Disasters do not occur out of context but are embedded in the political structures, economic systems and social orders of the societies in which they take place. Above all, they are historical events in that hazards are diachronic happenings, they occur as part of a sequence or process that determines a particular person’s or people’s vulnerability. But endowing disasters with histories and societies with hazardous pasts also requires a concomitant re-evaluation of the means by which such events have been conceptualised and their consequences analysed. Self-evidently disasters can no longer be viewed as merely meteorological or seismic phenomena divorced from social and cultural systems; neither can they simply be reduced to ‘laboratory studies’ of individual or group behaviour during extreme situations. Nor, too, can they be limited to or contained within small-scale ethnographic studies of the origins of communities’ coping mechanisms under stress. All these deal with important aspects of the occurrence and effect of natural hazard on societies but a more holistic approach is still demanded if disasters are to be appreciated in the wider context of the human–environment interplay over time.