Previous - Next
This is the old United Nations University website. Visit the new site at http://unu.edu
Richard Longhurst
Institute of Development Studies, University of Sussex,
Brighton, UK
The recent famine in Ethiopia, occurring after one ten years previously, has focused an unprecedented amount of attention on food distribution problems in African countries and the capacity of the international and national agencies to prevent famines. This paper examines the main nutritional issues in the genesis and establishment of famine, defining nutritional aspects in the broadest possible terms to include food consumption and the proximate causes of famine, as well as the nutritional status of persons in the affected areas.
It is well known that famines are the result of an interaction of social factors - distribution of income and political power- that affect food availability, with the physical factors of climatic change as they influence food production. Although this article concentrates on nutrition matters, these other vital influences should be borne in mind. A shift in thinking over the last ten years, accelerated by the work of Sen [1], proposes that famines are caused not necessarily or only by a physical shortage of food but by people's inability to command resources to obtain food in what has become known as entitlement.
This shift in thinking also requires that the traditional role of the nutrition professional be expanded. A recent debate in the United Kingdom included a statement to the effect that famine was a nutritionist's problem only in the sense that a road accident is a police problem. Nutritionists, it was suggested, could describe its effects and suggest ways to use food resources better but could not prevent famine. With this new understanding of the causes of famine, however, nutritionists can take a far more positive role. The traditional professional attention to how poor people go about obtaining food, the identification of vulnerable target groups, and the study of how food-acquisition behaviour differs among target groups and how food-consumption patterns can be undermined are important elements in designing policies and projects to prevent, or at least significantly reduce, the impact of famine.
Famines can no longer be attributed solely to irresistible natural phenomena. They are also affected by adaptive mechanisms employed by poor rural people in famine-prone areas when food availability fluctuates. Some of these mechanisms are in response to annual, expected seasonal fluctuations and include agricultural practices, patron-client relationships, and off-farm income sources to provide food in the hungry season. When food shortages extend beyond seasonal severity, vulnerable groups can, among other things, change the composition of their diet, sell off assets, or migrate. During a famine, this behaviour becomes more extreme, leading to the break-up of family or community, possibly to the eventual point of communities being unable to recombine or ecological areas being unable to support crops.
The nutrition profession can play a significant role in translating into policy the ways in which people ensure a sufficient supply of food. This information can be marshalled and gaps identified. Means by which current policies might be undermining household food strategy can also be identified.
DEFINITIONS OF FAMINE
Famine has been defined both by researchers and by those who suffer from it. "The regional failure of food production or distribution systems, leading to sharply increased mortality due to starvation and associated disease" [2] and "an abnormal breakdown in access to food which leads to mass starvation among vulnerable groups or classes of people" [3] are definitions proposed by two researchers. As Seaman and Holt [4] have pointed out, most definitions emphasize food shortages and starvation.
The local culture in Bangladesh defines three types of famine: scarcity is when times are bad, famine is when alms are scarce, and nation-wide famine is when the epoch changed [5]. In his description of the Bangladesh drought and food crisis of 1978-1979, Cutler [6] suggested that the word ''famine" is best used to describe a complete breakdown of social and moral order accompanying mass starvation, and "famine conditions" to refer to the pre-famine situation in which widespread starvation is prevented only by outside intervention by the state or other agencies. Cutler admitted, however, that these distinctions become academic in reality.
CAUSES OF FAMINE
The causes of famine are several. To date, famines have usually been studied individually, and attempts to generalize from them are only just beginning [7-9].
Drought
In terms of physical phenomena, drought is usually associated with most famines.
The famine conditions of 1966-1967 generated in Bihar, India, followed failures in the monsoon of 1965 and 1966 and hence reductions in paddy production [10]. A grain shortfall of nearly 30 million tons from 1964/65 to 1966/67 depleted food stocks [11].
In 1972 the failure of the long rains caused a severe crop failure in Wollo Province, Ethiopia, although provinces to the west and south had good harvests. Since then, parts of northern Ethiopia have suffered successive years of crop failure as a result of unfavourable (i.e., untimely as well as reduced) rainfall [12].
In the Bas-Fleure sub-region of Zaire a drought in November 1977 was associated with a fall in food-crop production of 60 per cent during 1970. "Famine conditions" were reported; all plantings of maniac between March and June 1978 were destroyed. The drought was believed to be a major precipitating factor in increasing levels of malnutrition in the population [13].
By the beginning of 1985 Sudan had suffered three to four years of rain failure, with the 1984 harvest yielding only 10 to 15 per cent of normal production [14]. Food shortages were such that huge numbers of people converged on the capital, Khartoum, seeking food.
Drought and a breakdown in civil order caused a famine in Karamoja, Uganda, during 1980 [3, 15].
Crude mortality and infant mortality during 1979 were increased by five and ten times respectively above 1969 census data.
The Bangladesh drought of 1978-1979 began with partial failure of the 1978 monsoon rains and a serious deficiency in the pre-monsoon rains of early 1979, with rainfall countrywide ranging between 40 and 80 per cent of average. Thereafter, the price of rice did not return to pre-crisis levels until December 1980 [6]. The mass starvation that characterized the famines of 1942-1943 and 1974-1975 did not occur, being averted by large-scale food distribution.
Prolonged drought in areas of West Africa beginning in 1969 culminated in low food intakes in 1973 and absence of grain in rural markets [4]. The greatest impact was believed to be on pastoralist groups.
Floods
Floods also have precipitated famine in India and China. This occurred most recently in Bangladesh in 1974, when a famine was declared. In 1984 flooding was also severe, but conditions of famine did not result. Rather, there was "intensification of stress on the vulnerable among the rural poor" [16].
One of the largest famines in history was that in Bengal in 1943, where deaths were reported to number 1.5 million. A combination of a cyclone, subsequent fungal disease, the Japanese occupation of Burma cutting off rice imports, and the needs of India to feed armies and refugees are all believed to have reduced food availability seriously [11, 17]. According to Sen [1], however, per capita food availability in Bengal in 1943 was similar to that in the previous two non-famine years. Prices rose and landless individuals could not afford to buy rice.
Disease
Disease has contributed to a number of famines, including crop failure (caused by blight fungus Phytophthora infestans) in the Irish potato famine of 1845 and 1846, and rinderpest outbreaks in East Africa in the 1890s [18].
Social Phenomena
Overlying the climatic phenomena are human-related phenomena that contribute to famine and may even be the primary cause, although famines can never be attributed to any single cause.
War is the most common, contributing to famine in Bengal in 1943, in Ethiopia in 1984-1985, and in Timor, Kampuchea, Laos, Vietnam, Afghanistan, Chad, and Uganda.
Poverty and the distribution of economic resources among groups are powerful determinants of who suffers and who does not. It has also been suggested that the move to a cash economy in both India and Nigeria with British colonization "caused" famines, where previously household and community food-security mechanisms existed to avoid them [19, 20]. Hoarding is a regular event, causing prices to rise and reducing the availability of food to the poor. Often famines can be extremely localized, existing close by areas where food is sufficient. Pockets of famine can exist where people cannot be reached by food distribution programmes, are dependent on crops hardest hit by climate, or are especially poor.
A great deal could be done to prevent famines by ensuring that nutritional considerations are integrated into development policies and projects and that the household food security of the poor is improved. Both suggestions are cost-effective and make the most sensible use of limited monies. Otherwise, for specific countries, famine will continue to result in high loss of life and funds, and the efforts of international agencies will be diverted to expensive attempts to extinguish brushfires.
CHANGES IN NUTRITION DURING FAMINE
It is self-evident that indicators of nutritional status such as child and infant mortality and anthropometry all worsen during famine. More people die than otherwise would, adult body weights fall, and the growth of children slows. Two qualifying statements are in order, however.
First, the health and nutrition statistics provided by most developing countries are usually so poor that it is quite difficult to measure, even in approximate terms, the incremental effects of famine [21]. Second, rural families can be extremely resilient, finding enough new sources of food to sustain them for several years. In Ethiopia, peasants have survived up to five years of crop failure before dying of starvation [22]. Therefore, the question of declines in nutritional status may not be whether, but when.
Some data from the literature suggest that the number of deaths from the 1973 Ethiopian famine might have been about 50,000 [23]. The prevalence of malnutrition was highest in famine areas [24]; in the drought in gas-Zaire in 1978, protein energy malnutrition (PEM) was also more common [13]. The 1980 famine in Karamoja, Uganda, resulted in very high mortality, with 21 per cent of the population dying in 12 months, most from starvation; infant mortality was around 600 per 1,000, with child mortality of 305 per 1,000 [3]. However, moderate and severe acute malnutrition was not as widespread as expected. In a survey of villages in the Bihar famine, the major nutritional-deficiency diseases were PEM in preschool children, and ocular signs of vitamin-A deficiency and oral lesions of vitamin-B-complex deficiency in all age groups. The frequency of PEM was highest in villages where feeding programmes were inadequate [25].
HOUSEHOLD SOURCES OF FOOD DURING FAMINE
Wild and Low-Status Foods
When food becomes in short supply, households turn to what are known as "famine foods." These are usually foods growing wild, such as vegetables, nuts and berries, and parts of trees. In normal times such items are eaten only by the very poor during seasonal shortages, and their consumption is usually a source of shame. Therefore in the Bangladesh famine of 1974-1975, people ate banana tree, wild arum (Araceae sp.), plantain saplings (Muse paradisiacal, leaves, and rice husk [5, 17]. In the Bihar famine greater amounts of wild green leafy vegetables were consumed in severely affected villages. Wild tubers were eaten in drought-affected parts of Andhra Pradesh [25].
In the Karamoja survey, 41 per cent of the population subsisted on wild weeds, fruits, and seeds collected in the bush or ate no food all day. These items had little nutritional value but could temporarily stave off the worst physical effects of hunger [3]. In northern Dafur, Sudan, in 1973 e most people collected the seeds of wild growing grasses [26]. The 1982-1983 drought in Mali resulted in a large increase in the consumption of the berry Boscia senegalensis; the flower of the wild tree mahua is a similar food in Bihar.
People spend a great deal more time procuring food during a famine than when they have stocks in a granary. Time is spent gathering foods from the bush and also in preparing them, as many require prolonged processing to extract toxic substances. Families try to reduce consumption overall by diluting gruel and adding weeds, grass, and greens to it.
Enhancing the Production and Use of These Foods
The importance of wild and previously low-status foods during famine is obvious. Thus it is necessary to ensure that they will continue to grow in their habitat and not be threatened by extinction due to misguided agricultural-development or natural-resources policy.
The vast majority of these crops will probably not respond to efforts to breed them to increase yields; nor would it be economical to attempt to modify them so that they are no longer toxic, especially as their use is occasional. When used, however, these foods are essential. More knowledge should be obtained about where they grow, how their cultivation could be encouraged to a limited extent, and how they might be prepared for consumption. Their role in the diet should be better understood, and they should not be regarded as low-status crops.
Additional Strategies for Survival
To secure food, households also adopt other strategies. Often they sell essential assets such as land, household utensils, farm implements, and old clothing [27]. Livestock, traditionally a buffer against hard times, appears on the market in large numbers, and its price starts to fall. There may be an unusual drop in the price of labour, and people enter into bonded-labour arrangements. In extreme circumstances, people turn to theft, prostitution, and even buying and selling children. Out-migration is common.
ADDRESSING THE SPECIAL PROBLEMS OF VULNERABLE GROUPS
The nutritionally vulnerable - young children, women, and very poor families - seem to suffer most in times of famine; the harsh effects are accentuated for them. Children are among the most powerless. Frequently, the family disintegrates as the male head goes off in search of food and income, and children are left with grandparents. They become isolated and may be sold. Children also are vulnerable to the effects of infection. which is often the greatest killer in a famine [28-32]. The continued malnutrition that is a characteristic of "normal" circumstances further increases their susceptibility.
The inequalities that exist in intra-household distribution of food can be expected to worsen during famine, although there is no hard evidence to show this at present. Evidence does suggest that terms of trade worsen against women and children in times of economic recession and during normal seasonal shortages.
Efforts should be made to ameliorate the problems of vulnerable groups before social disintegration sets in and these persons become lost. For example, the traditional system protects children in the beginning but then breaks down.
In the early stages of a famine, human lactation is relatively unaffected [33]; this is the protective effect of breast-feeding. Thus famine-relief programmes must distribute appropriate foods rich in calories, protein, and vitamins to children and to pregnant and lactating women in order to protect breast milk. Also, as deaths can occur from a mixture of malnutrition and infection, it is necessary to set up simple adaptive maternal and child health care clinics capable of recognizing and providing basic treatment of the main infections, including appropriate immunization [34, 35].
When children are in some kind of institution such as a clinic or hospital, targeting special foods for them is relatively simple. If they are in their homes, providing special foods will be useless if these items are likely to be eaten by other family members without freeing up food for the children as in normal times.
Finally, it is important to get food to children to meet overall energy deficits rather than divert time and effort to provide commodities of higher nutritional value and thus possibly impede this single objective. Foods should be simple and easy to prepare with a minimum of utensils, fuel, and water.
EFFECTIVE SYSTEMS OF WARNING OF NUTRITIONAL PROBLEMS
One of the most important issues relating to famine is how to prevent it. With adequate warning linked to effective action, many famines can be prevented. We must examine the appropriate indicators in conjunction with the timeliness with which they can be generated.
What has so far been proposed in terms of nutritional indicators has been too top-heavy with data requirements. Also, data collection has not been linked to administrative response. The capacity of developing countries to collect, analyse, interpret, and disseminate information effectively is very limited at best. Therefore proposals to set up systems that involve collecting large quantities of data and statistical "correctiveness" {e.g., representative sampling and derivation of complex indicators) are not likely to be successful. This is especially pertinent when stress on the administrative system is increasing.
A balance can be struck, however, on how much nutritional data and which indicators are to be used. There have been many false warnings of famines in the past, and these can cause panic, instability, and hoarding of food, all of which can themselves lead to famine.
Constant nutrition surveillance may not be appropriate for a number of reasons [36]. First, there are not likely to be continuous famine-forecasting variables with predictive value. For example, nutritional-status indicators such as anthropometry and blood haemoglobin levels do not often respond immediately to changes in food intake; in the beginning of a famine, the sources of food may change but not overall levels of consumption. Food-consumption surveys require enormous amounts of resources and cannot be carried out on a continuous basis. The consumption of certain key foods such as wild plants, fruits, or berries, however, may be a good indicator of shortages, especially if such consumption is unseasonal. Because eating these items is regarded as shameful, it will be very difficult to obtain accurate data on their usage.
Nutritional-status indicators may not pinpoint the location of the famine. Families often migrate and disperse; sometimes those in worst condition may be found on the edge of a famine rather than at its centre. In view of the close link between poverty and famine, a large cross-section of the population would have to be sampled, and the links between economic status and nutritional status are never very clear. Also, focusing on nutritional status puts an emphasis, probably excessively in the situation, on hard, quantified data, which may assume a property of prediction that is not supported by the method of collection.
Second, famine is not a frequently occurring event, like the seasonal shortages of rainfall, for which a surveillance system can be installed. India has gone through long periods without suffering famines. Third, it is likely that any new data system will be imposed on an existing, usually inadequate, system that has failed to detect famines in the past.
Systems to detect signals showing that a famine may be looming should be based more on communities' knowledge of, response to, and attitudes about declining food availability. Rural people themselves have most knowledge about whether a famine is developing, but they are rarely consulted. In an ideal system their views would be incorporated, but, as second best, their response can be monitored. Rural families know their own capacity to withstand shocks in terms of assets, food supplies from the previous season, and calls for assistance from the community. They know how much work is available and how long the local ecology can provide food for support.
A number of indicators could be suggested based on our knowledge of changes in food consumption and availability that precipitate a famine. The use of famine foods can be observed, if not precisely measured. If these items are being eaten on a regular seasonal basis at times when food is not usually short, this is a good indicator of a problem. Sharp, unseasonal price rises of staples in the markets can be detected from statistics that are collected regularly within a government's statistical service. Generally, people are more likely to become dependent on the markets for food during a famine. By collecting food-price data by region, it should be possible to produce maps that show where prices are rising [22].
A similar indicator is the volume of livestock sales and the prices at the time of trading. Prices of less than 50 per cent of normal suggest a mass movement in selling assets. Mass migration is a famine-related phenomenon that does not require elaborate data collection. By the time it has occurred, the famine is usually well established.
FEEDING PROGRAMMES AND EMERGENCY RELIEF
Most of the professional expertise of nutritionists concerning famine has been channelled into relief - cure rather than prevention. If nutritionists define their mandate broadly, a whole range of is sues related to feeding and emergency relief will come within this remit. To cover all the issues relating to emergency relief is not possible here, or appropriate.
The nutritional aspects of famine relief, as strictly defined, have concerned the nature and composition of food supplements, links between feeding programmed and health care systems, targeting of rations to specific groups, the period for which supplements should be provided, and food quality control. Also, debate has taken place over the use of food-for-work projects and whether there should be a work element in emergency relief. More recently, the nutritional effectiveness of providing food versus a cash payment has been discussed. Feeding programmes could be expanded in famine-prone areas to include aspects other than relief.
IMPLICATIONS FOR POLICY
In theory as well as practice, policies and programmes relate to preventing famine, to identifying its causes so that its frequency and severity might be reduced, and to providing relief or rehabilitation to its victims. Separate policies can be regarded as having implications for the short, medium, and long term.
Alleviating Rural Poverty
The new orientation toward understanding the causes of famine, based on entitlement to food rather than irresistible physical phenomena, points to the fact that economic development that alleviates rural poverty will be the most important preventive measure. Policies that strengthen household food security, provide basic health care (especially immunization), and generate employment for the landless are especially important. Previous neglect of rural areas, resulting in inadequate infrastructure and in food price policies that favour urban consumers rather than rural producers, has led to an inequitable distribution of resources and pockets of low entitlement to food.
Effective policies to provide resources to the poor, especially in rural areas, will reduce the impact of famine. These include policies for the provision of new varieties of seed, especially for sorghum and millet in semi-arid areas, and new breeds of small livestock that can tolerate adverse conditions, especially water stress; the development of better all-weather roads, small-scale irrigation, markets, storage, and market information; and better access to well-stocked clinics. These policies must be so implemented as to not weaken the ways in which poor people adapt to stress.
Some mechanisms that have already been monitored include means of food storing and sharing, selling assets (especially livestock - sometimes small stock first, but not always!, switching to drought-resistant crops, and drawing on famine reserve crops such as cassava. Other forms of "agronomic risk aversion" such as crop mixes, intercropping, and various domestic and community strategies are employed routinely in anticipation of famine. These routine mechanisms should form the basis of agricultural interventions to reduce famine.
A major reason that families and children have been able to weather some famines has been the women's considerable adaptive capacity, together with their willingness to assume a significant extra workload. The women gather foods, which involves large amounts of extra work. Often they assume the role of primary decision-makers and food-producers for their households. Helping women process food more easily and providing them access to agricultural developments in food production are important elements in avoiding famine.
Adapting Existing Programmes to Famine Conditions
Attention should be paid to how existing nutrition interventions that are implemented in normal conditions can be modified to adapt to famine conditions. Feeding programmes, of course, are especially relevant and can have a wider role than just feeding hungry people.
Food-for-work projects should be seen as having three functions and in famine-prone areas should be designed accordingly. The first function is to assist migrants as they arrive from famine areas. The second is to provide work and help; projects can be regarded as glue that holds together distressed families and communities. The third function is to generate assets as a result of the work programme, such as roads, anti-erosion devices, and replanted fields, that might contribute to the long-term solution.
In famine-prone areas, all agricultural and rural development projects should have some components that help to alleviate famine and assist survivors.
In drought areas, such components might include tree planting, well-digging, and small-scale irrigation. Nutrition education projects should include information on how to survive bad years as well as on how to improve nutrition in good ones.
Food Aid
Food aid must play an expanded role in famine policy. As already mentioned, it can be useful within a coherent policy framework in more ways than helping people survive in the short term. Wherever possible, emergency food aid should equally address long-term rehabilitation. In many countries, however, where emergencies are recurring, it is not possible either in theory or in practice to link short- and long-term aspects.
Children must receive extra protection, as their nutritional needs are greater than those of adults: projects within projects are required, such as supplementary feeding within food-for-work. As infection can spread rapidly, linking food aid to health care by providing medicines and building clinics is necessary.
Within the context of a famine, food aid is versatile. It can be used in many ways: income transfer, incentive effect for community development action, release of funds from government budgets, creation of funds, balance-of-payments support, price stabilization, and food reserves. All of these can be envisaged as playing a role before and after famine. For example, it has been seen that feeding vulnerable groups can be linked with activities to generate income for women. In other situations women have been trained to provide basic health and nutrition services. With food-aid support, family self-sufficiency can be reestablished with the development of gardens. The cash from food-aid sales can be used in a variety of ways to help the survivors of famine become self-sufficient without recourse to prolonged feeding programmes.
Income Supplements
The theory of famine as a breakdown of access to food through lack of resources suggests that in some situations, supplementing income may be a sufficient response, rather than increasing the availability of food. In this context, international agencies and national governments should expand the pilot cash-for-food projects tried recently in Ethiopia. These projects consist essentially of giving cash (which might well come from food-aid sales) to families affected by famine to avoid their disintegrating and migrating from communities. With a small monthly payment, families are able to buy food and, in addition, small stock, farm tools, and seed for the next planting season. Such resources enable families to enter local credit markets. In some projects, the distribution of monthly cash payments has been linked to community development works of the kind already mentioned: roads, anti-erosion devices, vegetable gardens, child-care centres, and health clinics. What are required are expert management skills, the availability of complementary assets such as fertile land, and food in the marketplace. The entrepreneurial talent of local people remains considerable even in a food emergency if they receive basic rations. The injection of cash into a famine area provides purchasers for goods being sold as the emergency spreads. This all makes the process of rehabilitation less difficult, because once families and communities have split up, the costs of reuniting them increase and refugee camps have to be set up, often with little chance of becoming self-supporting.
Linking cash for food with food aid is obvious and has an important political dimension. International donors may be reluctant for a variety of reasons to give cash for direct distribution to famine victims; food aid is far more acceptable. This can then be converted into non-food resources by the variety of means already mentioned.
Concentration on Famine-Prone Areas
Re-examination of early warning systems, providing nutritional surveillance, and monitoring and evaluating projects in famine-prone areas are required. Current research in Ethiopia, Sudan, and northern Kenya is proving of immense value in this regard. Several countries in which there is a potential for famine do not have any form of monitoring system. The reasons a famine occurred in Ethiopia despite an early-warning system might not be deficiencies in the system per se; however, improvements in methodologies are needed.
Rapid but reliable survey information should be available, with the data going directly to those who can make a response. Some indicators are more relevant in the early development of a famine than others, and can be derived from data that are collected routinely. Price mapping is one obvious example. Surveillance systems must be simple, as data-processing and management resources are likely to become less available during a food crisis. The systems should be durable also, so that they are continued during periods when no crisis is detected. In the past, systems were established in a flurry of post-crisis activity, only to wither away during the normal years that followed.
Surveillance systems must be modified to include household food-security strategies to incorporate responses and changes at the family level far more than is done at present. Indicators of changes of entitlement are important, together with data from satellite technology and estimates of total food availability at the national level.
CONCLUSIONS
These five policy issues are all ones that countries vulnerable to famine, or those that have recently suffered a food crisis, can incorporate into their national planning systems to avoid future crises or reduce the negative impact. Various UN agencies have an essential role to play in encouraging governments to adopt these policies. The pressing need is for policy measures that reduce rural poverty and build up rural infrastructure. Semi-arid areas that suffer drought-linked famine have been neglected in terms of technology development for modern varieties of plants and animals, control of environmental degradation, and provision of infrastructure. Pastoralists are transitory populations and have been difficult to assist.
Rural development is a long-term process; however, some of the other recommendations can be implemented in the short term and do require a commitment of resources by national governments backed by international assistance.
ACKNOWLEDGMENTS
This paper was prepared with the financial assistance of the ACC/SCN. I am grateful to Peter Cutler and Wendy McLean of the Food Emergencies Research Unit of the London School of Hygiene and Tropical Medicine for their advice and assistance, but the views expressed here are my own.
REFERENCES
1. A. K. Sen, Poverty and Famines. An Essay in Entitlement and Deprivation (Clarendon Press, Oxford, 1981).
2. G. W. Cox, "Famine Symposium. The Ecology of Famine: An Overview," Ecology of Food and Nutrition, 6: 207-220 (1978).
3. R. J. Biellik and P. L. Henderson, "Mortality, Nutritional Status, and Dietary Conditions in a Food Deficit Region: North Teso District, Uganda, December 1980," Ecology of Food and Nutrition, 11: 163- 170 (1981).
4. J. Seaman and J. Holt, ''Markets and Famines in the Third World," Disasters, 4 (3): 283-298 (1980).
5. B. Currey, "Famine Symposium Report. The Famine Syndrome: Its Definition for Relief and Rehabilitation in Bangladesh," Ecology of food and Nutrition, 7: 87-98 (1978).
6. P. Cutler, ''Detecting Food Emergencies: Lessons from the 1979 Bangladesh Crisis," Food Policy, 10: 207-224 (1985).
7. W. R. Aykroyd, The Conquest of Famine (Chatto & Windus, London, 1974).
8. B. Currey and G. Hugo, Famine as a Geographical Phenomenon (D. Reidel Publishing Co., Dordrecht, Netherlands, 1984).
9. J. R. Robson, ed., Famine: Its Causes, Effects and Management (Gordon and Breach Science Publishers, New York, 1981).
10. V. Ramalingaswami et al., "Studies of the Bihar Famine of 1966-67," in G. Blix, Y. Hofvander, and B. Valquist, eds., Famine: A Symposium Dealing with Nutrition and Relief Operations in Times of Disaster (AImqvist and Wiksells, Uppsala, Sweden, 1971).
11. A. Berg, "Famine Contained: Notes and Lessons from the Bihar Experience," in G. Blix, Y. Hofvander, and B. Valquist, eds., Famine: A Symposium Dealing with Nutrition and Relief Operations in Times of Disaster (Almqvist and Wiksells, Uppsala, Sweden, 1971).
12. P. Cutler, "Famine Forecasting: Prices and Peasant Behaviour in Northern Ethiopia," Disasters, 8 (1): 4856 (1984)
13. K. Nkamanay, "The Consequences of the Drought in Bas Zaire, 1978," Disasters, 4 (1): 55-64 (1980).
14. J. Shoham, "Report on a Visit to Sudan, November 1984," Disasters, 9 (1): 12-14 (1985).
15. R. J. Biellik and P. L. Henderson, "Mortality, Nutritional Status, and Diet during the Famine in Karamoja, Uganda, 1980," Lancet, 1981, pp. 1330-1333.
16. E. Clay, "The 1974 and 1984 Floods in Bangladesh: From Famine to Food Crisis Management,'' Food Policy, 10: 202-206 (1985).
17. M. Mujibur Rahaman, "The Causes and Effects of Famine in the Rural Population: A Report from Bangladesh," Ecology of Food and Nutrition, 7: 99- 102 (1978).
18. G. B. Masefield, Food and Nutrition Procedures in Times of Disaster, FAO Nutritional Studies, no. 21 (FAO, Rome, 1967).
19. B. Bhatia, Famines in India (Asian Publishing House, Bombay, 1967).
20. M. Watts, Silent Violence: Food, Famine and Peasantry in Northern Nigeria (University of California Press, Berkeley, Calif., USA, 1983).
21. R. V. Garcia and J. C. Escudero, eds., Drought and Man. The 1972 Case History, vol. 2, The Constant Catastrophe: Malnutrition, Famines and Drought (International Federation of Institutes for Advanced Study, Pergamon Press, Oxford, 1982).
22. P. Cutler, "The Use of Economic and Social Information in Famine Prediction and Response," report to ODA (London, 1985).
23. D. S. Miller and J. Holt, "The Ethiopian Famine," Proc. Nutr. Soc., 34: 167-172 (1975).
24. J. Seaman, J. Holt, and J. Rivers, "The Effects of Drought on Human Nutrition in an Ethiopian Province," Int. J. of Epidemiology, 7 (1): 31-39 (1978).
25. M. C. Swaminathan, K. Visweswara Rao, and D. Hanumantha Rao, "Food and Nutrition Situation in the Drought Affected Areas of Bihar," Ind. J. Nutr. Diet., 6: 209-217 (1969).
26. L. Holy, "Drought and Change in a Tribal Economy: The Berti of Northern Dafur," Disasters, 4 (1): 65-71 (1980).
27. N. S. Jodha, "Famine and Famine Policies: Some Empirical Evidence," Economic and Political Weekly, 1975, pp. 16091623.
28. F. B. Bang, "Famine Symposium. The Role of Disease in the Ecology of Famine," Ecology of Food and Nutrition, 7: 1 -15 (1978).
29. J. M. Bengoa and G. N. Beaton, Nutritional Aspects of Disasters, WHO Monograph Series. no. 62 (WHO, Geneva, 1976).
30. A. K. M. Alauddin Chowdhury and L. C. Chen, "The Interaction of Nutrition, Infection and Mortality during Recent Food Crises in Bangladesh," Food Res. Inst. Studies, 16 (2): 47-61 (1977).
31. S. Dunbar, "Malnutrition and Anaemia among Somali Refugee Children in Long Term Camps," Disasters, 8 (3): 174-177 (1984).
32. Protein Advisory Group, A Guide to Food and Health Relief Operations for Disater (UN, New York, 1977).
33. D. B. Jelliffe and E. F. Jelliffe, "Breastfeeding: A Key Measure in Large Scale Disaster Relief," Disasters, 1 (3): 199-204 (1977).
34. S. Peel, "Practical Relief and Preventative Methods: Selective Feeding Procedures," Disasters, 1 (3): 179-198 (1977).
35. C. de Ville de Goyet, J. Seaman, and U. Geijer, The Management of Nutritional Emergencies in Large Populations (WHO, Geneva, 1978).
36. B. Currey "Fourteen Fallacies about Famine," Ceres. Mar-Apr. 1981.