John Robinson's pages on
Luck, duty and benevolence|
How much should a rich person give to prevent a poor person from dying prematurely?
A tangle of political, moral and social possibilities surrounds this question, so I will make it more specific. First I will assume that we live and act in our present world, where diverse political systems interact, economies compete and fluctuate, but there is no immediate possibility of a change in the global order that will repair the damage of extreme poverty. The question of the rich, the poor, and their premature death, is therefore addressed to us as individuals, and the response will be voluntary, not mandated through public policy. Second, to distinguish "rich" from "poor", I will concentrate only on the obligations of richest 20% to the poorest 20%. The World Bank defines absolute poverty as living on less than $1 a day (for definitions and details see their web address http://www-esd.worldbank.org/html/esd/env/publicat/mep/mep13.htm), and a fifth of the world's population is in this state. The next richest fifth -- the moderately poor -- consume between $1 and $2 a day, while the fifth from $2 to $6 a day are average, and the fifth from $6 to $16 a day are moderately rich. Finally, those consuming over $16 a day are extremely rich. The extremely poor band includes most people in sub-Saharan Africa and southeast Asia, and the extremely rich band includes most North Americans (a family of four, each consuming $16 a day, can be supported by a single net annual income of $24,000). I will assume that my readers belong with me in the extremely rich category.
One motivation for concentrating on the obligations of the extremely rich to the extremely poor is the desperate plight of the latter, but I am also concerned, so far as possible, to avoid the complicated issue of desert. Some of the rich may deserve to be better off than others (of the rich, the average, and even, perhaps, some of the poor), because they have worked harder, saved, invested wisely. It is possible to move between income bands, and, arguably, the success of those who move upwards should be granted by virtue of their industry or talent, while those who move downwards may, sometimes, have themselves to blame. However, if we concentrate on the extremely rich and the extremely poor, not only must we be sure that starting life in one or the other circumstance is a contingency, but also the very rare movement from one extreme to the other is very hard to attribute solely to desert. Indeed, such economic mobility is simply impossible for most of the extremely poor, however deserving. It is their bad luck not only to be, but also to remain, extremely poor. This luck cannot be hedged: the extremely poor did not take a gamble on chance, voluntarily accepting the risk; they received their lot as initial bad fortune. The question is whether the lucky have any duty to compensate the unlucky.
One way to think about the question of extreme wealth and poverty is to consider different classes of "luxuries" that the rich may enjoy, and determine what "essentials" could be obtained for the poor with the same money. An example is weekly piano lessons for a wealthy person's child. This "luxury" has the illustrative benefit of being neither essential nor frivolous. At $20 per week for 40 weeks, the rich parent spends $800 per year. A donation of $200 to UNICEF is plausibly the minimum necessary to take a dying two-year-old in one of the poorest countries and provide them with the immediate treatment, plus the continuing health care that will see them through to adulthood. Peter Ungar  has detailed calculations that support this. Now, if I were to make a $800 donation to UNICEF today, my money would go into a large pot with many other contributions, and only some of that would go towards healing children. However, if the pot's effects were divided in a reasonable way between the donors, $200 would probably be enough to claim the saving of a life. This means that I could legitimately say, if I diverted my (hypothetical) $800 from children's piano lessons to UNICEF this year, that four people somewhere will be alive in 20 years time, who otherwise would die very soon. Probably we would consider the preservation of these four lives a good thing, perhaps a better thing than the partial musical education of a single child.
The essentials/luxuries comparison seems to have motivating force, particularly when we realize that we enjoy many lesser luxuries than the children's piano lessons. But it is value-free and irrational. An ethical analysis questions whether donating to (for example) UNICEF is required or even praiseworthy; it seeks reasons for a level of giving based on moral principles, rather than the arbitrary cost of a year's piano lessons, or any other luxury. But we should note that there are considerations that could lead you or me to make a donation, which have no moral basis. You may, for example, feel that a spontaneous $200 gift reinforces your impulsive self-image; you may like the idea of giving to charity for non-moral reasons - or for no reasons at all; you may feel a twinge of pride in benevolence, though this is rather unlikely. (This is an issue to which I'll return.) Moreover, if you feel sentimental, aesthetic, irrational sympathy with the idea that a life will be saved through your donation, then please write your cheque now before you change your mind. It can be made out to UNICEF and send to UNICEF Canada, 443 Mount Pleasant Road, Toronto, Ontario, M4S 2L8. (Note that such a donation would not be the first step on a slippery slope: "If $200, why not $400 - or $10,000? If one life saved, why not 10?". These extrapolations may be valid when we have moral reasons for decisions, but, for now, we are concerned only with one donation, for it's own sake, for fun, or for whatever reason appeals to you.)
The major aid organisations do careful studies of what works in persuading people to give, and we see some of the answers on our TV screens. Sophisticated viewers tend to watch the adverts of organisations like World Vision with cynicism, but this is only because their appeal to sympathy is aimed at maximum yield - motivating the most people in the most effective known way. We might find aid appeals that fixate on the rosy prospect to a tattered third-world child of sponsorship by wealthy north American families tacky, or even indecent, but presumably they work. The point is that moral reasons for giving, such as we are about to explore, are not necessarily or essentially the ones that count in improving people's lot. Your impulsive $200 gift, or the $80 per year it takes to sponsor a child, need not be and probably aren't motivated by an ethical imperative. But that needn't stop us looking for one. Can we find rational reasons based on a compelling system of morality for giving to the relief of starvation? Can we determine how much it is appropriate to give?
There are at least three basic types of ethical argument about obligations from the world's rich to the world's poor. These are utilitarian, rights-based and contractarian arguments. I will consider each of these in turn.
In contemporary applied ethics, the utilitarian point of view is often prominent. This is especially so in consideration of world hunger, relief and development. Peter Singer's 1972 article, "Famine Affluence and Morality" , is perhaps the most anthologized text in applied ethics. Singer is a utilitarian, arguing that we should base our ethics on maximizing the benefits of our actions. It is the consequences of what we do that count, not our motives. Singer argues from two premises. The first, which he claims is uncontroversial, is that dying from hunger is a bad thing. The second premise is that if we can prevent something bad from happening, without sacrificing something of comparable moral worth, then we ought to do so. Although disguised, this is a utilitarian principle and is worked out in utilitarian terms by Singer. We can prevent people from dying by giving money for their relief. Therefore we should give. This is an obligation of justice, not a matter of charity. Singer's conclusion is that the rich should give for the relief of the poor up to the point of marginal utility. That is, my giving should not stop until the suffering I am inflicting on myself and my dependents by directing resources towards the world's poor balances the suffering that the poorest are experiencing. In these terms, paying $20 per week for my child's piano lesson is unjustifiable, since that money could be better used in relieving extreme suffering. Not only should I give up more frivolous luxuries too (as well as many worthwhile ones), but I should also devote every spare moment (formerly used for leisure) to the same ends. Jan Narveson calls this view "The ethics of the hair shirt" , and points out that on Singer's terms, Beethoven should have been helping the peasants he depicted in the Pastoral symphony instead of wasting time on composing tone poems about them.
Singer really does not take the implications of his view far enough. A strict consequentialist would advise us to engage in any strategy we can to alleviate the suffering of the worst off. A more extreme view is explored by Peter Ungar in "Living High and Letting Die" . Ungar accepts the fact of moral relativity, but insists on the reality of ethical absolutes. For him, we can condemn the "Great Virginians", Washington and Jefferson, because they kept slaves and that is an absolute wrong, but we can also understand their conformance to the spirit of their age. In a similar way, we can understand our indifference to the plight of the starving, but there is no doubt that our indifference is an absolute wrong, worthy of condemnation. Ungar's conclusion is an uncompromising obligation to help the dying. For him, we are justified if we steal from the rich to give to the poor; any action (presumably including robbery with violence) is acceptable if the rewards are directed in the right direction and result in overall increased utility. Singer would perhaps temper such advice with some rule utilitarianism that aims at fairness with integrity.
The Singer point of view has been attacked as being unrealistic and idealistic. Ethics, as John Arthur has commented, is for humans not for angels. From an unexpected utilitarian quarter comes the argument that relief for the world's poor today will cause increased suffering tomorrow (Hardin, 1974 ). However, the main problems with the Singer approach are those faced by all pure utilitarian ethics - the impossibility of calculating total utility, and the denial of a role for agency or freedom in our choices. We can never know all the consequences of our actions, and to think that we can do such calculations is not only arrogant but dangerous. Further, reducing our decisions to calculations seems to deny something very important about us as free responsible agents.
Against the consequentialist presumption stand the other major approaches - rights based and contractarian - both of which owe much to the Kantian tradition. A rights-based approach says that the problem with consequentialism can be fixed if we start off by saying that there are certain inviolable human rights, and only when these are guaranteed, can we start talking about maximizing utility. These rights are things I would like guaranteed protected for myself, which I therefore ought to extend as rights to all persons. For example, on utilitarian grounds, perhaps I ought to give up a kidney so that someone can live (indeed, perhaps I ought to be actively seeking out someone to give my spare kidney to!); but a rights-based approach would override this "ought" with a stronger "right". In the Locke tradition, I could be said to own my body, and therefore my kidneys, and property rights (in this tradition) are inviolable. Or I may claim a certain list of rights to myself - life, liberty and security of person, for example - and insist that these be granted to me, and to all humans. Then no-one could be required to give up a kidney so that another person could live. Similarly, the poor cannot hold the rich to ransom because of their relative bad luck. In a rights-based ethics, the problem becomes ordering rights, deciding on precedence between rights, and if possible, reducing a system to a minimum number of "self-evident" or at least humanly compelling fundamental rights.
The most used categorization of rights divides them into negative and positive rights. Negative rights are those which define what other people should not do to us, and therefore what we should not do to others. Positive rights are those which define what people should do (for each other). So, for example, we can say that a right not to be killed is a negative right which imposes a duty on us not to kill other people. But a right to life is a positive right, and if we have this right as humans, then we have a corresponding duty to ensure/enable/foster life for other people. Negative rights are easier to define than positive rights, and some people (for example, libertarians) view them as the only legitimate rights or the only rights that can be enforced .
Perhaps the most persuasive rights-based argument for a strong duty towards the starving is that given by Onora O'Neill in "Lifeboat Earth" . She argues that the right not to be killed applies in hunger/relief situations, as it does for occupants of a lifeboat with sufficient resources to support all its passengers. (The lifeboat metaphor is adapted and extended from Hardin.) The right not to be killed is a very strong right which overrules lesser property rights. The argument is important because O'Neill wants to show that relief of world hunger is a matter of negative rather than positive rights. Having established that the right not to be killed applies, O'Neill does not try to argue that the rich do the killing (though it may be that as beneficiaries of an unjust world economy, they/we are culpable). However she does maintain we have a duty to support the legitimate right of self-defence, and that therefore, we (the rich) are required to intervene to protect the poor from being killed prematurely. This seems to turn a negative right into a positive right, that the poor be helped, and that therefore the rich have a duty to help. Although stated persuasively, O'Neill's self-defence argument looks like sleight of hand. Our normal thinking about self-defence relates to individual acts of danger, and the derivation of a duty to help in these cases may be correct; but applying the same reasoning to the chronic problem of poverty generates a perpetual positive right, and corresponding duty. Even if we allow self-defence to apply to chronic cases, it is still not clear why a derived positive right to the help of others in avoiding being killed should be counted stronger than the negative right not to have property that has been rightfully earned taken away and given to others.
Contractarian approaches to ethics attempt to settle the issue of rights by a thought experiment in which all persons engage in a "social contract" to decide the fundamental bases of justice. The foremost contemporary exponent of contractarianism is Rawls, whose "A Theory of Justice" renewed and envigorated the point of view.
In Rawls' presentation of the social contract , his object is to determine the principles of justice that free and rational persons would agree on if they were all in an initial position of equality. In this hypothetical initial state, everyone is behind a "veil of ignorance" regarding the particular lot or station they will have in life. So the decisions that are made about justice will involve no special interests. Rawls' imaginary initial position can be thought of as dramatizing the idea of universality. We do not know who we will be, and therefore must seek principles of justice that apply impartially and universally. Alternatively, the initial state can be thought of as trying to remove the effects of luck on our judgements. Rawls argues that people in the hypothetical original position would settle on a system of justice that provides liberty and opportunity to all, and ensures that the economic minimum (i.e. the wealth of the worst off) is as high as possible. This is based on the idea that people will adopt a maximin rule, as applied in game theory, to maximize the minimum payoff.
Rawls' informal statement of the principle of justice that follows from his thought experiment is: "All social values - liberty and opportunity, income and wealth, and the bases of self-respect - are to be distributed equally unless an unequal distribution of any, or all, of these values is to everyone's advantage." This is a remarkably compact description of a system of justice which Rawls spends several hundred pages justifying and applying. He develops more formal principles from this "general conception", and is careful to justify the ordering of rights, so that, for example, liberty has priority over economic justice. (Note that this ordering cannot be used to reject economic levelling on libertarian grounds, because the primacy of liberty means that all have access to the most extensive total system of equal liberties compatible with similar liberty for all. If this is not the case, then I cannot argue that my greater liberty is compromised by having to provide enhanced liberty, or sustenance, to the less advantaged.)
Rawls is quite modest in his claims for contractarian theory. He applies it to justice within a single society, and though he hints that it might be extendable to individual ethics, he is more concerned with the political, systemic impact of his theory. However, others have applied contractarian ideas more broadly, in particular to the question of global justice and wealth . In Thomas Pogge's "Realizing Rawls" (1989), for example, everyone in the world is included in the initial position, and global justice is defined as the principles that would be selected from this vantage point. Unsurprisingly, Pogge arrives at a view that involves redistribution of wealth and natural resources, though this is only one thread in his system which includes a redefinition of nationalism, and mechanisms for instituting and maintaining peace.
The chief argument against contractarian ethics is that advanced by Hume in "A Treatise of Human Understanding" as criticism of Hobbes. The hypothetical contract is never an actual contract; therefore the most that contractarian arguments can be is analogies. We can conduct imaginary deal-making sessions, but there is no logical basis for maintaining that these dreams should form the basis for our morality. The matter is even worse for globalized versions of contractarianism, like Pogge's. The contractual argument at least implicitly supposes that our obligations to each other come about because of social intercourse. While this may be indirect, when rich and poor live in the same society, it is conceivable that they could contract together. However, when the rich and poor are parts of different societies, that may be totally isolated from each other, it is harder to see why there should be any contract at all.
The three types of ethical theory so far discussed - utilitarianism, rights-based, and contractarian - belong in one of the two major streams of ethical thought. They all attempt to give guidance on what we should do, which, of course, is the orientation of most work on applied ethics. The other major stream of ethical thought emphasizes how we should be. Beginning with Aristotle, and revisited from time to time by secular philosophers, this stream includes much of religious ethics. In the Christian tradition, for example, the moral emphasis is on being like Jesus and the virtue of love. Recent academic work in virtue ethics and feminist ethics has returned to this emphasis. In Aristotelian terms, the point of ethics is not prescriptions for justice, but development of the character that will allow one to thrive as a human being.
The particular virtue that is relevant when thinking about wealth and poverty is generosity. Aristotle, the Apostle Paul, Thomas Aquinus and David Hume all view a generous disposition as praiseworthy. However, merely being aware of the virtue does not tell us how far it extends. It is assumed by these writers that the rich will remain rich though generous. Even Jesus, who on one occasion advised a rich person to sell everything and give the proceeds to the poor, allowed extravagance, and commented that "The poor you will always have with you". Furthermore, benevolence to people we never meet is not easily understood in terms of virtue ethics. However, as I hope to show in the final part of this paper, the thought experiments of contractarianism can be adopted to the service of virtue ethics, as tools that help us see what generosity means in our world of wealth and poverty.
As we have seen, all the major approaches have difficulties with the question posed at the beginning of this paper. Utilitarianism and the rights-based approach of writers like O'Neill logically lead to an obligation on the rich to give up nearly all they have to relieve the suffering of the poor. It is hard indeed to view this as a legitimate requirement of justice: such benefactors may be saints, but surely the rest of us are not all deficient. This answer seems wrong because it asks too much. A rights-based approach that accepts only negative rights would probably yield the advice that so long as our wealth is not the direct cause of other people's suffering, then we are free to enjoy it as we wish. Indeed, there is nothing wrong with a tycoon lighting cigars with $100 bills: it is the tycoon's own money and no-one else has a right to it. This answer seems wrong because it asks too little. A contractarian rights-based approach seems to push us towards advocating systems that improve the lot of the worst off, but it doesn't directly say what our obligations, as extremely rich people, should be, given that we're stuck in the present system. Virtue ethics suggests that I will thrive as a human being if I am generous, but does not advise how this is to be worked out in terms of saving the starving. Are we then no closer to answering the question?
I believe we can obtain an ethical answer, which is founded not on a moral theory, nor simply on sentiment, but on observation, imagination and action, and draws in insights from most of the views explored earlier.
First, observation. Determining what the true facts of hunger and aid are, is harder than it may seem. Although, for example, UNICEF is willing to provide data about the incidence of hunger, disease and want among children in the poorest countries, on the variety of treatments and aid offered, and on development successes and failures, it does not comment on the political conditions that give rise to poverty. Plenty of other organisations are ready to comment on the political questions, but they may be less informed on the practicalities of saving lives. Everyone has an axe to grind, and filtering the propaganda of global economics is not a skill that most of us have. For example, is Jan Narveson  correct to say, "All of the incidence of substantial starvation (as opposed to the occasional flood) has been due to politics, not agriculture"? His advice: "The cure isn't to have Western countries send over boatloads of western wheat. Even if the local government will let people have this bounty (they often don't - corrupt officials have been known to go out and privately resell the grain elsewhere instead of distributing it to their starving subjects), providing it indiscriminately hooks them on Western charity instead of enabling them to regain the self-sufficiency they enjoyed in earlier times, before modern Western benefits like 'democracy' enabled incompetent local governments to disrupt the food supply." Similarly, the Jubilee 2000 project is working for the cancellation of third-world debt. Much documentation has been gathered (for example at web address http://www.oneworld.org/jubilee2000) to persuade people merely to part with their signatures. Yet Anthony Hovey writes (from Namibia) , "Equating child mortality with the lack of progress on debt relief is naive. In practice, many governments don't spend what resources they have wisely - corruption, prestigious building, massive spending on the armed forces; the list is endless - and reducing the debt burden may well make hardly any difference to child poverty. Debt relief, along with development aid, must go hand in hand with better governance."
What can we say about these conflicting accounts? It is clear that becoming informed is important, but should we not act until we are convinced of all the facts? We could easily become paralysed through want of decisive information. In my view this is the worst of all cases because lack of action leads to boredom and eventual withdrawal from any engagement with the subject. It is better to be partially informed, finding organisations that you can trust, and then acting with vigilance. For this reason, my examples have used UNICEF, which has a very good record, and a willingness to monitor for and correct mistakes. There are many other aid agencies with strong records. I can recommend some to interested readers.
Observation can also help us make some pragmatic decisions. Most ethical theorists put the wealth/poverty question in the domain of justice. This is probably where it belongs, but justice is a societal matter, and its application to personal ethics is murky. If we tell people, as Singer tells us, that we should give so much to the poor because that is demanded by justice, then we condemn their society's system of justice which makes no such demand. We also take from them the opportunity to be benevolent, casting their donation as a duty rather than a freely-willed act. For these reasons, I think the pragmatic approach is to view giving to relief as generosity and therefore praiseworthy. An interesting aside here is the reluctance of our society to honour benevolence. I mentioned earlier that you are unlikely to feel proud of your charity as you donate to UNICEF. Conspicuous acts of giving (such as the donation last year of $1billion by Ted Turner) are often viewed cynically rather than admiringly. Aristotle would be amazed that we have managed to accord a virtue the indifference and even shame that we give benevolence. Perhaps this is because we are suspicious of givers' motives perhaps, even, wary of others' suspicion about our own. On the other hand, moralists from Jesus to Dickens have suggested that the benevolence of the rich can be tantamount to condescension, so perhaps it's not surprising that it has such a bad image. In any case, I wish to leave behind here any idea that we have a duty to give. If there were such a duty, it would require us to specify how much. Instead, we will consider how we should give if we wish to be virtuous, that is, to thrive as humans in our small world.
Let us imagine ourselves in Rawls' world of deal-making thought experiments, but instead of a single initial condition, consider a series of other possibilities. All of these will involve us meeting on equal terms with other persons to make social contracts. I will describe my imagined versions of these various congresses of souls, which I assume take place in a single conference room capable of seating the world's population in disembodied form, where all can speak, all are reasonably polite, and we have only a couple of weeks to make our deals. We may suppose that several million participants will sit the talks out - perhaps because they are bewildered, have nothing to say, or aren't used to being polite. Also, because of failure of imagination, too many of them may sound like ourselves. But this does not really matter; the important thing is that we're free rational agents seeking universal agreement.
The first case is where we imagine that everyone is plucked from their present status and situation in the world. At the end of the congress, we will all be shuffled and our souls returned to different bodies in random situations. Because of the preponderance of poor people in the world, the chances are that most of us are going to be poor when we return to the world. But it's purely a matter of luck. The question is, can we make a deal now that we can stick to when we're back in reality?
There will of course be many debates as we fumble towards to the sort of wisdom Rawls laid out in "A Theory of Justice". Some will be delighted at the opportunity to start afresh with a new political order; some will be skeptical about the ability of the newly rich to keep their side of any deal; there will be arguments over methods and timescales; but the clock will tick and we will gradually understand a little more about conditions in every part of the world that will help us develop a contract.
My guess is that the initial congress of souls would establish agreement on immediate redistribution of resources. The possibility that I could be starving in a fortnight is a great motivator. There would also, doubtless, be much accord on removing the reasons for conflict and war. The problem that would have to be faced, though, is how to enforce the deal. Some of the more optimistic souls would have no doubts, but the pessimists would point to the difficulty of getting the rich to support the poor in the old world, and the new rich are just as human as the old ones were. Perhaps the prudent solution to this would be to agree on a redistribution that preserves the ordering of wealth. So the richest person in the world will remain the richest; they will just have less in absolute terms. One way to do this might be to collect a tithe from everyone (a tenth of each person's wealth) and redistribute it equally. This seems like a very modest solution which does not greatly inconvenience the extremely rich, but significantly benefits the extremely poor. The more egalitarian members of the congress may count this as disappointing and paltry, but in terms of improving on the old world, it is spectacular.
The second imaginary form of the congress of souls is like the first, but is a repeated event. Everyone knows that the congress will meet every five years, and each time we will all be assigned new bodies and places in the world. I mention this case mainly as a stepping stone to later cases, but also for interest. Exploring its ramifications in terms of the motivations to produce wealth within one's current incarnation could be interesting in comparing capitalism and socialism, but this is beyond the realm of this paper. One thing we may suppose though is that with five-yearly cycles, contracts are more likely to be kept: the rich in the current cycle have a continuing interest in the status of the poor, for that might be their lot next time around.
The first two cases are similar to Rawls' argument in that they put everyone on equal terms behind a veil of ignorance. Now we consider a case that is more fanciful and therefore further from the rationality of Rawls, and closer to imagination for the purpose of developing sympathy. In the next case, assume that only 10% of the population of the world are chosen to participate in the congresses. These congresses will take place every five years, and the same persons will always be involved: they are trapped in the recycling system. The original participants are taken evenly from all societies and all social strata. You and I are among those chosen from the extremely rich. At least, that is what we were in the old world. We now have two weeks to negotiate with our fellow congresspersons about what we will do in the new world.
The talks this time will likely have an air of excitement (among the old poor), urgency and panic (among the old rich), because the problem to be solved is rather different. When we return to the world, 90% of the people there are going to be acting just as they always have. Only a few of us in the congress are going to be among the rich, and though a few of those will have economic and political power, they will be heavily outnumbered if they propose any global structural change. The majority of us are going to be poor, living among the already poor, with the additional despair that if everyone had been involved in a congress of the first type, our lot would probably begin to improve. Because we were unlucky enough to be in the 10% congress, we can't expect things to change very much. Such will be the despair at the start of the congress.
But then the more determined participants will start looking for strategies whereby the new rich, though only a small fraction of the rich population, will be able to make an impact. One can imagine the possibilities on the flipchart at the front of the congress.
The possibilities would be debated, and the strategies assessed. My guess is that a mixed strategy will be favoured, with some of the new rich going to Singerian extremes and others being generous, but ensuring they have enough to conduct (or at least participate in) political campaigns. (Incidentally, a very few of the new rich will wield considerable power in poor countries, and their opportunities for action will be considerable.) So far as the ideas about stealing etc., are concerned, my hope is that the congress will include sufficient deontologists (including members of the major religions as well as the odd Kantian philosopher) to rule out blanket consequentialist tactics like theft.
We can be sure that the third type of congress will be very focused on possible solutions and finding a workable agreement between the parties. Almost certainly a heavy expectation will be put on the new rich: not only are they proportionally few, but also some of them may be expected to break ranks and merely indulge their five years of luxury.
The third type of congress has many analogies to the situation we are actually in. We wish to act morally, and may be persuaded that a contractarian approach is as good as any, but we will be in a minority, not a system-changing totality. The congress of souls is now an imagination pump that can fuel our ideas, give us a sense of the urgency of the needy and stimulate us to learn more. It is a matter for our own imaginations as to how far we're persuaded of the need to respond.
The fourth and final congress of souls takes us closer once again to reality. Completely removed from the Rawlsian framework, it now has only sympathic value. Like the third type, the congress involves only 10% of the population; also, it meets every five years. The difference is that after each congress, each soul returns to its former body. There is no shuffling of stations; we merely meet then go back to our old lives. We're all aware of the possibilities of the other types of congress, but the rich participants smile with relief that because they're in the fourth type, they will be returning home after the meeting.
Yet the conversation at the fourth congress is on the same matters as the other three. Now the poor have no leverage with the rich. They can only present their cases, show facts, offer ideas, just as freely as before. But the rich don't need a contract, and the poor would be naive to expect one.
The later meetings of the fourth congress are more important than the first, for at these we would expect both the rich and the poor to report back on how things have changed. The later congresses will reveal what kinds of characters the participants have. Those who sit in stony silence will have decided not to act to help the souls that they meet for two weeks of purgatory every five years. Those who talk, share and go on learning, and more important go on acting while they are away, will have accepted the situation and the need to respond. Thus, in the end, our thought experiment has transformed from a contractarian formalism into a kind of virtue ethics yardstick. We will be able to identify the generous in the final congress of souls, for they will keep trying, giving and encouraging. In the end, all the congress has done for these people is enlarged the sphere in which their generosity acts. They will undoubtedly be involved in sharing their good fortune, raising money and awareness, and some kind of political action. They will be cheerful givers, and in the quagmire of conflicting moral theories will be happy to run with anything that saves lives and makes people glad to be generous.
So, how much should a rich person give to prevent a poor person from dying prematurely? A virtue ethic does not give an answer --- it merely says enlarge your generosity. But it also rests on knowing yourself, understanding and engaging with your psychology. When I do this I find a character that values rules. Theoretically, perhaps, I accept rules as channels leading to virtue; but psychologically I'm constituted to value rules in themselves. My rational rejection of deontology is coupled with a sentimental attachment to it. So I make for myself a law: I tax my good luck. I tell myself that I'm responsible for 80% of the high level (in global terms) of my income. This is a wild overestimate of course, leaving only 20% down to the luck of living when and where I do. But it does my self esteem good to believe I'm responsible for so much. Then I take half my luck income and give it away. Very generous you may think, but once I've admitted it's luck, it's only like buying a round when you've won the raffle. And yet the act also has a kind of nobility: redistributing luck is dealing a humane blow to fate's cruel favouritism. So I end up diverting 10% of my total income to development agencies. The tithe has advocates from Leviticus to Peter Singer, but because I'm not a real deontologist I don't think this act has anything to do with an imperative of duty, and because I'm not a utilitarian, I don't have to say ask why 10% instead of more. At least I know that my reason for a tenth is based on fiction. However I do recommend the practice. I recommend it in the way I would a matter of taste: "If you want a good laugh you should see The Wrong Trousers", "If you want good gardening advice you should go to Highland Nurseries", "If you want to make a personal moral response to poverty, you should give away 10% of your income". Stronger imperatives don't come from a virtue ethic. But, yes, you should try it.
 Peter Ungar, "Living High and Letting Die", Oxford University Press, 1996.
 Peter Singer, "Famine, Affluence and Morality", Philosophy and Public Affairs, Vol 1 No 3, (Spring 1972), anthologized in C Sommers, F Sommers (eds) "Vice and Virtue in Everyday Life", Harcourt Brace College Publishers, Fort Worth, Third edition, 1993.
 Jan Narveson, "Moral Matters", Broadview Press, Peterborough, Ontario, 1993.
 Garrett Hardin, "Lifeboat Ethics: The Case against helping the poor", Psychology Today, September 1974, anthologized in C Sommers, F Sommers (eds) "Vice and Virtue in Everyday Life", Harcourt Brace College Publishers, Fort Worth, Third edition, 1993.
 John Hospers, "The Libertarian Manifesto" (abridged from "What Libertarianism Is" in The Libertarian Alternative, ed Tibor Machan, Nelson-Hall Inc, 1974) in James P Sterba (ed.) "Morality in Practice", Wadsworth Publishing Company, Belmont, California, Fourth edition, 1994.
 Onora O'Neill, "Lifeboat Earth", Philosophy and Public Affairs, Vol 4 No 3, (Spring 1975), anthologized in Jan Narveson (ed), "Moral Issues", Oxford University Press, Toronto, 1983.
 John Rawls, "A Social Contract Persective" (abridged from A Theory of Justice, Harvard University Press, 1971), in James P Sterba (ed.) "Morality in Practice", Wadsworth Publishing Company, Belmont, California, Fourth edition, 1994.
 Paul B Thompson, "The ethics of aid and trade", Cambridge University Press, Cambridge, 1992.
 Anthony Hovey, Letter to the Guardian Weekly, 26 July 1998.