For absorbing aerosols like soil (or mineral) dust, radiative forcing at the surface differs substantially from the value at the top of the atmosphere (TOA). The climate response depends not only upon the TOA forcing, but its difference with respect to the surface value, which represents radiative heating within the atmosphere. Surface forcing alters evaporation and the hydrologic cycle, which feeds back upon the aerosol burden through the efficiency of wet deposition. We calculate the surface forcing by soil dust aerosols and its global sensitivity by varying aspects of the dust distribution that are poorly constrained by observations. Ignorance of the global dust burden corresponds to a forcing uncertainty of over a factor of two, with smaller uncertainties due to imprecise knowledge of particle optical properties and the particle size distribution. While global evaporation and precipitation are reduced in response to surface radiative forcing by dust, precipitation increases locally over desert regions, so that dust emission can act as a negative feedback to desertification. The effect of the global reduction in precipitation is to lengthen the particle lifetime by reducing the efficiency of wet deposition, representing a positive feedback upon the global dust burden. For the current climate, we calculate the reduction in wet deposition by dust radiative forcing and find that the aerosol burden is increased only modestly. However, the global dust burden and associated radiative forcing are substantially higher during glacial climates, so that the amplification of the dust load by this feedback is larger. By extrapolating from its radiative forcing in the current climate, we estimate that dust reduces precipitation during glacial times by as much as half the reduction due to the colder climate alone.