Responding to Catastrophes: A Public Health Perspective

I. INTRODUCTION

From a public health perspective, it is difficult to define exactly what a catastrophe is. Catastrophes can be of sudden onset or they can develop slowly; they can be the result of natural causes, such as hurricanes, droughts, or earthquakes, or they can be man-made, a consequence

of war or of terrorist acts. Some would say that the distinction between these is not totally clear-even in earthquakes, for example, mortality rates are predictably higher among the poor, those who live in housing that does not conform to local construction standards. The Indian Ocean tsunami of December 2004 did not spare anyone on the basis of socioeconomic status, but those with means were able to rebuild, rehabilitate, and reconstruct their lives much more rapidly and more completely than those who had only minimal assets. Finally, the plight of the poor left behind in the wake of Hurricane Katrina, one of the world's most recent major catastrophes, was visible on televisions around the world. The blurriness of the lines between these categories, acute versus slow onset and natural versus man-made disaster, has led some to coin the term "complex emergency." The global response to complex emergencies has become a subject of relatively recent study and many of its medical, engineering, and even legal ramifications are still being refined.

Unfortunately, there is no universally accepted definition of exactly what constitutes a complex emergency. One common description, initially promulgated by the Centers for Disease Control and Prevention, the lead public health agency of the United States government, contends that complex emergencies are "situations affecting large civilian populations that usually involve a combination of factors including war or civil strife, food shortages, and population displacement, resulting in significant excess mortality."1 This definition is obviously imprecise, but it is clear that events such as the outpouring of hundreds of thousands of Hutus into neighboring Tanzania and Zaire (now the Democratic Republic of Congo) in the wake of the 1994 genocide in Rwanda generally fit the definition. The displacement of villagers in the Darfur region of Sudan, fleeing both to more southern parts of that country and over the border to Chad, is another clear-cut example of a complex emergency.

Other aspects of these catastrophes are highlighted in a different definition of a complex emergency: "a humanitarian crisis in a country, region or society where there is total or considerable breakdown of authority resulting from internal or external conflict and which requires an international response that goes beyond the mandate or capacity of any single agency and/or the ongoing United Nations country programs."2 This definition does not characterize the situation as much as it does the nature of the response, which needs to be "humanitarian" and requires a team of government, nongovernmental, and United Nations agencies to work in the absence of a functional state (such as in Somalia and Kosovo) or in the face of an unwilling governmental partner (as is the case in Sudan or Burma).

Because of the imprecision of these and other definitions, the response to complex emergencies is somewhat unpredictable. Some have gone so far as to suggest that a reasonable definition of a catastrophe that requires response on the part of the "humanitarian community"3 is an event that is widely covered by CNN and/or other news media. Certainly, the political response to complex emergencies, especially when concerns of national security are not paramount, can be heavily influenced by the media. The 1991 US military incursion in Somalia is widely cited as a case in point. Without clear objectives, until about twenty years ago, emergency relief was based primarily on good intentions. For example, generous people and organizations were moved to action by photographs of starving babies in Europe after the Second World War; in Biafra during the war of secession in Nigeria in the late 1960s; and in Thailand, Somalia, and Ethiopia during the late 1970s and 1980s. Humanitarian assistance to people in those settings was based on an outpouring of generous, kindhearted, and charitable support from both the public and private sectors. However, it was not always effective. At least to a degree, it was not effective because humanitarian assistance was a profession without standards, without benchmarks against which the performance of the humanitarian community could be measured. In short, there was a lack of accountability both to those who provided funds for relief efforts and, perhaps more importantly, to those who were ostensibly being helped. One of the leading myths of disaster relief was that any kind of assistance, in any amount, was always helpful. This approach has now been proven, time and again, to be unwarranted, while a more scientifically oriented approach to the provision of humanitarian assistance in complex emergencies has evolved. The latter approach and the problems that have emerged in association with its development are explained further in this Article.

II. MORTALITY IN COMPLEX EMERGENCIES

Following an analysis of complex emergencies that included the refugee crisis engendered by the murderous Pol Pot regime in Cambodia, the flight of Ogaden-based Somali ethnic peoples from Ethiopia to Somalia following the overthrow of Emperor Haile Selassie, and the largely man-made Ethiopian famine of the mid-1980s, it became clear that in most complex emergencies, crude mortality rates (the number of deaths occurring in a defined population over a specified period of time) could be substantially elevated. From a public health perspective, it is logical to suggest that a primary objective of humanitarian relief should be to lower the crude mortality rate to below a specific level as rapidly as possible. Even recently, it has been stated that "mortality is the prime indicator by which to assess the impact of a crisis, the magnitude of needs and the adequacy of the humanitarian response."5 Accordingly, a threshold level of "acceptable" mortality was established on the basis of prevailing rates in developing countries, as reported by the United Nations Statistics Office. Because a typical mortality rate in sub-Saharan Africa and South and Southeast Asia might range between 0.3-0.5 deaths per ten thousand people per day, a threshold for mortality that might be useful in distinguishing a public health emergency from a non-emergent situation was suggested to be one death per ten thousand people per day, representing a doubling or trebling of the baseline rate. It was proposed that whenever the crude mortality rate in a population rose above the threshold level, a humanitarian response should be triggered. Intensive emergency relief efforts should be aimed at lowering the mortality to below the threshold. Finally, an emergency could be said to have subsided when mortality dips below the threshold rate.6 As can be seen in Figure 1,7 several quite different scenarios have been observed.

In Thailand in late 1979, for example, a large group of refugees from Cambodia was rapidly settled in a number of what became known as "death camps" because of what was, at that time, an extremely high crude mortality rate for the region of about 3.3 deaths per ten thousand people per day. A massive intervention ensued, characterized by a veritable invasion of Western medical personnel, the rapid construction of hospitals and clinics, and the provision of food, water, and other essential commodities. The rapid and favorable evolution of the situation, with mortality falling to the baseline level, well below the "emergency threshold" within a few months, was felt to be indicative of a highly successful relief effort.

The 1985 situation in Sudan was somewhat different. Initially measured crude mortality was, for that time, the highest ever recorded, at seven to eight deaths per ten thousand people per day. Massive publicity in the West, including the celebrated "Band-Aid" and "We are the World" efforts of popular entertainers in the United Kingdom and the United States triggered an unprecedented response from the entire humanitarian community. However, as can be seen from the graph above, the crude mortality rate fell far more slowly than had been the case in Thailand. In fact, one year after the opening of the camps in which refugees were settled, the crude mortality rate was just reaching the threshold level.

The last line in this graph depicts the unusual case of the Hartisheik B refugee camp in Eastern Ethiopia, where refugees from Somalia were temporarily settled in the late 1980s. In this situation, the crude mortality rate, which had actually been below the emergency threshold at the time the camp was opened, rose steadily for nine months after the relief effort by the international humanitarian community was initiated. Although multiple factors were responsible, the fundamental explanation for this surprising finding is that insufficient food was reaching the camp. Obviously, increasing mortality rates in a population that is dependent upon external assistance for survival should be indicative of an unacceptable relief effort.

Finally, no discussion of mortality in complex emergencies should omit the extreme catastrophe that occurred in refugee camps near Goma in the former Zaire. As alluded to earlier, between five hundred thousand and eight hundred thousand Hutu citizens of Rwanda fled to four refugee camps in July of 1994 in order to escape possible retribution for the Hutu role in the genocide of Rwandan Tutsis that had occurred in the Spring. Shortly after their arrival in the vicinity of Lake Kivu, a virulent epidemic of cholera swept through the entire population, resulting in approximately forty-five thousand deaths over a four week period. In connection with this outbreak, a number of salient points should be discussed.

First, the number of deaths that occurred was determined with unusual accuracy. Several of the camps were hastily established on ground composed of volcanic rock. Because graves could not be dug, most of the dead were brought to the side of the road where trucks hired by international authorities-in this case, the Office of the United Nations High Commissioner of Refugees ("UNHCR")-loaded the corpses and hauled them to where they could be buried in mass graves. Counting the bodies became part of the job description of those hired for this grisly task.

Second, despite the fact that an accurate numerator (number of dead) was determined, mortality rates in this population can only be given in approximate terms because the size of the population was never clearly determined. Best estimates placed the population at between five hundred thousand and eight hundred thousand so that by some estimates the epidemic caused a literal decimation of the initial population in only one month's time. In fact, estimated crude mortality rates in the camps ranged from 29.4 to 41.3 deaths per 10,000 people per day, by far the highest ever recorded in a complex emergency to that time.

Third, most deaths occurred out of the reach of the international community. One can only imagine that an attitude of fear and trepidation pervaded a population that had survived the unimaginable trauma of the genocidal period. These refugees must have been especially wary of the largely white and Western relief community that had gathered to build clinics and hospitals in an attempt to establish food and water supplies and to distribute materials for shelter. In any case, fewer than 10 percent of those who died were apparently seen for medical care by the humanitarian community. In fact, although 45,435 bodies were collected by UN trucks, only 4,335 deaths were reported by health relief agencies. According to some informed estimates, not a single cholera death was prevented by the humanitarian community during the entire relief effort.8

The situation was even worse for children. As death spread through the population, many children were orphaned and many more abandoned by the surviving parents or relatives to the care of the humanitarian community. For security reasons, expatriate relief workers did not stay in the camps at night. Upon their return each morning, however, groups of unaccompanied children would be waiting outside their clinics. These children, many of them babies and many of them sick, were taken to makeshift orphanages in the town of Goma where they were left, frequently unsupervised, to fend for themselves. Mortality rates in these children, admittedly based on small numbers, reached as high as several hundred per ten thousand children per day.9

A few additional words are in order concerning mortality in complex emergencies in order to set the stage for further discussion. These concern not the levels of mortality, but its causes. If a significant number of deaths in these settings were due to intractable diseases or to diseases for which the only cures were expensive and relatively ineffective, then the poor performance of the humanitarian community, as shown in several of the examples presented above, might be excusable, at least to a degree. In fact, however, this was not the case. Because armed conflict is frequently a part of the definition of a complex emergency, one might expect violence to be a frequent cause of death. Most often, though, the vast majority of deaths result from indirect causes, principally infectious diseases whose occurrence is favored by the crowding of large populations into temporary, substandard settlements with compromised water supplies and food of insufficient quantity and quality.

As a result, deaths in most complex emergencies are due to common conditions like diarrhea (including cholera and bacillary dysentery), pneumonia, malaria in areas where it is endemic, and vaccine-preventable diseases like measles and meningitis.10 These can be frequently compounded by malnutrition, a condition that does not affect the occurrence of infectious diseases but can increase the likelihood that an illness will result in death. For example, although many in Europe and the United States associate the Sudan refugee crisis of the mid-1980s with drought, crop failure, and starvation on the basis of the pathetic images of starving children that were ubiquitous on our television sets, more than one-half of all deaths were due to measles. Measles rarely causes death-and rarely even occurs-in industrialized countries because a safe, effective, and highly affordable vaccine has been routinely administered to children since 1964. In developing countries, however, measles vaccine coverage is comparatively low. Although policies recommending mass campaigns with the measles vaccine were adopted following the Sudan episode, they are not always implemented. Thus, during a 2000 famine in southern Ethiopia, investigators found that 22 percent of the deaths in children less than 5 years-old that they sampled were attributable to measles.11

III. INTERNATIONAL LEGAL IMPLICATIONS OF MORTALITY IN COMPLEX EMERGENCIES

What can be learned from this discussion of mortality in complex emergencies that might be of interest to students and practitioners of international law? To start with a relatively trivial point, it should be noted that in the course of the international response to complex emergencies, many health care professionals from countries around the world descend upon the territory of a sovereign state to deliver health services to a population in need. Few, if any, of these practitioners are licensed to practice in the country that is hosting them. In the wake of Hurricane Katrina, more than ninety-five countries around the world offered almost $1 billion in assistance to the United States. Among them, many developing countries with significant experience in dealing with disasters offered financial and material aid, including Bangladesh, whose offer of $1 million was accepted.12 But one can only imagine what the response of the US government, or of the American Medical Association, might have been should Bangladesh have offered the services of a team of diarrheal disease control specialists with the same expertise as the team that was sent to help quell the cholera epidemic in Goma.13 Questions of licensing, competence, and liability would surely have been raised. In international disaster settings, however, there is rarely an authority that assumes the responsibility of verifying credentials and approving health service providers.

More important than credentials, however, is the question of ability-are those who respond qualified, on the basis of knowledge and experience, to provide an adequate level of care to the affected population? Approaches to the prevention and control of the leading causes of disease and death in complex emergencies have been studied and described. The provision of health care for those suffering from the leading causes of morbidity and mortality in complex emergencies is not overly complicated from a technical standpoint, but it does require some familiarity with current policies and recommendations, as well as with the appropriate medicines and their potential adverse and secondary effects. Nevertheless, many of the nongovernmental organizations that send relief teams to work in the health sector in emergencies do not have training programs, nor do they necessarily require that their personnel demonstrate knowledge and technical competence prior to being assigned to the field. This lack of familiarity with required skills can result in unnecessary loss of life. As the Bangladeshi physicians cited above state in their article about the cholera epidemic in Goma, "the slow rate of rehydration, inadequate use of oral rehydration therapy, use of inappropriate intravenous fluids, and inadequate experience of health workers in the management of severe cholera are thought to be some of the factors associated with the failure to prevent so many deaths during the epidemic."14

Interestingly, to a large degree as a result of the failed humanitarian relief effort in Goma, a substantial number of training programs, both short courses and graduate degree offerings in the practice of public health in complex emergencies have been developed. It should be possible for nongovernmental organizations, either voluntarily or as a condition of receiving government funding, to ensure that at least an acceptable proportion of health teams have demonstrable knowledge of the skills most likely to be of benefit to disasteraffected populations. To date, however, no such requirements have been adopted.

The affected populations have little recourse in cases of incompetent medical care. They are distressed, vulnerable, and expected to be docile and compliant. Their source of care is determined by various means, but usually by a United Nations coordinating agency, which could be UNHCR in the case of a refugee crisis or, when designated as the lead UN agency by the secretaryGeneral, the World Health Organization or the United Nations Children's Fund ("UNICEF"). It is becoming increasingly common for the UN Office for the Coordination of Humanitarian Affairs ("OCHA") to function as the overall coordinating agency for disaster relief and to appoint one of the other UN agencies as the lead agency for the health sector. This agency, in turn, is responsible not only for the formulation and promulgation of appropriate health sector policies and the prioritization of programs, but also for assigning specific relief agencies to camps, settlements, or sites in a way that ensures that all of the disaster-affected population has access to appropriate health services. When this effort is not coordinated properly, high-priority health services will not be available to all and it is conceivable that preventable deaths will occur. One commentator has gone so far as to suggest that poor coordination be listed among the diseases mentioned above as a leading contributor to excess preventable mortality in complex emergencies.15

Another nagging issue with legal ramifications related to the management of common diseases concerns whether or not the humanitarian community must conform to national policies governing health care in the countries where it is providing relief. The problem presented here is that the national policy formulation process can be slow and cumbersome, and inadequate or inappropriate policies can remain in place for many years, especially in countries where the implementation of more technically correct health care policies may exceed the financial means of government, but not of the humanitarian community. A case in point concerns a confrontation between M?decins Sans Fronti?res ("MSF"), one of the leading humanitarian agencies working in the health sector, and the Ministry of Health of Burundi, a country in which civil strife had been responsible for the displacement of tens of thousands of individuals and for the compromised security of hundreds of thousands of others. Malaria is a leading cause of death among children in Burundi and, during an epidemic that occurred there from 2000-2001, over three million cases occurred among the country's population of 6.5 million. One reason for the high burden of illness may have been that the malaria-causing parasite had developed resistance to the drug of choice designated by national policy. MSF, working in a circumscribed area of the country, found it unethical to offer treatment it knew to be ineffective for "its" patients. While advocating for the adoption and use of a new, albeit more expensive, but more effective, antimalarial drug combination (artemisinin plus amodiaquine), MSF began to dispense it. This action was deemed illegal by government authorities and MSF's malaria-related work in Burundi was suspended. Subsequently, the government, under continued pressure from MSF and other organizations, agreed to adopt a new policy by July 2003. While MSF has been providing the new treatment in the provinces where it is working, the government of Burundi remains unable to implement the treatment on a wider scale for a variety of reasons, including the relatively high cost of artemisinin and its limited supply on the world market.16

As a result, only patients in MSF's service areas had access to reliably effective anti-malarial treatment. This dual health care system is difficult for a fledgling government attempting to establish legitimacy to tolerate. On the other hand, of course, it could easily be argued that were it not for MSF's ability to provide effective treatment, at least after the policy change was effectuated, and even before, no Burundian would have benefited from the technological advances that the humanitarian agency was able to provide. The humanitarian tradition has long been more concerned with basic human needs, including health care, than it has been with accepting legal constraints to effective humanitarian action.17 Conflicts between the humanitarian community and the laws of sovereign states occur frequently. In some instances there are undoubtedly good reasons for government policy to be strictly enforced. In others, logic and devotion to a cause may dictate that state concerns be considered secondary to the humanitarian imperative. More study, critical analysis, and consideration of some of these thorny issues would help guide appropriate action.

IV. INDIVIDUAL AND PUBLIC RIGHTS TO HEALTH CARE IN CONFLICT-RELATED DISASTERS

To this point, the discussion has concerned large populations. But what about the interests of the individual caught up in conflict-related disasters? Yet another crucially important tension exists between two fundamental objectives of humanitarian relief: the protection of the right of an individual to the highest attainable state of health possible and the preservation of the health of the public. There need not be conflict between these principles as long as there is a clear understanding that, in the end, the fundamental objective of public health interventions in complex emergencies is utilitarian: to do the most good for the most people. Implementing interventions that are primarily aimed at achieving this objective may, at times, mean that individuals must be denied care that they might receive in other, more stable circumstances.

When mass casualties occur and the number of patients overwhelms the human and material resources available, emergency room physicians use one of a number of systems of triage to decide who will receive care in what order. A patient whose life could possibly be saved, but only by taking time and care away from a number of others who require equally life-saving but less time-consuming care, can be left to die. Imagine a situation where a four year old child has suffered a severe head wound and requires major surgery in order to live. The parents are frantic for their only child to be treated, but the health staff is attending to other seriously wounded patients. After a few hours of neglect the child dies, but others have been saved. At the end of the day, what has been achieved is not a good outcome by any means, but it is the best possible outcome.

In an all too real sense, complex emergencies are the emergency rooms of public health. Appropriately, an equivalent system of triage is used to decide what programs should be prioritized.18 Although the needs of a disaster-affected population are specific to each setting and should be rapidly evaluated prior to initiating action, enough evidence has been accumulated from the study of complex emergencies over the last few decades to develop a suggested list of priority interventions.19 The need for information, though, cannot be overlooked, and an assessment of the local situation followed by the earliest possible establishment of a health surveillance system is essential. Immediate attention should be given to ensuring that the population has access to those things that are necessary to sustain life-adequate quantities of appropriate food, sufficient water of acceptable quality (although quantity is much more important than quality), acceptable sanitation facilities, and shelter. Health care is secondary to these concerns, although those working in the health sector should certainly be establishing their programs while specialists in the other areas are implementing theirs. Based on past experience, the first health intervention is usually a measles vaccination campaign aimed at children between the ages of six months through twelve or fifteen years of age. Only after this has been completed should other interventions be instituted.

The control of communicable diseases is, in most settings, high on the list of health sector priorities, as these are clear dangers to the public. When epidemics occur, as is often the case for reasons cited above, they must be dealt with to the abandonment of other activities. Conditions that can be addressed in the short-term will be given greater attention than those that require longer-term care. For example, time and human resources are rarely, if ever, accorded to disaster-affected patients with HIV/AIDS or tuberculosis until control measures for diarrhea, malaria, and other diseases of epidemic potential have been fully instituted. As a general rule, programs aimed at these and other chronic diseases will not be established until the emergency has subsided, until it is clear that population movement will be minimal for at least six months, and until an agency working in the health sector has assumed a coordinating role.

In other words, in disaster response, appropriate health care is not made available to everyone, nor should it be. It may not seem fair to suggest that patients with some diseases should be given treatment while those with others should not, but it is necessary if the most lives are to be saved. The ethical implications of triage have been discussed and debated, but in emergency settings it is a well-accepted and generally recommended process.20 Although the notion of triage has not been as carefully developed for prioritizing public health programs as it has for clinical decision-making, the analogy is clear and the principles remain the same.

V. EMERGENCY RELIEF AND QUALITY OF LIFE

To this point, I have concentrated almost exclusively on mortality reduction as the end point of humanitarian assistance. It is important, however, to look not only at the quantity of lives saved, but also at how emergency relief impacts the quality of those lives. The Goma relief effort was an important milestone in this regard as well. As the cholera epidemic and other epidemics of communicable diseases subsided after August 2004, the humanitarian community began to take stock of what it had done and of what it should continue to do. It became increasingly clear that as the emergency relief effort had been imperfectly implemented, so too was the post-emergency assistance program peculiarly flawed. Hutu militia exercised substantial control, by force and intimidation, over the majority of the population in the camps. By November 1994, a group of nongovernmental organizations threatened to stop providing assistance to the camps, claiming that under the prevailing conditions it was impossible to protect and assist the refugees, that the lives of aid workers were being threatened, and that, at least figuratively, the refugees were being held hostage. Aid was being hijacked by the same strongmen who had engineered the genocide, and the relief efforts could not continue without violating the principles of humanitarian assistance. The end result was, they asserted, that their "humanitarian" actions were not helping those in need, but were, in fact, fueling the conflict.21 By the end of the year, the French section of MSF and the American NGO International Rescue Committee had both terminated their operations in Goma.22

With time, it became clear that the Goma situation was not unique. Relief workers eventually realized that what came to be called "humanitarian space" had been shrinking as more and more complex emergencies came to be characterized by conflict-particularly as the populations being served came to include armed thugs and instigators out to enrich themselves. Incidents occurred in Liberia and Sudan where the intended beneficiaries of humanitarian relief were attacked and robbed after receiving relief rations. In other words, the humanitarian community found itself inadvertently helping people, either government authorities or community leaders, or both, who were more a part of the problem than a part of the solution. The political na?vet? of the humanitarian community that had been understood by a relative few during the Ethiopian famine of the 1980s became the subject of broad public scrutiny and of severe criticism after Goma.

Stung by the public impression that it had been "had," and reeling from a harshly critical multi-donor evaluation report, the humanitarian community undertook a period of introspection.23 After Goma, the nature of humanitarian discourse changed considerably and there was talk of a "new humanitarianism."24 This new approach to humanitarian assistance was often referred to as "smart relief." It was characterized by attempts to be more politically astute and to focus more on the social, economic, and political consequences of emergency relief efforts, and not just on what was felt to be the strictly technical interventions aimed at saving lives.25 As one commentator elegantly put it, the humanitarian community began to question its values and to reflect on the "apparent clash between humanitarianism and human rights . . . [between being] forced to choose between responding to the right to life or the right to justice and the broader values of civil and political rights. . . . Striking a balance between the two is at once the art and the agony of true humanitarianism."26

As a result of seeking to achieve this balance, humanitarian relief changed from a more or less chaotic and undirected set of interventions guided by charitable motivations to an emerging discipline that drew increasingly heavily on interpretations of human rights law combined with a more scientific approach to the provision of assistance to disaster-affected, and usually conflictaffected, populations. Because some of the failings of the relief effort in Goma were attributed to a proliferation of inexperienced nongovernmental organizations that responded to the compelling tragedy of overt genocide and were assisted by relatively easily accessed funding from both public and private sources, mainstream humanitarian organizations began to promote the development of professional standards for relief agencies and their personnel. One of the first of these sets of standards to appear was the "Code of Conduct for the International Red Cross and Red Crescent Movement and NGOs in Disaster Relief."27 Written in the hope that a wide variety of nongovernmental organizations would sign on to its principles, it is worth presenting here:

This Code of Conduct seeks to guard our standards of behaviour. . . . It is a voluntary code, enforced by the will of organisation accepting it to maintain the standards laid down in the Code.28

1. The humanitarian imperative comes first. (The right to receive humanitarian assistance, and to offer it, is a fundamental humanitarian principle which should be enjoyed by all citizens of all countries . . . we recognize our obligation to provide humanitarian assistance wherever it is needed.)

2. Aid is given regardless of the race, creed or nationality of the recipients and without adverse distinction of any kind. Aid priorities are calculated on the basis of need alone.

3. Aid will not be used to further a particular political or religious standpoint.

4. We shall endeavour not to act as instruments of government foreign policy.

5. We shall respect culture and custom.

6. We shall attempt to build disaster response on local capacities.

7. Ways shall be found to involve programme beneficiaries in the management of relief aid.

8. Relief aid must strive to reduce future vulnerabilities to disaster as well as meeting basic needs.

9. We hold ourselves accountable to both those we seek to assist and those from whom we accept resources.

10. In our information, publicity and advertising activities, we shall recognize disaster victims as dignified humans, not hopeless objects.29

Subsequent to its dissemination, more than one hundred fifty NGOs signed a pledge to respect the Code. Some donor agencies went so far as to make signing the Code a condition for NGOs to receive funding.

A full discussion of the Code of Conduct is beyond the scope of this Article, but a few comments are relevant. The Code is as clear a statement of the right to receive aid as is possible. People in need are not simply the beneficiaries of charity, dependent on the good will of those better off, but also have the right to assistance; the provision of humanitarian relief is not a choice; it is a duty. Furthermore, the Code insists that those who can provide aid should do so on the basis of need and not on the basis of national or regional security interests or other reasons of self-interest. Nevertheless, despite this clear expression of what humanitarian assistance should be, many relief efforts are financed by national governments, not NGOs, and the motives of the donors do not necessarily conform to the principles enunciated in the Code.30

In fact, by virtue of accepting funds from donors that have commercial or security interests in the country or region in which an emergency is evolving, NGOs can be said to "act as instruments of government foreign policy" no matter how much they protest this appellation. Indeed, in clearly politicized relief settings such as Afghanistan or Iraq, it is becoming increasingly difficult to separate the supposedly impartial humanitarians from the overtly political government authorities who frequently couch their interventions as primarily "humanitarian." Indeed, secretary of State Colin Powell, in remarks to the National Foreign Policy Conference for Leaders of Nongovernmental Organizations in October 2001, said, much to the embarrassment of those leaders: "I want you to know that I have made it clear to my staff here and to all of our ambassadors around the world that I am serious about making sure we have the best relationship with the NGOs who are such a force multiplier for us, such an important part of our combat team."31 This statement is clearly in contradiction to what NGOs signed in the Code of Conduct.

Building on the ICRC Code of Conduct, more than one hundred NGOs participated, in 1997, in the preparation of what is arguably the most important and most detailed enunciation of the contemporary approach to humanitarian assistance-the Sphere Project. The Sphere Project consists of two parts: a "Humanitarian Charter" and "Minimum Standards in Disaster Response." In a bold attempt to achieve a balance between the spirit of humanitarianism and a respect for human rights, the Sphere Project is based on two core beliefs: "first, that all possible steps should be taken to alleviate human suffering arising out of calamity and conflict, and second, that those affected by disaster have a right to life with dignity and therefore a right to assistance."32

The Humanitarian Charter, the first part of the Handbook, defines the responsibilities of states and parties to guarantee the right to assistance and protection. It is an attempt to summarize the legal principles on which the Sphere Project is based. The Charter draws from the Universal Declaration of Human Rights, the Geneva Conventions (international humanitarian law), the Convention on the Status of Refugees of 1951, and other international documents to establish the framework to which it hopes to hold the humanitarian community accountable. In doing so, it encapsulates the rights-based approach in three legal principles: the right to life with dignity, the distinction between combatants and non-combatants, and the principle of non-refoulement.33

But, in contrast to the Code of Conduct, which explicitly states that it is not about operational details, such as how one should calculate food rations or set up a refugee camp, the Sphere Project sets out to apply a quantitative construct to the rights it promotes. In order to do so, it sets out a number of standards that, taken together, represent the minimum levels to be attained in order for a relief effort to be deemed successful. Finally, it establishes quantifiable indicators of when those standards can be said to have been reached. In other words, it says that people in need of humanitarian assistance have not only a right to life but also a right to life with dignity, and a life with dignity means that, at a minimum, "all people have safe and equitable access to a sufficient quantity of water for drinking, cooking and personal and domestic hygiene."34 It then defines a "sufficient quantity of water" as "at least 15 liters per person per day."35 Standards and indicators are presented in this fashion for water, sanitation, and hygiene promotion; food security, nutrition, and food aid; shelter, settlement, and non-food items; and health services. For the most part, the standards and indicators adopted by the Sphere Project were not developed de novo-they are drawn from previously published recommendations or from reviews of the best practices of experts in each of the fields addressed. The great accomplishment of the Sphere Project has been to draw these technical standards together in an accessible format in such a way that they represent, at least to a limited degree, the "science" of humanitarian assistance while asserting that the science exists to support a human rights-based approach to disaster relief.

The Sphere Project has its shortcomings. It recognizes that the ability to achieve the minimum standards is frequently beyond the control of the humanitarian community. The political context in which an emergency unfolds may restrict the ability of nongovernmental organizations to act; the "humanitarian space" may be limited. Concerns for personal security due to a violent environment may similarly restrict access of the agencies to the affected population.36 Finally, limited financial resources can be a major obstacle to reaching the standards. In the face of any of these or other limiting factors, as discussed above, certain needs of the disaster-affected population need to be prioritized and certain aspects of the relief effort emphasized more than others. The Sphere Project suggests that those offering humanitarian assistance "should strive to meet [the standards] as well as they can."37

Furthermore, by its own admission, the Sphere Project does not address all aspects of humanitarian assistance. No standards pertaining to the physical protection of vulnerable populations, for example, are addressed. Also, no technical guidelines are presented for the education sector. The Sphere Project does not purport to be "all things to all people" and it encourages others to fill in its gaps. Instead, the Sphere Project is an initial effort to define humanitarian assistance as a right of disaster-affected populations and as the duty of those able to provide it. There are more profound objections to the Sphere Project as well. These revolve mostly around what is perceived to be the inflexibility of the standards and indicators that the Sphere Project promotes-particularly that its quantitative requirements stifle the creativity of NGOs and overly constrict the humanitarian response. Some have pointed out that, as was seen in the example of malaria in Burundi, attaining the minimum standards would leave a disasteraffected population in a much better condition than the non-affected surrounding areas, potentially fostering strife between groups.38

Despite the objections, most of which are at least partially valid, the Sphere Project has been quite successful.39 It has certainly raised the level of discourse concerning humanitarian assistance among donors and implementing agencies alike. It has become a centerpiece for training those entering the humanitarian arena for the first time, and it has profoundly influenced the way that relief agencies design and monitor their programs. The "rights-based approach" that it presents has become the norm against which the humanitarian community measures its work. The standards that the Sphere Project has laid down, imprecise and incomplete though they might be, have become the most commonly used yardsticks for determining whether or not a relief effort can be deemed successful.

VI. CONCLUSION

This Article discusses aspects of the evolution of the nature of the humanitarian response to complex emergencies. Beginning with an attempt to define an emergency in epidemiological terms as situations accompanied by elevated rates of morbidity and mortality, it documents the shift from a philanthropic to a rights-based approach to relief. It is important to realize, though, that no matter how organized, how intelligent, or how effective a humanitarian relief effort may be, the humanitarian community always arrives too late. The most important task of humanitarians should not be to minimize death and suffering, but to prevent them. Given the crises that exist in Afghanistan, Iraq, the Democratic Republic of Congo, Chechnya, Somalia, Louisiana, and South Asia as this Article is being written, it is clear that the prevention of the unnecessary loss of life with dignity for millions of people, whether it be due to natural or man-made disasters of slow or sudden onset, has not yet been achieved.