Phase 1: Cultural diversification: Project 037|
Archaeology and Language: the origin and dispersal of Tupi speakers in lowland South America
Stephen Shennan (Institute of Archaeology, University College London)
Fabiola da Silva
In recent years the relationship between patterns in the archaeological record and the history of languages and the people that spoke them has attracted renewed interest on the part of archaeologists and historical linguists alike. Central to the issue is the extent to which language and material culture based practices were transmitted through time in parallel with one another, and whether they can be taken as proxies for histories of populations. This research is the initial stage of an interdisciplinary project to address these issues for the case of the languages of the Tupi stock in lowland South America. Apart from the general theoretical importance of the topic, the project is of great substantive significance for the prehistoric archaeology of this large region because languages of this stock are spread over large parts of Brazil, Argentina and Paraguay and the history of this dispersal and the processes that produced it continue to be the subject of intense debate (e.g. Noelli 1998).
The project involves linguistics and archaeology, as well as making use of ethnoarchaeologicl data, and builds on existing work by Brazilian and international scholars. It is envisaged that if the first season is successful the project will last for five years.
On the linguistics side, the Comparative Tupi Project is a collaborative project whose aim is to collect comparable data for languages representing each of the 10 language families of the Tupi stock, with the aim of reconstructing proto-Tupi, believed to have been spoken 4-5000 years ago. On linguistic grounds – 6 out of the 10 extant families are represented there – the region where it existed is believed to have been Rondonia, in southwest Amazonia.
The initial stage of the archaeology project is funded by the British Academy and involves a comparative analysis of pottery assemblages from known Guarani sites in southern Brazil, known Tupi sites in the Lower Tocantins valley in Amazonia and from sites of unknown linguistic affiliation from the putative homeland area of Rondonia. The comparative analysis requires a quantitative description of the forms, fabric and decoration of a representative sample of the pottery from these sites. Samples will be sent to the Museum of Archaeology and Ethnology, University of Sao Paulo, for direct radiocarbon dating.
The Rondonia assemblages come from sites excavated in the 1970s and 1980s by Dr E. Miller. This material, together with representative Guarani material, is now in the museum at Takwara (MARSUL), Rio Grande do Sul. The Tocantins material is at the Emilio Goeldi Museum, Belem. The Brazilian archaeological co-director for this project is Dr. Eduardo Neves, University of Sao Paulo. The work is being carried out by Prof. Stephen Shennan, Institute of Archaeology, University College London, and Mr Fernando Almeida, University of Sao Paulo.