This paper explores some of the reasons why the well-laid plans of the 1970's failed to be an effective bulwark against hunger. It is reflective rather than critical because we are faced with the certainty that just as surely as the famines of the 1980's followed the famines of the 1970's, the 1990's will again see drought, crop failure and, unless things change a great deal, famine as well. The analysis of the causes of hunger current in the 1970's can be summarized somewhat brutally as follows. Either there is not enough to eat, or what is available is poor in nutritional quality. Poor nutrition is synergistic with disease. Together they result in increasing debility and finally death. Famine is an unusual event, precipitated by this same triad of factors, on a catastrophic scale. The strategies which emerged from this analysis can be placed similarly under three broad headings. The first is that food production must be increased so that there is more available for everyone. The second is that national food security strategies should be developed and implemented. The third is that nutritional quality of people's diets should be improved. These are three major goals which have dominated international thinking for a decade. This paper argues that this analysis is at the very least incomplete and that the strategies based upon it have failed to make a marked impact on the risk vulnerable households face to famine. It goes on to suggest that, irrespective of the quantity and quality of food generally available, the households, poorer communities and poorer countries are able to lay claim to a share of what is available. These claims are mediated by a hierarchy of relationships--households within communities, communities within countries and countries in the world at large--and the nature of the relationships constitutes the 'political economy' within which famines arise and must be analysed.