Eduardo Rivail Ribeiro
University of Chicago & Museu Antropológico/UFG, Brazil
This paper presents a reconstruction of the phonology, morphology, and a sample lexicon of Proto-Jê, the common ancestor of more than a dozen languages spoken in Brazil. The Jê language family comprises three subgroups (Northern, Central, and Southern), whose members are spoken from the Amazon region to the southernmost state of Rio Grande do Sul. Although a reconstruction of Proto-Jê has been proposed earlier by Davis (1966), a number of issues remain to be solved. Although Davis made some lasting contributions (such as demonstrating that the Southern Jê languages are indeed part of the family, a matter which had until then been the subject of controversy), his reconstructions are, in many cases, flawed. This can be attributed in part to the lack of accurate descriptions of most Jê languages at the time. Since then, a number of additional descriptive materials (including grammars and dictionaries) have appeared, thus providing the basis for a more comprehensive reconstruction of the proto-language.
The paper starts with a critical assessment of Davis’ work, pointing out some of the factors which may have contributed to hinder his reconstruction (including the lack of morphological information, which lead to the postulation of erroneous phonological correspondences). In certain cases, the revision of Davis’ proto-forms is prompted by data from a ‘new’ language, Panará (Northern Jê), the historical descendant of Southern Kayapó (which, until the 1960s, was believed to be extinct). Since Davis’ reconstruction has provided the basis for comparisons between Jê and other families (such as Karajá, Maxakalí, Ofayé, Rikbaktsá, and Krenák), this review has immediate consequences for our understanding of the Macro-Jê stock. In addition, the paper presents for the first time a reconstruction of Proto-Jê morphology, with an account of the factors which lead to the restructuring of the original paradigms in certain languages (such as the role of analogical leveling and phonological processes leading to morphological erosion and homonymy). Furthermore, the paper offers a considerably larger number of reconstructed forms, substantially improving the possibility of finding additional cognates in other families and providing a glimpse into Proto-Jê culture (revealing, for instance, lexical evidence for the existence of weaving among the speakers of the proto-language).