This paper reviews the evidence and arguments for reconstructing a person–number agreement paradigm for the Proto-Trans-Himalayan (=Sino-Tibetan) verb, and assesses the counter-arguments which have been presented in the literature. We demonstrate the cognacy of verb agreement paradigms across the family, and show that there is no plausible subclassification of the family which would place all the attesting languages in a single branch of the family, and no case for a “Rung” branch. The agreement systems of Jinghpaw and Northern Naga and the archaic postverbal paradigms of South Central/Kuki-Chin are demonstrably cognate to those of Rgyalrongic and Kiranti, and these languages have no common ancestor more recent than PTH.