There is growing concern that in response to global warming, ice sheets may contribute significantly to sea level rise over the coming centuries Despite this concern, many critical ice sheet processes are not well understood and not included in prognostic simulations of future ice sheet changes. Iceberg calving, for example, is one of the most prominent and least understood ice sheet processes, accounting for between half to two-thirds of the mass lost from the Greenland and Antarctic ice sheets. My group has been developing theories of iceberg calving, based on stochastic physics that begin to explain the many different styles of iceberg calving observed in nature. The idea underlying this approach is that the precise timing of calving events are difficult to fathom and depend on difficult to predict variables such as the arrival of big pulses of ocean swell or the precise arrangement of fractures within the ice. This suggests that individual calving events may be viewed as a random or stochastic. We explore this approach here and show how to derive a continuous and slowly varying calving-law from an underlying stochastic and discrete process. A consequence of this approach is that we can predict both the mean terminus position and the magnitude of fluctuations about the mean position.