Reflex prediction

Abstract While analysing lexical data of Western Kho-Bwa languages the Sino-Tibetan or Trans-Himalayan family with help a computer-assisted approach for historical language comparison, we observed gaps in where one more varieties lacked forms certain concepts. We employed new workflow, combining manual and automated steps, to predict most likely phonetic realisations missing our data, by making systematic use information on sound correspondences words that were potentially cognate forms. This procedure yielded list hypothetical reflexes previously identified sets, which first preregistered as an experiment prediction unattested word then compared actual elicited during secondary fieldwork. In this study describe workflow used process elicitation present results reflex experiment. Based experiment, identify four general benefits comparison. These comprise (1) increased transparency linguistic research, (2) efficiency field source work, (3) educational aspect offers teachers learners wide plethora phenomena, including regularity change, (4) possibility kindling speakers’ interest their own heritage.