Raymo and Ruddiman (1992) advocated mountain building as a driver of climate change. Molnar and England (1990) pointed to the ambiguities of the sedimentary record (increased sediment flux could be triggered by uplift or climate change) and suggested that erosion in response to climate change could drive peak uplift. In Nature this week, Willenbring and von Blanckenburg present a new contribution to the debate on the relationship between climate (specifically late Cenozoic cooling), weathering, erosion and atmospheric CO2 . They use the marine 10Be/9Be isotope system as a proxy for global weathering rates, and infer steady erosion rates for the last ~12 Myr. Hence, according to their evidence, weathering is not responsible for global cooling, but neither does global cooling trigger enhanced erosion.