This paper presents an internal classification of Tupí-Guaraní based on lexical data from 30 Tupí-Guaraní languages and 2 non-Tupí-Guaraní Tupian languages, Awetí and Mawé. A Bayesian phylogenetic analysis using a generalized binary cognate gain and loss model was carried out on a character table based on the binary coding of cognate sets, which were formed with attention to semantic shift. The classification shows greater internal structure than previous ones, but is congruent with them in several ways.
January 1, 2015
Michael, Lev, N. Chousou-Polydouri, Keith Bartolomei, Erin Donnelly, Vivian Wauters, and Zachary O'Hagan. 2015. A Bayesian Phylogenetic Classification of Tupí-Guaraní. LIAMES, 15. UC Berkeley. http://escholarship.org/uc/item/1331v899