a b s t r a c tTwo of the primary external factors influencing the variability of major river systems, over river reach scales, are human activities and tectonics. Based on the rivers Karun and Dez in south-west Iran, this paper presents an analysis of the geomorphological responses of these major rivers to ancient human modifications and tectonics. Direct human modifications can be distinguished by both modern constructions and ancient remnants of former constructions that can leave a subtle legacy in a suite of river characteristics. For example, the ruins of major dams are characterised by a legacy of channel widening to 100's up to c. 1000 m within upstream zones that can stretch to channel distances of many kilometres upstream of former dam sites, whilst the legacy of major, ancient, anthropogenic river channel straightening can also be distinguished by very low channel sinuosities over long lengths of the river course. Tectonic movements in the region are mainly associated with young and emerging folds with NW eSE and NeS trends and with a long structural lineament oriented EeW. These earth surface movements can be shown to interact with both modern and ancient human impacts over similar timescales, with the types of modification and earth surface motion being distinguishable. This paper examines the geomorphological evidence and outlines the processes involved in the evolution of these interactions through time. The analysis shows how interactions between earth surface movements and major dams are slight, especially after ancient dam collapse. By contrast, interactions between earth surface movements and major anthropogenic river channel straightening are shown to be a key factor in the persistence of long, near-straight river courses. Additionally, it is suggested that artificial river development, with very limited river channel lateral migration, may promote incision across an active fold at unusually long distances from the fold "core" and may promote markedly increased sinuosity across a structural lineament.