I

The language of rights now dominates political debate. Does our government respect the moral and political rights of its citizens? Or does the government’s war policy, or its race policy, fly in the face of these rights? Do the minorities whose rights have been violated have the right to violate the law in return? Or does the silent majority itself have rights, including the right that those who break the law be punished? It is not surprising that these questions are now prominent. The concept of rights, and particularly the concept of rights against the government, has its most natural use when a political society is divided, and appeals to cooperation or a common goal are pointless.

The debate does not include the issue of whether citizens have some moral rights against their government. It seems accepted on all sides that they do. Conventional lawyers and politicians take it as a point of pride that our legal system recognizes, for example, individual rights of free speech, equality, and due process. They base their claim that our law deserves respect, at least in part, on that fact, for they would not claim that totalitarian systems deserve the same loyalty.

Some philosophers, of course, reject the idea that citizens have rights apart from what the law happens to give them. Bentham thought that the idea of moral rights was “nonsense on stilts.” But that view has never been part of our orthodox political theory, and politicians of both parties appeal to the rights of the people to justify a great part of what they want to do. I shall not be concerned, in this essay, to defend the thesis that citizens have moral rights against their governments; I want instead to explore the implications of that thesis for those, including our present government, who profess to accept it.

It is much in dispute, of course, what particular rights citizens have. Does the acknowledged right to free speech, for example, include the right to participate in nuisance demonstrations? In practice the government will have the last word on what an individual’s rights are, because its police will do what its officials and courts say. But that does not mean that the government’s view is necessarily the correct view; anyone who thinks it does must believe that men and women have only such moral rights as government chooses to grant, which means that they have no moral rights at all.

All this is sometimes obscured in the United States by our constitutional system. The Constitution provides a set of individual legal rights in the First Amendment, and in the due process, equal protection, and similar clauses. Under present legal practice the Supreme Court has the power to declare an act of Congress or of a state legislature void if the Court finds that the act offends these provisions. This practice has led some commentators to suppose that individual moral rights are fully protected by our system, but that is hardly so, nor could it be so.

The Constitution fuses legal and moral issues, by making the validity of a law depend on the answer to complex moral problems, like the problem of whether a particular statute respects the inherent equality of all men. This fusion has important consequences for the debates about civil disobedience; I have described these elsewhere¹ and I shall refer to them later. But it leaves open two prominent questions. It does not tell us whether the Constitution, even properly interpreted, recognizes all the moral rights our citizens have, and it does not tell us whether, as many suppose, citizens would have a duty to obey the law even if it did invade their moral rights.

Both questions become crucial when some minority claims moral rights which the law denies, like the right to run its local school system, and which lawyers agree are not protected by the Constitution. The second question becomes crucial when, as now, the majority is sufficiently aroused so that Constitutional amendments to eliminate rights, like the right against self-incrimination, are seriously proposed. It is also crucial in nations, like England, that have no constitution of our form.

Even if the Constitution were perfect, of course, and the majority left it alone, it would not follow that the Supreme Court could guarantee the individual rights of citizens. A Supreme Court decision is still a legal decision, and it must take into account precedent and institutional considerations like relations between the Court and Congress, as well as morality. And no judicial decision is necessarily the right decision. Judges stand for different positions on controversial issues of law and morals and, as the recent fights over Nixon’s Supreme Court nominations showed, a President is entitled to appoint judges of his own persuasion, provided that they are honest and capable.

So, though the constitutional system adds something to the protection of moral rights against the government, it falls far short of guaranteeing these rights, or even establishing what they are. It means that, on some occasions, a department other than the legislature has the last word on these issues, which can hardly satisfy someone who thinks such a department profoundly wrong.

It is of course inevitable that some department of government will have the final say on what law will be enforced. When men disagree about moral rights, there will be no way for either side to prove its case, and some decision must stand if there is not to be anarchy. But that piece of orthodox wisdom must be the beginning and not the end of a philosophy of legislation and enforcement. If we cannot insist that the government reach the right answers about the rights of its citizens, we can insist at least that it try. We can insist that it take rights seriously, follow a coherent theory of what these rights are, and act consistently with its own professions. I shall try to show what that means, and how it bears on the present political debates.

II

I shall start with the most violently argued issue. Does an American ever have the moral right to break a law? Suppose someone admits a law is valid; does he therefore have a duty to obey it? Those who try to give an answer seem to fall into two camps. The conservatives, as I shall call them, seem to disapprove of any act of disobedience; they appear satisfied when such acts are prosecuted, and disappointed when convictions are reversed. The other group, the liberals, are much more sympathetic to at least some cases of disobedience; they sometimes disapprove of prosecutions and celebrate acquittals. If we look beyond these emotional reactions, however, and pay attention to the arguments the two parties use, we discover an astounding fact. Both groups give essentially the same answer to the question of principle that supposedly divides them.

The answer that both parties give is this. In a democracy, or at least a democracy that in principle respects individual rights, each citizen has a general moral duty to obey all the laws, even though he would like some of them changed. He owes that duty to his fellow citizens, who obey laws that they do not like, to his benefit. But this general duty cannot be an absolute duty, because even a society that is in principle just may produce unjust laws and policies, and a man has duties other than his duties to the state. A man must honor his duties to his God and to his conscience, and if these conflict with his duty to the state, then he is entitled, in the end, to do what he judges to be right. If he decides that he must break the law, however, then he must submit to the judgment and punishment that the state imposes, in recognition of the fact that his duty to his fellow citizens was overwhelmed but not extinguished by his religious or moral obligation.

Of course this common answer can be elaborated in very different ways. Some would describe the duty to the state as fundamental, and picture the dissenter as a religious or moral fanatic. Others would describe the duty to the state in grudging terms, and picture those who oppose it as moral heroes. But these are differences in tone, and the position I described represents, I think, the view of most of those who find themselves arguing either for or against civil disobedience in particular cases.

I do not claim that it is everyone’s view. There must be some who put the duty to the state so high that they do not grant that it can ever be overcome. There are certainly some who would deny that a man ever has a moral duty to obey the law, at least in the United States today. But these two extreme positions are the slender tails of a bell curve, and all those who fall in between hold the orthodox position I described—that men have a duty to obey the law but have the right to follow their conscience when it conflicts with that duty.

But if that is so, then we have a paradox in the fact that men who give the same answer to a question of principle should seem to disagree so much, and to divide so fiercely, in particular cases. The paradox goes even deeper, for each party, in at least some cases, takes a position that seems flatly inconsistent with the theoretical position they both accept. This position is tested, for example, when someone evades the draft on grounds of conscience, or encourages others to commit this crime. Conservatives argue that such men must be prosecuted, even though they are sincere. Why must they be prosecuted? Because society cannot tolerate the decline in respect for the law that their act constitutes and encourages. They must be prosecuted, in short, to discourage them and others like them from doing what they have done.

But there seems to be a monstrous contradiction here. If a man has a right to do what his conscience tells him he must, then how can the state be justified in discouraging him from doing it? Is it not wicked for a state to forbid and punish what it acknowledges that men have a right to do?

Moreover, it is not just conservatives who argue that those who break the law out of moral conviction should be prosecuted. The liberal is notoriously opposed to allowing Southern school officials to go slow on segregation, even though he acknowledges that these school officials think they have a moral right to do what the law forbids. The liberal does not often argue, it is true, that the desegregation laws must be enforced to encourage general respect for law. He argues instead that the desegregation laws must be enforced because they are right. But his position also seems inconsistent: can it be right to prosecute men for doing what their conscience requires, when we acknowledge their right to follow their conscience?

We are therefore left with two puzzles. How can two parties to an issue of principle, each of which thinks it is in profound disagreement with the other, embrace the same position on that issue? How can it be that each side urges solutions to particular problems which seem flatly to contradict the position of principle that both accept? One possible answer is that some or all of those who accept the common position are hypocrites, paying lip service to rights of conscience which in fact they do not grant.

There is some plausibility to this charge. A sort of hypocrisy must be involved when public officials who claim to respect conscience deny Muhammad Ali the right to box in their states. If Muhammad Ali, in spite of his religious scruples, had joined the Army, he would have been allowed to box even though, on the principles these officials say they honor, he would have been a worse human being for having done so. But there are few cases that seem so straightforward as this one, and even here the officials do not seem to recognize the contradiction between their acts and their principles. So we must search for some explanation beyond the truth that men often do not mean what they say.

The deeper explanation lies in a set of confusions that often embarrass arguments about rights. These confusions have clouded all the issues I mentioned at the outset and have crippled attempts to develop a coherent theory of how a government that respects rights must behave.

In order to explain this, I must call attention to the fact, familiar to philosophers but often ignored in political debate, that the word “right” has different force in different contexts. In most cases when we say that someone has a “right” to do something, we imply that it would be wrong to interfere with his doing it, or at least that some special grounds are needed for justifying any interference. I use this strong sense of right when I say that you have the right to spend your money gambling, if you wish, though you ought to spend it in a more worthwhile way. I mean that it would be wrong for anyone to interfere with you even though you propose to spend your money in a way that I think is wrong.

There is a clear difference between saying that someone has a right to do something in this sense and saying that it is the “right” thing for him to do, or that he does no “wrong” in doing it. Someone may have the right to do something that is the wrong thing for him to do, as might be the case with gambling. Conversely, something may be the right thing for him to do and yet he may have no right to do it, in the sense that it would not be wrong for someone to interfere with his trying. If our army captures an enemy soldier, we might say that the right thing for him to do is to try to escape, but it would not follow that it is wrong of us to try to stop him. We might admire him for trying to escape, and perhaps even think less of him if he did not. But there is no suggestion here that it is wrong of us to stand in his way; on the contrary, if we think our cause is just, we think it right for us to do all we can to stop him.

Ordinarily this distinction, between the issues of whether a man has a right to do something and whether it is the right thing for him to do, causes no trouble. But sometimes it does, because sometimes we say that a man has a right to do something when we mean only to deny that it is the wrong thing for him to do. Thus we say that the captured soldier has a “right” to try to escape when we mean, not that we do wrong to stop him, but that he has no duty not to make the attempt. We also use “right” this way when we speak of someone having the “right” to act on his own principles, or the “right” to follow his own conscience. We mean that he does no wrong to proceed on his honest convictions, even though we disagree with these convictions, and even though, for policy or other reasons, we must force him to act contrary to them.

Suppose a man believes that welfare payments to the poor are profoundly wrong, because they sap enterprise, and so declares his full income tax each year but declines to pay half of it. We might say that he has a right to refuse to pay, if he wishes, but that the government has a right to proceed against him for the full tax, and to fine or jail him for late payment if that is necessary to keep the collection system working efficiently. We do not take this line in most cases; we do not say that the ordinary thief has a right to steal, if he wishes, so long as he pays the penalty. We say a man has the right to break the law, even though the state has a right to punish him, only when we think that, because of his convictions, he does no wrong in doing so.²

These distinctions enable us to see an ambiguity in the orthodox question: Does a man ever have a right to break the law? Does that question mean to ask whether he ever has a right to break the law in the strong sense, so that the government would do wrong to stop him, by arresting and prosecuting him? Or does it mean to ask whether he ever does the right thing to break the law, so that we should all respect him even though the government should jail him?

If we take the orthodox position to be an answer to the first—and most important—question, then the paradoxes I described arise. But if we take it as an answer to the second, they do not. Conservatives and liberals do agree that sometimes a man does not do the wrong thing to break a law, when his conscience so requires. They disagree, when they do, over the different issue of what the state’s response should be. Both parties do think that sometimes the state should prosecute. But this is not inconsistent with the proposition that the man prosecuted did the right thing in breaking the law.

The paradoxes seem genuine because the two questions are not usually distinguished, and the orthodox position is presented as a general solution to the problem of civil disobedience. But once the distinction is made, it is apparent that the position has been so widely accepted only because, when it is applied, it is treated as an answer to the second question but not the first. The crucial distinction is obscured by the troublesome idea of a right to conscience; this idea has been at the center of most recent discussions of political obligation, but it is a red herring drawing us away from the crucial political questions. The state of a man’s conscience may be decisive, or central, when the issue is whether he does something morally wrong in breaking the law; but it need not be decisive or even central when the issue is whether he has a right, in the strong sense of that term, to do so. A man does not have the right, in that sense, to do whatever his conscience demands, but he may have the right, in that sense, to do something even though his conscience does not demand it.

If that is true, then there has been almost no serious attempt to answer the questions that almost everyone means to ask. We can make a fresh start by stating these questions more clearly. Does an American ever have the right, in a strong sense, to do something which is against the law? If so, when? In order to answer these questions put in that way, we must try to become clearer about the implications of the idea, mentioned earlier, that citizens have at least some rights against their government.

I said that in the United States citizens are supposed to have certain fundamental rights against their government, certain moral rights made into legal rights by the Constitution. If this idea is significant, and worth bragging about, then these rights must be rights in the strong sense I just described. The claim that citizens have a right to free speech must imply that it would be wrong for the government to stop them from speaking, even when the government believes that what they will say will cause more harm than good. The claim cannot mean, on the prisoner-of-war analogy, only that citizens do no wrong in speaking their minds, though the government reserves the right to prevent them from doing so.

This is a crucial point, and I want to labor it. Of course a responsible government must be ready to justify anything it does, particularly when it limits the liberty of its citizens. But normally it is a sufficient justification, even for an act that limits liberty, that the act is calculated to increase what the philosophers call general utility—that it is calculated to produce more over-all benefit than harm. So, though the New York City government needs a justification for forbidding motorists to drive up Lexington Avenue, it is sufficient justification if the proper officials believe, on sound evidence, that the gain to the many will outweigh the inconvenience to the few. When individual citizens are said to have rights against the government, however, like the right of free speech, that must mean that this sort of justification is not enough. Otherwise the claim would not argue that individuals have special protection against the law when their rights are in play, and that is just the point of the claim.

Not all legal rights, or even Constitutional rights, represent moral rights against the government. I now have the legal right to drive either way on Fifty-seventh Street, but the government would do no wrong to make that street one-way, if it thought it in the general interest to do so. I have a Constitutional right to vote for a congressman every two years, but the national and state governments would do no wrong if, following the amendment procedure, they made a congressman’s term four years instead of two, again on the basis of a judgment that this would be for the general good.

But those Constitutional rights that we call fundamental, like the right of free speech, are supposed to represent rights against the government in the strong sense; that is the point of the boast that our legal system respects the fundamental rights of the citizen. If citizens have a moral right of free speech, then governments would do wrong to repeal the First Amendment that guarantees it, even if they were persuaded that the majority would be better off if speech were curtailed.

I must not overstate the point. Someone who claims that citizens have a right against the government need not go so far as to say that the state is never justified in overriding that right. He might say, for example, that although citizens have a right to free speech, the government may override that right when necessary to protect the rights of others, or to prevent a catastrophe, or even to obtain a clear and major public benefit (though if he acknowledged this last as a possible justification he would be treating the right in question as not among the most important or fundamental). What he cannot do is to say that the government is justified in overriding a right on the minimal grounds that would be sufficient if no such right existed. He cannot say that the government is entitled to act on no more than a judgment that its act is likely to produce, over-all, a benefit to the community. That admission would make his claim of a right pointless, and would show him to be using some sense of “right” other than the strong sense necessary to give his claim the political importance it is normally taken to have.

But then the answers to our two questions about disobedience seem plain, if unorthodox. In our society a man does sometimes have the right, in the strong sense, to disobey a law. He has that right whenever that law wrongly invades his rights against the government. If he has a moral right to free speech, that is, then he has a moral right to break any law that the government, by virtue of his right, had no right to adopt. The right to disobey the law is not a separate right, having something to do with conscience, additional to other rights against the government. It is simply a feature of these rights against the government, and it cannot be denied in principle without denying that any such rights exist.

These answers seem obvious once we take rights against the government to be rights in the strong sense I described. If I have a right to speak my mind on political issues, then the government does wrong to make it illegal for me to do so, even if it thinks this is in the general interest. If, nevertheless, the government does make my act illegal, then it does a further wrong to enforce that law against me. My right against the government means that it is wrong for the government to stop me from speaking; the government cannot make it right to stop me just by taking the first step.

This does not, of course, tell us exactly what rights men do have against the government. It does not tell us whether the right of free speech includes the right of demonstration. But it does mean that passing a law cannot affect such rights as men do have, and that is of crucial importance, because it dictates the attitude that an individual is entitled to take toward his personal decision when civil disobedience is in question.

Both conservatives and liberals suppose that in a society which is generally decent everyone has a duty to obey the law, whatever it is. That is the source of the “general duty” clause in the orthodox position, and though liberals believe that this duty can sometimes be “overridden,” even they suppose, as the orthodox position maintains, that the duty of obedience remains in some submerged form, so that a man does well to accept punishment in recognition of that duty. But this general duty is almost incoherent in a society that recognizes rights. If a man believes he has a right to demonstrate, then he must believe that it would be wrong for the government to stop him, with or without benefit of a law. If he is entitled to believe that, then it is silly to speak of a duty to obey the law as such, or of a duty to accept the punishment that the state has no right to give.

Conservatives will object to the short work I have made of their point. They will argue that even if the government was wrong to adopt some law, like a law limiting speech, there are independent reasons why the government is justified in enforcing the law once adopted. When the law forbids demonstration, then, so they argue, some principle more important then the individual’s right to speak is brought into play, namely the principle of respect for law. If a law, even a bad law, is left unenforced, then respect for law is weakened, and society as a whole suffers. So an individual loses his moral right to speak when speech is made criminal, and the government must, for the common good and for the general benefit, enforce the law against him.

But this argument, though popular, is plausible only if we forget what it means to say that an individual has a right against the state. It is far from plain that civil disobedience lowers respect for law, but even if we suppose that it does, this fact is irrelevant. The prospect of utilitarian gains cannot justify preventing a man from doing what he has a right to do, and the supposed gains in respect for law are simply utilitarian gains. There would be no point in the boast that we respect individual rights unless that involved some sacrifice, and the sacrifice in question must be that we give up whatever marginal benefits our country would receive from overriding these rights when they prove inconvenient. So the general benefit cannot be a good ground for abridging rights, even when the benefit in question is a heightened respect for law.

But perhaps I do wrong to assume that the argument about respect for law is only an appeal to general utility. I said that a state may be justified in overriding or limiting rights on other grounds, and we must ask, before rejecting the conservative position, whether any of these apply. The most important—and least well understood—of these other grounds invokes the notion of competing rights that would be jeopardized if the right in question were not limited. Citizens have personal rights to the state’s protection as well as personal rights to be free from the state’s interference, and it may be necessary for the government to choose between these two sorts of rights. The law of defamation, for example, limits the personal right of any man to say what he thinks, because it requires him to have good grounds for what he says. But this law is justified, even for those who think that it does invade a personal right, by the fact that it protects the right of others not to have their reputations ruined by a careless statement.

The individual rights that our society acknowledges often conflict in this way, and when they do it is the job of government to discriminate. If the government makes the right choice, and protects the more important at the cost of the less, then it has not weakened or cheapened the notion of a right; on the contrary it would have done so had it failed to protect the more important of the two. So we must acknowledge that the government has a reason for limiting rights if it plausibly believes that a competing right is more important.

May the conservative seize on this fact? He might argue that I did wrong to characterize his argument as one that appeals to the general benefit, because it appeals instead to competing rights, namely the moral right of the majority to have its laws enforced, or the right of society to maintain the degree of order and security it wishes. These are the rights, he would say, that must be weighed against the individual’s right to do what the wrongful law prohibits.

But this new argument is confused, because it depends on yet another ambiguity in the language of rights. It is true that we speak of the “right” of society to do what it wants, but this cannot be a “competing right” of the sort that might justify the invasion of a right against the government. The existence of rights against the government would be jeopardized if the government were able to defeat such a right by appealing to the right of a democratic majority to work its will. A right against the government must be a right to do something even when the majority thinks it would be wrong to do it, and even when the majority would be worse off for having it done. If we now say that society has a right to do whatever is in the general benefit, or the right to preserve whatever sort of environment the majority wishes to live in, and we mean that these are the sort of rights that provide justification for overruling any rights against the government that may conflict, then we have annihilated the latter rights.

In order to save them, we must recognize as competing rights only the rights of other members of the society as individuals. We must distinguish the “rights” of the majority as such, which cannot count as a justification for overruling individual rights, and the personal rights of members of a majority, which might well count. The test we must use is this. Someone has a competing right to protection, which must be weighed against an individual right to act, if that person would be entitled to demand that protection from his government on his own title, as an individual, without regard to whether a majority of his fellow citizens joined in the demand.

It cannot be true, on this test, that anyone has a right to have all the laws of the nation enforced. He has a right to have enforced only those criminal laws that he would have a right to have enacted if they were not already law. The laws against personal assault may well fall into that class. If the physically vulnerable members of the community—those who need police protection against personal violence—were only a small minority, it would still seem plausible to say that they were entitled to that protection. But the laws that provide a certain level of quiet in public places, or that authorize and finance a foreign war, cannot be thought to rest on individual rights. The timid lady on the streets of Chicago is not entitled to just the degree of quiet that now obtains, nor is she entitled to have boys drafted to fight in wars she approves. There are laws—perhaps desirable laws—that provide these advantages for her, but the justification for these laws, if they can be justified at all, is the common desire of a large majority, not her personal right. If, therefore, these laws do abridge someone else’s moral right to protest, or his right to personal security, she cannot urge a competing right to justify the abridgment. She has no personal right to have such laws passed, and she has no competing right to have them enforced either.

So the conservative cannot advance his argument much on the ground of competing rights, but he may want to use another ground. A government, he may argue, may be justified in abridging the personal rights of its citizens in an emergency, or when a very great loss may be prevented, or, perhaps, when some major benefit can clearly be secured. If the nation is at war, a policy of censorship may be justified even though it invades the right to say what one thinks on matters of political controversy. But the emergency must be genuine. There must be what Oliver Wendell Holmes described as a clear and present danger, and the danger must be one of magnitude.

Can the conservative argue that when any law is passed, even a wrongful law, this sort of justification is available for enforcing it? His argument might be something of this sort. If the government once acknowledges that it may be wrong—that the legislature might have adopted, the executive approved, and the courts left standing, a law that in fact abridges important rights—then this admission will lead not simply to a marginal decline in respect for law, but to a crisis of order. Citizens may decide to obey only those laws they personally approve, and that is anarchy. So the government must insist that whatever a citizen’s rights may be before a law is passed and upheld by the courts, his rights thereafter are determined by that law.

But this argument ignores the primitive distinction between what may happen and what will happen. If we allow speculation to support the justification of emergency or decisive benefit, then, again, we have annihilated rights. We must, as Learned Hand said, discount the gravity of the evil threatened by the likelihood of reaching that evil. I know of no genuine evidence to the effect that tolerating some civil disobedience, out of respect for the moral position of its authors, will increase such disobedience, let alone crime in general. The case that it will must be based on vague assumptions about the contagion of ordinary crimes, assumptions that are themselves unproved, and that are in any event largely irrelevant. It seems at least as plausible to argue that tolerance will increase respect for officials and for the bulk of the laws they promulgate, or at least retard the rate of growing disrespect.

If the issue were simply the question whether the community would be marginally better off under strict law enforcement, then the government would have to decide on the evidence we have, and it might not be unreasonable to decide, on balance, that it would. But since rights are at stake, the issue is the very different one of whether tolerance would destroy the community or threaten it with great harm, and it seems to me simply mindless to suppose that the evidence makes that probable or even conceivable.

The argument from emergency is confused in another way as well. It assumes that the government must take the position either that a man never has the right to break the law, or that he always does. I said that any society that claims to recognize rights at all must abandon the notion of a general duty to obey the law that holds in all cases. This is important, because it shows that there are no short cuts to meeting a citizen’s claim of right. If a citizen argues that he has a moral right not to serve in the army, or to protest in a way he finds effective, then an official who wants to answer him, and not simply bludgeon him into obedience, must respond to the particular points he makes, and cannot point to the draft law or a Supreme Court decision as having even special, let alone decisive, weight. Sometimes an official who considers the citizen’s moral arguments in good faith will be persuaded that the citizen’s claim is plausible, or even right. It does not follow, however, that he will always be persuaded or that he always should be.

I must emphasize that all these propositions concern the strong sense of right, and they therefore leave open important questions about the right thing to do. If a man believes he has the right to break the law, he must then ask whether he does the right thing to exercise that right. He must remember that reasonable men can differ about whether he has a right against the government, and therefore the right to break the law, that he thinks he has; and therefore that reasonable men can oppose him in good faith. He must take into account the various consequences his acts will have, whether they involve violence, and such other considerations as the context makes relevant; he must not go beyond the rights he can in good faith claim, to acts that violate the rights of others.

On the other hand, if some official, like a prosecutor, believes that the citizen does not have the right to break the law, then he must ask whether he does the right thing to enforce it. In the article I mentioned earlier I argued that certain features of our legal system, and in particular the fusion of legal and moral issues in our Constitution, mean that citizens often do the right thing in exercising what they take to be moral rights to break the law, and that prosecutors often do the right thing in failing to prosecute them for it. I will not repeat those arguments here; instead I want to ask whether the requirement that government take its citizens’ rights seriously has anything to do with the crucial question of what these rights are.

III

The argument so far has been hypothetical: if a man has a particular moral right against the government, that right survives contrary legislation or adjudication. But this does not tell us what rights he has, and it is notorious that reasonable men disagree about that. There is wide agreement on certain clear-cut cases; almost everyone who believes in rights at all would admit, for example, that a man has a moral right to speak his mind in a non-provocative way on matters of political concern, and that this is an important right that the state must go to great pains to protect. But there is great controversy as to the limits of such paradigm rights, and the so-called “anti-riot” law involved in the Chicago Seven trial is a case in point.

The defendants were accused of conspiring to cross state lines with the intention of causing a riot. This charge is vague—perhaps unconstitutionally vague—but the law apparently defines as criminal emotional speeches which argue that violence is justified in order to secure political equality. Does the right of free speech protect this sort of speech? That, of course, is a legal issue, because it invokes the free speech clause of the First Amendment of the Constitution. But it is also a moral issue, because, as I said, we must treat the First Amendment as an attempt to protect a moral right. It is part of the job of governing to “define” moral rights through statutes and judicial decisions, that is, to declare officially the extent that moral rights will be taken to have in law. Congress faced this task in voting on the anti-riot bill, and the Supreme Court will face it if the Chicago Seven case goes that far. How should the different departments of government go about defining moral rights?

They should begin with a sense that whatever they decide might be wrong. History and their descendants may judge that they acted unjustly when they thought they were right. If they take their duty seriously, they must try to limit their mistakes, and they must therefore try to discover where the dangers of mistake lie.

They might choose one of two very different models for this purpose. The first model recommends striking a balance between the rights of the individual and the demands of society at large. If the government infringes on a moral right (for example, by defining the right of free speech more narrowly than justice requires), then it has done the individual a wrong. On the other hand, if the government inflates a right (by defining it more broadly than justice requires) then it cheats society of some general benefit, like safe streets, that it is perfectly entitled to have. So a mistake on one side is as serious as a mistake on the other. The course of government is to steer to the middle, to balance the general good and personal rights, giving to each its due.

When the government, or any branch, defines a right, it must bear in mind, according to the first model, the social cost of different proposals, and make the necessary adjustments. It must not grant the same freedom to noisy demonstrations as it grants to calm political discussion, for example, because the former causes much more trouble than the latter. Once it decides how much of a right to recognize, it must enforce its decision to the full. That means permitting an individual to act within his rights, as the government has defined them, but not beyond, so that if anyone breaks the law, even on grounds of conscience, he must be punished. No doubt any government will make mistakes, and will regret decisions once taken. That is inevitable. But this middle policy will ensure that errors on one side will balance out errors on the other over the long run.

The first model, described in this way, has great plausibility, and most laymen and lawyers, I think, would respond to it warmly. The metaphor of balancing the public interest against personal claims is established in our political and judicial rhetoric, and this metaphor gives the model both familiarity and appeal. Nevertheless, the first model is a false one, certainly in the case of rights generally regarded as important, and the metaphor is the heart of its error.

The institution of rights against the government is not a gift of God, or an ancient ritual, or a national sport. It is a complex and troublesome practice that makes the government’s job of securing the general benefit more difficult and more expensive, and it would be a frivolous and wrongful practice unless it served some point. Anyone who professes to take rights seriously, and who praises our government for respecting them, must have some sense of what that point is. He must accept, at the minimum, one or both of two important ideas. The first is the vague but powerful idea of human dignity. This idea, associated with Kant, but defended by philosophers of different schools, supposes that there are ways of treating a man that are inconsistent with recognizing him as a full member of the human community, and holds that such treatment is profoundly unjust.

The second is the more familiar idea of political equality. This supposes that the weaker members of a political community are entitled to the same concern and respect of their government as the more powerful members have secured for themselves, so that if some men have freedom of decision whatever the effect on the general good, then all men must have the same freedom. I do not want to defend or elaborate these ideas here, but only to insist that anyone who claims that citizens have rights must accept ideas very close to these.³

It makes sense to say that a man has a fundamental right against the government, in the strong sense, like free speech, if that right is necessary to protect his dignity, or his standing as equally entitled to concern and respect, or some other personal value of like consequence. It does not make sense otherwise.

So if rights make sense at all, then the invasion of a relatively important right must be a very serious matter. It means treating a man as less than a man, or as less worthy of concern than other men. The institution of rights rests on the conviction that this is a grave injustice, and that it is worth paying the incremental cost in social policy or efficiency that is necessary to prevent it. But then it must be wrong to say that inflating rights is as serious as invading them. If the government errs on the side of the individual, then it simply pays a little more in social efficiency than it has to pay; it pays a little more, that is, of the same coin that it has already decided must be spent. But if it errs against the individual, it inflicts an insult upon him that, on its own reckoning, it is worth a great deal of that coin to avoid.

So the first model is indefensible. It rests, in fact, on a mistake I discussed earlier, namely the confusion of society’s rights with the rights of members of society. “Balancing” is appropriate when the government must choose between competing claims of right—between the Southerner’s claim to freedom of association, for example, and the black man’s claim to an equal education. Then the government can do nothing but estimate the merits of the competing claims, and act on its estimate. The first model assumes that the “right” of the majority is a competing right that must be balanced in this way; but that, as I argued before, is a confusion that threatens to destroy the concept of individual rights. It is worth noticing that the community rejects the first model in that area where the stakes for the individual are highest, the criminal process. We say that it is better that a great many guilty men go free than that one innocent man be punished, and that homily rests on the choice of the second model for government.

The second model treats abridging a right as much more serious than inflating one, and its recommendations follow from that judgment. It stipulates that once a right is recognized in clear-cut cases, then the government should act to cut off that right only when some compelling reason is presented, some reason that is consistent with the suppositions on which the original right must be based. It cannot be an argument for curtailing a right, once granted, simply that society would pay a further price in extending it. There must be something special about that further cost, or there must be some other feature of the case, that makes it sensible to say that although great social cost is warranted to protect the original right, this particular cost is not necessary. Otherwise, the government’s failure to extend the right will show that its recognition of the right in the original case is a sham, a promise that it intends to keep only until that becomes inconvenient.

How can we show that a particular cost is not worth paying without taking back the initial recognition of a right? I can think of only three sorts of grounds that can consistently be used to limit the definition of a particular right. First, the government might show that the values protected by the original right are not really at stake in the marginal case, or are at stake only in some attenuated form. Second, it might show that if the right is defined to include the marginal case, then some competing right, in the strong sense I described earlier, would be abridged. Third, it might show that if the right were so defined, then the cost to society would not be simply incremental, but would be of a degree far beyond the cost paid to grant the original right, a degree great enough to justify whatever assault on dignity or equality might be involved.

It is fairly easy to apply these grounds to one problem the Supreme Court has recently faced, and must face soon again. The draft law provides an exemption for conscientious objectors, but this exemption, as interpreted by the draft boards, has been limited to those who object to all wars on religious grounds. If we suppose that the exemption is justified on the ground that an individual has a moral right not to kill in violation of his own principles, then the question is raised whether it is proper to exclude those whose morality is not based on religion, or whose morality is sufficiently complex to distinguish among wars. The Court has just held that the draft boards are wrong to exclude the former, and it will soon be asked to decide whether they are wrong to exclude the latter as well.

None of the three grounds I listed can justify either of these exclusions. The invasion of personality in forcing men to kill when they believe killing immoral is just as great when these beliefs are based on secular grounds, or take account of the fact that wars differ in morally relevant ways, and there is no pertinent difference in competing rights or in national emergency. There are differences among the cases, of course, but they are insufficient to justify the distinction. A government that is secular on principle cannot prefer a religious to a non-religious morality as such. There are utilitarian arguments in favor of limiting the exception to religious or universal grounds—an exemption so limited may be less expensive to administer, and may allow easier discrimination between sincere and insincere applicants. But these utilitarian reasons are irrelevant, because they cannot count as grounds for limiting a right.

What about the anti-riot law, as applied in the Chicago trial? Does that law represent an improper limitation of the right to free speech, supposedly protected by the First Amendment? If we were to apply the first model for government to this issue, the argument for the anti-riot law would look strong. But if we set aside talk of balancing as inappropriate, and turn to the proper grounds for limiting a right, then the argument becomes a great deal weaker. The original right of free speech must suppose that it is an assault on human personality to stop a man from expressing what he honestly believes, particularly on issues affecting how he is governed. Surely the assault is greater, and not less, when he is stopped from expressing those principles of political morality that he holds most passionately, in the face of what he takes to be outrageous violations of these principles.

It may be said that the anti-riot law leaves him free to express these principles in a non-provocative way. But that misses the point of the connection between expression and dignity. A man cannot express himself freely when he cannot match his rhetoric to his outrage, or when he must trim his sails to protect values he counts as nothing next to those he is trying to vindicate. It is true that some political dissenters speak in ways that shock the majority, but it is arrogant for the majority to suppose that the orthodox methods of expression are the proper ways to speak, for this is a denial of equal concern and respect. If the point of the right is to protect the dignity of dissenters, then we must make judgments about appropriate speech with the personalities of the dissenters in mind, not the personality of the “silent” majority for whom the anti-riot law is no restraint at all.

So the argument that the personal values protected by the original right are less at stake in this marginal case fails. We must consider whether competing rights, or some grave threat to society, nevertheless justify the anti-riot law. We can consider these two grounds together, because the only plausible competing rights are rights to be free from violence, and violence is the only plausible threat to society that the context provides.

I have no right to burn your house, or stone you or your car, or swing a bicycle chain against your skull, even if I find these natural means of expression. But the defendants in the Chicago trial were not accused of direct violence; the argument runs that the acts of speech they planned make it likely that others would do acts of violence, either in support of or out of hostility to what they said. Does this provide a justification?

The question would be different if we could say with any confidence how much and what sort of violence the anti-riot law might be expected to prevent. Will it save two lives a year, or two hundred, or two thousand? Two thousand dollars of property, or two hundred thousand, or two million? No one can say, not simply because prediction is next to impossible, but because we have no firm understanding of the process by which demonstration disintegrates into riot, and in particular of the part played by inflammatory speech, as distinct from poverty, police brutality, blood lust, and all the rest of human and economic failure. The government must try, of course, to reduce the violent waste of lives and property, but it must recognize that any attempt to locate and remove a cause of riot, short of a reorganization of society, must be an exercise in speculation, trial and error. It must make its decisions under conditions of high uncertainty, and the institution of rights, taken seriously, limits its freedom to experiment under such conditions.

It forces the government to bear in mind that preventing a man from speaking or demonstrating offers him a certain and profound insult, in return for a speculative benefit that may in any event be achieved in other if more expensive ways. When lawyers say that rights may be limited to protect other rights, or to prevent catastrophe, they have in mind cases in which cause and effect are relatively clear, like the familiar example of a man falsely crying fire in a crowded theater.

But the Chicago story shows how obscure the causal connections can become. Were the speeches of Hoffman or Rubin necessary conditions of the riot? Or had thousands of people come to Chicago for the purposes of rioting anyway, as the government also argues? Were they in any case sufficient conditions? Or could the police have contained the violence if they had not been so busy contributing to it, as the staff of the President’s Commission on Violence said they were?

These are not easy questions, but if rights mean anything, then the government cannot simply assume answers that justify its conduct. If a man has a right to speak, if the reasons that support that right extend to provocative political speech, and if the effects of such speech on violence are unclear, then the government is not entitled to make its first attack on that problem by denying that right. It may be that abridging the right to speak is the least expensive course, or the least damaging to police morale, or the most popular politically. But these are utilitarian arguments in favor of starting one place rather than another, and such arguments are ruled out by the concept of rights.

This point may be obscured by the popular belief that political activists look forward to violence and “ask for trouble” in what they say. They can hardly complain, in the general view, if they are taken to be the authors of the violence they expect, and treated accordingly. But this repeats the confusion I tried to explain earlier between having a right and doing the right thing. The speaker’s motives may be relevant in deciding whether he does the right thing in speaking passionately about issues that may inflame or enrage the audience. But if he has a right to speak, because the danger in allowing him to speak is speculative, his motives cannot count as independent evidence in the argument that justifies stopping him.

But what of the individual rights of those who will be destroyed by a riot, of the passer-by who will be killed by a sniper’s bullet or the shopkeeper who will be ruined by looting? To put the issue in this way, as a question of competing rights, suggests a principle that would undercut the effect of uncertainty. Shall we say that some rights to protection are so important that the government is justified in doing all it can to maintain them? Shall we therefore say that the government may abridge the rights of others to act when their acts might simply increase the risk, by however slight or speculative a margin, that some person’s rights to life or property will be violated?

Some such principle is relied on by those who oppose the Supreme Court’s recent liberal rulings on police procedure. These rulings increase the chance that a guilty man will go free, and therefore marginally increase the risk that any particular member of the community will be murdered or raped or robbed. Some critics believe that the Court’s decisions must therefore be wrong.

But no society that purports to recognize a variety of rights, on the ground that a man’s dignity or equality may be invaded in a variety of ways, can accept such a principle. If forcing a man to testify against himself, or forbidding him to speak, does the damage that the rights against self-incrimination and the right of free speech assume, then it would be contemptuous for the state to tell a man that he must suffer this damage against the possibility that other men’s risk of loss may be marginally reduced. If rights make sense, then the degrees of their importance cannot be so different that some count not at all when others are mentioned.

Of course the government may discriminate and may stop a man from exercising his right to speak when there is a clear and substantial risk that his speech will do great damage to the person or property of others, and no other means of preventing this are at hand, as in the case of the man falsely shouting fire in a theater. But we must reject the suggested principle that the government can ignore rights to speak when life and property are in question. So long as the impact of speech on these other rights remains speculative and marginal, it must look elsewhere for levers to pull.

IV

I said at the beginning of this essay that I wanted to show what a government must do that professes to recognize individual rights. It must dispense with the claim that citizens never have a right to break its law, and it must not define citizens’ rights so that these are cut off for supposed reasons of the general good. The present government’s policy toward civil disobedience, and its campaign against vocal protest, its enforcement of the anti-riot law, may therefore be thought to count against its sincerity.

One might well ask, however, whether it is wise to take rights all that seriously after all. America’s genius, at least in her own legend, lies in not taking any abstract doctrine to its logical extreme. It may be time to ignore abstractions, and concentrate instead on giving the majority of our citizens a new sense of their government’s concern for their welfare, and of their title to rule.

That, in any event, is what the Vice President seems to believe. In a recent policy statement on the issue of weirdos and social misfits, he said that the liberals’ concern for individual rights was a headwind blowing in the face of the ship of state. That is a poor metaphor, but the philosophical point it expresses is very well taken. He recognizes, as many liberals do not, that the majority cannot travel as fast or as far as it would like if it recognizes the rights of individuals to do what, in the majority’s terms, is the wrong thing to do.

The Vice President supposes that rights are divisive, and that national unity and a new respect for law may be developed by taking them more skeptically. But he is wrong. Our country will continue to be divided by its social and foreign policy, and if the economy grows weaker the divisions will become more bitter. If we want our laws and our legal institutions to provide the ground rules within which these issues will be contested, then these ground rules must not be the conqueror’s law that the dominant class imposes on the weaker, as Marx supposed the law of a capitalist society must be. The bulk of the law—that part which defines and implements social, economic, and foreign policy—cannot be neutral. It must state, in its greatest part, the majority’s view of the common good. The institution of rights is therefore crucial, because it represents the majority’s promise to the minorities that their dignity and equality will be respected. When the divisions among the groups are most violent, then this gesture, if law is to work, must be most sincere.

The institution requires an act of faith on the part of the minorities, because the scope of their rights will be controversial whenever they are important, and because the officers of the majority will act on their own notions of what these rights really are. Of course these officials will disagree with many of the claims that a minority makes. That makes it all the more important that they take their decisions gravely. They must show that they understand what rights are, and they must not cheat on the full implications of the doctrine. The government will not re-establish respect for law without giving the law some claim to respect. It cannot do that if it neglects the one feature that distinguishes law from ordered brutality. If the government does not take rights seriously, then it does not take law seriously either.