This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-commercial and no modifications or adaptations are made.ABSTRACT: Landslides contribute to dismantle active mountain ranges and faults control the location of landslides. Yet, evidence of the long-term, regional dependency of landslides on active faults is limited. Previous studies focused on the transient effects of earthquakes on slope stability in compressive and transcurrent regimes. Here we show that in the Peloritani range, NE Sicily, Italy, one of the fastest uplifting areas in the Mediterranean, a clear geographical association exists between large bedrock landslides and active normal faults of the Messina Straits graben. By interpreting aerial photographs, we mapped 1590 landslides and sackungs and 626 fault elements and their facets in a 300 km 2 area in the eastern part of the range. We used the new landslide and fault information, in combination with prior geological and seismic information, to investigate the association between bedrock landslides and faults. We find that the distribution and abundance of landslides is related to the presence of large active normal faults, and matches the pattern of the local historical seismicity. Landslide material is more abundant along the East Peloritani Fault System where the long-term activity of the faults, measured by the average yearly geological moment rate, is larger than in the West Peloritani Fault System where landslides are less abundant. Along the fault systems landslide material concentrates where the cumulated fault throws are largest. We conclude that large landslides and their cumulated volume are sensitive to local rates of tectonic deformation, and discriminate the deformation of the single fault segments that dissect the Peloritani range. Our findings are a direct test of landscape evolution models that predict higher rates of landslide activity near active faults. Our work opens up the possibility of exploiting accurate landslide and fault maps, in combination with geological and seismic information, to characterize the long-term seismic history of poorly instrumented active regions.