The Himalayan orogen provides a type example on which a number of models of the causes and consequences of crustal deformation are based and it has been suggested that it is the site of a variety of feedbacks between tectonics and erosion. Within the broader orogen, fluvial drainages partly reflect surface uplift, different climatic zones and a response to crustal deformation. In the eastern Himalaya, the unusual drainage configuration of the Yarlung Tsangpo-Brahmaputra River has been interpreted either as antecedent drainage distorted by the India-Asia collision (and as such applied as a passive strain marker of lateral extrusion), latest Neogene tectonically-induced river capture, or glacial damming-induced river diversion events.Here we apply a multi-technique approach to the Neogene paleo-Brahmaputra deposits of the Surma Basin (Bengal Basin, Bangladesh) to test the long-debated occurrence and timing of river capture of the Yarlung Tsangpo by the Brahmaputra River. We provide U-Pb detrital zircon and rutile, isotopic (Sr-Nd and Hf) and petrographic evidence consistent with river capture of the Yarlung Tsangpo by the Brahmaputra River in the Early Miocene. We document influx of Cretaceous-Paleogene zircons in Early Miocene sediments of the paleo-Brahmaputra River that we interpret as first influx of material from the Asian plate (Transhimalayan arc) indicative of Yarlung Tsangpo contribution. Prior to capture, the predominantly Precambrian-Paleozoic zircons indicate that only the Indian plate was drained. Contemporaneous with Transhimalayan influx reflecting the river capture, we record arrival of detrital material affected by Cenozoic metamorphism, as indicated by rutiles and zircons with Cenozoic UPb ages and an increase in metamorphic grade of detritus as recorded by petrography. We interpret this as due to a progressively increasing contribution from the erosion of the metamorphosed core of the orogen. Whole rock Sr-Nd isotopic data from the same samples provide further support to this interpretation. River capture may have been caused by a change in relative base level due to uplift of the Tibetan plateau. Assuming such river capture occurred via the Siang River in the Early Miocene, we refute the "tectonic aneurysm" model of tectonic-erosion coupling between river capture and rapid exhumation of the eastern syntaxis, since a time interval of at least 10 Ma between these two events is now demonstrated. This work is also the first to highlight U-Pb dating on detrital rutile as a powerful approach in provenance studies in the Himalaya in combination with zircon U-Pb chronology.