Contents - Next

This is the old United Nations University website. Visit the new site at

Hunger and society

Understanding hunger and predicting starvation
Social structure, land use, and food availability in the Caribbean
Village food and nutrition planning in Tanzania
Methodological issues in nutritional anthropology


Understanding hunger and predicting starvation

Lars Bondestam
Centre of Interdisciplinary Studies of the Human Condition, Göteborg University, Göteborg, Sweden


"Hunger" is defined here-somewhat vaguely-as lack of sufficient food in a society or lack of resources to produce a sufficient amount of food. This article maintains that hunger is basically caused by non-natural factors in both pre-capitalist and capitalist societies and that, although the direct causes of starvation may be natural (floods, drought, etc.), the basic ones are man-made. These hypotheses are supported by examples from Ethiopia in the 1960s and 1970s. The choice of a particular country, in this case Ethiopia, does not imply any appraisal of its existing politics. In fact, the present government of Ethiopia cannot be held responsible for the foundations of most of the economic problems of today. These foundations were laid during the time of the previous imperial government, and thus the explanations for them are historical.

The causes of starvation can be divided into two categories: basic and direct causes. Basic causes are always non-natural and most often policy-dependent, i.e., they are created by man, either by the person directly affected (the one who starves), or more often by characteristics inherent in the economic structure (mode of production) or by external penetrating forces. In the absence of basic causes, the direct ones are less harmful; i.e., the force of the effect of direct causes depends on the existence of basic ones. So, for instance, a crop failure due to drought (direct cause) has less disastrous consequences in more egalitarian societies than it does in class societies with a high degree of exploitation (basic cause). An implication of this approach is that the effects on nutrition (including the effects on food production, supply, distribution, and prices) of certain events can be predicted much better than is often acknowledged. However, this requires an acceptance that political and economic realities have something to do with human survival.

The early-warning systems (EWS's) which are being used in some countries to warn of disasters usually ignore these realities. The information collected is always related to natural phenomena and usually includes data on rainfall- areas particularly deficient in precipitation are identified. The data rarely include political causes of starvation. This weakness in the EWS's is perhaps understandable, but it can be fatal. Another drawback, related to the former, is the lack of economic analysis in the process of data collection. Hence, the technically advanced and impressive EWS of Ethiopia in 1977 failed to predict the 1978 famine in time. An economic analysis of a particular society does not have to be as complicated and costly as is often indicated by various experts. Indeed, analyses which are too sophisticated and pretentious are rarely ready before famine is a fact.

In the five examples given below, the political and economic causes of hunger and starvation are analysed. The first two examples are of a general nature and show how both pre-capitalist and capitalist modes of production hamper economic growth, thereby making hunger inevitable in the long run and famine always a risk. They are fundamental to the understanding of the next two-more concrete-examples, which refer to transition from one mode of production to another. Irrespective of the possible economic and social gains of such a transition, certain groups may suffer to the extent of actually starving. The fifth example refers to a war economy and its effects on food production and nutrition; specific examples of the effects of a war economy on food production are discussed.


Pre-capitalist economic systems are all characterized by stagnation, because of the obstruction of food production. Basic common features of these primitive community, tributary, feudal, and other modes of production are a lack of investment in machinery and other such inputs- i.e., in higher productivity-and a lack of wage employment. In the feudal system, for instance, the tillers are allowed to keep for their own reproduction only some of the food they produce. The surplus is usually handed over as land rent to the owner of the land and of the means of production. The tiller can rarely afford to invest with a view to higher production. Nor are the returns on the surplus reinvested, but rather they are spent on luxurious consumption or staked in land speculation, trade, buildings, and other non-productive activities. Thus, these rigid relations of production between owner and tiller prevent any development of the productive forces. Production stagnates or at least does not increase faster than the population grows. The minimum standard of peasant survival in combination with the misuse of the surplus result in poverty and in a poor physical and social infrastructure, high illiteracy rates, insufficient health and medical facilities, low nutritional status, high mortality, no margins for safety, and no resistance to crop failures. Small changes in the nature-dependent conditions of life, even temporary changes, may have devastating effects.

The well-known and much-discussed famine in Ethiopia in 1973 struck both agriculturalists in the northern highlands and cattle-breeders in the lowlands to the east (whose case is discussed later). The former were typical victims of a stagnating system similar to the one just described. Their living conditions had always been marginal, and they had no spare resources. Hunger was normal, severe undernutrition was seasonal, and starvation never came as a surprise. Ecologists and many others blamed bad soils and absolute overpopulation for the famine, but the soil would have been more fertile and the relative overpopulation (not absolute, in the Malthusian sense} would not have existed if the imperial government had not "forgotten" this part of the country in favour of southern and central Ethiopia.


In private or state capitalist agriculture, labour productivity increases faster than the market grows (total demand for agricultural products). This is partly due to stagnant relative labour wages, which in concrete terms means that the labourers cannot afford to increase their consumption at the same rate as they increase production. This leads to overproduction and relative overpopulation {cf. the collapse of the "Brazilian miracle"). There are plenty of examples of how capitalist penetration of the agriculture of the third world results in rural eviction and urbanization. Manpower surplus is found in towns, on top of seasonal underemployment in the rural areas. Urban industry wages are kept low ("labour market economy"), the total purchasing power in urban areas does not grow in proportion to urban population growth, consumption above minimum needs is automatically blocked, and so is further food production for domestic sale. Hence, there is nothing surprising in the combination of poverty, hunger, and even starvation on the one hand and food exports on the other -a reality in many third-world countries, where a section of the population is too poor to afford even a minimum daily intake of food.

In the long run, traditional, stagnant agriculture in combination with population growth has to end up in more and more frequent famines. Replacing it by an advanced capitalist mode of production may solve the immediate low productivity problems. However, as many studies have shown-not least, studies of the so called green revolution -marginalized people and members of the lower classes often have to pay a high price for this economic growth. Moreover, the transition from one system to another itself is usually painful, as the two following examples from Ethiopia will show. In both cases, the famine caused by transition was predicted.


This example shows some of the causes of the 1973 famine; these causes had been analysed and explained to the imperial government as early as 1971.

North-eastern Ethiopia is inhabited by the Afar people, who breed cattle, camels, sheep, and goats. They depend on the naturally irrigated plains along the Awash River for pasture. The river overflows and floods these plains during the rainy season, May-September. In the 1960s the river was regulated by a dam and concessions were given to private investors to start commercial agriculture in the area (mainly cotton and sugar). The river flooded less land than before, and much of the best pasture was occupied by plantations. When international finance capital entered, the Afar people lost land; in some areas they lost access to the river, where they water the animals, and the quality of the pasture declined because of the regulation of the river. In December 1971 I wrote to the government:

The region under this rapid agricultural development is inhabited by people who graze their livestock in areas which are now more and more being taken over by the commercial farms. This leads to such consequences as overgrazing of the few areas left for the pastoralists, livestock starvation, decrease in milk production, undernutrition of the pastoralists, etc. If the pure economic development continues at the same pace and with the same objectives without any improvements for the pastoralists, the whole business will end up in a predictable situation. [1]

A little more than one year after this was written, the famine among the Afar people was a fact. Some had been evicted to rainfed lands and others were forced to crowd on small plots of land along the river. When rainfall decreased in 1972/1973, they found themselves in an entirely new situation which they had never experienced before. They could not escape starvation, basically caused by the penetration of capitalist agriculture in a region which used to be entirely dominated by a stagnant and defenceless pastoralist economy and directly caused by rain failure.


The important transition discussed in this example- including the badly needed land reform of April 1975- explains the 1978 famine to some extent.

Six years after the above-mentioned paper had been presented to the imperial government (with no immediate effect), another similar document was submitted to the new military government. This paper was an analysis of the situation in each region of the country with respect to political and economic realities, such as the effects of the land reform and the effects of the wars on third parties. The following extracts from the paper primarily refer to the peasant economy and urban-rural contradictions:

The 1973 famine in Wollo and Tigrai did not come as a surprise to those who predicted it or who saw the gradual introduction to it. Nor did the ravaging effects of the famine come as a surprise to those who withheld information about it. This paper, which is an expression not only of the observations of the author, is an attempt to show that there is an inevitable risk of a famine in Ethiopia, possibly of the same magnitude as the one of 1973. By acknowledging this, the risk may be diminished and a disaster may be avoided....

Before the land reform, peasants in the southern and central parts of Ethiopia were forced to give up a surplus for the market without compensation and had to survive on a low calorie intake. With the cessation of this type of exploitation they now decide over the utilization of their total produce. So, without increasing production, they can eat more and better than before, which obviously results in a smaller surplus for the market....

"The Ethiopian scissors" are very similar to the Russian ones, though important differences should not be denied [of the decline in productivity in the Soviet Union after 1917, and the New Economic Policy of 1921]. The low purchasing power of the rural population and the underdeveloped market relations between towns and countryside hampered domestic industry production growth of consumer and capital goods before 1974/75. This is one important explanation to the massive urban unemployment in Ethiopia. Before the political change [of 1974], the resultant urban poverty was, however, to some extent neutralized by comparatively low food prices-made possible with the heavy exploitation of the peasantry. Thus, in exchange for this cheap food the towns produced almost nothing for the peasants. The urban areas as a whole lived as a parasite on the rural areas. After the land reform, the producing peasant got his rightful compensation for his labour (or at least for most of it), peasant incomes grew, food prices automatically raised and the previous imbalance in favour of the urban dwellers turned in favour of the peasants. Still unable to produce goods for the increasing demands in rural I areas, the lower classes of the urban population are those who most intensively will feel the cost of change. With no or little access to manufactured goods, and due to lack of social security, which many may feel under the present circumstances, peasants are reluctant to sell their surplus. Their new economic position, with reference to control over land and produce, permits them to sell a smaller surplus....

The new imbalance between urban and rural areas implies a transfer of the hunger, and maybe famine, from the countryside to the towns, and a transfer of money from the towns to the countryside, where banking is non existent (or, when banks are available, peasants tend to mistrust them). Unless consumer goods (specifically salt, sugar, oil, clothes) and agricultural implements, all preferably locally made, are offered the peasants, this state will continue and will in fact mean a loss of both health and money....

Grain stocks, especially of wheat, maize and barley, are now rapidly declining, which results in lack of food for non-peasant consumption and in increasing prices of grain. This decrease in the exchange value of money hits primarily the vulnerable majority of the non-peasants, namely the lumpen-proletariate, the permanent, seasonal and daily labourers, and the poorer part of the petty-bourgeoisie. The results are malnourishment, in some cases undernourishment, and in extreme cases famine. The upper and higher brackets of the middle classes manage to satisfy their demands by hoarding of grain, which, together with the inevitable emergence of a widespread black market, further aggravates the supply of food for the urban poor-now more alienated from the basic necessities than ever before. [3]

The famine of 1978 was severe but not comparable to the disaster of five years earlier. Again, this year, there are reports of famine in the Horn of Africa, particularly in Ethiopia. An international charity organization, World Vision, even claims-an exaggeration-that one-fifth of Ethiopia's population is threatened with death due to drought (the same kind of reports were made by others in 1978 and proved to be exaggerations). But more important is the fact that the drought would not have had the same impact on food production if the basic causes had not been present. However, an important factor, one which is mentioned in the World Vision report, is the ravaging effect of the wars on the supply of food. This effect was also predicted in 1977, when war was the dominant factor in the economy.


In 1977 resources in the form of capital, manpower, and means of transportation were allocated primarily to the various wars in Ethiopia. This discussion will be limited to manpower and transportation, to see the effects on food production.

"The local and regional disturbances lead to a withdrawal of manpower from agricultural production: peasants are forced to flee before the enemies, others are recruited in the local cadres [of the many thousands of peasant associations] or in the national militia [a huge army of recruited peasants, who fought the various external enemies] " (3). The withdrawal of manpower from agriculture implied a decrease in food production. Under certain fairly realistic assumptions (constant peasant labour productivity, an increase of between 5 and 10 per cent in peasant food intake after the land reform, and a mobilization of some 5 per cent of the productive male peasants), it was calculated that the marketed amount of grain per capita (i.e. per non-producer) decreased by some 25-35 per cent. "In order to balance this loss, the average peasant labour productivity has to increase by some 10-13%, which corresponds to an additional labour time of 1-1.5 hours per day during the peak seasons (ploughing, sowing, weeding, harvesting, threshing and marketing). It is not realistic to expect the peasants to increase their labour time that much: first, simply because they already work some 10-12 hours daily during the peak seasons, and second, because it is not in their interest to do so" (3; see the reference to "the Ethiopian scissors" above).

This withdrawal of manpower later proved to be an important factor in explaining the food shortage not only in towns but in certain rural areas as well.

Moreover, due to the wars, there were difficulties in importing new trucks to renew the fleet, and at the same time the military demand for civilian trucks increased. Although a government decree of September 1977 said that 35 per cent of all means of transport from the port should be used for emergency grain and fertilizers, this rule could not possibly be followed; most of the transports had to concern direct military activities.

The gravity of the problem is obvious, not least manifested in the congestion of goods in the port of Assab [the main harbour by the Red Sea]..... The lack of sufficient transport facilities may turn out to be a decisive factor in a coming famine. Transport of agricultural implements, like fertilizers and seeds, and transports from Assab of imported grain, of grain from surplus areas to central stores, and of emergency grain to deficit rural areas will all be affected. [3]

An example of the problems of transportation in 1977 was the effect on the distribution of fertilizer. The amount of fertilizer dispatched went down by almost 16,000 tons from 1976 to 1977. This may have caused a production decrease of as much as 90,000-100,000 tons of grain. Only about a third of the requested fertilizer was dispatched in 1977. The figure for seeds was even lower.

Here only some of the links between the wars and the nutritional condition of the people are mentioned. Taking all the factors into account, it wasn't difficult to predict the famine of 1978. Irrespective of the weather that year, the impact of the wars was sufficient to cause disaster.


The examples show that, in the particular case of Ethiopia, hunger and famine during, say, the last two decades have been basically "man-made." Of course, this does not imply a denial of the fact that the extremely bad weather conditions during the last years were important direct causes of the insufficient amount of food produced within the country. The examples are of such a nature that their validity can be extended to many other third-world countries and to other periods of time. Thus, the examples should not be interpreted as peculiar to Ethiopia.

Finally, it should also be stressed that, whereas famine was ignored and even denied by representatives of the imperial government prior to 1974, food aid within Ethiopia today is comparatively effective thanks to the work of the Ethiopian Relief and Rehabilitation Commission (RRC). The success of the RRC in avoiding disasters depends, however, to a large extent on international emergency aid and external supply of food in sufficient quantities.


1. Lars Bondestam, "Agricultural Development in Awash Valley"
(mimeographed, Addis Ababa, Ethiopia, 1971).

2. L. Rudebeck, Guinea-Bissau: A Study of Political Mobilization
[Scandinavian Institute of African Studies, Uppsala, Sweden,
1974), p. 1980.

3. Lars Bondestam, "Expected Famine in Ethiopia" [Disaster
Preparedness Planning Programme, Relief and Rehabilitation
Commission, Addis Ababa, Ethiopia, 1977).

Contents - Next