Triangulation reduces the polygon of error for the history of Transeurasian

Abstract

In a recent study we used an interdisciplinary approach combining linguistics, archaeology and genetics to analyse the Transeurasian languages1. Our analysis concluded that the early dispersals of these languages were driven by agriculture. A preprint published on this server presents objections to the Transeurasian hypothesis and its association with farming dispersals2. However, close inspection of that text reveals numerous misinterpretations and inconsistencies. In the interest of furthering scientific debate over Transeurasian language and population history, we address the critiques, revising datasets and fine-tuning approaches. The linguistic critique questions the quantity and quality of our datasets. Here we show that the number of surviving cognate sets for Transeurasian is in line with that for well-established language families. In addition, we find that Tian et al.s failure to reject a core of regularly corresponding cognates in the basic vocabulary creates ground for a consensus about the genealogical relatability of the Transeurasian languages. The archaeological critique attempts a re-analysis of one Bayesian test using re-scored data only for northern China. Over half of the suggested re-scorings contain inconsistencies and it is not explained why the re-analysis retains the original data for sites outside northern China, comprising almost 60% of the total. More importantly, the sweeping claim that there is no evidence supporting the prehistoric migrations analysed in our study is not backed by any discussion of the archaeological record. With respect to genetics, the preprint claims a re-analysis showing that the data do not conclusively support the farming-driven dispersal of Turkic, Mongolian, and Tungusic, nor the two-wave spread of farming to Korea. In fact, the only genetic re-analysis presented is limited to samples from Korea and Japan and does not contradict our original conservative modelling of Neolithic individuals with Hongshan and our Bronze Age ones with Upper Xiajiadian. In sum, in bringing multiple lines of evidence together through triangulation, we gained a more balanced and richer understanding of Transeurasian dispersals than each discipline could provide individually. Our research doubtless leaves room for improvement but we remain confident that triangulation did not fail, but rather brought us a step closer to understanding the history of the Transeurasian languages.