Apart from the second group, which relies on lexicostatistic methodology, the tree topologies in these proposals are based on an investigator’s perception of relative proximities between branches, with no quantification of uncertainty.
A search for linguistic innovations uniting several branches of the family is ongoing; the limited results so far are consistent with the first group of hypotheses (9, 17). Here we combine classical historical linguistics with cutting-edge computational methods and domestication studies.
These difficulties place some uncertainty on cognate identification and, in turn, affect our ability to identify shared innovations. The Sinitic languages, whose ancestor was spoken about 2,000 y ago, form a homogeneous group in the eastern half of the Sino-Tibetan area.The list of most appropriate concepts was established through careful evaluation of concept lists used in similar studies (, section 3), and lexical cognates were identified by experts in Sino-Tibetan historical linguistics using the comparative method supported by state-of-the-art annotation techniques.Second, we apply Bayesian phylogenetic methods to these data to estimate the most probable tree, outgroup, and timing of Sino-Tibetan under a range of models of cognate evolution; similar methods have been applied to several other families of languages, including Indo-European (18–20), Austronesian (12), Semitic (21), and Bantu (22).The area with the most diverse Sino-Tibetan languages is in northeastern India and Nepal.This has suggested to some authors that the family’s homeland was located there (10).
Search for dating sino:
The Sino-Tibetan family comprises about 500 languages (1) spoken across a wide geographic range, from the west coast of the Pacific Ocean, across China, and extending to countries beyond the Himalayas, such as Nepal, India, Bangladesh, and Pakistan (map, , section 2).