Rock slope failures in the Lake District, UK, have been associated with deglacial processes after the Last Glacial Maximum, but the controls and timing of the failures remain poorly known. A cirque headwall failure was investigated to determine failure mechanisms and timing. The translated wedge of rock is thin and lies on a steep failure plane, yet the friable strata were not disrupted by downslope movement. Fault lines and a failure surface, defining the wedge, were used as input to a numerical model of rock wedge stability. Various failure scenarios indicated that the slope was unstable and would have failed catastrophically if not supported by glacial ice in the base of the cirque. The amount of ice required to buttress the slope is insubstantial, indicating likely failure during the thinning of the cirque glacier. We propose that, as the ice thinned, the wedge was lowered slowly down the cirque headwall, gradually exposing the failure plane. A cosmogenic 10Be surface exposure age of 18.0±1.2 ka from the outer surface of the wedge indicates Late Devensian de-icing of the backwall of the cirque, with a second exposure age from the upper portion of the failure plane yielding 12.0±0.8 ka. The 18.0±1.2 ka date is consistent with a small buttressing ice mass being present in the cirque at the time of regional deglaciation. The exposure age of 12.0±0.8 ka represents a minimum age, as the highly fractured surface of the failure plane has experienced post-failure mass-wasting. Considering the chronology, it appears unlikely that the cirque was reoccupied by a substantial ice mass during the Younger Dryas stadial.