AREAL TYPOLOGY OF PROTO-INDO-EUROPEAN: THE CASE FOR CAUCASIAN CONNECTIONS By RANKO MATASOVIC ´ University of Zagreb ABSTRACT This paper re-examines the evidence for early contacts between Proto-Indo-European (PIE) and the languages of the Caucasus. Although we were not able to find certain proofs of lexical borrowing between PIE and North Caucasian, there are a few undeniable areal-typological parallels in phonology and grammar. Some features generally attributed to PIE are not found in the majority of languages of North and Northeastern Eurasia, while they are common, or universally present, in the languages of the Caucasus (especially North Caucasus). Those features include the high consonant-to-vowel ratio, tonal accent, number suppletion in personal pronouns, the presence of gender and the morphological optative and, possibly, the presence of glottalized consonants and ergativity. 1. INTRODUCTION 1 More than a century ago, Dutch linguist Christian Uhlenbeck suggested that Proto-Indo- European (PIE) was substantially influenced by Caucasian languages. He believed that PIE was a language genetically related to Uralic, but that it underwent areal influences from unknown, but presumably Caucasian languages (Uhlenbeck 1901). Like most of Uhlenbeck’s output, his hypothesis about PIE and Caucasian connections remained largely ignored, but many of his arguments have been revived over the last couple of decades (for a re-evaluation of Uhlenbeck’s work see Kortlandt 2009). Much discussed at the time were the proposals by Trubetzkoy, who in the 1930s claimed that PIE was, in a sense, a mixed language, one component of which came from the Caucasus (Trubetzkoy 1939). Although his denying of the ‘purity’ of PIE was understandable in the period of intensive ideological debates about the significance of ‘Aryan’ languages and ‘races’, Trubetzkoy’s ideas were not well received by many linguists. In the 1960’s and 1970’s, a number of typological parallels between PIE and Kartvelian, chiefly in the verbal system, was investigated by Schmidt (1967). Some of his rather inconclusive findings were gathered by one of his pupils, Shimomiya (1978). In 1984, Gamkrelidze & Ivanov (1984) argued that the PIE homeland was located to the south of the Caucasus mountains, and tried to support this theory by adducing a number of arguments from areal typology, showing structural convergences between PIE and the Caucasian languages (especially Kartvelian). They also discussed a set of lexical isoglosses connecting PIE and Semitic, Kartvelian and ancient Mesopotamian languages, which were, in their opinion, only possible if PIE was spoken south of the Caucasus. However, the majority of 1 Earlier versions of this paper were read in the Russian State University for the Humanities in January 2011, at the meeting of the Philological Society in Oxford (in March 2011), and in the E ´ cole Normale Supe´rieure in Paris (in April 2011). I thank all the audiences who commented on it and helped me express my ideas more clearly, as well as two anonymous reviewers of TPhS. Transactions of the Philological Society Volume 110:2 (2012) 283–310 Ó The author 2012. Transactions of the Philological Society Ó The Philological Society 2012. Published by Blackwell Publishing, 9600 Garsington Road, Oxford OX4 2DQ and 350 Main Street, Malden, MA 02148, USA.