The Problem of Political Authority   —   Part I: The Illusion of Authority

7. What if There Is No Authority?

  1. Some policy implications
    1. Prostitution and legal moralism
    2. Drugs and paternalism
    3. Rent seeking
    4. Immigration
    5. The protection of individual rights
    6. Taxation and government finance
  2. The case of aid to the poor
    1. Welfare and drowning children
    2. The utility of antipoverty programs
    3. Are poverty programs
      properly targeted?
    4. A clash of analogies: drowning
      children and charity muggings
    5. In case the foregoing is wrong
  3. Implications for agents of the state
  4. Implications for private citizens
    1. In praise of disobedients
    2. On accepting punishment
    3. On violent resistance
    4. In defense of jury nullification
  5. Objections in support of rule-worship
    1. May everyone do as they wish?
    2. Process versus substance
    3. Undermining social order?
    4. The consequences of the doctrine
      of content-independence
  6. A modest libertarian foundation

If there is no authority, does it follow that we ought to abolish all governments? No. The absence of authority means, roughly, that individuals are not obligated to obey the law merely because it is the law and/or that agents of the state are not entitled to coerce others merely because they are agents of the state. There might still be good reasons to obey most laws, and agents of the state might still have adequate reasons for engaging in enough coercive action to maintain a state. If the arguments of the preceding chapters are correct, the circumstances and purposes that would justify coercion on the part of the state are just the circumstances and purposes that would justify coercion on the part of private agents. It remains to be seen whether some organizations are justified in engaging in enough statelike activities to qualify as states. In the terminology of contemporary political philosophy, I have so far defended philosophical anarchism (the view that there are no political obligations), but I have yet to defend political anarchism (the view that government should be abolished).[1]

The aim of the present chapter is to discuss the practical implications of a philosophical but not political anarchism. That is, suppose one accepts the arguments of the preceding chapters, but one believes (contrary to the arguments to follow in later chapters) that government is necessary to a decent society. In that case, what practical conclusions should one draw?

7.1 Some policy implications

If there is no political authority, then the vast majority of laws are unjust, because they deploy coercion against individuals without adequate justification. There are too many laws of this kind to mention them all. Here I briefly touch on some of the more prominent examples.

7.1.1 Prostitution and legal moralism

Moralistic laws prohibit some behavior on the grounds that the behavior is 'immoral', even though it does not harm anyone or violate anyone's rights. The most obvious examples are the laws against prostitution and gambling. How should we view these laws?

Political authority is a special moral status, setting the state above all nonstate agents. If we reject this notion, then we should evaluate state coercion in the same manner as we evaluate coercion by other agents. For any coercive act by the state, we should first ask what reason the state has for exercising coercion in this way. We should then consider whether a private individual or organization would be justified in exercising a similar kind and degree of coercion, with similar effects on the victims, for similar reasons. If the answer is no, then coercion by the state is not justified either.

Consider a story about three private individuals. Jon wants to have sex with Mary. But Mary does not like Jon as much as he likes her. What she does like is money, which Jon happens to have. So Mary tells Jon that she is willing to have sex with him, provided that he gives her $300. That will make it worth it to her. Jon agrees, and they complete the transaction. Later, one of their neighbors, Sam, finds out about what happened. Sam thinks people should only have sex for procreation or sensory pleasure; the thought of people having sex for money makes him angry. So Sam goes over to Mary's house with his gun. He points the gun at Mary and orders her to accompany him to his house. Once there, he locks her in his basement for the next six months.

As it turns out, Mary was not from the neighborhood; Jon had convinced her to travel from out of town to have sex with him. When Sam learns of this, he is incensed. He kidnaps Jon at gunpoint and locks him up in the basement for the next 20 years.

Whatever one thinks of Jon and Mary, Sam's behavior in this story is clearly wrong. Perhaps Mary and Jon are doing something bad (though it is unclear how); if so, it would be appropriate for Sam to explain to them what he sees as problematic about their behavior, in an effort to persuade them to stop. If he cannot persuade them, however, coercion and kidnapping are not appropriate responses.

Sam's behavior in this story is analogous to that of the government in countries where prostitution is illegal. Mary's six- month imprisonment is not unlike what a prostitute might expect to suffer. Admittedly, johns are rarely prosecuted and rarely serve prison time. Jon's 20-year imprisonment is, however, an allusion to actual U.S. federal law, which provides for a prison sentence of up to 20 years for 'enticing' someone to cross state lines for the purpose of prostitution.[2] It is worth pointing out how absurdly punitive some laws are. But the main point is not that the sentences are too high; the main point is that no coercion is justified at all merely to stop a pair of individuals from voluntarily exchanging sex for money.

7.1.2 Drugs and paternalism

Paternalistic laws restrict individuals' behavior for their own good. Certain drugs, for example, are outlawed, mainly because they are harmful to the user. They may damage the user's health or relationships with other people; they may cause the user to lose his job, drop out of school, or otherwise have a less successful life.

Are these adequate reasons for prohibiting drug use? The prohibition of drugs means that users and sellers are subject to coercive threats by the state. Those who are caught are often forced to spend years of their lives in prison. For most readers, being sent to prison would probably be the worst thing they ever experienced. This is of particular concern in the United States, where over half a million people are locked up for drug offenses.[3] To justify the imposition of such a large harm, the reasons for prohibition would have to be very strong.

Consider another story about Sam. Sam is opposed to cigarette smoking due to its severe health harms. Not content merely to avoid cigarettes himself, he issues a proclamation to his community that no one may smoke. After the proclamation, Sam catches you smoking, kidnaps you at gunpoint, and locks you in his basement. You share the basement with thieves, rapists, and murderers for the next year, until you are released. The person who sold you the cigarettes is locked in the basement for the next six years.

Did Sam act rightly? It is hard to imagine anyone saying so. The desire to prevent others from harming their health in this manner hardly seems an adequate justification for coercion and kidnapping, let alone for stealing months or years of someone's life. But Sam's action was no worse than what the government presently does to drug offenders. Tobacco is about seven times more deadly (on average, per user) than illegal drugs, so Sam has much stronger justification for what he does than the government has for what it does.[4]

Some advocates of prohibition stress the harmful nonmedical effects drugs can have on one's life. To take account of this, imagine that Sam also watches out for people in these other respects; when he learns about someone who has damaged her relationships with others without good reason, he kidnaps that person and holds her captive in his basement. Similarly for those who lose their jobs or drop out of school due to their own fault. (Add any other negative life events of the sort that drug abuse might cause.) Sam explicitly warns people against these behaviors, and he only punishes people who knowingly and willfully violate his orders. Sam's reason for punishing these people would be stronger than the reason the state has for punishing drug offenders, since drugs only have a chance of causing one to damage one's relationships, lose one's job, and so on, whereas Sam punishes only people who have in fact knowingly damaged their relationships, lost their jobs, and the like. Yet Sam's behavior seems outrageous. The desire to prevent people from damaging their own lives in these ways does not constitute adequate grounds for coercion.[5]

There are many other paternalistic laws about which similar arguments might be made. In general, paternalism is justified only in extreme circumstances - for instance, if a person is about to throw himself off a bridge, one might be justified in coercively preventing him from doing so, at least long enough to find out why he wants to kill himself and whether he is of sound mind. Coercion is not justified merely because another person wishes to make an unwise choice of the sort that normal people frequently make in ordinary life. Here are some other examples of legal paternalism.

As these examples illustrate, legal paternalism is quite widespread in modern Western society. All of these are unjustified laws.

7.1.3 Rent seeking

Rent seeking is behavior designed to extract wealth from others, especially through the vehicle of the state, without providing compensatory benefits in return.[6] The most straightforward example is a company lobbying the government for subsidies. But many of the policies that exemplify legal paternalism are also motivated partly by rent seeking. Consider the following.

What is the moral status of such laws? Suppose we extend the story of Sam as follows. Sam happens to have a friend named Archer Midland. Archer asks Sam for some financial assistance, so Sam goes out, mugs some people, and gives the money to Archer. Obviously, this action is wrong. The desire to profit at others' expense is not an adequate justification for coercion.

7.1.4 Immigration

Marvin is in need of food, without which he will suffer from malnutrition or starvation.[8] He plans to travel to a nearby marketplace, where he will be able to trade for food. But before he can reach the marketplace, he is accosted by Sam, who does not want Marvin to trade in the marketplace, for two reasons. First, Sam's daughter is going to be shopping in the marketplace, and Sam fears that Marvin might bid up the price of food. Some vendors might even run out of bread if too many people come to the marketplace. Second, Marvin comes from a different culture from most people presently at the marketplace, and Sam fears that Marvin might influence other people and thus alter the culture of the marketplace. Sam decides to solve the problem by force. He points his gun at Marvin and orders Marvin to turn around. The starving Marvin is thus forced to return home empty-handed.

Sam's reasons for coercing Marvin in this story are clearly inadequate. Furthermore, Sam will be culpable for whatever harms Marvin suffers as a result of being unable to reach the marketplace; they will be harms that Sam inflicted upon Marvin. If Marvin starves to death, then Sam will have killed him. This is true even though Sam was not responsible for Marvin's initial situation of being hungry and out of food; it is true because Sam actively prevented Marvin from obtaining more food. If a person is starving, and you refuse to give him food, then you allow him to starve. But if you take the extra step of coercively interfering with his obtaining food from someone else, then you do not merely allow him to starve; you starve him. The same point applies to lesser harms: if, for example, Marvin merely suffers malnutrition as a result of being unable to reach the marketplace, Sam will have inflicted this harm upon him.

The behavior of Sam in the story is analogous to that of the government of any modern country that excludes poor immigrants. Potential immigrants from developing nations come to participate in the marketplaces of wealthier countries. The governments of the wealthier countries routinely forcibly exclude these potential immigrants. As a result, many suffer greatly diminished life prospects. The government does not merely allow harms to befall these would-be immigrants. If the government merely stood by passively and refused to give aid to potential immigrants, then it would be allowing harms to occur. But it does not stand by passively; the government of every wealthy country in the world deliberately hires armed guards to forcibly exclude or expel unwanted persons. This coercive intervention constitutes an active infliction of harm upon them, just as Sam inflicts harm on Marvin in the story above.

The most common reasons given for immigration restriction are twofold. First, that new immigrants compete with existing Americans in the labor market, thus driving down wages for unskilled labor and making it more difficult for American workers to find jobs. Second, that if too many immigrants enter the country, they will alter the country's culture. The first concern is analogous to Sam's concern about Marvin's competing with Sam's daughter in the marketplace. It is not permissible to use force against another person simply to prevent a third party from suffering economic disadvantage through normal marketplace competition. The second concern is analogous to Sam's concern about the culture of the marketplace. It is not permissible to use force against another person simply to prevent that person from influencing the culture of one's society in undesired ways.

7.1.5 The protection of individual rights

Are there any government policies immune to the style of criticism deployed in the preceding subsections?

The policies exempt from my criticisms are generally policies that serve to protect the rights of individuals. For instance, people who wish to commit murder are subject to coercive threats by the state. Murderers who are caught are forced to spend years in confinement. But this is not at all unjust. Individuals have a right not to be murdered, and it is appropriate to defend that right by force.

Why can one not deploy here the same style of argument used in earlier subsections? Imagine that a private individual, Sam, issues a proclamation to his community that no one may murder anyone. One day, Sam discovers that someone has committed a murder. Sam takes the murderer captive at gunpoint and confines the murderer in his basement for a period of years. Has Sam acted rightly?

In this case, unlike the earlier episodes involving Sam, it seems to me that Sam's behavior is permissible, even praiseworthy.

Some are uncomfortable with this sort of vigilante action, for either of two reasons. First, one might worry about Sam's reliability in identifying the guilty. When private vigilantes exact justice, they may misidentify criminals and wind up punishing the innocent. Second, in most societies, Sam's vigilantism would be unnecessary since there are police forces and courts to punish the guilty. These are the most important reasons for opposing vigilante justice in most circumstances.

To put these concerns to rest, let us stipulate that there are no other established mechanisms for dealing with murderers in Sam's society. If Sam does not pursue the murderers, then murderers will have more or less free rein. Assume also that Sam has careful procedures for verifying the guilt of the parties whom he punishes. He has a lengthy process of reviewing evidence, in which the accused is given every opportunity to question the evidence against him and to present evidence in his own favor. The process is careful, reliable, and open to public scrutiny. In this case, I see no objection to Sam's behavior.

Sam's behavior in this last story is analogous to that of a government that pursues murderers, gives them fair and public trials, and imprisons them. There is nothing objectionable in such a practice. The same goes for policies aimed at protecting society from a number of other sorts of criminals, such as thieves, rapists, and other violent criminals.

Asimilar point can be made about military defense. Invaders from a foreign country are simply a large and well-organized group of thieves and murderers, and it is appropriate to use force against them in defense of oneself and one's neighbors.

I shall not attempt here to catalog all the actions that the state may justly use coercion to prevent. Particular kinds of action must be judged using our ordinary ethical intuitions and applying the general principle that it is permissible for the state to prohibit some action if and only if it would be permissible for a private individual to use force to prevent or retaliate for that sort of action, assuming the individual used reliable methods of identifying guilty parties and had no better remedies available.

7.1.6 Taxation and government finance

How may a government finance its activities? The main method now used is coercive extraction of money from the population (taxation). The prevalence of this method of finance is most likely due to the fact that it is a very reliable method of collecting very large amounts of money. But it is not normally permissible to coercively extract money from others, even if you have a very good use for the money. On the face of it, therefore, taxation appears impermissible.

That inference, however, seems to presuppose that individuals are justly entitled, prima facie, to their pretax incomes. Thomas Nagel and Liam Murphy have disputed this assumption. They believe that property rights are created by governmental laws and therefore that one only has property rights in those things to which the state's laws grant one ownership. By creating tax laws, the state shapes the property rights that individuals have such that individuals own only their after-tax incomes.[9]

In response, there are three views one might hold regarding property rights. First, one might hold that property rights are natural, that is, moral rights that exist prior to the state. John Locke, for example, held that individuals are justly entitled to the fruits of their labor, even in a pregovernmental society.[10] On this view, taxation would seem to be a prima facie injustice, for whatever the ethically correct way of acquiring property may be, it presumably is not forcible extraction of goods held by others.

Second, one might hold that property rights are partly natural, in that there are certain broad principles of property that are valid independent of governmental laws, but that there are many details of a regime of property rights that are not settled by these general moral principles. For instance, perhaps our inherent moral rights determine that we are justly entitled to the fruits of our labor, but these rights do not determine at what altitudes one may fly one's airplane over someone else's land. One might hold that state-created laws are needed to settle such matters of detail. This view still offers little comfort to a defender of taxation, for the entitlement of one agent to coercively extract vast quantities of resources from the rest of the population is not the sort of matter of detail (like the altitude at which one may overfly others' property) that is plausibly taken to be left indeterminate by the basic moral principles of property.

Third, one might hold that there are no natural property rights. Nagel and Murphy assume that this means that property rights are created by governmental decree. This is plausible only for one who presupposes a strong doctrine of political authority. Nagel and Murphy ascribe to the state a moral entitlement, arising from its power to create property rights, to coercively enforce its chosen distribution of resources. Since no nongovernmental agent may declare a distribution of resources and a regime of property rights and then coercively enforce them, the state's right to do so would require political legitimacy. At the same time, the state's creation of a regime of property rights would presumably impose obligations on the part of citizens to respect that regime. These would be political obligations. If, therefore, the state has no authority, it has no such power of creating property rights as Murphy and Nagel suppose.

The result would seem to be that even after the state has made its laws, there still are no property rights. (If one finds this conclusion implausible, one ought to return to the view that there are natural property rights.) One might think the rejection of property rights leaves the way open for taxation: since taxpayers have no right to 'their' wealth, the seizure of some of that wealth will no longer appear as a rights violation. But by the same token, the state will have no right to that wealth either, and thus citizens do no wrong by withholding it. Meanwhile, there are the harms the state coercively imposes on those who fail to pay taxes, and these would seem to be prima facie injustices.

In short, the defender of taxation must hold that the state, rather than the taxpayers, is justly entitled to the tax revenues that the state collects. There is no plausible way to defend this view unless one assumes a doctrine of political authority.

How could a government finance its activities without taxation? One alternative is for the state to charge fees for its services. The state might charge for each service it provides or set a single fee to cover all government services. Suppose the state set a single, annual fee for its services. Those who did not pay the fee would be excluded from most government services over the course of the year - for instance, they might be unable to file lawsuits in government courts and be unable to call the government's police to protect them or to investigate crimes against them. Police protection might be provided for buildings and neighborhoods that had paid the appropriate fee, with homeowners associations collecting the funds to pay for a given neighborhood's protection. The state could establish a policy that, if a crime was committed in a building or neighborhood that had not paid the fee for governmental security, then the police and courts would do nothing about it. Provided that the state was reasonably good at its job and its fees were reasonable, most citizens, for obvious reasons, would choose to pay.

Some individuals, when first exposed to this idea, think that the proposal amounts to coercive extraction of funds from citizens, just as surely taxation does, for individuals who did not pay the state's fees would be subject to a serious risk of violence. This is a mistake. Under the present taxation scheme, the state itself inflicts harm on those who fail to pay their taxes. Under the fee-for-service scheme I have proposed, the state fails to protect from harm those who refuse to pay the necessary fee, but it does not itself inflict harm on them. Consider an analogy. Doctors provide medical care for a price. They usually do not provide care to those who do not pay them, but they are not coercing everyone who does not buy their service; if you don't hire a doctor, a doctor will not come and infect you with a disease. The fee-for-service model of government finance is like the system in which doctors provide medical care only to those who hire them. The taxation system is like a system in which doctors give diseases to those who don't hire them.

Because this alternative model of government finance has not been tried, there will undoubtedly be many questions raised about it. I cannot explore the proposal in detail here, but I will briefly mention three obvious issues. One concerns how much money the government could hope to raise through voluntary fees. In fiscal 2010, the U.S. federal government spent approximately $3.7 trillion, or about one quarter of GDP.[11] A voluntary payment scheme may be unable to support such large expenditures. The best solution to this problem would be to drastically reduce government expenditures, consistent with the very limited range of government activities we have found justified.

Another concern is that poor persons might be unable to pay the government's fees and would therefore be left with even less protection than they presently have. However, the state need not charge the same price to all citizens. Differential pricing schemes often occur even in the free market, as in the case of movie theaters that charge lower prices for senior citizens and students. More to the point, those with expensive homes routinely pay more for property insurance than those with inexpensive homes. Along the same lines, the wealthy would be willing to pay more for protection of their persons and property than the poor would or could pay.

Another question is whether the state would be entitled to prohibit nonstate individuals or organizations from selling services similar to those of the state. For instance, could private security companies provide security for people who did not pay the government's fees? If such competition were allowed, many citizens might opt for private security, perhaps to save money or to obtain better service. If enough people behaved in this way, the government could conceivably be driven out of business. In my view, such competition should be allowed, and this provides the key to the political anarchist proposal of later chapters. However, the present chapter is for those who believe that any scheme of competing security provision with no single central authority would be socially disastrous. On that assumption, the state could justly prohibit private provision of security. It is normally wrong to coerce others, but such coercion can be justified when it is necessary to prevent something much worse from happening.

A similar point would apply if it turned out that the fee-for-service model of government finance was for some reason unworkable. If taxation were necessary to prevent a societal catastrophe, then the state would be justified in taxing. However, the fee-for-service model, if workable, has the advantage in terms of justice, since it reduces the amount of government coercion. For this reason, governments ought to at least attempt to implement this model and should resort to taxation only if serious good-faith efforts at voluntary financing fail.

7.2 The case of aid to the poor

7.2.1 Welfare and drowning children

Many government policies serve to redistribute wealth from the rich to the poor. This class of policies looms large in contemporary social theory, overshadowing every other kind of policy in discussions of social justice. I devote the present section to addressing what I consider the strongest argument in favor of wealth redistribution. This is a broadly humanitarian rather than an egalitarian argument - that is, it focuses on the problem that some people's basic needs are unsatisfied rather than on the alleged problem that people have disparate levels of wealth and income.[12]

Imagine that you are passing by a pond where you see a drowning child. If you can save the child at slight cost to yourself, then it would be wrong not to do so. This example is often deployed in the ethics literature to motivate the principle that, if one can prevent something very bad from happening at little cost to oneself, one is obligated to do so. In particular, it is often said that if we have the opportunity to save poor people from suffering from starvation, malnutrition, or other serious harms at small cost to ourselves, we must do so.[13]

But now imagine that for whatever reason, you are unable to save the child in the pond yourself. There is, however, another bystander who could save the child at slight cost to herself. This individual, however, does not care enough about the child to do so voluntarily. The only way to cause the child to be saved is to threaten the bystander with violence unless she saves the child. You do so, and she saves the child. Call this the Drowning Child case. In this case, regrettable as the resort to coercion may be, it seems justified.

This appears to show that it is permissible to coerce others to assist those in distress, provided that they can do so at modest cost and that there is no other way of causing the people in distress to be helped. By analogy, one may argue, the state is justified in using coercion to induce citizens to aid the poor, as in the case of government social welfare programs. In the following subsections, I suggest three objections to this conclusion.

7.2.2 The utility of antipoverty programs

Consider a variation on the Drowning Child case. Call this the Incompetent Bystander case: as before, there is a drowning child whom you are unable to help directly, but you can coerce a reluctant bystander into taking action. This time, however, assume that even if you coerce the bystander into entering the pond to pull the child out, it is unclear whether the child will actually be saved (whether because the child may already be too far gone, because the bystander is incompetent, or for some other reason). Second, assume that there is a fair chance that, on his way to trying to save the drowning child, the bystander will accidentally knock one or more other children into the pond who will then drown. You find it difficult to assess these probabilities, so it is quite unclear to you whether the net expected benefit of forcing the bystander to 'help' is positive or negative. Nevertheless, you cannot stand the thought of doing nothing, and so you whip out your trusty pistol and force the bystander to go after the drowning child.

In this case, you act wrongly. There must be some presumption against coercion. In the scenario as just described, there is no compelling case in favor of getting the bystander to take action, so the presumption against coercion stands. The conclusion is even clearer if the example is specified such that you would be justified in taking the expected benefit of coercing the bystander to be negative (that is, taking the expected harms to outweigh the expected benefits). Government antipoverty programs are justified, then, only if their expected benefits are positive and this fact is reasonably clear (that is, we have strong all-things-considered justification for believing it).[14]

There is a simple and well-known argument for thinking that antipoverty programs are overall beneficial: antipoverty programs redistribute money from wealthier people to poorer people. According to the well-known principle of the diminishing marginal utility of money, a given quantity of money will usually give more benefit to a poorer person than to a wealthier person (the poor need the money more). These redistributive programs should therefore do more good than harm.[15] This theoretical argument has clear prima facie plausibility. It rests on a very widely accepted and plausible economic principle, that of the diminishing marginal utility of wealth.

There are also a number of prima facie plausible arguments for the opposite conclusion. Charles Murray, the most influential critic of government antipoverty programs, argues that these programs create a moral hazard problem.[16] They lower the costs of or create benefits for certain social conditions, such as unemployment and out-of-wedlock pregnancy. This lowers people's aversion to those conditions, leading more people to behave in ways more likely to lead to those conditions. Rather than helping the poor get on their feet, Murray contends, the government programs create a cycle of dependency, making it easier in the short term to engage in behaviors that are self-destructive in the long term. The general thrust of his empirical argument is that, as government antipoverty programs enjoyed enormous increases in funding and scope between the 1960s and 1980, poverty, unemployment, illegitimacy, crime, deficient education, and other social problems persisted: 'In some cases, earlier progress slowed; in other cases mild deterioration accelerated; in a few instances advance turned into retreat.'[17] Other social scientists, however, have strongly disputed Murray's empirical argument.[18]

Other arguments focus on the effects of wealth redistribution on overall economic productivity. One argument often heard in the popular discourse contends that high taxes on the wealthy reduce the incentive for people to be productive. A related, more subtle argument begins from the observation that high-income people tend to invest a much larger proportion of their income than low-income people do. Therefore, redistribution of wealth from high-to low-income persons will reduce a society's total investment rate in favor of near-term consumption. This will reduce a society's rate of economic growth. Changes in the growth rate have exponentially larger total impacts as one compounds them over longer periods of time. So the lower growth rate will make a very large difference to the material wealth of future generations.[19]

Finally, it should be remembered that government programs are not frictionless machines. It may be true that a given dollar would do more good for a poor person than for a wealthy person, but once account is taken of administrative costs and waste, government programs that take a dollar from a wealthy person are unlikely to give the poor person anything close to the whole dollar.

All of these arguments have some validity: each identifies one relevant factor tending to either promote or diminish social welfare. One important factor speaks in favor of government antipoverty programs, while other important factors speak against these programs. My guess is that in the long term, the argument from the rate of investment wins out.

I cannot attempt here to resolve the very complex question of the net effects of government antipoverty programs. On that question, I have nothing of import to add to the existing literature (see the notes). Nevertheless, the foregoing discussion should help to explain why the issue is controversial and why it is at best unclear that these programs are overall beneficial rather than harmful. But given the presumption against coercion, the programs are justified only if it is clear that they have a net positive expected benefit.

something very bad. It need not be clear that the coercive act in fact prevents the bad event; however, it must at least be reasonably clear that the coercive act reduces the risk. If the coercive act creates some other risk, it must also be reasonably clear that the reduction in the original risk outweighs the newly created risk.

It is conceivable that in the future someone will devise government antipoverty programs that have clear net benefits. At that time, coercion might become justified, depending on how large the benefits were, how much coercion was required, and so on. However, bearing in mind that the theoretical arguments for the harmfulness of government antipoverty programs are based on very broad features of those programs, it is also likely that no one will devise programs devoid of these problems. I suspect that the only programs that would genuinely produce large net benefits are politically unfeasible due to widespread biases against foreigners, as suggested in the following subsection.

7.2.3 Are poverty programs properly targeted?

I turn now to another variation on the needy children theme. Call this the Cold Child case: there is a child who is uncomfortable because of the cold night air. She needs a jacket, but you have no jacket to give her. You do, however, have a gun, and you see a bystander nearby who is wearing both a jacket and a sweater. The bystander does not want to give up any of his clothing. You pull out your gun and compel the bystander to give his jacket to the girl.

Meanwhile, farther away, there is a boy who is drowning in a shallow pond. You are aware of this second child, and you could coerce the bystander to help the drowning child; however, this would interfere with his helping the cold child (the bystander has to leave shortly for very important reasons and only has time to help one child). You like the cold child better; she looks more like you, she is from your home town, and so on. So you secure help for the cold child, leaving the other child to drown.

Your behavior in this case is morally unacceptable. There are two obvious problems: first, the desire to keep a child warm is, in the absence of a medical emergency, not an adequate justification for armed robbery. Second, if you are going to use coercion to help someone, it has to be the drowning child, whose needs are far more urgent.

The government of a wealthy nation is in a similar position. Some of its citizens are somewhat needy. But there are much needier citizens in other countries. The American poor, for example, are only poor relative to other Americans; they often nevertheless own automobiles, color televisions, microwave ovens, and so on. They may be in danger of, for example, being unable to afford new clothes or being unable to send their children to college. The poor in the developing world, however, are absolutely poor. They are in danger of dying because of starvation, malnutrition, or easily preventable diseases. Yet for the most part, the governments of wealthy countries, such as the United States, choose to use their funds to aid people in their own countries, mostly ignoring the much needier people elsewhere. The two activities are related, because the funds that are spent on domestic poor could have been spent on foreign poor. The state could of course raise its tax rate so as to have more money available, but no matter how high the state raised its revenues within the range of plausible possibilities, it would still be the case that all or nearly all of that money would have to be spent on the foreign poor if funds were allocated in anything like a need-based manner.

As in the case of the Cold Child, most actual government wealth-redistribution programs appear to suffer from two problems. First, the needs they aim to address are not sufficiently urgent to justify coercion. The need to save a person from death or serious injury may justify a moderate level of coercion and moderate violations of property rights. But the desire to supply a person with quality clothing, a college education, or an air conditioner typically does not suffice to justify coercive seizure of the necessary funds from innocent third parties.

Second, if the government is to institute coercive aid programs at all, it surely must direct its efforts toward people whose lives are in grave danger yet who could be saved at minimal cost rather than toward people with much less urgent needs that are much more expensive to address. For example, it has been estimated that programs of vitamin Asupplementation in the developing world can save lives at a cost of between $64 and $500 per life saved.[20] For comparison, when performing cost-benefit analyses, the U.S. EPA uses a figure of $6.9 million for the value of a statistical life in the United States.[21] The government could give its antipoverty funds to charity groups implementing extremely cost-effective, lifesaving programs in less developed countries. Surely programs of this kind must come before giving money to an American family whose income, while low by American standards, is many times greater than that of most inhabitants of the developing world.

Some would argue that the state's seemingly perverse priorities are justified because the state bears special responsibilities to its own citizens that it does not bear to foreigners.[22] This strikes me as an inadequate reply. Suppose we add to the Cold Child case the stipulation that the cold child is actually your daughter, whereas the drowning child is a stranger from some other country. If governments have special duties to their own citizens, parents have even clearer and stronger special duties to their own children. So if it were a question of saving one of the two children's lives, it would be proper to save your daughter. But you may not choose securing a jacket to keep your daughter warm over saving the life of a stranger.

The argument of this subsection does not attempt to show that no coercive antipoverty programs could be justified. What it shows is that if the state is morally justified in adopting any such programs, they would have to be very different from the programs actually found in wealthy countries. They would focus on extremely needy yet easily helped people in foreign countries. Existing programs are almost entirely aimed at the wrong people and the wrong problems.

7.2.4 A clash of analogies: drowning children and charity muggings

Is the Drowning Child case the closest analogy we can find to government antipoverty programs? Consider the Charity Mugging case: you have started a charity to provide monetary assistance to the poor. To collect the needed funds, you take to mugging people on the street.

This seems clearly impermissible. Now suppose for the sake of argument that your coercive action in the Charity Mugging case is impermissible, yet your coercive action in the Drowning Child case is permissible. Which case provides a closer analogy to government antipoverty programs?

On the face of it, if we are to pick one of the cases as a closer analogy, it has to be the Charity Mugging. In the Charity Mugging, the coercive action is taken in service of exactly the same sort of program as the government programs at issue, a program of direct economic aid to the poor. The coercive act is also of the same kind as in the government programs: forcible extraction of money. Neither of these things is true of the Drowning Child case. So, if we accept ordinary intuitions about both the Drowning Child and the Charity Mugging, we should conclude that government antipoverty programs are impermissible.

Some philosophers would argue, however, that the Drowning Child and the Charity Mugging cases have no morally significant differences: both are cases in which one coerces another person to secure aid for a needy third party, and this is all that matters. Because these cases are clearly analogous to each other, these philosophers would say, our intuitions about one of the two cases must simply be wrong.[23] Since the intuition about the Drowning Child is stronger than the intuition about the Charity Mugging, we should stick with the Drowning Child intuition, and thus, we should endorse government antipoverty programs.

Are there any relevant differences between the two cases? For those who have not previously done so, it is worth taking a moment to reflect on this question before reading on.

There are at least three differences that might be thought, either individually or in conjunction, to be morally significant:

  1. In the Charity Mugging, the problem you seek to address is a chronic social condition, whereas in the Drowning Child, the problem is an acute emergency. The examples in the literature that draw forth the strongest intuitions about duties to assist others are examples of acute emergencies. The cases in which we lack strong intuitions of a duty to assist but in which philosophers try to argue us into accepting such duties are typically cases of chronic social conditions.
  2. In the Drowning Child case, one can easily and quickly solve the problem, whereas in the Charity Mugging case, one can realistically hope only to alleviate the problem.
  3. In the Drowning Child case, the coercion required to address the problem is a one-time intervention, whereas in the Charity Mugging, it is an ongoing program of coercion.[24]

Government antipoverty programs align with the Charity Mugging case in all these respects. There may be other interesting differences between the two kinds of cases, perhaps including some that no one has yet identified. This is likely, since it is in general very difficult to identify the sources of our intuitions, and most people have difficulty even coming up with points (a)-(c).

There seem now to be four philosophical views about the Drowning Child and Charity Mugging cases that are worth considering:

  1. The two kinds of case are relevantly alike, and in neither case is coercion permissible.
  2. The two kinds of case are relevantly alike, and in both cases coercion is permissible.
  3. One-time coercive interventions to solve acute emergencies are permissible, but ongoing programs of coercion to alleviate chronic social conditions are not. Thus, coercion is justified in the Drowning Child case but not in the Charity Mugging case.
  4. The cases are disanalogous for some other reason, and coercion is justified in the Drowning Child case but not in the Charity Mugging case.

Only on option (ii) would we conclude that government antipoverty programs are permissible. But at first glance (iii) seems much more plausible than (ii). Option (iv) also seems much more plausible than (ii), despite the failure to specify the relevant difference between the cases (it is not at all implausible to think that there may be a relevant difference that has escaped notice). Views (i) and (ii) strike me as of comparable implausibility, with (ii) less plausible than (i), though reasonable thinkers will differ on this.

Thesis (ii) has more implausible implications than just that one may extort money from others to support charity. The bystander in the Drowning Child has a stringent moral duty to assist the drowning child. If the Charity Mugging case is relevantly similar, then individuals must have stringent moral duties to donate to charity, comparable to the duty to assist a drowning child. If they did not have such duties, this would be a morally relevant difference between the two cases (it is morally relevant that in Drowning Child, one coerces the bystander only to do her duty).

Now imagine another case; call this the Overworked Philanthropist case. Suppose that you regularly donate 80 percent of your paycheck to charities helping poor children. On the way to work, you see a child drowning in a shallow pond. Considering how much sacrifice you have already made for others, you wonder whether you must really get your clothes wet to save yet another child.

Intuitively, the answer is yes. Even after donating 80 percent of your income to charity, you are still obligated to save a drowning child when given the chance. Now if the duty to donate to charity is comparable to the duty to save a drowning child, then it seems that we may make the same claim about donating to charity; that is, even after donating 80 percent of your income to charity, you are still obligated to donate (more) to charity when given the chance. If this is not so, then the obligation to give to charity must be somehow less stringent than the obligation to assist a drowning child. Hence, if we accept thesis (ii), it seems we must conclude that we are obligated to donate more than 80 percent of our income to charity.[25] Furthermore, in the Overworked Philanthropist case, you would not be just slightly blameworthy if you failed to save the child. Failure to save the child would be extremely blameworthy, perhaps not much better than murder. Therefore, if the obligation to give to charity is morally comparable to the obligation to assist a drowning child, then one who fails to give away over 80 percent of his income to charity is also extremely blameworthy, perhaps not much better than a murderer. We might have to conclude that the behavior of nearly everyone, including, for example, philanthropists who give away only 75 percent of their income, is utterly despicable.

Some philosophers embrace just that sort of extremely demanding morality, along with its harsh judgment of nearly everyone's conduct. These philosophers point out that our strong aversion to giving away almost all our money is no proof that we are not obligated to do so. They may say that our reluctance to accept their demanding morality is simply due to our self-interested bias - we don't want to do what morality actually requires of us, and so we close our eyes to our obligations.[26] The hypothesis of self-interested bias might serve as a plausible debunking explanation of one isolated datum - our reluctance to accept extremely demanding obligations of charity. But the hypothesis fares worse in explaining the larger pattern of moral attitudes that cohere with that reluctance. If we simply suffered from an egoistic bias, then we might expect this fact to be evidenced by a shift in our intuitions when we directed attention away from ourselves and toward the behavior of others or when we imagined ourselves in different positions. But this does not seem to be the case. We do not exempt ourselves from a duty of charity that we recognize for others: when we hear of someone else who gives large amounts of money to charity, praiseworthy and supererogatory; we do not react as if we had been told of someone who merely refrains from murdering as many people as he could.

Even when we ourselves are in economic need - if, for example, we lose our job - we do not think of strangers as obligated to donate money to support us. Even those who are chronically poor do not consider strangers obligated to help them (though they may consider the state obligated to help them).

Nor do our intuitions about most other situations follow the direction of self-interest. We do not generally consider ourselves entitled to harm or exploit others for our own benefit. Even those who would be particularly good at exploiting others in a particular way do not typically hold that it is permissible to exploit others in that way.

Lastly, even those philosophers who accept extremely demanding ethical systems do not have the emotional reactions that would seem to cohere with belief in such demanding ethical systems. Utilitarian philosophers do not react with horror when you tell them that you spent $40 on dinner at a restaurant instead of sending the money to famine relief, yet they would surely react with horror if you told them that you left a child to drown in a pond because you didn't want to get your clothes wet.

None of these observations entail that some extremely demanding morality is not correct. But they illustrate the fact that our attitudes are coherent and can be parsimoniously explained by the hypothesis that we are not in fact obligated to donate very large amounts of money to charity. It remains possible that we are suffering from a self-interested bias that blinds us to our extremely demanding obligations of charity, but this hypothesis does a poor job of explaining the pattern of judgments and attitudes that most people evince. In moral philosophy, as well as the rest of human intellectual inquiry, it is reasonable to assume that things are the way they seem until proven otherwise.[27]

The above arguments should not, however, be taken as a license for selfish disregard of those in need. Regular donation to charity groups aiding the world's least fortunate is the compassionate and decent thing to do. Virtually no one doubts this.[28] An average member of a prosperous society may be able, over the course of his lifetime, to save literally hundreds of lives by donating a small fraction of his income.[29] In view of this, it is plausible to view regular donation as a requirement of decent respect for human life (see the footnote for recommendations).[30]

7.2.5 In case the foregoing is wrong

It is often valuable to consider what the most likely alternative is, in case one's own view is mistaken. I think the most credible alternative to the position taken above is that it is permissible for the state (or a private agent) to coercively collect funds to alleviate world poverty. In doing so, the state would be obligated to prioritize people with very serious problems that can be addressed reliably and at low cost. All or nearly all of the people satisfying that description are inhabitants of the developing world. Once the state had properly targeted its poverty relief efforts, some of the objections of Section 7.2.2 would also be obviated.

The view would challenge the distinction suggested above between acute emergencies and chronic conditions, arguing that some emergencies are chronic social conditions or components thereof. Imagine that you have been lost in the woods for several days without food, and you are in danger of starvation. You come upon a cabin in the woods. The owner is not home, but there is plenty of food inside. It seems permissible to take some food to preserve your life despite the violation of the owner's property rights. (This is permissible even if you know you will not be able to compensate the owner afterwards and even if you doubt that the owner would consent to your taking his food.) This illustrates the fact that extreme hunger can count as an emergency situation sufficient to justify the violation of another person's property rights. And if your hunger is an emergency of that sort, then the extreme hunger of a child in the Third World is an emergency of the same sort for that child. It happens that there are at any given time a great many people in that situation, so the existence of such emergencies is itself a chronic social condition. But why should that make a difference? If a certain type of rights violation would be justified when necessary to save one person from starvation, then shouldn't a program consisting of many such rights violations also be considered justified when it is necessary to save many people from starvation?

I am unsure what to make of this. Perhaps there is an ethical difference between committing an isolated theft to save oneself and starting up a regular program of extortion designed to save third parties wherever they may be. Or perhaps the conclusion of the preceding paragraph is simply correct.

Nevertheless, I want to insist on two points. First, the actual antipoverty programs in wealthy countries are unjustified. They deploy coercion with inadequate justification, they are not focused on the neediest people, and they cannot be defended by appeal to analogies of drowning children and cabins in the woods. Second, the state has no special moral status. If the state may coercively seize funds for poverty relief, it is because the state would only be doing the same thing as the private parties in the examples of the Drowning Child and the cabin in the woods. If so, the same argument could be used to show why a private party would be justified in coercively seizing funds for poverty relief. One could even rob the government to provide funds to help the poor. The state has no special authority here, although the state may enjoy practical advantages in its efforts at seizing funds.

7.3 Implications for agents of the state

Government officials who are responsible for making policy ought to take account of the observations of the last two sections and avoid making unjust policies. What about government employees who do not make policy but are instructed to help enforce policies, some of which are unjust? Police officers, for example, are asked to arrest drug users and sellers. Judges are asked to sentence them. Soldiers are asked to fight in aggressive wars. What should these government employees do?

The police officer should refuse to arrest drug offenders. If he sees someone using drugs, he should leave that person alone or perhaps stop to give the user tips on how to avoid being seen by police officers. To arrest the drug user would be to initiate an unjustified act of coercion. The state has no right to commit unjust acts of coercion nor to order such acts to be committed, so it cannot confer on its employees a moral entitlement to perform such acts.

Of course, it is not as though police officers simply decide on their own to coerce drug users; they are required to do so as part of their job. If they refuse to enforce unjust laws, this fact will in all probability become known, and they will be reprimanded or fired. But this provides no excuse for violating the rights of others. Imagine that I have hired a chauffeur to drive me around town. Periodically, I ask my driver to perform unjustified acts of coercion. One day, for example, we see some children playing on the sidewalk. I tell the driver to stop and beat up one of the children for my entertainment. I warn the reluctant driver that if he does not follow my orders, I will fire him. So the driver proceeds to beat can-achieve.php.

In this case, I have acted wrongly by ordering the child to be beaten. But the driver has clearly also acted wrongly by following this command. Perhaps I am more blameworthy than the driver, but this does not change the fact that the driver should refuse such commands, even if it results in his losing his job.

Some would deny that the driver is really just doing his job, because his job is to drive the car, not to beat up children. This is a red herring; it does not matter whether beating up children is part of his job description. Suppose that my original Help Wanted ad had read, 'Wanted: Person with clean driving record and strong muscles to drive car and beat innocent children.' My inclusion of the 'beat innocent children' clause in the job description does not afford the driver any ethical justification for beating up the child. The only ethical difference it might make is that it might render it wrong for the driver to have accepted the job in the first place. Having accepted the job, he still has no justification for beating up innocent children.

Similarly, it does not matter whether the job of a police officer includes the enforcement of unjust laws; this does not create any justification for enforcing unjust laws. The only difference it might make is that it might render it wrong to be a police officer in the first place.

Some would object that if all police officers took my arguments to heart, then all would either quit or get themselves fired, which would be much worse for society than having police who enforce both just and unjust laws. But surely, long before all police officers had resigned or been fired, the government would accede to the need for reform and repeal the unjust laws that were causing it to lose its police force, or at least allow the police to refrain from enforcing those laws. Thus, if all police were to adopt the view I have advanced, society would in fact be much better off.

For similar reasons, a judge in a case involving the violation of an unjust law should do his best to secure the minimum possible punishment. The judge should order the defendant released if feasible - that is, if this will not simply result in the defendant's being arrested again and brought before a more punitive judge. If a judge finds himself conducting a trial for something that ought not to be illegal and for which it would be unjust to punish the defendant, then the judge should do everything in his power to bias the outcome in favor of the defendant. If the judge winds up having to sentence the defendant, he should order the smallest sentence possible. If a judge finds himself routinely required, by the demands of his job, to participate in injustice, he should probably resign in search of a more just profession.

Asoldier should likewise refuse to fight in an unjust war. Not to put too fine a point on it, to fight in an unjust war is to participate in murder. In joining the military, one volunteers to fight in whatever wars one's country may enter into. Therefore, if one cannot be assured that one's nation will not enter any unjust war, one should refrain from joining the military; if one is already in the military, one should resign as soon as possible.

Similar recommendations apply to all other government employees who are called upon to implement unjust policies. They should do their best to undermine those policies or, if that is not feasible, refuse to serve.

These recommendations are hardly ever followed. Government employees almost always enforce whatever policies, just or unjust, they are ordered to enforce. One reason is that they mistakenly believe in political authority; they believe that the state has the right to coercively impose these policies, even when the policies are in themselves wrong. They consider themselves permitted, perhaps even obligated, to help to enforce those policies in accordance with their job requirements. How does this affect our evaluation of their behavior?

We can distinguish the evaluation of an individual's character from the evaluation of the individual's actions. Often one of these evaluations is much more positive or negative than the other, particularly when the agent is ignorant of important facts about his behavior. Soldiers who fight in an unjust war, for example, are typically much better people and are typically much less blameworthy than private murderers. This is compatible with the fact that there are very strong objective reasons for refusing to serve in an unjust war, reasons about as strong as the reasons for refusing to participate in a private conspiracy to commit murder. As a rule, the fact that government employees believe themselves to be acting rightly makes them less blameworthy than they would otherwise be. It does not render them completely blameless; they may still be blameworthy if, as is probably the case, they have not exercised sufficient effort to find out where their true moral duty lies. In any case, government employees' ignorance of their ethical duty does not alter the appropriate assessment of what they really ought to do. It does not alter the fact that they have no right to enforce unjust laws.

7.4 Implications for private citizens

7.4.1 In praise of disobedients

If there is no authority, then disobedience to governmental edicts is justified far more often than is generally recognized.

Suppose that Sam has been issuing demands to his neighbors that he has no right to make, backed up by threats of punishment. Agang of followers helps him to forcibly impose punishments that they have no right to impose. Sam issues demands as to what his neighbors may eat, the terms of the contracts they may make with each other, which of them may provide medical care to others, how much money they must pay Sam's gang, and so on. He also issues some morally justified commands (though his issuing them is ethically redundant): he demands that no one murder, steal, and so on. Now imagine that you are one of Sam's neighbors. You wish to ingest a certain herb with psychoactive properties, but you are aware of Sam's demand, backed up by a threat of physical force from Sam's gang, that you not do so.

It seems clear that you have no ethical reason not to eat the herb, though you may of course have a strong prudential reason, arising from fear of Sam's gang, to avoid the herb. If anything, you have an ethical reason to eat the herb, as a way of standing up to Sam's bullying. To submit to a bully's demands is at best excusable. To defy Sam privately would be perfectly acceptable; to defy Sam publicly would be a praiseworthy act of courage.

Similarly, there is no question of its being wrong to defy unjust laws. The only ethical question is whether defiance is obligatory or supererogatory. In view of the severity and credibility of the threats commonly issued by the state to lawbreakers, I believe that in most cases, defiance of unjust laws is supererogatory. In some cases, defiance is foolhardy, as it would be foolhardy, when a mugger points a gun at you, to refuse to hand over your wallet. But it is not ethically wrong.

7.4.2 On accepting punishment

On some contemporary accounts, those who engage in civil disobedience must do so publicly and must accept the punishment that the state prescribes.[31] These accounts, however, are drawn against the backdrop of an assumed political authority. If there is no political authority, are there still reasons for submitting to legal punishment for acts of justified disobedience?

To disobey a law publicly is to disobey it in such a way and under such circumstances that one's action will become widely known (among those who follow such affairs), and it will be known that the action broke the law. In many cases, it will be possible to disobey a law publicly in this sense without revealing one's identity. (Imagine peace activists painting graffiti on a military factory in the middle of the night, then sneaking away.) When feasible, this form of disobedience affords obvious advantages: one may avoid suffering unpleasant punishments and remain free to perform further acts of disobedience in the future, while still communicating rejection of the unjust law.

Sometimes it is said that those who engage in civil disobedience should accept punishment for their actions to prove their sincerity and seriousness to others.[32] For instance, some hold that in case of a draft, conscientious objectors should voluntarily go to prison - rather than, for example, escaping to another country - to prove that their objection to going to war is principled and unselfish.

There are a number of salient objections to this thinking. First, the requirement to accept punishment is excessively demanding. No doubt there is some value in communicating one's sincerity, moral seriousness, or other admirable traits to others. But one is not typically obligated to communicate this sort of information to others, even when the costs of doing so are minimal. For instance, suppose that I recently found a wallet, and I went out of my way to return it to its owner. I am not then morally obligated to relate this episode to other people, just to communicate my honesty and virtuousness, even if I can do so at no cost. Still less would I be obligated to communicate this information if doing so would require my spending some months or years in prison. It is unclear why the case should be different for acts of civil disobedience. If I have broken the law, I should no doubt prefer that others know that I had virtuous motives for doing so. But I am not obligated to communicate this information, even if I could do so for free, still less if doing so requires my spending months or years in prison.

One might argue that this case is different, because in disobeying the state, I might lead others to disobey other laws, including laws that should be obeyed, if others fail to understand the moral reasons behind my disobedience. This suggestion is contrived and implausible; it is in most cases highly unlikely that my act of disobedience will cause other people to disobey some unrelated law. In addition, it is typically not obligatory for a person to undertake extremely large sacrifices, such as spending time in jail, to prevent other people from irrationally choosing to do wrong.

Second, voluntarily accepting the state's punishment for an act of disobedience may (either instead of or in addition to communicating one's moral seriousness) communicate false and destructive ideas - most notably, that the state has the right to punish people for disobeying unjust laws. If a law is unjust, then the enforcement of that law through punishment of those who disobey is also unjust. Why, then, should one facilitate this injustice by submitting oneself for punishment? For example, suppose one's government is engaged in an unjust war, for which it has instituted a draft. In such a case, no one is obligated to participate in the war; if anything, citizens are obligated to refuse to participate. Now in addition to the injustice of the war itself, there will also be the injustice of the state's punishing those who virtuously refuse to participate in that war. And just as no one is obligated to facilitate the war itself, no one is obligated to facilitate or cooperate with the unjust punishing of those who refuse to participate in the war.

Consider an analogy. A homophobic gang in your neighborhood is beating up homosexuals.[33] If you are gay, should you present yourself at the gang's headquarters and announce your sexual orientation so that they may beat you? Obviously not. Among other things, to submit to a beating would wrongly communicate that you have done something that deserves punishment and that the gang has the right to punish you. Even if you believe that by submitting to a beating you would increase the probability that public outrage would eventually lead to a change in the gang's behavior, you still would not be obliged to submit to a beating.

I conclude that in most cases, those who disobey unjust laws are both ethically permitted and well advised to conceal their identity or otherwise evade punishment by the state.

7.4.3 On violent resistance

If there is a central premise of this book, it is the moral seriousness of coercion. But the resort to physical force is not always wrong. It is often justified for purposes of self-defense or defense of innocent third parties. It is not implausible, therefore, that violent resistance may often be justified in response to unjust coercion by the state.

To assess this thought, let us begin with some general principles governing the defensive use of force:

  1. The use of force is justified only when necessary to prevent some serious wrong. That is, there must be no alternatives available that would prevent the wrong without using comparable levels of force, committing some other equally serious prima facie wrong, or demanding unreasonable sacrifices of the agent.
  2. The use of force must have a reasonable chance, on the agent's evidence, of preventing the wrong from occurring. Unless this condition is satisfied, the use of force will not count as a defensive measure. (It may instead count as a retaliatory measure; however, the conditions for justified retaliatory force are beyond the scope of this discussion.)
  3. The expected harm caused by the use of force may not be out of proportion to the expected harm averted. For example, it is not permissible to kill another person merely to prevent that person from stealing your stereo. It what Rawls labels 'civil disobedience' and what he labels 'conscientious refusal' (1999, section 56).
  4. It is usually not permissible to harm innocent third parties in the course of defensive violence. Such harm to innocent third parties can sometimes be justified, but this will generally require expected benefits much greater than the expected harms.

Historically, there are two main forms of armed resistance to the state: terrorism and (attempted) revolution. In developed modern societies, attempted armed revolution is unlikely to be justified, for three reasons. First, there are usually nonviolent options available, which have shown surprising success in some cases, such as the well-known cases of Gandhi and Martin Luther King Jr. Second, the probability of successful revolution in most modern, developed societies is very close to zero. Third, in the case of attempted revolution, harm to innocent third parties is likely to be very great.

Terrorist attacks are no more likely to be justified. The same three points apply: nonviolent methods are typically available, terrorist methods are ineffective, and expected harms to innocent parties are excessive. A2006 study examining 28 terrorist groups found that, using generous criteria of success, these groups achieved their policy objectives only 7 percent of the time. Later studies using larger samples have found success rates under 5 percent, and in many cases the terrorists' political goals were actually set back.[34] Why is terrorism so ineffectual? When terrorists attack civilians, populations tend to increase their support for right-wing political candidates proposing aggressive responses. These hard-liners are not frightened by terrorism, nor should they be; they are extremely unlikely to be personally victimized by terrorism, and in fact their political careers are strongly advanced by terrorism and the opportunity it provides them for aggressive posturing.[35] All of these points are illustrated by the infamous terrorist attacks of 11 September 2001 in the United States, which prompted the U.S. government to vastly increase its military presence in the Middle East, killing hundreds of thousands of Muslims. While this response was irrational and reprehensible, it was also predictable.

As a rule, therefore, terrorist attacks are morally wrong. The question of when one may harm innocent third parties in the course of bringing an end to oppression and injustice is a matter for debate. But surely one may not harm innocent third parties for the sake of ineffectual or counterproductive gestures.

7.4.4 In defense of jury nullification

Most readers will at some point be required to serve on a jury in a criminal trial. Many of these trials will be for crimes that genuinely deserve to be punished. But many others will be for the violation of unjust laws, such as the laws mentioned in Section 7.1. It is therefore of great practical interest what a juror should do in the latter sort of case.

When the law is unjust, the juror should vote to acquit, regardless of the evidence. Briefly, the argument is this: in general, it is wrong to knowingly cause unjust harm to another human being. To convict the defendant for violating an unjust law will, as a rule, result in his suffering significant, unjust harm at the hands of the state. Therefore, it is prima facie wrong to convict such a defendant.

Two objections must be considered. To begin with, it might be argued that a juror who votes for conviction would not be culpable for the defendant's suffering, because the juror did not make the unjust law nor is the juror himself directly imposing the punishment. Assuming that the prosecutor proves his case, the jury member who votes to convict is merely correctly reporting the fact that the defendant performed a certain action. What officials of the state do with that information is up to them; the juror is not telling them to punish the defendant (even though he knows that they will do so). A related objection is that one has a duty to tell the truth. To vote for an acquittal, in a case where the evidence shows that the defendant did in fact violate the unjust law, would be dishonest. It would be tantamount to a false assertion that the defendant was not shown to have violated that law.[36]

Both objections may be addressed by the following analogy. You are walking down the street with one of your more flamboyantly dressed friends. You run into a gang of hoodlums. The leader of the gang asks you whether your friend is gay. You are convinced that these are gay bashers and that if you either answer 'yes' or refuse to answer, they will beat up your friend. The two of you have the best chance of being left unharmed if you answer 'no'. You know, however, that your friend is in fact gay. Thus, by answering 'no', you would be lying. Should you, therefore, either refuse to answer or answer 'yes'?

None but a fanatical Kantian would say so. Granted, lying is usually wrong, but not when the person to whom you are lying is someone who would use a truthful answer as a pretext for bringing serious and unjust harm upon another human being. If you should tell the hoodlums the truth, will you later be able to patch things up with your friend, as you visit him in the hospital, by reminding him that you didn't make the hoodlums hate gays to begin with nor did you personally pound on your friend with your own fists? Could you plead that all you did was report on a factual matter and that what the hoodlums did with that information was up to them?

Jurors in most courts are instructed that they must render a verdict based upon the evidence and that they must not choose to nullify the law. They may even be asked to take an oath to this effect, where refusal to do so results in dismissal from jury service. This does not alter the juror's true moral duty. Suppose that, in the above scenario, the homophobic gang leader asks you to promise to tell the truth about your friend. Suppose he also instructs you, with a great air of confidence and solemnity, that you must tell him the truth and that you have no right to lie because you disagree with his gay-bashing predilections. Then would you be obligated to tell him the truth? Again, no. Homophobic hoodlums do not have a right to know who is and is not gay. You should promise to tell the truth and then immediately lie.

In the United States, jurors who vote to acquit a defendant on the grounds that the law is unjust are subject to no punishment, and their verdicts cannot be overturned. Thus, despite what they may be told, jurors certainly can nullify laws, in the relevant senses of 'can'. The aversion to lying and promise breaking (if that is what jury nullification involves) is a trivial consideration next to the importance of preventing a person from suffering severe and unjust harms.

7.5 Objections in support of rule-worship

7.5.1 May everyone do as they wish?

People have their own opinions as to exactly which laws are just. It might therefore be thought that the philosophical view I have advanced affords carte blanche for individuals simply to do whatever they want, citing idiosyncratic interpretations of justice in their defense.

This would of course be a mistake. My philosophical position does not imply that individuals may break whatever laws they want to break. Suppose that Sally wants to steal money from her company so that she can live at others' expense. Sally therefore dishonestly claims to find the laws governing property 'unjust', and she uses this to rationalize her behavior. In this case, Sally's .behavior is wrong. Her mere assertion that the property laws are unjust does nothing, ethically, to excuse her.

Suppose that Mary is also stealing money from her company. Mary, however, sincerely believes that the laws governing property are unjust, for she has been taken in by a misguided political ideology that rejects private property. In this case, is Mary's behavior right? No, it is not. Mary is mistaken in thinking that the property laws are unjust, so she is also mistaken in taking her own behavior to be ethically permissible. Depending on how understandable her error is, Mary may be less blameworthy than Sally, but her action is just as wrong. Thus, for example, it would be appropriate for third parties to use coercion to stop Mary from taking more money and to compel her to compensate her employer.

This is consistent with everything I have said earlier in this chapter. If a law is unjust, one may break it. But it is not the case that if one merely believes a law to be unjust, one may break it; it depends upon whether one's belief is correct.

There are many cases in which we cannot tell whether a law is just or unjust; justice is a difficult subject. What ought we to do then? In cases where we do not know whether the law is just, we will simply not know whether it is permissible to break that law. I can say nothing here that will cause readers to be able to know in all cases what is just or what they ought to do. My only advice for such situations is that one do further research on the topic (perhaps in the ethical and political philosophy literature) and then exercise one's best judgment.

To some, this view will be unsatisfying. A more satisfying view would be one that provides a simple, more or less mechanical rule for what to do in all cases. For instance, if we could say, 'When in doubt, always obey the law', many would find this a more satisfying position than the position that we sometimes cannot tell whether we should obey the law or not.

But satisfyingly simple and convenient rules are not therefore correct. In particular, there is no reason to think that whenever there is doubt as to the justice of a law, it is better to obey than to disobey that law. Suppose a soldier has been ordered by his government to fight in a war. The soldier is unsure whether this order is just, because he is unsure whether the war itself is just. Nothing in this description of the case enables us to infer that it would be right or good for the soldier to fight in the war. If he fights, he may be participating in mass murder. We do not know enough to say whether this is the case. The crucial information we would need, before we could advise the soldier as to what he ought to do, is a piece of moral information: we need to know whether the war is just. The fact that this knowledge may be difficult or even impossible to obtain does not prevent it from being the relevant and necessary knowledge for addressing the question at hand, nor does it enable some other, more easily knowable fact to settle the question. It simply is the human condition that our ethical questions frequently have no easy answers.

7.5.2 Process versus substance

In an early article defending the Fair Play Account of political obligation, John Rawls takes as his central question, 'How is it possible that a person, in accordance with his own conception of justice, should find himself bound by the acts of another to obey an unjust law ... ?' And he answers: 'To explain this ... we require two hypotheses: that among the very limited number of procedures that would stand any chance of being established, none would make my decision decisive ... ;and that all such procedures would determine social conditions that I judge to be better than anarchy.'[37]

My reading of that passage is that Rawls is assuming (1) that we must rely on some procedural criterion for deciding which laws are legitimate or ought to be obeyed, and (2) that an individual who disobeys a law on the grounds that the law is unjust is applying the following procedural rule: that a law is to be rejected if it conflicts with that individual's sense of justice. He finds the latter rule inadequate and inferior to democratic procedures. Therefore, Rawls believes, if a law has been made according to democratic procedures, an individual should not disobey that law on the grounds that (he believes) the law is unjust.

But Rawls does not justify these assumptions; he does not explain why grounds for obeying or disobeying particular laws must be procedural. Instead, a law may be accepted or rejected on substantive grounds. When I say that the drug laws may be violated because they are unjust, I am not saying that the drug laws were made according to the wrong procedure. I am saying they are substantively unjust; they violate a substantive moral right, the right to control one's own body, that individuals possess regardless of the decisions of the state. This would be true regardless of how the law was made (except, of course, in the unlikely event of unanimous consent to the law, which would render it no longer a rights violation). I am not proposing a procedure according to which my personal opinion is decisive; if I did not exist or if I had sanctioned the drug laws, they would still be unjust. If I objected to some law that is actually just - for instance, if I objected to the murder statutes - my objection would not convert the just law into an unjust one. In other words: when I object to the drug laws, my grounds for objection are not simply that I object. My ground for objection is the right of self-ownership, the right of individuals to control their own bodies. Rawls's argument has nothing to say about whether this is a legitimate ground for rejecting, and hence disobeying, a law.

Why might it be thought that we must rely on procedural rules rather than substantive moral principles? Perhaps because it is thought that we do not know what is substantively morally correct, whereas we do know what a desirable procedure is. If this is the reasoning behind Rawls's proceduralist assumption, it is doubly mistaken. First, it is mistaken because it is false that in general we do not know what is substantively morally correct. Sometimes we do not know what is substantively just. But often we do know. I do not know, for example, whether a ban on abortion would be unjust. But I know that the Jim Crow laws were unjust. When we know that a law is unjust, our opposition to it can and should be based on the fact that it is unjust, not on the fact that it conflicts with our personal opinions or preferences.

Second, if it were true that we never knew what was substantively just, then we also would not know what was procedurally just. There is no reason to think that knowledge of just procedures would somehow evade the reach of a moral skepticism strong enough to rule out all knowledge of just outcomes. If, for example, we cannot know that laws that treat citizens grossly unequally on the basis of morally irrelevant characteristics are unjust, then why would we know that lawmaking procedures that fail to give citizens an equal voice are unjust?

7.5.3 Undermining social order?

Imagine that the views I have defended become widespread and, in particular, that the notion of political authority is widely rejected. Citizens thus feel free to violate any laws they find ethically objectionable whenever they can evade punishment. Government agents refuse to enforce laws they find ethically objectionable. Juries refuse to convict defendants under statutes that the juries find objectionable. Wouldn't this render our legal system too chaotic and unpredictable? Might social order not collapse entirely?

The suggestion of the preceding paragraph is simply that it may be very harmful to propagate the views advanced in this book, so much so that perhaps I should not have published the book. That is compatible with the possibility that everything I say is actually true.

The dire warnings about the collapse of social order, however, are ill taken. The views I advance are more likely to be socially beneficial than harmful. We have imagined citizens violating laws they find unjust, police refusing to enforce laws they find unjust, and juries refusing to convict under laws they find unjust - all because a general skepticism of authority has taken hold of society. In addition, we must assume that lawmakers themselves have absorbed the same philosophy of skepticism about political authority. In this case, there would be far fewer laws - and far fewer unjust laws in particular. Most of the laws that would have occasioned widespread civil disobedience would not exist, because legislators would not make them or would have repealed them during the period in which skepticism about authority was taking hold of society.

But suppose a particular law survives that some people consider unjust. If the number who find the law unjust is very small, there is no difficulty. For instance, suppose a tiny number of people consider the laws against theft unjust. Since the overwhelming majority of society considers those laws just, there would still be enough police officers, judges, and jurors willing to enforce the antitheft laws. The government would only encounter a problem when a large portion of society considers some law unjust. For instance, a significant number of people presently consider the drug laws unjust. If it were widely accepted that one should not help to enforce an unjust law, then some police officers would refuse to arrest drug offenders. Some judges would refuse to sentence them. And many juries would refuse to convict them. Trials under the drug laws and other controversial statutes would repeatedly result in hung juries. Once this pattern became clear, the state would probably give up trying to enforce such laws.

Is this the social disaster we should worry about? On the contrary, this would be a much better situation than the status quo. When the justice of a law is controversial, it is better to err on the side of freedom than on the side of restriction. Perhaps some just laws would, unfortunately, go unenforced. But the reduction in the number of people wrongly punished under unjust laws would more than compensate for this disadvantage. It is widely held that it is better for ten guilty people to go free than for one innocent person to be punished. If this is true, then it is also better for ten people to fail to be convicted under just laws than for one person to be convicted under an unjust law. Our present system, however, errs very much in the opposite direction: even when the moral status of a law is in doubt, police officers, judges, and juries almost always enforce the law without question.

On a realistic note, the picture of ordinary people as perched on the verge of disorder, waiting for an excuse to run rampant in disregard of law and order, flies in the face of everything we know about the psychology of authority (see Chapter 6). Evidence such as the Milgram experiments, the Holocaust, and the My Lai Massacre leave little doubt that the average human being is far more likely to commit heinous crimes in the name of obedience to authority than he is to rashly disobey justified commands of an authority figure. Literally millions have died because of the widespread disposition to obey unjust commands. So even if my skepticism about authority goes too far, it will more likely serve as a valuable corrective to our excessive tendency to obey rather than posing a danger of destroying social order.

7.5.4 The consequences of the doctrine of content-independence

I have argued that the state has the right to make and enforce only ethically correct laws. Some think that this asks too much of the state; any government run by fallible human beings will sometimes make mistakes, including moral mistakes.[38] If the agents of the state are thought to have no leeway, no entitlement to make mistakes, then they may be paralyzed into inaction by the fear of doing wrong. One might therefore think that the state should be accorded at least some leeway in the form of a content-independent entitlement to make rules, as long as its rules are not too unreasonable.

Parallel reasoning could be applied to private agents. It is also unrealistic, for example, to expect a large private corporation to be perfect; any such corporation will sometimes make mistakes, including moral mistakes. But no one thinks this means that we must ascribe to large corporations a moral entitlement to periodically perform unjust or wrongful actions just as long as they are not too unreasonable. We recognize that a large corporation will sometimes do wrong, but we do not acquiesce in those wrongs. We condemn them when they happen and demand that the corporation make amends. In the same way, we should not acquiesce in wrongdoing by the state, however predictable it may be; we should condemn it when it happens and demand that the state make amends. This attitude will not make it impossible to maintain a state, any more than the analogous attitude toward corporations makes it impossible to maintain a corporation.

What are the likely social consequences of the belief in content-independent authority? Christiano tells us that 'the democratic assembly has a right to do wrong, within certain limitations.' Rawls observes, 'It is, of course, a familiar situation ... that a person finds himself morally obligated to obey an unjust law.'[39] Does this idea increase or decrease the state's likelihood of achieving the social aims for the sake of which the state is supposed to be needed?

Consider an analogy. You have hired a gardener to take care of the plants in your yard. You want him to care for all the plants, and you want him not to do anything else, such as entering the house and stealing your jewelry. Which of the following two instructions should you give the gardener?

  1. You must take care of all the plants. You must not enter the house and steal jewelry.
  2. Ideally, you should take care of all the plants, but you have some leeway; you are entitled to periodically damage or neglect a few of them. It would also be best if you did not enter the house and steal jewelry. But you may do that occasionally, as long as it doesn't get out of hand.

Rawls, Christiano, and other defenders of content-independent political authority are in effect giving the gardener instruction (B). I would tell the gardener (A). Which is really the socially dangerous philosophy?

7.6 A modest libertarian foundation

Libertarianism is a minimal government (or, in extreme cases, no government) philosophy, according to which the government should do no more than protect the rights of individuals.[40] Essentially, libertarians advocate the political conclusions defended in this chapter. But this position is very controversial in political philosophy. Many readers will wonder if we are really forced to it. Surely, to arrive at these radical conclusions, I must have made some extreme and highly controversial assumptions along the way, assumptions that most readers should feel free to reject?

Libertarian authors have indeed frequently relied upon controversial assumptions. Ayn Rand, for example, thought that capitalism could only be defended by appeal to ethical egoism, the theory that the right action for anyone in any circumstance is always the most selfish action.[41] Robert Nozick is widely read as basing his libertarianism on an absolutist conception of individual rights, according to which an individual's property rights and rights to be free from coercion can never be outweighed by any social consequences.[42] Jan Narveson relies on a metaethical theory according to which the correct moral principles are determined by a hypothetical social contract.[43] Because of the controversial nature of these ethical or metaethical theories, most readers find the libertarian arguments based on them easy to reject.

I have appealed to nothing so controversial in my own reasoning. I reject the foundations for libertarianism mentioned in the preceding paragraph. I reject egoism, since I believe that individuals have substantial obligations to take into account the interests of others. I reject ethical absolutism, since I believe an individual's rights may be overridden by sufficiently important needs of others. And I reject all forms of social contract theory, for reasons discussed in Chapter 2 and 3.

The foundation of my libertarianism is much more modest: common sense morality. At first glance, it may seem paradoxical that such radical political conclusions could stem from anything labeled 'common sense'. I do not, of course, lay claim to common sense political views. I claim that revisionary political views emerge out of common sense moral views. As I see it, libertarian political philosophy rests on three broad ideas:

  1. A nonaggression principle in interpersonal ethics. Roughly, this is the idea that individuals should not attack, kill, steal from, or defraud one another and, in general, that individuals should not coerce one another, apart from a few special circumstances.
  2. A recognition of the coercive nature of government. When the state promulgates a law, the law is generally backed up by a threat of punishment, which is supported by credible threats of physical force directed against those who would disobey the state.
  3. A skepticism of political authority. The upshot of this skepticism is, roughly, that the state may not do what it would be wrong for any nongovernmental person or organization to do.

The main positive ethical assumption of libertarianism, the nonaggression principle, is the most difficult to precisely articulate. In truth, it is a complex collection of principles, including prohibitions on theft, assault, murder, and so on. I cannot completely articulate this set of principles. Fortunately, it is not the locus of disagreement between libertarians and partisans of other political ideologies, for the 'nonaggression principle', as I use the term, is simply the collection of prohibitions on mistreating others that are accepted in common sense morality. Almost no one, regardless of political ideology, considers theft, assault, murder, and so on morally acceptable. We do not need a complete list of these prohibitions, since we have been able to construct the arguments of this book by relying on intuitions about specific cases. I have made no particularly strong assumptions about these ethical prohibitions. I do not, for example, assume that theft is never permissible. I simply assume that it is not permissible under normal circumstances, as dictated by common sense morality.

The second principle, that of the coercive nature of government, is equally difficult to dispute. The coercive nature of government is commonly forgotten or ignored in political discourse, in which the justification for coercion is seldom discussed. But virtually no one actually denies that the state regularly relies upon coercion.

It is the notion of authority that forms the true locus of dispute between libertarianism and other political philosophies.

Libertarians are skeptical about authority, whereas most accept the state's authority in more or less the terms in which the state claims it. This is what enables most to endorse governmental behavior that would otherwise appear to violate individual rights: nonlibertarians assume that most of the moral constraints that apply to other agents do not apply to the state.

I have therefore focused on defending skepticism about authority by addressing the most interesting and important theories of authority. In defending this skepticism, I have, again, relied upon no particularly controversial ethical assumptions. I have considered the factors that are said to confer authority on the state and found that in each case, either those factors are not actually present (as in the case of consent-based accounts of authority) or those factors simply do not suffice to confer the sort of authority claimed by the state. The latter point is established by the fact that a nongovernmental agent to whom those factors applied would generally not be ascribed anything like political authority. I have suggested that the best explanation for the inclination to ascribe authority to the state lies in a collection of nonrational biases that would operate whether or not there were any legitimate authorities. Most people never pause to question the notion of political authority, but once it is examined, the idea of a group of people with a special right to command everyone else fairly dissolves.

These three ideas - the nonaggression principle, the coercive nature of government, and skepticism about authority - together call for a libertarian political philosophy. Most government actions violate the nonaggression principle - that is, they are actions of a sort that would be condemned by common sense morality if they were performed by any nongovernmental agent. In particular, the government generally deploys coercion in circumstances and for reasons that would by no means be considered adequate to justify coercion on the part of a private individual or organization. Therefore, unless we accord the state some special exemption from ordinary moral constraints, we must condemn most government actions. The actions that remain are just the ones that libertarians accept.

How might one avoid the libertarian conclusion? Only by rejecting one of the three core principles I have identified. It is extremely unpromising to question the coercive nature of government, and I doubt that any theorist will wish to take that tack. Some theorists will question common sense morality. I have not undertaken a general defense of common sense morality in this book, and I shall not do so now. Every book must begin somewhere, and beginning with such assumptions as that under normal conditions one may not rob, kill, or attack other people seems reasonable enough. This is about the least controversial, libertarianism.least dubious starting point for a book of political philosophy that I have seen, and I think few readers will feel happy about rejecting it.

The least implausible way of resisting libertarianism remains that of resisting the libertarian's skepticism about authority. I have addressed what strike me as the most interesting, influential, or promising accounts of political authority - the traditional social contract theory, the hypothetical social contract theory, the appeal to democratic processes, and appeals to fairness and good consequences. But I cannot address every possible account of authority, and a fair number of thinkers may react to my performance by proposing alternative accounts of authority.

I suspect, however, that the general strategy I have relied upon will be able to be extended to such alternative accounts. A theory of authority will cite some feature of the state as the source of its authority. My strategy begins by imagining a private agent who possesses that feature. Of course, this will not be possible if the feature in question entails statehood - but so far, the features that have been alleged as the source of the state's political authority have not entailed statehood (no one, for example, has proposed that authority is conferred simply by the property of being a state). For instance, the property of being something that would be agreed to by all reasonable people, the property of being actually accepted by the majority of society, and the property of producing very good consequences are all properties that a nongovernmental organization or the policies of such an organization could possess. As I say, then, we imagine a nongovernmental agent with the relevant feature. We then realize that intuitively we would not ascribe to that agent anything like a comprehensive, content-independent, supreme entitlement to coerce obedience from other people. And so we conclude that the proposed feature fails as a ground of political authority.

Notes

1 Though I have followed the established terminology here, it should be noted that the terminology is misleading, since it falsely suggests that one of the doctrines is philosophical but not political, while the other is political but not philosophical. In fact, both kinds of 'anarchism' are philosophical and political claims.

2 U.S. Code, Title 18, section 2422: 'Whoever knowingly persuades, induces, entices, or coerces any individual to travel in interstate or foreign commerce, or in any Territory or Possession of the United States, to engage in prostitution, or in any sexual activity for which any person can be charged with a criminal offense, or attempts to do so, shall be fined under this title or imprisoned not more than 20 years, or both.'

3 See Huemer 2010a, 361-2.

4 See Huemer 2010a, 356-7.

5 See Huemer 2010a for elaboration.

6 Tullock 1987.

7 See Friedman 1989, 42-4, for discussion.

8 This example is from Huemer 2010b, which defends the argument of this subsection at length.

9 Murphy and Nagel 2002, 173-7. Compare Holmes and Sunstein 1999, chapter 3.

10 Locke 1980, chapter 5.

11 U.S. Census Bureau 2011b, 310, table 467.

12 See Huemer 2003 and forthcoming for arguments against egalitarianism.

13 Singer 1993, chapter 8; Unger 1996.

14 The formulation in terms of 'expected benefit' is intended to allow for the possibility that a coercive act might be justified by virtue of its merely reducing the risk ofof something very bad. It need not be clear that the coercive act in fact prevents the bad event; however, it must at least be reasonably clear that the coercive act reduces the risk. If the coercive act creates some other risk, it must also be reasonably clear that the reduction in the original risk outweighs the newly created risk.

15 Lerner 1944, chapter 3; Nagel 1991, 65.

16 Murray 1984. See also Olasky 1992; Schmidtz 1998.

17 Murray 1984, 8-9.

18 See Jencks 1992, chapter 2; Murray and Jencks 1985; Cowen 2002, 39-44.

19 Schmidtz 2000; Cowen 2002, 44-9.

20 Horton et al. 2009. Other extremely cost-effective programs include zinc supplementation, iron and folate fortification, salt iodization, and deworming - all in the developing world (Bhagwati et al. 2009).

21 Borenstein 2008. In other words, the EPAconsiders a regulation to be worthwhile if it imposes a cost of no more than $6.9 million for every American life that it is expected to save.

22 See Goodin 1988 (but note the last sentence of the article, which comes close to taking back what the rest of the article seems to say). See also Wellman 2000.

23 Though Unger does not discuss the Charity Mugging case directly, his remarks about other cases (1996, chapter 3) suggest that he would endorse the argument mentioned in the text.

24 Unger (1996, chapter 2) discusses a similar pair of examples and considers several potentially relevant differences, including essentially (a) and (b) above. He finds (a) morally irrelevant (42). Roughly, he holds that the distinction between an 'emergency' and a 'chronic problem' consists merely in the fact that the victims of the latter have been suffering for a longer time; but this surely cannot lessen the reasons for helping victims of chronic problems. It is not obvious, however, that Unger's account of the we treat this as distinction must be accepted.Unger considers point (b) 'confused' (41). Roughly, he would argue that in the Charity Mugging case, we can satisfy the needs of some victims of poverty. The only reason we say that our aid cannot 'solve the problem' is that we are grouping together all people suffering from poverty, and we cannot satisfy all these people's needs. But in the Drowning Child case, we could conceptually group together that particular drowning child with all the other people suffering from anything bad anywhere in the world. So here, too, our aid cannot 'solve the problem' because we cannot stop all bad things in the world. Therefore, there really is no difference between the Charity Mugging and the Drowning Child. Unger's argument here depends upon the assumption that there is no distinction between more and less natural groupings.

25 Compare Unger 1996, chapter 6. Of course, one need only give up to the point at which further giving threatens one's survival or one's ability to give in the future.

26 Norcross 2003, 461; Shaw 1999, 286-7.

27 See Huemer 2005, chapter 5; 2007.

28 See Hardin (1974) for the inevitable exception, but see Sen (1994) for refutation of Hardin.

29 For pertinent statistics, see www.givingwhatwecan.org

30 It is plausible that one ought to give an amount that one feels is respectful. For a review of the most cost-effective charities, see Give Well (www.givewell.org). As of this writing, Give Well gives its highest ratings to the Against Malaria Foundation (https://www.againstmalaria.com/Donation.aspx) and Schistosomiasis Control Initiative (http://www.imperial.ac.uk/schistosomiasis-control-initiative). Both take credit card donations through the Internet.

31 King (1991, 74) saw himself as expressing respect for law by practicing his civil disobedience openly and with a willingness to accept the legally prescribed punishments. Rawls (1999, section 55) seeks to build these conditions into the definition of 'civil disobedience'. Here I consider disobedience to unjustified state commands, including both the child. As he does so, he regretfully tells the child, 'I am just doing my job. I don't make the rules.'

32 Rawls 1999, 322.

33 Though beatings are rarely prescribed by the state, there have been a number of laws with similar motivations and effects to the actions of this hypothetical gang. Until a 2003 Supreme Court Case (Lawrence v. Texas, 539 U.S. 558), sodomy was outlawed in a number of U.S. states. Many other countries still have such laws (see www.glapn.org/sodomylaws/world/world.htm), which seem to be aimed at harming homosexuals.

34 Abrahms 2006; 2011, 587-8.

35 Abrahms 2011, 589. Abrahms notes that attacks on military targets are more effective and account for most of the successes terrorists have had.

36 This argument depends on the assumption that a jury verdict is an assessment solely of whether the defendant performed the actions ascribed to him. Duane (1996) contends instead that a jury verdict is an assessment of the justice or appropriateness of punishing the defendant

37 Rawls 1964, 11-12. The context also contains some social contract imagery that I find unhelpful.

38 Estlund 2008, 157-8; Christiano 2008, 239-40; Klosko 2005, 116.

39 Christiano 2008, 250; Rawls 1964, 5. By 'the democratic assembly', Christiano means the legislature in a representative democracy.

40 Terminological note: capitalistic anarchism counts as an extreme form of libertarianism.

41 Rand 1964, 33; 1967, 195-6, 200-1.

42 Nozick 1974, 28-35. Nagel (1995, 148) reads Nozick as an absolutist, though in fact Nozick (1974, 30n) expresses some doubt about absolutism.

43 Narveson 1988, chapters 12-14.

Previous Chapter       Next Chapter

The Problem of Political Authority

An Examination of the Right to Coerce and the Duty to Obey

MichaelHuemer
Michael Huemer

Index

Previous
Chapter

Next
Chapter

AnarchoDollar-sm-tr Hogeye's Anarchism Page

books
Library of Liberty