New data from Chapel Beck in the southwest Howgill Fells combined with existing data support the identification of three phases of extensive hillslope gullying after 2200, between 1250–700 and after 500 cal. BP. Examination of cumulative probability density function (CPDF) plots of the distribution of calibrated radiocarbon dates confirms the identification of these phases. There are differences in the spatial distribution of geomorphic activity between these phases; with the c. 2200–1800 cal. BP episode limited to the western Howgill Fells. In comparison, the 1250–700 and post-500 cal. BP episodes are more widespread cross-region phenomena, and affected the more remote central fells. Pollen and geochemical data reveal several phases of increased human activity in these uplands, with small-scale temporary clearances during the late Bronze Age / early Iron Age and more substantial clearances during the late Iron Age and Romano-British times (2300–1500 cal. BP), and these were followed by woodland expansion and decline in human activity during both the mid- to late Iron Age and the period after the Roman withdrawal from Britain (after 1500 cal. BP). The final substantial woodland clearances occurred from c. 1100–900 cal. BP, after which there has been little woodland recovery. A range of pollen sites across the Howgill Fells indicate that there is spatial variability in the level of human activity and impact. Iron Age / Romano-British woodland clearances appear to have affected the west of the region more than the remote central valleys, whereas the later 1100–900 cal. BP woodland reduction affected the entire region. This pattern is similar to that identified from the geomorphology, and demonstrates the importance of pollen sites that record local palaeo-vegetation information and the significance of using a series of sites to explore intra-regional variability in the vegetation history. We suggest that human activity is a critical factor in mediating the late Holocene geomorphic record, but climate and particularly individual storms are the trigger mechanism responsible for erosion and incision on hillslopes and within the wider fluvial system. Palaeoclimate data (humification and plant macrofossil) from peat deposits at Archer Moss identify shifts to wetter conditions: c. 3100–2700; 2400–1700; 1300–1100; and after 650 cal. BP, and probably reflect increases in summer wetness. Of these, the 2400–1700, 1300–1100 and post-650 cal. BP wetter phases coincide with increased geomorphic activity, and perhaps encouraged incision on hillslopes rendered more susceptible to erosion by the activities of people.