The existence of subtropical deserts, such as the Sahara, has often been attributed to the annual-mean, zonal-mean Hadley circulation which shows strong descent in the subtropics. However, the zonal-mean Hadley circulation shows considerable evolution over the course of the year with very strong subtropical descent during winter, but practically no zonal-mean subtropical descent during summer when rainfall over the eastern Sahara and the Mediterranean is least. Charney (1975) proposed a biosphere-albedo feedback mechanism whereby local anthropogenic effects related to over-grazing could affect the radiative balance, enhancing summertime diabatic descent and leading to desertification of the subtropics in general. The present study, which uses an idealized model, suggests a monsoon-desert mechanism for desertification whereby remote diabatic heating in the Asian monsoon region can induce a Rossby-wave pattern to the west. Integral with the Rossby-wave solution is a warm thermal structure that interacts with air on the southern flank of the mid-latitude westerlies causing it to descend. This adiabatic descent is localized over the eastern Sahara and Mediterranean, and over the Kyzylkum desert to the south-east of the Aral Sea, by the mountains of north Africa and south-west Asia. Trajectories indicate that the monsoon-desert mechanism does not represent a simple ‘Walker-type’ overturning cell. Instead, the descending air is seen to be mainly of mid-latitude origin. It is speculated that the monsoon-forced adiabatic descent may result in clear air and, therefore, a local diabatic enhancement which effectively doubles the strength of descent. With this mechanism, desertification can be forced by remote changes in monsoon strength rather than by local effects. This conclusion is supported by the observed dramatic strengthening of descent over the Mediterranean and east Sahara during the onset of the Asian monsoon and, on the longer timescale, by relating prehistoric lake-levels to Milankovitch-monsoon forcing. The latter may help to explain the perceived discrepancies between the palaeoclimate of the eastern Sahara and the strength of a ‘tropic-wide’ monsoon. The monsoon-desert mechanism may not be confined to the Asian monsoon alone and the existence of other monsoon-climate regions over the globe may, in a similar way, explain the observed summertime strengthening of the oceanic sub-tropical anticyclones and the existence of western continental deserts and ‘Mediterranean-type’ climate regions.