' of the Vedic language (more properly to be regarded as old uninflected forms) the author develops a theory of the origin of the case-endings, which are traced to earlier postpositional elements still sometimes found occurring elsewhere in a free state. The thematic vowel in nominal formations is identified as a postposed definite article of pronominal origin to which case-endings were subsequently added. The identity of nominative and genitive singular endings in this class which is normal in Hittite and can be traced in residual forms in other languages is regarded as original, but the theory that the thematic class arose from the reinterpretation of the genitive singular of thematic nouns in -e/os as a nominative is rejected for the good reason that such a reinterpretation would presuppose the existence of thematic nominatives in -os. Parallels for the expansion of adjectival forms by the addition of pronominal elements are cited from Baltic, Slavonic and Germanic, and comparisons are also drawn with the 'formal' use of the relative pronoun in such phrases as Vedic maruto ye sahasah, where, it is argued, the insertion of the relative pronoun acts as a 'distanciateur' between noun and adjective rather than strengthening the link between them. Many of the common suffixes found in noun-formation are explained as the result of 'hypostasis': formerly case-endings, they had as a consequence of the accretion of further postpositions been reinterpreted as stems. Where these new stems had been incompletely generalised, heteroclitic inflexion could result. Complex suffixes consisting of more than one case-ending could be built up in this way; the author tries to relate their functions to those of the underlying case-endings. (https://archive.org/details/@arbomhard)