Children are the real victims of world hunger: at least 70% of the malnourished people of the world are children. By best estimates forty thousand children a day die of starvation (FAO 1989: 5). Children do not have the ability to forage for themselves, and their nutritional needs are exceptionally high. Hence, they are unable to survive for long on their own, especially in lean times. Moreover, they are especially susceptible to diseases and conditions which are the staple of undernourished people: simple infections and simple diarrhea (UNICEF 1993: 22). Unless others provide adequate food, water, and care, children will suffer and die (WHO 1974: 677, 679). This fact must frame any moral discussions of the problem.

And so it does -- at least pre-philosophically. When most of us first see pictures of seriously undernourished children, we want to help them, we have a sense of responsibility to them, we feel sympathy toward them (Hume 1978: 368-71). Even those who think we needn't or shouldn't help the starving take this initial response seriously: they go to great pains to show that this sympathetic response should be constrained. They typically claim that assisting the hungry will demand too much of us, or that assistance would be useless and probably detrimental. The efforts of objectors to undermine this natural sympathetic reaction would be pointless unless they saw its psychological force.

We want to explain and bolster this sympathetic reaction -- this conviction that those of us in a position to help are responsible to the malnourished and starving children of the world. We contend that we have this responsibility to starving children unless there are compelling reasons which show that this sympathetic reaction is morally inappropriate (Ibid.: 582). This requires, among other things, that we seek some "steady and general point of view" from which to rebut standard attempts to explain away this instinctive sympathetic response. By showing that assistance is neither too demanding nor futile, we think more people will be more inclined to act upon that pre-philosophical sense of responsibility. And, by philosophically championing that sense of responsibility, we will make most people feel more justified in so acting.

Vulnerability and innocence
Our initial sense of responsibility to the starving and malnourished children of the world is intricately tied to their being paradigmatically vulnerable and innocent. They are paradigmatically vulnerable because they do not have the wherewithal to care for themselves; they must rely on others to care for them. All children are directly dependent on their parents or guardians, while children whose parents cannot provide them food -- either because of famine or economic arrangements -- are also indirectly dependent on others: relief agencies or (their own or foreign) governments. Children are paradigmatically innocent since they are neither causally nor morally responsible for their plight. They did not cause drought, parched land, soil erosion, and over-population; nor are they responsible for social, political, and economic arrangements which make it more difficult for their parents to obtain food. If anyone were ever an innocent victim, the children who suffer and die from hunger are.

Infants are especially vulnerable. They temporarily lack the capacities which would empower them to acquire the necessities of life. Thus, they are completely dependent on others for sustenance. This partly explains our urge to help infants in need. James Q. Wilson claims that our instinctive reaction to the cry of a newborn child is demonstrated quite early in life.

As early as ten months of age, toddlers react visibly to signs of distress in others, often becoming agitated; when they are one and a half years old they seek to do something to alleviate the other's distress; by the time they are two years old they verbally sympathize ... and look for help (Wilson 1993: 139-40).

Although this response may be partly explained by early training, available evidence suggests that humans have an "innate sensitivity to the feelings of others" (Wilson 1993: 140). Indeed, Hans Jonas claims the parent-child relationship is the "archetype of responsibility," where the cry of the newborn baby is an ontic imperative "in which the plain factual `is' evidently coincides with an `ought'" (1983: 30).

This urge to respond to the infant in need is, we think, the appropriate starting point for discussion. But we should also explain how this natural response generates or is somehow connected to moral responsibility.

The purpose of morality
The focus of everyday moral discussion about world hunger is on the children who are its victims. Yet the centrality of children is often lost in more abstract debates about rights, obligations, duties, development, and governmental sovereignty. We do not want to belittle either the cogency or the conclusions of those arguments. Rather, we propose a different way of conceptualizing this problem. Although it may be intellectually satisfying to determine whether children have a right to be fed or whether we have an obligation to assist them, if those arguments do not move us to action, then it is of little use -- at least to the children in need. So we are especially interested in philosophical arguments which are more likely to motivate people to act. We think arguments which keep the spotlight on starving children are more likely to have that effect.

Moreover, by thinking about hunger in these ways we can better understand and respond to those who claim we have no obligation to assist the starving. For we suspect that when all the rhetoric of rights, obligations, and population control are swept away, what most objectors fear is that asking people to assist the starving and undernourished is to ask too much. Morality or no, people are unlikely to act in ways they think require them to substantially sacrifice their personal interests. Thus, as long as most people think helping others demands too much, they are unlikely to provide help.

John Arthur's critique of Peter Singer (both essays reprinted here), highlights just this concern. Arthur objects to moral rules which require people to abandon important things to which they have a right.

Rights or entitlements to things that are our own reflect important facts about people. Each of us has only one life and it is uniquely valuable to each of us. Your choices do not constitute my life, nor do mine yours.....It seems, then, that in determining whether to give aid to starving persons...[agents must assign] special weight to their own interests (1977: 43).

Thus, people need not assist others if it requires abandoning something of substantial moral significance. Since what we mean by "substantial moral significance" has an ineliminable subjective element (Ibid.: 47), some individuals may conclude that sending any money to feed the starving children would be to ask too much of them. Arthur thereby captures a significant element of most people's worries about assisting the needy. The concern for our own projects and interests is thought to justify completely repressing, or at least constraining, our natural sympathies for children in need.

At bottom, we suspect that what is at issue is the proper conception and scope of morality. Some philosophers have argued that morality should not be exceedingly demanding; indeed, one of the stock criticisms of utilitarianism is that it is far too demanding. On the other hand, some theorists, including more than a few utilitarians, have bitten the proverbial bullet and claimed that morality is indeed demanding, and that its demandingness in no way counts against it's cogency (Parfit 1984; Kegan 1988; Cullity, this volume). On the former view, morality should set expectations which all but the most weak-willed and self-centered person can satisfy; on the latter view, morality makes demands which are beyond the reach of most, if not all, of us.

We wish to take the middle ground and suggest that morality is a delicate balancing act between Milquetoast expectations which merely sanctify what people already do, and expectations which are excessively demanding and, thus, are psychologically impossible -- or at least highly improbable. Our view is that the purpose of morality is not to establish an edifice which people fear, but to set expectations which are likely to improve us, and -- more relevant to the current issue -- to improve the lot of those we might assist. Morality would thus be like any goal which enables us to grow and mature: they must be within reach, yet not easily reachable (LaFollette 1989: 503-6). Of course, what is within reach changes over time; and what is psychologically probable depends, in no small measure, on our beliefs about what is morally expected of us. So by expecting ourselves to do more and to be more than we currently do and are, we effectively stimulate ourselves to grow and improve. But all that is part of the balancing act of which we speak.

Thus, we frame the moral question in the following way: what should responsible people do? Our initial sympathetic response is to help the starving children. Are there any compelling reasons to think our compassion should, from some "steady and general point of view," be squelched? We think the answer is "No." Are there additional reasons which bolster this initial reaction? We think the answer is "Yes." In short, we think our initial conviction that we are responsible to malnourished children is not only undefeated, it is also rationally justified.

Moral Responsibility
We "instinctively" respond to the needs of starving and malnourished children. But are we, in fact, morally responsible for their plight? There are, of course, two different questions intermingled here: 1) Are we causally responsible for their condition -- did we, individually or collectively, cause their hunger or create the environment which made their hunger and malnourishment more likely? 2) Are we morally responsible to these children, whether or not we are causally responsible for the conditions which make them hungry?

It is a commonplace of moral argument that people are morally responsible to those to whom they cause harm. If I run a stoplight and hit your auto, then I must pay any medical bills and either repair or replace your auto. If I trip you, causing you to break your arm, then I am expected to carry any resulting financial burden. The principle here is that we should respond to those whose cry for help results from our actions. If others are contributing causes to the harm, we may be jointly responsible to you (Hart and Honore 1959: 188-229). Or, if my action was itself caused by the actions of some other agent -- e,g., if someone shoved me into you -- then this other person is both causally and morally responsible for the harm. But, barring such conditions, a person is morally responsible for harms he or she causes.

Some commentators have argued that the affluent nations, especially colonial powers, are morally responsible to the starving because they created the conditions which make world-wide starvation possible, and perhaps inevitable (O'Neill 1993: 263-4). We find such claims plausible. But, such claims, although plausible, are contentious. Hence, for purposes of argument, we will assume that we in affluent nations are in no way causally responsible for the plight of the starving. If we can show we are (morally) responsible to the children, even if we are not (causally) responsible for their plight, then our responsibility to them will be all the stronger if, as we suspect, these causal claims are true.

Shared responsibility
If we are the cause of harm, then we are responsible to the "victim" because we are responsible for their condition. For instance, we assume biological parents have some responsibility to children because they were responsible for bringing them into the world. However, being the cause of harm is not the only condition which creates a responsibility to someone. We are also responsible to those whom we have explicitly agreed or promised to help. For instance, by assuming a job as a lifeguard, I have agreed to care for those who swim at my beach or pool, even if they, through lack of care or foresight, put themselves into jeopardy.

More important for the current argument, responsibilities also arise from actions which, although not explicit agreements, nonetheless create reasonable expectations of care. For example, although some of the parents responsibilities to their children is explained by their being the cause of the children's existence, this clearly does explain the full range of parental responsibilities. For even when an agent is indisputably responsible for the harm to another, we would never think the agent is obliged to change the "victim's" soiled pants, to hold her at night when she is sick, or to listen patiently as her recounts her afternoon's activities. Yet we do expect this -- and much more -- of parents.

Our ordinary understanding of parental responsibilities makes no attempt to ground specific responsibilities to the child on any causal claims about the parents' responsibility for the child's condition. Rather, this understanding focusses on the needs of the child, and the fact that the parents are in the best position to respond to those needs. This is exactly where the focus should be.

Although for any number of reasons these responsibilities typically fall to the child's biological parents, the responsibilities are not limited to the parents. Others of us (individually or collectively) have a responsibility to care for children whose parents die or abandon them. It matters not that we neither brought these children into the world nor did we voluntarily agree to care for them. Rather, as responsible people we should care for children in need, especially since they are paradigmatically vulnerable and innocent. This is our natural sympathetic reaction. "No quality of human nature is more remarkable, both in itself and in its consequences, than the propensity we have to sympathize with others" (Hume 1978: 316).

This helps explain our shared moral responsibility to care for children who are not being cared for by their parents. Since the range of parental responsibilities cannot be explained either by the parents' being the cause of the child's existence or by their explicitly agreeing to care for the child, it should not be surprising that our shared responsibility likewise does not depend on an explicit agreement or an implicit assumption of responsibility. We assume responsible people will, in fact, care for abandoned children. This shared responsibility springs from our common vulnerability, and from our ability to respond to others who are similarity situated.

Acute Need
Until now we have spoken as if all starvation and malnutrition were created equal. They are not. The hunger with which we are most familiar -- the hunger whose images often appear on our television sets -- is hunger caused by famine. And famines tend to be episodic; often they are unpredictable. An extended drought or a devastating flood may destroy crops in a region, so that the people of that region can no longer feed themselves. (Or, as is more often the case, these environmental catastrophes may not destroy all crops, but primarily that portion of the crop which is used to feed the local population; crops used for export may be protected in some way.) In these cases the problem may emerge quickly and, with some assistance, may disappear quickly. Such need is acute.

The nature of our responsibility to the starving arguably depends on the nature of their need. Peter Singer offers a vivid example of acute need and claims his example shows we have a serious moral obligation to relieve world starvation.

If I am walking past a shallow pond and see a child drowning in it, I ought to wade in and pull the child out. That will mean getting my clothes muddy, but this is insignificant when the death of the child would presumably be a very bad thing (1971: 231)

This case, Singer claims, illustrates the intuitive appeal of the following moral principle: "if it is in our power to prevent something bad from happening, without thereby sacrificing something of comparable moral importance, we ought, morally, to do it." In the case in question, this is sage moral advice. If muddying my clothes saves the life of an innocent child, then it is time for me to send the cleaners some additional business.

Singer's example vividly illustrates our fundamental moral responsibility to meet acute need, especially the acute need of children -- those who are paradigmatically vulnerable and innocent. In Singer's example, the child is in immediate danger; with relatively little effort we can remove her from danger. As we argued earlier, we have a shared moral responsibility which arises from our common vulnerability. None of us has complete control over our lives. All of us are vulnerable to circumstances beyond our control: floods, hurricanes, droughts, etc. Through no fault of our own, our lives and welfare may be jeopardized. Admittedly some acute need results from our ignorance or stupidity. Even so, others should assist us when feasible, at least if the cost to them is slight. After all, even the most careful person occasionally makes mistakes. When need is caused by natural disaster or personal error, we each want others to come to our aid. Indeed, we think they should come to our aid. If, upon reflection, our desire for assistance is reasonable when we are in need, then, by extension, we should acknowledge that we should help others in similar need. Shared responsibility and sympathy conspire to create the sense that we should go to the aid of those who cannot alleviate their own acute needs.

Although we are here emphasizing responsibility rather than justice (narrowly defined), it is noteworthy that the conditions which generate responsibility to help others in acute need resemble the conditions Hume cites as generating our sense of justice: "... 'tis only from selfishness and confin'd generosity of man, along with the scanty provision nature has made for his wants, that justice derives its origin" (1978: 495; emphasis his). Our common vulnerability to circumstances and to the "scanty provision nature has made" leads us to seek ways to protect ourselves against misfortune and error. Natural disasters occur. They may occur where I live; they may not. Prudent people will recognize that we are all more secure, and thus, better off, if we recognize a shared responsibility to assist others in acute need.

As we have suggested throughout this essay, this responsibility is all the more apparent when those in need cannot care for themselves and are in no way responsible for their plight. In short, the responsibility is greatest (and less contentious) when children are the victims. In fact, when children are in acute need, especially when many are in a position to help, there's little moral difference between the responsibility of biological parents and others. If a child is drowning, then even if the parents (or some third party) tossed the child into the pond (and are thus singularly responsible for the child's plight), we should still rescue her if we can. Likewise, if a child is starving, and her need is acute, then even if the child's parents and its government have acted irresponsibly, we should still feed the child if we can.

Arguably the problem is different if the acute need is so substantial and so widespread as to require us to make considerable sacrifices to help those in need. In this case our responsibilities to the children in acute need may resemble our responsibilities to children in chronic need.

Chronic need
Acute need arises once (or at least relatively infrequently). It requires immediate action, which, if successful, often alleviates the need. But most hunger is not acute, it is chronic. Chronic hunger is the hunger of persistently malnourished children, where the causes of hunger are neither episodic nor easily removed. If the need can be met at all, it can be met only through more substantial, sustained effort, and often only by making numerous (and perhaps fundamental) institutional changes, both within our countries, and the other countries in need of aid.

That is why Singer's case is disanalogous with most world hunger. The drowning child is in acute need. Suppose, however, that Singer's fictional child lives on the edge of a pond where she is relatively unsupervised. We cannot protect this child by simply dirtying our clothes once. Rather, we must camp on the pond's edge, poised to rescue her whenever she falls or slips into the water. However, can we reasonably expect anyone to devote her entire life (or even the next six years) as this child's lifeguard? It is difficult to see how. The expectation seems even less appropriate if there are many children living beside the pond.

Likely the only sensible way to protect the child from harm is to relocate her away from the pond. Or perhaps we could teach her to swim. But are we responsible to make these efforts? Do we have the authority to forcibly relocate the child or to erect an impregnable fence around the pond? Can we require her to take swimming lessons? Can we force her government to make substantial internal economic and political changes? In short, even though we are morally responsible to assist those in acute need (and especially children), we cannot straight-forwardly infer that we must assist those (even children) in chronic need.

For instance, if we try to save a child from famine, we may have reason to think that quick action will yield substantial results. Not so with chronic hunger. Since we are less likely to see the fruits of our efforts and, we may be less motivated to assist. Moreover, some have argued that we can alleviate chronic need only if we exert enormous effort, over a long period of time. If so, expecting someone to respond to chronic need arguably burdens her unduly. Responsible people need not spend all their time and resources helping those in chronic need, especially if there is only a small chance of success. This is surely the insight in Arthur's view.

Consider the following analogy which illuminates that insight. Suppose an adult builds a house by the side of a river that floods every few years. After the first flood we may help them, thinking we should respond to someone who appears to be in acute need. However, after the second or third flood, we will feel it is asking too much of us to continue to help. We would probably conclude that this adult has intentionally chosen a risky lifestyle. They have made their own bed; now they must sleep in it. Although this case may well be disanalagous to the plight of starving adults -- since most have little control over the weather, soil erosion, or governmental policy -- nonetheless, many people in affluent nations think it is analogous.

What is indisputable, however, is the case is totally disanalogous to the plight of children. Children did not choose to live in an economically deprived country or in a country with a corrupt government. Nor can they abandon their parents and relocate in a land of plenty, or in a democratic regime. Hence, they are completely innocent -- in no sense did they cause their own predicament. Moreover, they are paradigms of vulnerability.

Since they are the principal victims of chronic malnutrition, it is inappropriate to refuse to help them unless someone can show that assisting them would require an unacceptable sacrifice. That, of course, demands that we draw a line between reasonable and unreasonable sacrifice. We do not know how to draw that line. Perhaps, though, before drawing the line we should ask: if it were our child who was starving, where would we want the line to be drawn?

A dose of reality
Evidence suggests, however, that this whole line of inquiry is beside the point. Although it would be theoretically interesting to determine how to draw the line between reasonable and unreasonable sacrifices, this is not a determination we need make when discussing world hunger. Doomsayers like Garrett Hardin claim we have long-since crossed that line: that feeding starving children requires more than we can reasonably expect even highly responsible people to do; indeed, Hardin claims such assistance is effectively suicide (1974; reprinted here). However, the doomsayers are mistaken. Current efforts to alleviate hunger have been far short of efforts which would require a substantial sacrifice from any of us. Nonetheless, even these relatively measly efforts have made a noticeable dent in the problem of world hunger. And these successes have been achieved with smaller than anticipated growth in population. According to the FAO:

The number of chronically undernourished people in developing countries with populations exceeding 1 million is estimated at 786 million for 1988-90, reflecting a decline from 941 million in 1969-71 and a lowering of their proportion of the population from 36 to 20 percent..." (FAO 1992b: 1)

During the same period, the average number of calories consumed per person per day went from 2430 to 2700 -- more than a 10% increase (FAO 1992b: 3).

Since the relatively meager efforts to assist the starving has made a noticeably dent in the incidence of world hunger, then, although enormous problems clearly remain, we have good reason to think that heightened efforts -- efforts still far short of those requiring substantial sacrifices from the affluent -- could seriously curtail, if not completely eliminate, world starvation. If so, we do not need to decide where the line should be drawn. We are still some distance from that line. Put differently, many of the world's poor are not like the unsupervised child who lives on the side of the lake. Even though their need may be chromic, their needs can be met short of the enormous efforts that would require us to camp next to the pond for the remainder of our days. To that extent, our responsibility to chronically starving children is, despite first appearances, similar to our responsibility to children in acute need.

How to Act Responsibly
Many people are already motivated to help others (and especially children) in need. Indeed, this helps explain the influence and appeal of Singer's essay more than two decades after its publication. Thus, the claim that we have a shared responsibility to meet the needs of others in acute need is psychologically plausible. Even so, it is often difficult to motivate people to respond to others in chronic need. Many in affluent nations feel or fear that aid just won't do anything more than line the pockets of charitable organizations or corrupt governments. Doubtless some money sent for aid does not reach its intended source. But that may simply reflect our inability to determine which relief agencies are most effective. Moreover, even if some aid does not reach those in need, it is even more obvious that most relief aid does reach its desired target. That is what the statistics cited in the last section demonstrate.

We suspect that the strongest barrier to helping those in chronic need is more psychological than philosophical: most people just don't feel any connection with someone starving half-way around the world (or, for that matter, in the ghetto across town). As Hume noted, most of us we do tend to feel more sympathy for what we see than for what we do not see. This at least partly explains why many of us are less willing to help starving children in foreign lands -- we don't see them, and thus, don't feel a tie or connection to them. As we have argued through the paper, this is the core insight in Arthur's view: moral obligations which require us to abandon what is important to us, especially in the absence of some connection with those in need, will rarely be met by many people -- and thus, will make no moral difference. Someone might argue, on more abstract philosophical grounds, that we should not need that link. Perhaps that is true. But, whether we should need to feel this connection, the fact is, most people do need it. And our concern in this paper is how to help meet the needs of the children. Thus, we want to know what will actually motivate people to act.

Of course, just as we should not take our initial sense of responsibility to children as determining our moral obligations, neither should we put too much weight on the unanalyzed notion of "normal ties." Doing so ignores ways in which our moral feelings can be shaped for good and for ill. So perhaps the better question is not whether we have such feelings, but whether we could cultivate them in ourselves and perhaps all humanity, and, if so, whether that would be appropriate. We suspect, though, that many of us cannot develop a sense of shared responsibility for every person in need. More likely we must rely on a more limited sense of shared responsibility; certainly that is not beyond the psychological reach of most of us. Indeed, it is already present in many of us. Thus, working to cultivate this sense of responsibility in ourselves and others would increase the likelihood that we could curtail starvation.

Since people have a natural sympathetic response to the cry of children, the best way to cultivate this connection is to keep people focussed on children as the real victims of starvation and malnutrition. If we keep this fact firmly in the fore of our minds, we are more likely, individually and collectively to feel and act upon this sense of shared responsibility.

But even if we acknowledge this responsibility, how should we meet it? Should we provide food directly? Perhaps sometimes. But this direct approach will not solve chronic starvation. More likely we should empower the children's primary caretakers so they can feed and care for their children. To this extent our shared responsibility to hungry children is mediated by the choices and actions of others. Thus, it might be best conceptualized as akin to (although obviously not exactly like) our responsibility to provide education. Our responsibility is not to ensure that each child receives an education (although we will be bothered if a child "slips through the cracks.") Rather, our responsibility is to establish institutions which make it more likely that all will be educated. By analogy, since it is virtually impossible to feed children directly, our responsibility is not to particular children, but a responsibility to change the circumstances which make starvation likely.

Changing those circumstances might occasionally require that we be a bit heavy-handed. Perhaps such heavy-handedness is unavoidable if we wish to achieve the desired results. OXFAM, for example, provides aid to empower people in lands prone to famine and malnutrition to feed themselves and their children. If the recipients do not use the aid wisely, then OXFAM will be less likely to provide aid again. This is only a bit Draconian, but perhaps not so much as to be morally objectionable.

Conclusion
In both cases of chronic and acute need, we must remember the children who are the real victims of world hunger. The suffering child is paradigmatically vulnerable and innocent. Since we can, without serious damage to our relatively affluent lifestyles, aid these children, we should help. We share a responsibility to them because we are well-placed to help them, and because we can do so without substantially sacrificing our own interests. This is so even if we in no way caused or sustained the conditions which make their hunger likely.

However, if the stronger claim that we caused their starvation (or created the conditions which made their starvation more likely) can be defended -- as we think it probably can -- this responsibility becomes a stronger imperative. Thus, if the views of Sen, Crocker, and Balakrishnan/Narayan (this volume) are correct -- and we suspect they are -- then most of our responsibility is to cease supporting national and international institutions which cause and sustain conditions which make hunger likely. And this responsibility could be explained much more simply as a responsibility to not harm others.1

1 We wish to thank William Aiken, John Hardwig, and Carl Wellman for helpful comments on earlier drafts of this paper.

Brown, L. and 1994: The State of the World 1994: A World Watch Institute Report on Progress Toward a Sustainable Society. New York: W.W. Norton.

------ 1974: In the Human Interest. New York: W.W. Norton.

Food and Agricultural Organization (FAO) 1992a: World Food Supplies and Prevalence of Chronic Undernutrition in Developing Regions as Assessed in 1992. Rome: FAO Press.

------ 1992b: "FAO News Release. Rome: FAO Press.

______ 1989: World Hunger. Rome: FAO Press.

Hardin, G. 1975: Exploring a New Ethics for Survival. New York: Penguin.

1974: "Lifeboat Ethics: The Case Against Helping the Poor." Psychology Today, 8: 38-43, 123-6. Reprinted here.

Hart, H. and Honore, A 1959: Causation in the Law. Oxford: Oxford University Press.

Hume, D. 1978: A Treatise of Human Nature, L.A. Selby-Bigge (ed.). Oxford: Oxford University Press.

Jonas, H. 1984: The Imperative of Responsibility. Chicago: University of Chicago Press.

Kegan, S. 1988: The Limits of Morality. Oxford: Oxford University Press.

LaFollette, H. 1989: "The Truth in Psychological Egoism." In J. Feinberg (ed.), Reason and Responsibility, Belmont, CA: Wadsworth Publishing Co.

May, L. 1996: Socially Responsible Self, forthcoming.

------ 1992: Sharing Responsibility. Chicago: University of Chic ago Press.

Mesarovic, M. and Pestel, E. 1974: Mankind at the Turning Point. New York: Signet Books.

O'Neill, O. 1993: "Ending World Hunger." In T. Regan (ed.), Matters of Life and Death, New York: McGraw Hill.

Parfit, D. 1984: Reasons and Persons. Oxford: Oxford University Press.

Singer, P. 1972: "Famine, Affluence, and Morality." Philosophy and Public Affairs, 1: 229-43. Reprinted here.

United Nations Children's Fund (UNICEF) 1993: The State of the World's Children 1993. Oxford: Oxford University Press.

Wilson, J. 1993: The Moral Sense. New York: The Free Press.

World Health Organization (WHO) 1974: Health Statistics Report. Geneva: World Health Organization.