2.20 Human disasters 188

2.20 Human disasters 188

ESSENTIALS Human disasters, as massive misfortunes long recorded over history, have great importance for medicine, rightly prompting the call for prevention, relief, and practical intervention by medical personnel. But why do human disasters happen? A sharp distinction is some- times drawn between natural disasters (e.g. earthquakes), and social disasters (e.g. wars), but detailed knowledge often shows that this contrast is not always clear: many disasters have mixed causes. Taking the example of famines. These are popularly understood in terms of food output decline, yet many famines have occurred without any decline in food production. Such misunderstanding has been responsible for the loss of millions of lives, mainly by under- mining the role of social intervention. Starvation is a characteristic of some people not having enough food to eat, not of there being not enough food in the economy. Even when nature plays a part in human disasters, society can make a huge difference. The death toll from so-​called natural disasters is much less in higher than lower-​income countries, even when they do not have any fewer disastrous natural events. The type of political rule can also be a factor of importance: authoritarian governments and military dictatorships are generally insensitive to vulnerable people. Interventions that can be effective include envir- onmental policies, those that mitigate the effects of disasters on the economy, and healthcare actions. Introduction ‘A sudden or great misfortune’ is the way the New Shorter Oxford English Dictionary defines a disaster. That is not a bad starting point in trying to understand the idea of ‘human disasters’. Disaster is a common enough term in public discussion. The addition of the ad- jective ‘human’ to the word points specifically to the misfortunes of human beings, rather than those of animals or of the perils of our natural surroundings (except to the extent that these non​human ­adversities yield human mishaps). It also guides us towards focusing particularly on the predicaments of groups of people with shared and interlinked predicament, rather than on the personal tragedy or sadness of particular individuals, seen on their own. The term could be used differently of course. There is nothing in- herently odd in talking, for example, about the human disaster that befell King Lear, and indeed Lear’s woeful elderly life can, of course, be of interest for a textbook of medicine. It would belong, however, to some other section of this book linked, for example, with psycho- logical adversity, the trauma of dejection and betrayal, or perhaps geriatric deprivation, with their own clinical challenges. In contrast, what are standardly called ‘human disasters’ have come to be seen as sudden developments of huge misfortunes, in the form of death or debilitation or displacement or impoverishment, of substantial groups of people simultaneously affected by such events as earthquakes, floods, epidemics, wars, or famines, which call for imperative social action for prevention and relief. That is the sense in which the idea of human disasters will be addressed in this chapter. It is, however, important to mention that the disastrous effects of terrorism and targeted killing are increasingly becoming another source of what can be called human disaster. Such specialized vio- lence in the contemporary world demands attention as well, and they take us in the direction of examining the influence of exploitation—​ and sometimes deliberate ‘planting’—​of a sharp and confrontational human identity in a highly belligerent form, used for exciting violent feelings towards other groups with different identities—​other reli- gions, other nationalities, other races, or other ethnicities. Studies are needed not only about the way identity politics can be made to serve the cause of human violence, but also about how organized in- flammation can become a source of danger for human security and an ingredient of a form of human disaster. Problems of a natural-​social dichotomy In the vast literature on the subject of human disasters, a sharp distinction is sometimes made between natural disasters, like earthquakes, floods or droughts, and social disasters, like wars or genocide. There is some rudimentary logic in that distinction: polit- ical sagacity can obviously have more success in averting genocide or wars than in, say, preventing earthquakes. Yet the contrast is not so 2.20 Human disasters Amartya Sen

2.20  Human disasters 189 total, since in the transformation of a natural calamity into a human disaster, social factors play a big role. For example, though floods are natural phenomena in an obvious sense, their effects—​and indeed even their incidence—​are also in- fluenced by social and economic events, such as the making of canals or drainage systems, not to mention the climatic effects of global warming that may be significantly influenced by preventable human activities. Even though in understanding the immediate parameters of such events as floods or droughts or storms, we can begin well enough by ‘rounding up the usual suspects’ in the natural world (such as rainfall, temperatures, gales), it is important that our ana- lysis should not end there. Some other examples of human disaster are quintessentially mixed bags, such as epidemics, which depend on the working of nature (e.g. on the properties of viruses, bacteria, and other conta- gions), and yet are massively influenced by human behaviour (like contacts and exposures) and social arrangements (like immuniza- tion and medical care). In such cases it would be very difficult even to begin the analysis of disasters of this mixed type—​epidemics and others—​only as phenomena of nature, since the social-​natural en- tanglement is rampant in every aspect of this type of human disaster. In some cases, the inescapably mixed causation of disaster has tended to escape attention because of the popularity of apparently ‘obvious’ causal explanations that fail to look far enough and are easily satisfied with finding a natural correlate. A good example of this kind of neglect can be seen in the popular understanding of the causation of famines in terms of food output decline, with much-​ repeated explanations that do not go beyond the proximate fea- tures of droughts, floods, or storms. This causal confusion, which is contradicted by the great many cases of famines that have occurred without any decline in food production, has been responsible for the loss of millions of lives, mainly through undermining the necessity and urgency of social intervention. The subject demands some discussion here, and to that I turn next. Indeed, by concentrating on famines, which have been much more extensively studied than other kinds of disasters, it is possible to illustrate some general points about the relationship between social and natural aspects of human disasters. Famines: Causes and prevention Famines are gigantic events of carnage in which millions of people die from starvation and from diseases linked with human debili- tation, social disruption, and movements of destitutes in search of something to eat, and the spread of communicable ailments asso- ciated with these phenomena. Even though a great many people die from a famine, the misfortune that a population experiences in a famine can go well beyond the mortality of those who succumb to the famine, since famines can leave an inheritance of huge disrup- tion for many years to come. The crude logic that if people are dying of starvation, there must be a shortage of food, has had much influence on traditional thinking across the world. That theory has had a particularly damaging effect in persuading governments to do nothing even when many people lose their livelihood (for one reason or another), which makes them unable to buy food. This hypothesis was relevant to the British Raj’s strange inactivity as the Bengal famine of 1943 slowly gathered momentum. It would eventually kill close to 3 million people. The of- ficial statistics of food output in Bengal indicated that there had been no significant change in the availability of food there, and sticking to their theory, the government did little to relieve the relentless emer- gence of a large famine. Instead, it tried to deny the existence of the unfolding famine, sustained by the censorship of the ‘native’ press, and the complicit silence of the British-​owned English papers. That gigantic cover-​up ended when the leading English paper of India, The Statesman (also British owned), broke rank with the decision of the highly agonized editor (Ian Stephens) that he could no longer be a party to the concealment of a huge human disaster. Stephens supplemented his factual reports of starvation by stinging editorials on official policy. Once the news went into the public do- main it got large attention in the British press and in Parliament in London, making the government admit the existence of a famine, which also led to the starting of much-​delayed relief work. However, ever loyal to the ‘theory’ that famines are caused only by food avail- ability decline, the same imperial government would later revise the statistics downward to bring the revised ‘facts’ into line with their deluded theory, even though the previous food statistics had been broadly right. What went wrong, of course, was the inability to understand that starvation is a characteristic of some people not having enough food to eat: it is not a characteristic of there being not enough food in the economy. The critical connection of the command over food—​what is sometimes called the ‘entitlement’ to food—​is with the purchasing power to buy food (in a market economy), linked to jobs and in- comes and relative prices of what people have to sell (their labour and the commodities they can make, such as services and crafts) to buy staple food. A famine is, in this sense, an economic phenom- enon, not just one of agricultural production. However, going further, famines are not only manifestations of an economic phenomenon, they also have political aspects and medical correlates as well. Famines are easy to prevent since only a relatively tiny proportion of the population is affected, and it is easy to regenerate the purchasing power of those who are destitute through emergency employment. This can provide incomes in a re- gion in which many people have lost their purchasing power be- cause of job loss or price rise. Indeed, the fact that independent India, despite considerable regular undernutrition at a chronic level, has not had a famine since the end of the Raj in 1947, is only partly linked with the progress of food production and agriculture in post-​independence years; it is also critically connected with the state’s immediate willingness to start relief programmes, mainly in the form of emergency projects, giving jobs and incomes to the af- flicted population. The urgency of preventive intervention is a political imperative for the government in office in a functioning democracy. If a famine is allowed to develop then no ruling party (or coalition) has much chance of winning a democratic election (or even to survive blis- tering criticism in the media and in parliaments or assemblies). Indeed, famines in the recent years, as in the past, have continued to occur only in countries without a functioning democracy (e.g. military dictatorships, one-​party states, or countries that do have elections but do not have other necessary features of a democracy such as a free media and room for public discussion and social criti- cism). Famines are thus manifestations of political as well as eco- nomic phenomena.

190 section 2  Background to medicine Famines also belong to the domain of medicine and public healthcare. This is because most people who die from famines die from diseases of the region, some of which are locally endemic. While starvation is the prime mover of famine mortality, actual deaths very often come from people getting ill and succumbing to their illness. Debilitation due to severe undernourishment plays a part in this in making people vulnerable to fall ill and sometimes to die of it. But no less importantly, hunger precipitates illness and death because of (1) the propensity of hungry people to eat what- ever scraps they can pick up from anywhere; (2) vast population movements induced by search for jobs and food that spread con- tagions with great speed; and (3) breakdown of essential services including healthcare and medical attention. Famines typically kill not just directly as a result of starvation but through intensifying and aggravating the forces of illness-​based mortality common to the region. Table 2.20.1 gives a breakdown of the causes of death in the ex- cess mortality in the Bengal famine of 1943, as presented in my 1981 book Poverty and Famines, for the period 1943–​1946, taking the prefamine 1941–​1942 average as the standard for comparison. More than four-​fifths of the death toll resulting from the Bengal famine, in this estimate, was directly connected with diseases common to the region, with pure starvation death accounting for no more than only a fifth of the total. A similar picture emerges from most other famines. This aspect of famines is the centre of attention in Alex de Waal’s important study of famines in Darfur in Sudan. This is certainly a rich and policy relevant perspective on famines, particularly be- cause even with the failure of food entitlements, the magnitude of deaths can be very substantially reduced through health inter- ventions. If the emergence and survival of a famine is largely an economic and political issue, the mortality it generates is also very significantly a matter of healthcare and medical attention (including timely prevention, particularly through immuniza- tion). The allegedly ‘natural’ phenomenon of famines tends to be intensely ‘social’ at many different levels—​economic, political, and medical. Before I move on from the subject of famines to other types of human disasters, let me briefly note the variety of circumstances in which famines have occurred and have killed with abandon across the world. It would be good to associate firm mortality es- timates with each famine, but this would be a difficult thing to do in many cases. Even though death tolls have been thoroughly studied for some famines, using reasonably good statistics, there is always an element of uncertainty in any such estimate, no matter how carefully it may have been prepared. Part of the difficulty lies in the underlying conceptual issues. The death toll of a famine has to be calculated by contrasting the actual number of deaths in that period with the number of deaths that would have taken place in the absence of that famine. That is, famine mortality is the difference between actual mortality in the famine period and the estimated number of people who would have died anyway, even if the famine had not occurred at all. Thus the uncertainties involved relate only partly to problems in getting good statistics of actual deaths, which is not an easy task as it is, especially at a time when many of the normal functioning of social institutions, including registration of deaths, are disrupted (sometimes because of death or migration of the staff involved in the relevant social institutions). They also relate to the uncertainties inherent in the ways and means of estimating—​ as a ‘counterfactual’—​what would have happened had the famine not occurred at all. One can nevertheless get some general impression of the size and reach of a famine by looking at the estimates of death tolls that we have, along with reports of other events and predicaments con- nected with the respective famines. Table 2.20.2 relates to many ‘recent’ famines that have occurred over the last couple of hundred years, beginning with the Irish famines of 1845–​1851 and drawing on a large number of sources. The association of famines with political authoritarianism and civil disruption is well illustrated by the incidence of such events in countries with military governments (Ethiopia, Sahel countries), or one-​party rules (Soviet Union, China, Cambodia, North Korea), or alien governance (preindependence India and politically dom- inated Ireland), or civil disruption (Nigeria, and to some extent Bangladesh, even though the death toll was kept in check, particu- larly in the latter, through organized famine relief). Disasters and policy intervention Human disasters can be of many types, varying from floods, droughts, wind storms, extreme temperatures, and earthquakes, where natural factors have some clear role, to wars and indus- trial accidents (like the Union Carbide disaster in Bhopal in 1984) where the story is almost entirely social in the broad sense. But the important point in the context of policy intervention is to recognize that, even when nature plays a part, society can make a huge difference. Consider natural calamities such as floods or droughts, which have been associated with processes that have led to great many disaster deaths. What can intervention achieve? There are at least three different ways of prevention or amelioration. • First, environmental policies can make floods, droughts, and other natural adversities less common. This is a subject that is being energetically discussed today—​at long last—​moved largely by the evidence of massive global warming and the related long-​ run environmental hazards. The recent ‘COP21 summit’ in Paris in December 2015 is a good example of attempting disaster pre- vention through concerted international efforts. There was, Table 2.20.1  Proportionate breakdown of excess mortality in the Bengal famine: 1943–​1946 over 1941–​1942 Ailment types Percentage contribution Malaria 36.7 Cholera 7.1 Smallpox 5.0 Dysentery, diarrhoea, and enteric group of fevers 5.0 Other fevers (often undiagnosed) 27.4 Respiratory diseases 0.4 Percentage share of ailment-​related deaths 81.6 Source: Sen A (1981). Poverty and Famines: An Essay on Entitlement and Deprivation. Oxford University Press, Table D.3, p. 204.

2.20  Human disasters 191 contrary to earlier fears, some limited success, and yet much was left unresolved in Paris—​what George Monbiot described in The Guardian as ‘by comparison to what it could have been, it’s a mir- acle’, but ‘by comparison to what it should have been, it’s a disaster’ (12 December 2015). There will no doubt be many efforts in future years to avert the climatic catastrophe-​in-​making. • Second, any human disaster that kills directly will also tend to disrupt the economy through its effects on jobs and incomes and prices, thereby affecting the entitlement to food and other es- sentials that people need. The discussion on famines above has already illustrated this general connection. That analysis can be extended. For example, an earthquake may not lead to a famine, but can leave an economy unsettled, making lives precarious for those who are not killed in the physical calamity itself. These ef- fects can be prevented or at least reduced in their impact through careful social intervention. The type of housing and the arrange- ments of city planning can also reduce the effects of earthquakes, floods, storms, and other hazards. Third, the morbidity and mortality associated with a human disaster can be large and even extend far beyond the immediately affected population who are directly hit by the physical events. However, taking note of the likely dangers from local conta- gions, the scale of both illness and death can be radically altered through a variety of preventive measures, such as immunization, ensuring safe water, influencing the routes of contact and the spread of infection, and also through health education and med- ical preparedness. The divisiveness of disasters The death toll from so-​called natural disasters reduces sharply as we move to higher and higher income countries, even when they do not have any fewer natural events of that type. The contrast can be very sharp indeed, as Kofi Annan, the former Secretary General of the United Nations, has noted (International Herald Tribune, 19 September 1999): ‘Ninety per cent of the disaster victims worldwide live in developing countries where poverty and population pressures force growing numbers of poor people to live in harm’s way on flood plains, in earthquake prone zones and on unstable hillsides. Unsafe buildings compound the risks. The vulnerability of those living in risk prone areas is perhaps the single most important cause of disaster casualty and damage.’ A similar remark can be made about the role of medical services and epidemiology, which are often rudimentary in the poorer countries. If low income is a predisposing condition for the penalty of disasters, the type of political rule can also be a factor of importance. The impact of the insensitivity of ruling governments in authori- tarian countries on the interests of vulnerable people was illustrated earlier in the specific context of famine prevention, but the problem of authoritarian insensitivity is a widespread phenomenon and goes well beyond the terrible record of authoritarian rules with famines. The neglect of medical services and healthcare that can be seen in many military dictatorships, for example in sub-​Saharan Africa, is Table 2.20.2  Some recent famines 1845–​1951, Ireland A series of famines linked with a blight in potato farming, with loss of jobs, incomes, and staple food; famine mortality around one million deaths (the largest proportionate mortality among all ‘recent’ famines). 1928–​1929, China Famines, particularly affecting farming population, especially in Shensi, Honan, and Kansu provinces, with mortality estimates that are in excess of 5 million deaths. 1932–​1934, Soviet Union Connected with agricultural turmoil, at least partly linked with collectivization, especially severe in Ukraine, with mortality estimates in excess of 5 million deaths. 1943, India In the province of Bengal in British India, with a fairly normal food availability; mortality numbers are now placed between 2 and 3 million. 1958–​1961, China Linked with the disastrous failure of the so-​called Great Leap Forward, there was a sharp decline in food production, disruption of normal economic processes, and huge chaos; the famine was well hidden from those not directly exposed to it. The death toll has been later estimated to be at least around 30 million (making it the largest absolute size of famine mortality in the recorded history of famines in the world). 1967–​1968, Nigeria Connected with civil war and the blockade of Biafra; no reliable estimate of deaths has been made. 1972–​1974, Sahel region Affecting several African countries in the Sahel belt (Burkina Faso, Chad, Mali, Mauritania, Niger, Senegal), with some reduction in agricultural production, but also affecting pastoralists both through animal death and through a fall in the relative prices of animal products against staple food from agriculture; no reliable mortality estimates can be made. 1973, Ethiopia The drought in the Wollo province reduced the purchasing power of the poor, and even though there was no significant reduction in food output at all for Ethiopia as a whole (the reduction was confined mainly to Wollo), food did not move into Wollo, and some moved out of it, because of the relatively larger purchasing power of the rest of Ethiopia; no reliable mortality estimates can be made. 1974, Bangladesh This famine, which followed shortly after the disruption of civil strife that led to the break-​up of Pakistan, occurred in a year of peak food availability. Floods that would reduce food output the following year immediately affected employment and incomes of rural labourers, and also an exaggerated anticipation of a coming food crisis made food prices rises very steeply (followed by a price decline later on, after the famine). Mortality estimates vary but it was kept very much in check by a huge food relief programme arranged by the government through which 4.35 million people (about 6 per cent of the total population) were fed. 1975–​1979, Cambodia Linked with the Khmer Rouge ravages of the rural economy, deportation of parts of the urban population, and systematic slaughter of unfavoured people; the death toll is estimated to be around one million. Since early 1990s, North Korea Intermittent famines with economic disruption by an authoritarian regime with a strong ideological agenda; death estimates are often placed around one million—​so far.

192 section 2  Background to medicine itself a problem for disaster management. There are political param- eters related to the medical, social, and economic interventions that can make disasters less likely, and less devastating when they do occur. The effects of similar natural phenomena can be widely different, depending on income levels, political reactions, medical arrangements, and other variable conditions. Human disasters are very divisive events, not only between coun- tries but also within them. Even in the same country, and sometimes even within the same region of a country, a disaster can ruin some groups of people while leaving others almost completely untouched. Famines, for example, tend to affect those who are already vulner- able because of poverty and the absence of safety nets on which they can rely, private or social. Policy intervention has to take note of the divisions along the lines of class and occupation groups. There are also differences related to age and gender. Children may be particularly vulnerable in some disasters, including in epidemics of some diseases. The need for immunization of children and arran- ging medical attention for children are important requirements in disaster management, and so is the institution of economic safety nets through state provision and other social arrangements for child support and income supplement at moments of dire need. In some disasters (for example in the present phase of AIDS epi- demic), women are increasingly more affected. It is usually the case that the opposite is true in wars, which typically kill more men than women or children, although women may be particularly vulner- able when there is a civil war targeted against a specific community where women are systematically raped and injured, as has happened for example in Darfur in Sudan. Even though the mortality rate may be higher among men, leaving a lower ratio of men in the surviving population, the huge incidence of rapes amounts to a gigantic vul- nerability for women in particular. There are other kinds of special vulnerabilities in disasters, like breakdown of families in famines, from which women tend especially to suffer. There is also a more subtle point that tends to be neglected in journalistic reports. The life expectancy of women is typically longer than that of men, unless this is reversed through gender bias in social care. There is some evidence that the life expectancy of women is comparatively more reduced by many types of so-​called natural disasters than that of men. This greater vulnerability of women is sometimes hidden by the fact that even after the larger reduction, women may still have a higher life expectancy than men, but—​as was discussed in the context of famine analysis—​the right comparison is between what actually happens with a disaster and what could have been expected to have happened in its absence. That distinction is as relevant for medical practitioners and public health experts as it is for social and economic investigators, as is the recognition that human disasters are typically highly divisive events. FURTHER READING Bose S (2009). Pondering poverty, fighting famines: towards a new history of economic ideas. In: Basu K, Kanbur R (eds) Arguments for a better world, Vol. II. Oxford University Press, Oxford. Depoortere E, et al. (2004). Violence and mortality in West Darfur, Sudan (2003–​2004): epidemiological evidence from four surveys. Lancet, 364, 1315–​20. Dreze J, Sen A (1989). Hunger and public action. Oxford University Press, Oxford. Guha-​Sapir D, Hargitt D, Hoyois P (2004). Thirty years of natural disasters 1974–​2003:  the numbers. Centre for Research on the Epidemiology of Disasters, Presses Universitaires de Louvain, Louvain-​la-​Neuve, Belgium. Kahn ME (2005). The death toll from natural disasters: the role of income, geography, and institutions. Review of Economics and Statistics, 87, 271–​84. Malik K (2014). Multiculturalism and its discontents: rethinking diver- sity after 9/​11. Seagull Books, Kolkata, India. Neumayer E, Plumper T (2007). The gendered nature of natural disasters: the impact of catastrophe events on the gender gap in life expectancy, 1981–​2002. Ann Assoc Am Geogr, 87, 551–​66. Ravallion M (1987). Markets and famines. Oxford University Press, Oxford. Sen A (1981). Poverty and famine: an essay on entitlement and depriva- tion. Oxford University Press, Oxford. Sen A (2006). Identity and violence:  the illusion of destiny. Norton, New York and Penguin, London. Stephens I (1966). Monsoon morning. Ernest Benn, London. Stern N (2007). The economics of climate change. Cambridge University Press, Cambridge. Vaughan M (1987, 2007). The story of an African famine. Cambridge University Press, Cambridge. de Waal A (2005). Famine that kills: Darfur, Sudan. Oxford University Press, Oxford. World Disaster Report 2005 (2006). International Federation of Red Cross and Red Crescent Societies, Geneva, Switzerland.