SUMMARY: This article presents a theory of obligation in the context of humanitarianism. Its foundational assumption is that there exists a moral imperative to assist the structurally dispossessed and functionally abused. It builds particularly on the cross-disciplinary work (both academic and applied) of anthropologists, but also of political scientists, sociologists, human rights specialists, and others. The links between human rights and humanitarianism are stressed, while suggesting principles that can guide humanitarian organizations as they serve those in need. Humanitarianism is defined as “crossing a boundary;” risk usually is encountered by the service provider as scarce resources are used to help the vulnerable. Obligation is defined, in part, as “what one should do.” A theory emerges as the “morally possible” and the “materially possible” intersect. Notions of human dignity are shown not to be appropriate in orienting the real-world work of humanitarians; notions of fairness are more appropriate as humanitarian work is organized and implemented. “Pragmatic humanitarianism” occurs as principled guidelines and achievable actions merge, and as non-neutral stances are taken as (for example) refugees are assisted. Humanitarian aid is shown to be fundamentally a moral relationship based on the obligation of “those who have” to address the felt needs of “those who have not.” Examples from Bosnia are featured.


This essay focuses on moral theory at the level of the humanitarian organization. In particular, we offer a theory of obligation as a moral framework suitable for organizing humanitarian assistance. Much of the case material on which the argument rests comes from the work of several persons, including the second author, with refugees and internally displaced persons in Bosnia.

While in one sense theoretic, this article is in an equally important sense pragmatic. It attempts to address difficult questions such as: What drives the humanitarian enterprise? Which organizations should be involved? What are possible negative ramifications of outreach activities which are deemed positive by those employing them? What are the obligations of one human to another? What are the relationships among rights, needs, morals, and obligations?

Anthropologists Address Obligation

Within anthropology, the contemporary discussion of moral obligation in the assistance of others can be traced to articles written by Carolyn Pope Edwards (1985) and Ronald Cohen (1989). In the first instance, Edwards creatively addressed the issue of ethical discourse. Couched within a discussion of ethical relativism, she stressed the importance of understanding and comparing discourses associated with rights and morals. She also was interested in the research methodologies that might be employed in assessing these. One key for us was her cogent statement of the following: ” ‘Ethical discourse’ can be defined as a string of…arguments containing ‘moral statements’ (statements about what actions or attitudes are obligatory or virtuous) and/or ‘ethical statements’ (statements about why those actions or attitudes are morally right or wrong)” (p. 319, emphasis in the original). In the second instance, Cohen (as with most of us since) built his argument upon considerations of human rights and cross-cultural variations in their interpretation. It is “irrelevant … to assert and defend simplistic polarities of relativism versus universal moral imperatives…. The answer to such questions [of what works and what should be done helping others] lies out in the hurly-burly amid the blooming, buzzing confusion of real world experience, where rights or a sense of what is just and fair emerge….”(p. 1016). As for us today, both these authors stressed early-on that the use of empirical, case-based data is essential.

Anthropologists often have addressed the notion of obligation, but rarely have done so in the ways we are suggesting in the present article. Based on ethnographic observation and field study, as well as “arm-chair theorizing,” research conducted in traditional societies and developing nations has focused on obligations based on personal status, verbal contracts, reciprocal relationships, ritual expectations, and shared ideology (see, e.g., Rappaport 1967; Wu 1974; Moore 1995). The topic also has been addressed in developed nations. Referring to community outreach activity in post-World War I Italy, Horn (1988) distinguished among three concepts: charity, social service, and security; the first term linked to a moral obligation consistent with teachings of Catholicism. Angrosino (2001) took a similar tact as he assessed Catholic social policy and its impact on U.S. health care reform during the 1990s. Viewed more reflexively, attention also has been paid to the moral obligations that anthropologists themselves have as they ply their trade in battling such atrocities as racism (Hill 1993) and genocide (Lewin 1993).

Most recently, cogent debates that bear upon moral obligations have been presented in the pages of Anthropology News. Again, the “framing field” has been human rights. Jane Cowan (2006) states that anthropologists do not have an ethical obligation to promote human rights; she worries that a “hegemonic human rights model” crowds out alternative ethical visions. Beware, she stresses, of undemanding morality tales “where victim and violator are clearly distinguished and where others are to blame” (p. 7). Closest to our own stance is that of Laura Graham (2006). Reminding that anthropologists are not neutral, she stresses that members of this discipline who are researching suffering and abuse have an ethical obligation to seek ways to improve the human condition. Those working with vulnerable and marginalized populations have a responsibility to advocate on their behalf, this being justified – and demanded – by virtue of the fact that the important information that they have gathered is not the anthropologists’ alone. Social justice interventions, in turn, are largely based on this knowledge.

American anthropology, with a rich tradition of applied and theoretic study of migration (including refugee issues), has not systematically addressed moral obligation, or attempted to develop a theory of obligation, in the context of humanitarianism, human rights application, and refugee service. Three European scholars of anthropology who recently have addressed obligation in the context of migration are Alessandro Monsutti (2005), Didier Fassin (2005), and Barbara Harrell-Bond (2002). Monsutti touches upon obligation in the context of the support networks used by Hazara refugees from Afghanistan; mutual assistance is paramount. Fassin touches upon obligation in the context of the ways humanitarian assistance has been provided Kurds and other refugees at France’s Sangatte Center; he places his analysis of resource disputes within the framework of a “moral economy.” Harrell-Bond does more than touch upon the topic, and (as will be demonstrated) has inspired part of the analysis we provide.

The scholars, activists, and everyday citizens who have battled genocide, ethnocide, and ethnic cleansing have been among the most outspoken in addressing the notion of obligation. Some have placed the term in counterpoint to terms suggesting more ominous modes of response, such as revenge (Hinton 1998). Haing Ngor, the Academy-Award winning actor (“The Killing Fields,” 1984), told Van Arsdale in 1992 that he had an obligation to assist fellow Cambodians whose families had experienced genocide under the ominous regime of Pol Pot. His obligation was to speak out, to advocate for change, and to bring perpetrators to justice. Ngor’s comment, as much as any other, has inspired us to tackle a theory of obligation.

The Conceptual Framework

Humanitarianism, as we define it, involves “crossing a boundary” to help a person in need. The boundary can be economic, cultural, ethnic, psycho-social, or geopolitical, but a metaphorical “stretch” is mandated. The humanitarian herself may be at risk. Whereas some analysts view “boundary crossing” as negative, we see it as positive; the overall notion of sovereignty still is respected. Humanitarianism suggests an understanding of principles associated with the use of scare resources in which a moral imperative is implied. Risk to both service provider and beneficiary is involved. An understanding of human rights is mandated.

A rich interdisciplinary discourse on the topic of humanitarianism is now underway involving academics, practitioners, and aid recipients; even members of the military are speaking out more openly than ever before. We cannot possibly cover all these diverse – and divergent – perspectives. What we can do is frame key elements of the discourse.

Humanitarianism connotes, but is not to be equated with, human rights. In one sense, the former signifies “what should be done to improve a deteriorating human condition” while the latter signifies “what must be promoted to preserve the human condition at its best.” While rights may be linked to needs, such as sustenance and security, the precise empirical relationship is unclear. As Donnelly (2006) stresses, human rights rest on humankind’s moral nature. They are not grounded on psycho-biological needs but on human possibility. Human rights are an evolving social project, not a cultural given. As the ideals they represent are put into practice, a kind of “self-fulfilling moral prophecy” emerges. In a sense, as he adds, to the degree to which an underlying moral vision of human nature represents achievable possibilities, the implementation of rights makes real the ideal. Thus debates regarding human rights are, in essence, debates over which morals should be universally recognized.

As the remainder of this article suggests, obligation can be defined (in part) as “what one must do” to assist someone in need. It is about humanitarians working to approximate what human rights discourse best suggests.

The Notion of Fairness

To understand the notion of fairness in the context of humanitarianism one must first understand a notion more widely discussed, that of dignity. There has recently been some debate as to the need to ensure the preservation of “human dignity” in humanitarian situations. For example, refugee specialist Barbara Harrell-Bond (2002: 61) argues: “There is an over-reliance on ‘fairness’ as a distributive measure rather than on ‘the inherent dignity of all human persons.'” She pushes for a shift from the language of fairness to one of dignity. Harrell-Bond’s work on the concepts of human dignity and the importance of including local actors as key elements in the service network for refugees have both been of great import to the field. We, however, find this notion of human dignity as an orienting principle in humanitarian assistance problematic. The issue raised here concerns finding the overlap between principled guidelines and achievable actions – pragmatic humanitarianism.

It is necessary, first, to take a hard look at the concept of dignity. For our purposes, we refer to dignity as the intrinsic quality of worth which commands respect, both by the self and by others. It suggests ennoblement in the face of adversity. It suggests the preservation of autonomy. While culturally nuanced, it nonetheless transcends cultural boundaries. In specifying “human dignity,” this quality is claimed as inherent to the human condition; it is to be conferred on the sole basis of one’s existence as a human and one’s opportunity to live autonomously.

The primary problem that arises is that, while it is relatively simple to give this textbook-style definition, the actualization of the concept as humanitarian aid is being contemplated or delivered remains elusive. What exactly constitutes this sense of self-worth depends upon the person in question and the crisis he or she is confronting.

A second, and related, problem is that “human dignity” is difficult to operationalize. If we cannot know the content of dignity, neither can we know specifically how to work towards it. Aid, in and of itself, does not “enhance dignity.” There are some actions that are clearly detrimental to or supportive of human dignity, but there are far more that are context dependent.

For example, one hot afternoon in July, 2000, as the second author was walking to his apartment near Sarajevo’s Grbavica Bridge, a man on crutches approached him. “Do you have a moment?” he asked. “I want to show you something important.” Although a complete stranger, Van Arsdale was intrigued and decided to follow. The two walked across the bridge, past a flea market, and into an enclosure framed on two sides by a dump, on one side by the back of a warehouse, and on one side by a crumbling stone fence abutting a major street. “My home,” he said.

Within blocks of Bosnia’s parliamentary building, yet tucked out of sight, within the enclosure this man had constructed a tiny house (made of cardboard and scraps of wood) and a vegetable garden. His home measured approximately 5 feet by 10 feet, and was sealed with a padlocked door. Inside was a simple bed, a chair, a cooking pan, a plate, silverware, a clock, and one change of clothes. Outside in the garden was a table and another chair. Everything was in its place, neat and tidy.

A lengthy discussion ensued. It turned out the man was a Muslim, internally displaced from the Serbian stronghold of Derventa during the recent war. His son, disabled and reliant upon a wheelchair, was being cared for in Italy. The remainder of his family was scattered. Although skilled, he had been unable to find any work upon his arrival in Sarajevo.

In viewing this scene, and in the subsequent discussions which followed, Van Arsdale had a clear sense of the man’s dignity. It was, in large part, reflected in his autonomy. Yet it was fairness, in the context of his extremely limited resources and the second author’s much larger reservoir of resources, which mandated that aid be rendered and that money be donated.

Effective operationalization of this concept, therefore, requires an intimate knowledge of the individual and his/her society that some humanitarian field workers are unlikely to possess. In effect, while certain actions obviously can be judged appropriate or inappropriate (e.g., regarding physical abuse), there remain a plethora of hidden traps into which an individual not familiar with the context might fall. The humanitarian must have the skills and the opportunity to be able to evaluate cultural context, and within that, socioeconomic circumstances.

A third problem with enshrining human dignity as the distributive goal of humanitarian aid is that some violations of it are inevitable and unavoidable. In the most basic sense, life itself is detrimental to human dignity: people live, they grow old, they die, and this basic process is fraught with indignity in even the best of circumstances. In the context of humanitarian aid work, the very institutionalization of this process can exacerbate indignity.

The example of hospice services in the United States serves to illustrate this point. The goal of such programs, which the second author helped found in Colorado in the 1970s, is stated in some publications as helping terminally ill people to “die with dignity” (e.g., Jaffe and Ehrlich 1997), but dying in fact is a process that tends to rob people of all dignity. Dying in hospice care is humane, but does not somehow “add dignity.” In a Western medical institution death is often a long, drawn out process, and one that is witnessed not only by the sufferer and his/her family, but also by various institution workers whose job it is to monitor, clean, and care for a long line of dying patients. The very intimacy of these job functions results in a situation that is intrusive and destructive of dignity. Dignity is, in part, reliant on individual freedom of choice, having “a little flex” in your relations with others; some types of hospital care take this from patients.

Thus, it is our contention that human dignity is not an appropriate goal to orient humanitarian work. This is not to say that the idea of human dignity is of no use. Rather, that its most proper role is not to be found in the actualization of humanitarian organizational work and outreach. Instead, it should be regarded as an abstract principle, one embedded in the human rights context as we consider it philosophically. In this abstract sense, human dignity is viewed as a good thing, and something which should not be knowingly or casually contravened. The problem is that it is all too easy to do so unwittingly or due to necessity as limited resources are being distributed.

We offer instead the principle of fairness. We believe that it is more suitable to inform the regular tasks undertaken in humanitarian aid work. Fairness, for our purposes, is defined as action and aid in conformation with objective criteria, based on ethical and regulatory standards, of relative merit without regard for personal biases or favoritism. Pragmatically, as will be demonstrated, it bridges both need and service, both circumstance and assistance. Therefore, in order to operationalize fairness, the concept must be set in a specific, situational (and at times regulatory) context as well as a general ethical context. While a theory of obligation, we argue later in this article, provides an adequate ethical structure, fairness as a concept also must take the distinctive aspects of each situation into account in order to maintain a sense of proportionality within the reference group and in relation to available resources. In this sense, fairness is a relative concept that must be measured not only in reference to an ethical position, but also in relation to the pool of people to be served and the resources available for serving them.

Fairness is often confused with two other terms: equality and equity. Fairness is similar to equity; in fact, in discussions of justice the two are often defined in terms of each other (Rawls 1999: 10-15). We have avoided equating the two terms here because of the legalistic connotations that equity sometimes carries. The key is to distinguish between fairness and equality. As stressed above, fairness is a relative principle and is internal to the situational context; equality is not. Instead, equality is characterized by sameness according to an objective, external measure.

The United States has had a recent demonstration of the difference between fairness and equality in the form of responses to Hurricane Katrina. In an interview with newsman Ted Koppel, then-FEMA director Mike Brown expressed surprise that so many New Orleans residents failed to evacuate when given the order. While the order applied equally to all New Orleans residents, the government did not take the varying levels of need into consideration. As a result, the evacuation strategy unfairly impacted the many poor residents in the area, as Koppel pointed out:

Mr. Brown, some of these people are dead. They’re beyond your help. Some of these people that have died because they needed insulin and they couldn’t get it. Some of the people died because they were in hospitals and they couldn’t get the assistance that they needed. You say you were surprised by the fact that so many people didn’t make it out. It’s no surprise to anyone that you had at least 100,000 people in the City of New Orleans where are dirt poor, who don’t have cars, who don’t have access to public transportation, who don’t have any way of getting out of the city simply because somebody says, “you know, there’s a force five storm coming, you ought to get out.” If you didn’t have buses there to get them out, why should it be a surprise to you that they stayed? (Nightline 2005).

Before Hurricane Katrina, the policy of issuing evacuation orders appeared equal: the orders were accessible to all and were given with sufficient time to evacuate the entire city. In retrospect it becomes clear that this resulted in an unfair situation in which those with access to transportation were able to leave the city while those who did not remained, for the most part, stranded. Disparities among people’s access to resources, in this case measured in terms of transportation, correlate with varying levels of need; when all people are treated exactly the same, regardless of real differences in their level of need, “equality” becomes “unfair.” In its simplest form, fairness means recognizing and responding proportionate to people’s differing needs.

For these reasons, we argue that both dignity and equality are unsuitable in orienting humanitarian aid; fairness remains a more reasonable and relevant principle to inform humanitarian action. Fairness emphasizes a practical element, which must be evaluated “on the ground” in the relevant context of human need. But fairness alone is insufficient. It must be placed within an ethical framework which transcends the situational context in order to frame our reaction to, evaluation of, and actions regarding the need for humanitarian assistance. These considerations must necessarily come prior to engaging in humanitarian action.

The Need for a Moral Framework in Humanitarian Organizations

It seems evident that humanitarian organizations should be organized around a moral framework. After all, their purpose is to build a better world and this is inescapably a moral task. But all too often mission statements or statements of purpose come to stand in the place of a conscious and articulated moral foundation at the organizational level. If there is a general consensus among agency personnel that the mission of a humanitarian organization is, in fact, moral it might be argued that developing the moral foundations is superfluous. It is not.

First, there is the obvious problem of determining what missions are or are not morally justified. As Van Arsdale (2006) stresses, addressing refugee human rights is an example of a mission (writ large) that is generally accepted to be moral – but not all humanitarian missions share such widespread acceptance. Are programs to end female genital mutilation morally justified? How about those that teach family planning methods that include birth control and abortion? What about those that teach family planning methods that rely on abstinence in countries with high AIDS infection rates? Most people have strongly held personal beliefs related to these issues; the problem is these personal perspectives can, legitimately, differ. Who gets to decide which arguments are the more persuasive?

To some extent consensus building determines the legitimacy of such missions; this is played out in the field, at the level where actual assistance is rendered. Thus forums for local consensus building (such as town councils) confer local legitimacy, while institutions for international consensus building confer international legitimacy. However, consensus building takes time and is subject to overt and covert political influence at each level. In addition, the consensus reached at one level may conflict with the consensus reached at another level. A cohesive and generally accepted moral framework would help to structure contentious debates and would therefore help make negotiations more likely to come to conclusions that are acceptable at multiple levels and in a more expeditious manner.

At the more immediate and more practical level of the specific organization, moral structure also plays a very important role. Even in the best of circumstances, an organization’s mission is not simply stated, then carried out. There are a variety of intervening steps (often expressed as goals, objectives, and tasks) that serve to turn mission into reality. Planning, organization, and implementation, we argue, are best conducted within a precisely articulated moral frame. A moral framework, in this context, should be constituted in a discourse among key stakeholders that makes clear the distinctions among the different levels of thought that go into transforming an ideal mission into a practical operation.

This is illustrated by what we call the “should, would, could” paradigm. We suggest that the first level of thought that ideally goes into a humanitarian mission is at the “should” level: what should the organization do from a moral perspective? This is the level at which a moral framework needs to be agreed upon and articulated. In Bosnia, World Vision International (WVI) engaged indigenous and expatriate staff in multidisciplinary fora where the psychosocial needs of war-traumatized children were considered. WVI staff believed they should assist, in part because of the organization’s general mission (itself grounded in Christian beliefs regarding assistance) and in part because traumatized children who are not helped remain at great risk. Moral arguments are presented regarding what should be done; they function as the basic assumptions of moral theory in context. This framework then structures subsequent layers of thought; it serves as a guide as to which missions and actions are or are not acceptable.

The second, “would” level of thought spells out what staff would want to do as the different “should-approved options” are weighed. This is an exercise in mission clarification, a chance to express the organization’s preferred strategy under ideal or near-ideal circumstances. The would-level stage often is engaged during organizational retreats and brainstorming sessions. This happened with the WVI initiative. Nonetheless, it must be recognized as an intermediary stage which builds upon the should-level moral structure.

The third level is the “could” level of thought. At this level what the organization should and would do is evaluated in the context of available resources to arrive at those options regarding what the organization could do. When evaluating this, material and practical (including logistical) considerations come to the fore, but it is also important to engage in critical evaluation of the probable and possible consequences, both negative and positive, of those actions. This is the stage at which the should-level morals and the would-level strategies and ideals are brought down to earth; they are situated within the practical environment that the organization must operate in. In the case of WVI, donor-targeted resources allowed outreach to certain rural children but not to urban children in Bosnia. These practical obligations operationalize the moral imperative.

Seen in this context, it becomes more obvious how each level of thought structures the subsequent level thus ensuring that moral and ideal content are not lost or ignored as the humanitarian organization moves to implement its program. This is particularly important today as we grapple with growing recognition that poorly designed humanitarian missions can actually cause harm to humanitarian workers and the intended recipients of aid, and can even be responsible for prolonging conflict situations (Session 2004; Prendergast 1996). Therefore, we believe it essential for a moral framework to structure humanitarian assistance. It is in this spirit that we propose a theory of obligation as an ethical system that blends moral and practical elements in a manner that is particularly appropriate for humanitarian work.

A Theory of Obligation:
Theoretical and Practical Implications

A theory of obligation can be centered around one foundational assumption: that there exists a moral imperative to aidthe structurally dispossessed and functionally abused. Other theories of obligation might emphasize different, more utilitarian approaches. We do not think that ours entertains a highly controversial assumption. Indeed, the moral imperative to assist others has been codified in international human rights laws which have been widely ratified and have garnered global (even if often rhetorical) attention and support (Donnelly 2006).

In turn, a theory of obligation structures our response to the moral imperative to give aid. At its simplest, obligation is found in basic human interactions and expectations. Friendship entails obligation, often accompanied by feelings of gratitude. Gratitude, in turn, is expressed for acts of kindness (Epstein 2006: 69). Such understandings provide us with a framework within which we can operate ethically and effectively not only in everyday interactions but to benefit the dispossessed and abused, and to guarantee their human rights. Thus our theory of obligation has two major components: a moral/ethical element which informs decisions as to which issues are appropriate for humanitarianism and which actions are morally permissible in pursuing them; and a pragmatic element which guides us in evaluating the most effective use of available resources. The overlap between the two components, the morally possible and the materially possible, consists of those actions that we are obligated, as representatives of humankind, to pursue.

Arsedale Figure 1

Figure 1: The intersection of the material and the moral, in the context of obligation

The Morally Possible

The moral component of a theory of obligation can be further broken down into four constituent parts: burden sharing, personal responsibility and institutional accountability, sympathy and compassion, and non-neutrality. Each of these concepts will be discussed in turn.

If we admit to a moral obligation to aid the dispossessed and disadvantaged, we must concurrently admit to the existence of a burden that we carry. In this sense, a burden is the positive load that stimulates response to people in need. It is our responsibility to act, to involve ourselves with people who are in need of our aid and care. Stated differently, we only have an option regarding which type of assistance to offer, not whether we will offer assistance at all.

Burden sharing implies that the burdens should be borne equitably among those who share in the moral obligation to give aid. It is not that each person should shoulder as much of the burden as their emotional, physical, and material abilities allow, but that the amount that each individual bears should be proportional to their own resources and the actual needs to be addressed. The obligation to assist is most appropriately actualized on an individual scale. A theory of obligation is fundamentally concerned with the actions and needs of individuals, both providers and beneficiaries. The second moral principle of this theory of obligation is personal responsibility: the idea that responsibility for the burden devolves onto individuals.

The dictates of effective data collection and comprehension require that we sometimes look at groups of individuals, at what might be termed “aggregate responsibility”. However, this should not shield us from the sometimes extraordinary efforts of individuals. In contrast to the seeming disinterest of the United States as a whole, certain American individuals in the United States come closer to the ideal of equitable burden sharing. Bill and Melinda Gates are examples ( 2005). We speculate that Americans, in particular, may be more willing to engage in humanitarian giving in a non-governmental context. For example, although subsequent donations dropped, the Red Cross received donations in excess of $1.3 billion in the wake of Hurricane Katrina.

The organizational counterpart to personal responsibility is institutional accountability. As aid organizations are the preferred structure for providing aid on the ground, they must be held accountable for policies and their effects, intended or not. Accountability relates to the actual attainment of pre-defined institutional goals. The question of who institutions are accountable to is currently a subject that is much talked about (Kristof 2006). It is our contention that aid organizations, across the board, are primarily accountable to aid recipients, to the people on whose lives they have the greatest impact and for whom their services are intended. This is also, in part, due to the “obligation” inherent in our theory of obligation: if we acknowledge a moral obligation to aid the disadvantaged, then beneficiaries become the “ultimate arbiters” (Smillie and Minear 2004: 18-21). The “obligation” explicit in our theory is clearly an obligation to aid recipients, to assist those disproportionately disadvantaged.

Our notion of obligation is not an abstract or distant concept. Instead, it is grounded in the actual experience and expression of sympathy and compassion. (For those service providers sharing experiences to those being served, empathy also is essential.) Both of these sentiments require us to become engaged on a personal and emotional level with those whom we are aiding. Sympathy is the emotive ability to appreciate the negative experience of another, usually despite the lack of full, explicit communication of circumstances, resulting in a (hopefully deep) awareness of that suffering. Compassion is closely related; it can be defined as a profound awareness and understanding of the suffering of others coupled with a real desire to alleviate that suffering.

Returning to the example of hospice care, we can see how sympathy and compassion are crucial to humanitarian work. Suffering is ubiquitous in hospice situations. The very pervasiveness of suffering can overwhelm hospice workers or inure them to the pan-human impact that suffering has. In this context, sympathy and compassion structure our response and ensure a humane connection between caregivers and recipients. The second author, who has worked extensively with the hospice movement, stresses that:

“…compassion entails the ability to genuinely assist others, but is far more than simply being helpful…. [It] centers on offering care…. It focuses on the provision of psychosocial support to a suffering person…. Most importantly, a compassionate person is one who actively listens to the other person, without treating him or her as ‘the Other.’ In listening, choices are considered” (Van Arsdale 2006: 9).

This personal type of connection by aid workers in relation to aid recipients implies non-neutrality, the final moral principle that orients our theory of obligation. The ability of humans to maintain an impartial objectivity is very much in doubt in any case. In the context of a sympathetic response to human suffering it is also inappropriate, as Prendergast (1996) has repeatedly stressed based on his work in Africa. It is our human capability for subjective relationships that allows us to identify with suffering, to approach the concept of humanitarianism not as a set of abstract principles, but as part of an obligation to every struggling individual. It is because we cannot remain neutral to the suffering of others that we are compelled to actively engage in humanitarian action.

At first glance, it appears that accepting non-neutrality might hinder the equitable distribution of aid. This is not the case. Effective humanitarian organizations, such as Doctors Without Borders, must acknowledge the non-neutral, subjective nature of humans in order to develop the capacity to see aid’s impact and channel it in an appropriate manner. Political circumstances must be evaluated and constraints to service discussed. In order to accomplish this non-neutrality, sympathy and compassion must be accompanied by a strong commitment to fairness, a concept described in depth above.

The Materially Possible

In the same way that we have broken down the moral component of the theory of obligation, we can also break down the material component. There are three constituent parts: pragmatism, felt needs, and the networks of service providers and associated infrastructure. Each is described below.

When evaluating what actions are materially possible, it is of primary importance to recognize the need for pragmatism: the acknowledgment of real-world conditions that constrain actions and impact the results of those actions, coupled with the acceptance of the idea that the value of ideas and actions is found in their real-world consequences. As it is obviously physically impossible to resolve every humanitarian issue at once, setting realistic, achievable goals that can be met “on the ground” is vital (Smillie and Minear 2004: 225-229). Sustainability is key in this respect: to be pragmatic, one must chose initiatives that are able to be supported and sustained over time. Sustainability depends on access to reliable resources, the presence of staff and/or volunteers who are sufficiently committed to the mission, and an “indigenous connective-ness” to the mission in order to address felt needs. Thus workable solutions are context-based.

Because an emphasis on pragmatic humanitarianism implies a need to acknowledge the inability to solve all of the world’s problems at once, there must be some mechanism determining where to apply scarce resources, where to “triage” our activity. Again, the ethical principle of obligation to assist can structure our response. In assessing the environment where assistance is to be delivered we look for the areas where the available resources can be of most assistance. This is what our colleagues working in Bosnia currently are doing. Here the comparison with the medical concept of triage is quite apt: deal with the worst of the fixable problems first.

But how do we determine which are the “most significant problems”? The assessment of the felt needs of intended beneficiaries is a key component of a theory of obligation. Central to this is the assumption that the needs and interests of the beneficiaries are of greater importance than those of the humanitarian organization, its board, its staff, or its donors. Therefore, recognizing and incorporating the voices of intended recipients through an assessment of their felt needs is fundamental to the theory. “Felt needs” are those needs that represent beneficiaries’ worldviews; felt needs are expressed in unfiltered fashion and may not conform to service providers’ visions. They often are not evident to a detached observer. They are exemplified in the needs expressed to the second author in 2004 by a poor woman in the Bosnian village of Brezik. They were not food, water, sanitation, clothing, or shelter, but seed money to buy chicks to start a micro-enterprise of chicken breeders, itself intended to raise enough money to run a women’s cooperative. Who would have known? What outsider could have guessed?

One key characteristic of humanitarian assistance, it is assumed, is to facilitate the empowerment of beneficiaries. Prioritizing felt needs above institutional concerns ensures that the individual remains the central concern of humanitarianism. While this seems to be a basic concept, it is one that is easily lost in the bureaucratization of many large humanitarian organizations. The work of Barbara Harrell-Bond highlights one important reason why the felt needs of refugees, in particular, are often considered secondary to institutional needs. As Prendergast did earlier (1996), in her article “Can Humanitarian Work with Refugees be Humane?” Harrell-Bond argues that humanitarian organizations compete against each other for donor funds and thus for their institutional survival; “the motivation for institutional survival in a competitive environment” (2002: 74) makes humanitarian organizations more sensitive to their responsibilities towards donors than towards beneficiaries:

“…the structure of humanitarian assistance programs is organized in such a way that it makes it almost inevitable that some people will act crassly and sometimes cruelly. Humanitarians, who control the distribution of aid, view themselves as accountable to the donors rather than to the beneficiaries. Because giving assistance is generally regarded as charity, humanitarians also assume the power to decide who is deserving” (2002: 68).

Harrell-Bond advocates moving responsibility, and funding, for refugees away from humanitarian organizations and into the national and local governments of host countries where it could be run in a fashion similar to other social services. However, we would argue, if international aid is still to be the primary source of refugee funding, the issue of external accountability is not necessarily relieved, it is merely shifted. Instead of humanitarian organizations competing for funding, national and local governments would do so.

So long as the supply and demand mindset rules the disbursement of aid from donors and its receipt by NGOs or host governments, such confusion regarding accountability is likely to continue. A shift in theoretical orientation is called for. Instead of viewing the transaction as economic – a transfer of scarce resources to the most efficient humanitarian use – it must be acknowledged that humanitarian aid is, fundamentally, a moral relationship based on the obligation of “those who have” to address the felt needs of “those who have not.”

If a key characteristic of humanitarian assistance is the empowerment of others, as we contend, then aid decisions must incorporate the needs and desires of the intended beneficiaries. This is evident. Today, however, much aid is structured in such a way that the needs and desires of donors are paramount, further disempowering aid recipients as they are forced into disadvantageous situations within the service structure of the humanitarian organizations themselves. We believe that humanitarian organizations are indeed best positioned to offer aid in most circumstances; the challenge is to make their programs flexible and focused on recipients‘ (i.e., intended beneficiaries’) felt needs.

Finally, when implementing programs and policies the network of service providers and associated infrastructure also must be considered. These networks are an integral part of the local context that structures humanitarian action. Verdirame and Harrell-Bond (2005: 8) argue that:

“The social context includes a multiplicity of actors, each standing in a particular relationship of power vis-à-vis the other, which the study assumed it was necessary to unravel. … We made the assumption that studying humanitarian organizations required not only examining the policy documents emanating from ‘headquarters’ but also the practices of regional and local offices down to the actions of individuals working at the [refugee] camp level.”

In many humanitarian situations, and especially in refugee situations, there are a multiplicity of aid organizations, governmental actors, and local actors that form a complex web of service providers and service obstructors. Over 400 were operating in Bosnia as recently as 2000. Unraveling this web requires a great deal of local knowledge including a solid understanding of the motivations of various actors and the likely real-world, pragmatic impact of proposed actions on the local population. It is a vital task. In the context of limited resources and a finite amount of infrastructure, understanding who has the capacity and the willingness to perform which tasks enables greater efficiency in resource distribution. It also allows humanitarians to identify the strengths and weaknesses of the existing networks so that they can be further developed to benefit people at risk.

When the network of service providers works correctly within the bounds of both the morally and materially possible, we are capable of achieving great things.


In a recent conversation with Nobel peace laureate José Ramos-Horta, the second author asked about his work with children in his homeland of East Timor (a.k.a. Timor-Leste). Ramos-Horta noted that humanitarian obligations are met one child at a time, and that a change in one life will be mirrored by changes in others. He believes in the value of humanitarian organizations, but stressed that individual humanitarians must demonstrate their commitments through everyday action.

A theory of obligation, as we conceive it, is useful because it justifies ways in which humanitarians can engage in non-neutral activities that benefit others. As pragmatic humanitarianism is pursued, it ensures that those serving do not treat those being served as “the Other.” It mandates fairness in the accessing of resources and the distributing of assistance. It also affords researchers and policy makers a clear-cut “operational platform” from which to embark on their analyses (and critiques) of these endeavors.

The humanitarian enterprise must be seen as an unfolding process. It must be perceived as issue-oriented and pursued as issue-driven. The institutions which engage this enterprise must not be idealized, nor seen as ends in themselves (Ignatieff 2001; Kennedy 2004). Like the human rights which humanitarianism reflects, it must not be allowed to take on “a life of its own” (Evans 2005). Humanitarianism is not achieved through the establishment of an institution, but through assistance to those in need. We believe that cross-cultural differences must be recognized as need is assessed and assistance is rendered, but that a pan-cultural moral obligation exists to do so.



Angrosino, Michael
2001 Catholic Social Policy and U.S. Health Care Reform: A Relationship Revisited. Medical Anthropology Quarterly 15(3):312-328.

Cohen, Ronald
1989 Human Rights and Cultural Relativism: The Need for a New Approach. American Anthropologist 91(4):1014-1017.

Cowan, Jane K.
2006 An Obligation to “Support Human Rights” Unconditionally is Misguided Moralism. Anthropology News 47(7):7.

Donnelly, Jack
2006. Human Rights. In Oxford Handbook of Political Theory. John Dryzek, Bonnie Honig, and Anne Phillips, eds. Oxford: Oxford University Press.

Edwards, Carolyn Pope
1985 Rationality, Culture, and the Construction of “Ethical Discourse:” A Comparative Perspective. Ethos 13(4):318-339.

Evans, Tony
2005 International Human Rights Law as Power/Knowledge. Human Rights Quarterly 27(3):1046-1068.

Epstein, Joseph
2006 Friendship: An Exposé. Boston: Houghton Mifflin.

Fassin, Didier
2005 Compassion and Repression: The Moral Economy of Immigration Policies in France. Cultural Anthropology 20(3):362-387.

Graham, Laura R.
2006 Anthropologists are Obligated to Promote Human Rights and Social Justice. Anthropology News 47(7):4-5.

Harrell-Bond, Barbara
2002 Can Humanitarian Work with Refugees Be Humane? Human Rights Quarterly 24(1):51-85.

Harrell-Bond, Barbara
1994 Pitch the Tents. The New Republic (September 19-26):18.

Hill, Jonathan D.
1993 Response to Comments by E.N. Anderson and Takami Kuwayama on the Contemporary Issues Forum. American Anthropologist 95(3):707-710.

Hinton, Alexander Laban
1998 A Head for an Eye: Revenge in the Cambodian Genocide. American Ethnologist 25(3):352-377.

Horn, David G.
1988 Welfare, the Social, and the Individual in Interwar Italy. Cultural Anthropology 3(4):395-407.

Ignatieff, Michael
2001 Human Rights as Idolatry. In Michael Ignatieff: Human Rights as Politics and Idolatry. Amy Gutmann, ed. Pp. 53-98. Princeton, NJ: Princeton University Press.

Jaffe, Carolyn and Carol H. Ehrlich
1997 All Kinds of Love: Experiencing Hospice. Amityville, NY: Baywood.

Kennedy, David
2004 The Dark Sides of Virtue: Reassessing International Humanitarianism. Princeton, NJ: Princeton University Press.

Kristof, Nicholas
2006 Aid: Can It Work? New York Review of Books 53(15):41-44.

Lewin, Carroll McC.
1993 Negotiated Selves in the Holocaust. Ethos 21(3):295-318.

Monsutti, Alessandro
2005 War and Migration: Social Networks and Economic Strategies of the Hazaras of Afghanistan. New York: Routledge.

Moore, Erin
1995 Moral Reasoning: An Indian Case Study. Ethos 23(3):286-327.
2005 Bill Gates Pledges $258 Million to Fight Malaria. Accessed 11/13/05.


2005 Transcript on Hurricane Katrina’s Aftermath. The Dallas Morning News. dnnatkattrans.1f527f23.html. Accessed 11/12/05.

Prendergast, John
1996 Frontline Diplomacy: Humanitarian Aid and Conflict in Africa. Boulder, CO: Lynne Rienner Publishers.

Rappaport, Roy A.
1967 Pigs for the Ancestors: Ritual in the Ecology of a New Guinea People. New Haven, CT: Yale University Press.

Rawls, John
1999 A Theory of Justice (revised edition). Cambridge, MA: Belknap/Harvard University Press.

Session, Noni
2004 Review of “The Dark Sides of Virtue: Reassessing International Humanitarianism,” by David Kennedy. PoLAR 27(2):150-153.

Smillie, Ian and Larry Minear
2004 The Charity of Nations: Humanitarian Action in a Calculating World. Bloomfield, CT: Kumerian Press.

Van Arsdale, Peter W.
2006 Forced to Flee: Human Rights and Human Wrongs in Refugee Homelands. Lanham, MD: Lexington Books.

Verdirame, Guglielmo and Barbara Harrell-Bond
2005 Rights in Exile: Janus-Faced Humanitarianism. New York: Berghahn Books.

Wu, David Y.H.
1974 To Kill Three Birds with One Stone: The Rotating Credit Associations of the Papua New Guinea Chinese. American Ethnologist 1(3):565-584.

Comments are closed.