A Reply to Reiman

Shortly after I published this essay, Jeffrey H. Reiman mounted a full-scale attack on its theses in an answering essay, In Defense of Political Philosophy. Harper & Row has graciously agreed to my proposal for a new edition of my book in which I reply to Reiman's critique, and I must now try to defend myself as best I can against his forceful onslaught. Usually, this sort of exchange is reserved for learned journals or the letters columns of journals of opinion, books being rather clumsy vehicles for genuine, as opposed to stage-managed, dialogues.

The reactions to my essay have been varied, as one might expect. Some critics have thought I was merely irrelevant, others that I was logically confused, and at least one has suggested that it was immoral of me to publish the book at all. Paul T. Menzel sums the situation up nicely in the opening paragraph of an essay designed to defend me, after a fashion, against my enemies:

If published reactions are any indication of the response of the philosophical community to a philosopher's argument, Robert Paul Wolff's In Defense of Anarchism has perhaps persuaded and impressed its readers as little as any recently published work.

As I understand Reiman, his defense of political philosophy proceeds in four stages, most of which are completed in the first two chapters of his book. Since my analysis of his argument will at some points become rather detailed, it will help to have the whole argument before us in outline form before we proceed to a criticism of it. The steps are as follows:

1. The definition of "authority" which I offer is logically confused, a contradictio in adjecto as he puts it. Such "moral authority" is directly inconsistent with the conditions of being a moral agent in general. Hence, if my aim is to argue that no one ever has authority in that sense of the term, I could more than adequately have handled the subject in a paragraph.
2. But from the impossibility of moral authority, nothing follows for politics (or indeed, for anything else), because it is political authority that states claim and anarchists deny. Political authority is the right to use coercion to compel compliance with commands. States do indeed claim that sort of authority, and the task of political philosophy is, in the first instance, to determine the circumstances under which such claims are valid.
3. Broadly speaking, the criterion of legitimacy of political authority -- that is to say, the standard for evaluating a state's claim to the right to compel compliance to its commands by coercion -- is the goodness of the consequences of the totality of laws which constitute the political system. In evaluating a state's claim to legitimate political authority, we must compare the probable consequences of its system of laws with those of other systems which might reasonably be considered possible substitutes, as well as with the probable consequences of an absence of any system of laws at all.
4. When a state passes the test of legitimacy, its possession of political authority creates a prima facie obligation on the part of the citizen to obey the law. That obligation is not overriding, but it is also not negligible. There will be a convergence of legitimate laws and the citizen's duty, but not (as I suggested) an absolute identity. In some real life situations, the prima facie obligation to obey the law will be overridden by other, weightier obligations. Therefore, legitimate political authority is compatible with moral autonomy. Wolff is wrong; political philosophy lives.

Reiman has some other things to say about the inadequacies of my analysis, including the historical observation that anarchists care about coercion, not about mere moral autonomy. (He is partly right and partly wrong about this.) But these four steps embody the heart of his refutation, and if I am to sustain my position, I must deal with them directly. As we shall see, Reiman's argument turns on two points: the definition of political authority, and the notion of a prima facie obligation to obey the commands of a legitimate state. Now let us take a look at the argument step by step.

Reiman begins by arguing that I have posed a pseudo-problem to which there neither could be nor need be a solution. How can the moral autonomy of the individual, I ask, be made compatible with the (moral) authority of the state? It can't, Reiman answers, because the notion of authority as the right to be obeyed is simply contradictory to the notion of moral obligation itself. A "philosopher of mathematics" might just as sensibly ask how one can make a triangle out of two straight lines.

Now, Reiman and I agree completely on this point, so why do we seem to be on opposite sides of the fence? I suggest that we are actually in serious disagreement about three other propositions, all of which are worth disputing. First of all, Reiman thinks it is obvious that what he calls moral authority is an incoherent notion. I agree that it is incoherent, but I don't think that fact is quite so obvious as he imagines. During most of the two and one half millennia that moral philosophy has been written in West, claims of moral authority have found acceptance among respectable thinkers. The Catholic Church in particular, and western religion in general, have demanded obedience either to God's commands or to the edicts of His church. Despite the presence of "protestant" strains in Christian thought before the modern era, it is only in the past four hundred years that anything like a doctrine of moral autonomy has gained currency. Immanuel Kant is the great spokesman for the thesis that every single rational agent must decide for himself what is right, and the books in which he so persuasively argued that claim are not yet two hundred years old. Reiman acknowledges that states make the sort of authority claim I describe, but he isn't impressed by that fact. As he puts it:

Let us not be deceived by the fact that all states attempt to engender in their subjects a feeling or belief that there is a duty to obey the law or the ruler simply because it is the law or he is the ruler. The question is, is the moral duty to obey essential to the concept of political authority? We shall argue that it is not. (p. 19)

This leads us to our second disagreement, which concerns the significance of classical democratic theory. I believe that social contract theory, with its central reliance upon a strict notion of the consent of the governed, is a direct attempt to overcome the conflict between the primary demand of moral agency, which is autonomy, and the primary demand of state authority, which is obedience. If the criterion of legitimacy is good effects, then what is the point of demanding that those who are ruled must consent to be ruled? Or -- to put the point in its stricter and more correct form -- that those who are commanded must, either in their own persons or through their representatives, be the authors of the commands? One might, of course, point out that men are more tractable when they imagine that they have had a role in the making of the laws, but that is merely an argument for social engineering, not for democracy. Reiman's failure to confront democratic theory's premise of popular sovereignty undermines his entire argument. Indeed, when he finally does speak of consent in Chapter Five, he contradicts everything he has said earlier in his analysis of political authority.

The third disagreement between Reiman and myself directly concerns this notion of political authority. Reiman is willing to treat the state's claim to obedience as a mere "attempt to engender ... a feeling or belief" because he thinks the case for legitimate authority can be made on independent grounds. The legitimate state can be distinguished from the prepolitical state of nature without any invocation of a duty to obey, according to Reiman. I shall show that he is wrong.

Finally, Reiman disputes my treatment of the special case of unanimous direct democracy. I claimed that when the citizens of a state collectively make all the laws by unanimous vote, the autonomy of the individual is compatible with the authority of the state. Reiman says that this is nonsense. He is quite right. It may be that men are bound by the collective commitments they make, but such commitments do not create the sort of political authority I was attempting to analyse. I stand corrected.

In my definition of authority, I focused on the state's claim of a right to be obeyed. Reiman, instead, focuses on the means which the state claims to have a right to use in order to secure compliance. "Political authority," he says, "is the right to make commands and use coercion to discourage noncompliance with them" (p. 18). Who is right, Reiman or I?

In a sense, this is just a pointless dispute about words. I want to use the word "authority" to mean "the right to be obeyed." Reiman wants to use the word "authority" to mean "the right to use coercion to secure compliance." Reiman agrees with me that no one ever has authority in my sense of the word; and I agree with Reiman (as we shall see) that men frequently have authority in his sense of the word. So except for some clout that the word itself may have acquired through long centuries of use and abuse, why do we bother to argue? The answer will emerge, but it will take a bit more discussion before it becomes clear.

First of all, let me grant a point and make a point. I agree with Reiman that states claim the right to use coercion to compel compliance. But Reiman, in invoking Max Weber's definition of authority (as I had also done), fails to notice a most significant clause in that definition. Here are Weber's words:

A compulsory political association will be called a "state" if and in so far as its administrative staff successfully upholds a claim to the monopoly of the legitimate use of physical force in the enforcement of its order.

In the absence of a legitimate political authority, men and women confront one another as moral agents simpliciter. Reiman and I agree with Locke that in such a prepolitical "state of nature," persons have moral obligations and rights, among which presumably are the obligation to respect the legitimate pursuits of others and the right of self-defense.¹ Each moral agent must decide for himself what he ought to do and that decision includes a choice of the means by which he shall do it. Although anyone in a state of nature is free to claim a monopoly of the legitimate use of force, no one can demonstrate such a claim in principle.²

Whether and when you think a man has the right to use force in the state of nature will depend on the moral principles you believe to be valid. A total pacifist might hold that no one in a state of nature has any right to use force against another person in pursuit of his ends, even in the extreme case of self-defense against an attack on his life. Those with different moral convictions might claim that one had the right to use force in order to defend one's own life or the lives of one's family, or the life of any person under unwarranted attack. Still others would hold that force may legitimately be used in defense of one's possessions or style of life. And some would assert that in a state of nature, each man has the right to use force to accomplish whatever ends he has set himself, even when such ends interfere with the lives or possessions of others. It makes no difference for the present dispute what particular views Reiman and I hold, so long as neither of us is an absolute pacifist. I am not, and the text of In Defense of Political Philosophy seems very strongly to suggest that Reiman is not either. So we can take it as agreed between us that in the prepolitical condition, in the absence of a legitimate state, there are at least some situations in which individuals have the moral right to use force to accomplish their ends.

Now, the common theme of classical contractarian theory is that men relinquish their natural right to self-protection and the use of force when they collectively bring the commonwealth or republic or state into existence. Indeed, that giving-up of the individual right to coercion is the defining mark of the social contract. If men do not renounce the private use of force, then no public body has been created by the compact. There may be a meeting of the minds or a community of interests or a substitution of negotiation for the war of all against all, but there is no state. This point seems too obvious to belabor, but since it goes to the heart of Reiman's confusion, let us devote a page or so to some passages from the familiar texts. First Locke:

Man being born, as has been proved, with a title to perfect freedom, and an uncontrolled enjoyment of all the rights and privileges of the law of nature equally with any other man or number of men in the world, hath by nature a power not only to preserve his property -- this is, his life, liberty, and estate -- against the injuries and attempts of other men, but to judge of and punish the breaches of that law in others as he is persuaded the offence deserves, even with death itself, in crimes where the heinousness of the fact in his opinion requires it. But because no political society can be nor subsist without having in itself the power to preserve the property, and, in order thereunto, punish the offences of all those of that society, there, and there only, is political society, where every one of the members hath quitted this natural power, resigned it up into the hands of the community in all cases that exclude him not from appealing for protection to the law established by it; and thus all private judgment of every particular member being excluded, the community comes to be umpire; and by understanding indifferent rules and men authorized by the community for their execution, decides all the differences that may happen between any members of that society concerning any matter of right, and punishes those offences which any member hath committed against the society with such penalties as the law has established; whereby it is easy to discern who are and who are not in political society together. Those who are united into one body, and have a common established law and judicature to appeal to, with authority to decide controversies between them and punish offenders, are in civil society one with another; but those who have no such common appeal -- I mean on earth -- are still in the state of nature, each being, where there is no other judge for himself and executioner, which is, as I have before shown it, the perfect state of nature. (Second Treatise Concerning Civil Government, chapter 7)

Now Rousseau. Rousseau takes it for granted that the formation of civil society involves each man's forfeiture of the right to use his force at his own discretion. The real problem for political philosophy, as Rousseau sees it, is how to ensure that each member of the body politic, while thus alienating his natural liberty in order to gain a measure of security, shall nevertheless "obey only himself and remain as free as before." The solution, he claims, lies in the terms of the social contract:

Each of us places in common his person and all his power under the supreme direction of the general will; and as one body we all receive each member as an indivisible part of the whole. (The Social Contract, book 1, chapter 6)

The effect of the social contract is to destroy the moral symmetry of the state of nature. Such a symmetry is anarchic just in the original sense that there is no rule, no party possessing rights of decision and enforcement which are not possessed by all. Whether a morally symmetrical society is orderly or chaotic depends on other factors which are not relevant to my dispute with Reiman. What is crucial is that civil society differs from the state of nature in exhibiting an asymmetrical distribution of rights and duties. Specifically, the state -- whoever that may be -- has the right to command and (in Reiman's interpretation) to use force to compel compliance, whereas individual citizens have an obligation to obey (my interpretation) or at least to refrain from using force to compel compliance to their particular wishes To put the point in quasi-economic terms which capture something of the logic of traditional liberal political theory: by means of the social contract, a moral monopoly is substituted for the moral free market of the state of nature.

But Reiman rejects the monopolistic claims of the traditional theory! He agrees that citizens may, under special circumstances, be morally justified in resisting, defying, or evading the coercive efforts of even a just state. So it would appear that the dispute between us has dissolved. Not so, for in place of the claim to monopoly, Reiman offers a modest claim to what might be called prima facie monopoly. According to Reiman, "legitimacy [of the state] suggests prima facie reasons for obedience [by the citizen], which can be overridden by other moral considerations" (p. 35). But this watered-down claim has the same logical force as the traditional claim of absolute monopoly, for the classical distinction between a state of nature and civil society rests not on the absoluteness of each citizen's renunciation of private force, but only on his renunciation ire some degree or to some extent of the right to use force. If the renunciation is absolute, then the state does indeed acquire a monopoly of the legitimate (i.e., rightful) use of force. And Reiman to the contrary notwithstanding, states everywhere and always claim such a monopoly. If, on the other hand, the renunciation is only partial, then the state acquires a prima facie right to the employment of force over and above the right which each individual had and to some extent still retains. In either case, the perfectly symmetrical moral situation of the state of nature, in which each person has the same rights and duties vis-a-vis each other, is exchanged for an asymmetrical moral relationship in which one party, the state, has a preponderant right to force vis-avis all the others. To put the point somewhat differently, a citizen will need better reasons to justify his defiance of a legitimate state than the state will need (aside from appeals to its legitimacy per se) to justify its coercion of him, for the state's appeal to its legitimacy per se will count as a reason of some weight all by itself.

In my original essay, I took the traditional theory as my target. Reiman thinks he can turn aside my attack on legitimacy by claiming for the legitimate state no more than a prima facie right to coercion which is not shared by each citizen. But as I shall show, his argument fails to establish even such a prima facie right. The truth is that there is no legitimate political authority either in the classical sense of a monopoly of the use of force compatible with the autonomy of the individual or in Reiman's sense of a prima facie right to the use of force possessed by the state but not by the individual. I need hardly add that for the purposes of this dispute between myself and Reiman, it makes not the slightest difference how large or small a weight one assigns to that supposed prima facie right. My position is that one must assign zero weight to it, and Reiman's position is that one must assign some nonzero weight to it. If I can refute even so modest a claim for the Legitimate State, then I will have considerably strengthened my original thesis.

Let us return to Reiman's definition of political authority as "the right to make commands and use coercion to discourage noncompliance with them." So far as making commands is concerned, of course, that is something anyone can do. As Hotspur says, after Glendower has boasted that he can call spirits from the vasty deep, "Why, so can I, or so can any man;/ But will they come when you do call for them?" So the essence of political authority, in Reiman's view, is the right to use coercion to enforce one's wishes.³ Whoever has that right can be said to have political authority.

But this commits Reiman to a very peculiar conclusion indeed. By his definition, in a state of nature every single person has actual or potential political authority over every other person. Assuming, as I have, that Reiman is not an absolute pacifist, suppose that I return home one night in a state of nature (so to speak) to find a burglar breaking into my house. I command him to get away from my house immediately, not, however, adding a threat to call the police since in a state of nature there are no police. Now if Reiman will agree that I have the moral right to use some measure of force to "discourage [the burglar's] noncompliance" with my command, then he thinks that I have "political authority" over the burglar! What is more, if Reiman also thinks that the burglar has the moral right to use force to defend himself against an attack by me that goes beyond what is required to drive him away from my house, then to that degree the burglar has "political authority" over me (if he has the foresight to "command" me to stop trying to kill him before he resorts to force in self-defense).

Something has obviously gone wrong here. Reiman and I were engaged in a serious debate about the grounds and limits of legitimate political authority, and all of a sudden I am drawing facetious conclusions about the political authority of burglars and householders. Reiman doesn't mean to impute political authority to every private individual who has a moral right to use force on some occasion or other. Indeed, he isn't concerned with individuals at all when he defines political authority. Reiman is talking about the putative political authority of states. It is the commands of the state which are to be complied with or not; it is the state which claims to have the right to use coercion to discourage noncompliance; states are said to have, or not to have, political authority. Individuals may have rights against the state; they may be morally justified in defying the state; they may even be right to deny entirely some actual state's claim to political authority. But individuals do not have, or claim, political authority -- states do.

And now we have come to the very center of Reiman's confusion. Like so many defenders of the state who have gone before him, Reiman suffers from what we may diagnose as a virulent case of mystification. By mystification I mean the tendency to use language so as to conceal the real nature of a thing, in particular to conceal the real moral or power relations among persons, their real interests, and the real status of their rights and obligations. Reiman's talk about the state thoroughly obscures its real nature. He encourages us to reify the state, to conceive it as having an ontological status different from that of individual persons, to blind ourselves to the real locus of moral responsibility in political relationships, and thereby to construe the relationship between the individual and the state in an entirely false way.

On the very first page of my essay, I define the state as "a group of persons who have and exercise supreme authority within a given territory." The state is a group of persons. It is not an institution, if by that is meant something other than the people who occupy the roles which constitute that institution. It is not a system of laws, if by that is meant something other than the persons who make, interpret, and enforce the laws. The state is a group of persons. Reiman says that he accepts this statement (p. 19), but over and over again his use of language obscures its force. Consider, for example, the following passage from the Introduction:

To put it somewhat crudely, what Wolff fails to see is that the proposition "Everyone should determine his own moral duty" (Wolff's anarchism) in no way implies the proposition "Everyone should be allowed to do what he determines as his moral duty" (political anarchism), (p. xxiii)

[P]olitical systems begin from the assumption that some areas of behavior are too crucial to the mutual well-being and survival of the community to be left to the consciences of its members, (p. 29)

Political systems start from the assumption that some forms of behavior must be prevented, even if they are conscientiously chosen. This is the logic of political systems, (p. 29)

The political or legal system . . . of the state can be expressed as the principles specifying when and in what form force will be applied to coerce behavior into or out of certain forms, (pp. 21-22)

In these passages, and in many others like them, a sort of conceptual conjuring trick has been performed. It is clear enough who is to be the object of the force and coercion: it is the citizen, the individual whose behavior is to be altered. But the other party to the coercive interaction has somehow vanished into thin air. Rather like the Cheshire Cat, nothing is left but the state, smiling benevolently and impersonally (or perhaps malevolently, but still always impersonally) .

The real fact, of course, is that when a citizen is coerced, some other real flesh and blood person does the coercing, in the active voice. Not "force will be applied" but "Jones, who is a policeman or a judge or a prison guard, will apply force to Smith, who is a citizen." The state is either a real group of persons or it is a fiction.

If we can hold the real nature of the state firmly in mind, we will recognize that Reiman's theory of political authority had failed to do the job for which it was intended. It does not distinguish between the relations among men in a prepolitical state of nature and the relations among them in civil or political society. Either all moral agents have the right, under some conditions or other, to use force to implement their purposes, or else none do. In the absence of a collective agreement to forfeit that right -- an agreement, I have argued, which would carry with it a forfeiture of autonomy -- men remain in a moral state of nature, regardless of the beliefs that some or all may have about the supposed special right to force of that small group who call themselves the state.

I think Reiman would actually agree with the major thesis of this argument, although he might take exception to my manner of expressing it, because I have so far omitted a factor on which he places very great weight. The confrontation between the individual and state is not, Reiman insists, merely a confrontation between a small group of people (the individual) and a large group of people (the state). Rather, it is a confrontation between an individual and a system of laws. A page or so ago I quoted three statements from Reiman, and in the third I omitted a parenthetical aside. Here is the passage again, with the parentheses restored:

The political or legal system (and the two are used here interchangeably, in that it is the legal system which makes the political system a "system," i.e., a rule-governed totality as opposed to a simple fact of domination) of the state can be expressed as the principles specifying when and in what form force will be applied to coerce behavior into or out of certain forms, (pp. 21-22)

Will [it] tend to yield more morally worthy results than can be expected from the absence of a political system, or more morally worthy results than -- or at least as much as -- can be expected from other possible and viable political systems? (p. 32)

We finally have Reiman's answer to the central question between himself and me: "Wherein does our moral situation as subjects of a legitimate state differ from our moral situation in a state of nature?" The classical democratic answer is, In a state of nature each of us is morally autonomous and each has the right to use force to compel compliance with what he judges to be his morally justified commands; but in a legitimate state, no private citizen has any right to use force for such a purpose -- only the state has. My answer is, In a state of nature each of us is morally autonomous and each has the right to use force to compel compliance with what he judges to be his morally justified commands; a special authority can be acquired by some group of persons called "the state" only at the price of the moral autonomy of the subjects, and that is a price that it is contrary to the nature of rational agents to pay. Reiman's answer is, In a state of nature each of us is morally autonomous and each has the right to use force to compel compliance with what he judges to be his morally justified commands (Reiman doesn't put it this way, but I think he would grant the claim) ; but under a system of laws which is rule-consequentially maximal, each subject has a new prima facie duty to obey the laws, over and above whatever obligations (and rights) he had in the state of nature. As Reiman says, in a legitimate state, the subject's legal obligation is "simply one morally relevant factor figuring in his determination of his moral obligation" (p. 38).

Reiman offers no demonstration of this claim, despite its centrality to his position.4 Now I do not in general approve of the philosophical tactic of playing the idiot, of claiming not to understand some statement that looks perfectly comprehensible on the face of it. But I must confess that I genuinely do not understand the usual appeals to prima facie obligations. In the writings of W. D. Ross, where it plays a central role, the notion is based upon a theory of a power of moral intuition which I find opaque and impossible either to explicate or to translate into some other, more comprehensible terms. We are supposed by Ross simply to know further how much weight their Tightness is to be given in a deliberation among conflicting prima facie duties. The intuitionist position, even in the careful and intelligent version offered by Ross, seems to me simply no position at all -- rather like Leibniz's pre-established harmony or a religious man's appeal to faith.

Sometimes, however, the notion of prima facie obligation is introduced as part of a "rational reconstruction" of our ordinary moral consciousness. That is to say, it is claimed that our moral reasoning actually makes appeal to such a notion, and that the deliberations of serious and intelligent moral agents can be construed to involve appeal to the notion of conflicting prima facie duties, whether or not anything quite that precise and explicit is actually expressed.

I reject the use of the notion of prima facie obligation as construed in this manner because in the end a rational reconstruction of moral consciousness is a description of men's moral convictions rather than a justification of them, and it is a justification that I seek. I readily grant that many mature, serious, reflective students of politics believe that we have a prima facie obligation to obey the valid laws of a constitutional democracy. I deny this belief. I have never heard any plausible argument for it, and it seems to me merely a superstitious submission to authority.

It might be thought that I am being unduly harsh with Reiman and inaccurate to boot, for he makes no appeal to an inexplicable power of moral intuition. He explains quite clearly what the grounds are for the imputation to the citizen of a prima facie duty of obedience, namely, the goodness of the general foreseeable results of the operation of the legal system as a system. Where is the mystery in that?

The answer is a bit tricky, but it is essential to the dispute between Reiman and myself, for the notion of a pritna facie rather than an overriding duty of obedience is the last refuge of the myth of state authority. Once we have exposed the emptiness of even this diminished claim, there will be no further justification to be offered for the state's pretensions. Reiman and the other defenders of the state will have to acknowledge that autonomous moral agents are always in a state of nature vis-a-vis one another. Only an autonomy-sacrificing contract can alter their moral relationship.

Reiman's invocation of a prima facie duty of obedience involves a sort of double counting which is entirely unjustified and which has the practical effect of tilting the scales in favor of the de facto state in any dispute between it and an individual. To see how this double counting conies about, let us develop an example in some detail.

Suppose that a friend comes to me and asks me to back up a lie he has told to Internal Revenue agents who are investigating his tax returns. He has claimed a number of business deductions for lunches and dinners at which he says that he and I discussed plans for a joint publishing venture. The deduction has been challenged, and he wants me to sign an affadavit swearing that I did eat with him on those occasions and that we did discuss business. All of this, let us further suppose, is quite false. Now I must engage in a moral deliberation, weighing a number of considerations which may count as reasons for and against complying with his request.

Just what I consider to be relevant considerations, and what weight I give, positive or negative, to each will depend on the particular moral theory or set of moral opinions which I hold. For example, I may take into account that he is a good friend, that on past occasions he has gone out of his way to help me when I was in need, that I am a husband and father with responsibilities to my family which may be compromised if I am caught out in the lie (a fact which I will properly discount by the degree of probability that I will be caught), and that the amount of money involved is so small as to make only a marginal difference to my friend. I will also take into account the purposes for which the government is likely to use the tax money it collects. I will weigh my evaluation of the tax structure in general, including its fairness or unfairness in distributing tax burdens among the citizens of this country. I will add into my calculation some estimate of the probable effects of my lie on the future law-obedience of myself, my friend, and any others who may learn of it (possibly giving positive weight to a negative effect if I believe that there is too high a level of obedience to law in myself and my friend, or a negative weight if I believe he and I are insufficiently law-abiding). I may also assign some negative weight to lying in and of itself.

At some point in my moral deliberations (which the reader may think have already gone on quite long enough), I will attempt to judge the constitutionality of the tax laws under which my friend is being investigated, and I will extend my examination of the subject so far as to consider the effect on the total system of laws of my proposed violation. If I judge the system of laws to be generally felicific in its consequences, then I shall discount their value as an actually operating system of laws by my estimate of the degree of effect which my violation will have upon them, and add that discounted weight on whichever side of the balance is appropriate. I might conclude, for example, that the legal system of the United States is highly beneficial, but that it would be even better if there were somewhat more defiance on the part of individual citizens. (I might believe, for instance, that obedience to law is, in the Aristotelian sense, a virtue, the excess of which is slavish servility and the defect of which is anarchic self-indulgence.) I might also judge that my defiance was likely to have very little effect one way or the other on the law-obedience of either myself, my friend, or my fellow-citizens. Nevertheless, all this (and more, of course) will go into my calculation.

Now, up to this point, no question of legitimacy of state authority has been raised. The several weights and their balance will presumably vary from state to state and from case to case, but the sorts of considerations relevant to the deliberation will be exactly the same in a monarchy, a popular dictatorship, a constitutional democracy, and a military autocracy. Some of the factors are particular, in the sense that they relate to the individual case alone. Such, in this instance, are my indebtedness to my friend for past favors, my own personal obligations, and so forth. Other factors, such as my estimate of the beneficence of the system of laws as a whole, are more general, and might enter in virtually the same way into many different calculations (although the details of each case may alter the weights to be assigned even to general factors by altering the probabilities of outcomes -- one act of law-defiance might carry virtually no likelihood of influencing the behavior of others, while a second might reasonably be predicted to affect the behavior of thousands or even millions of other citizens).

In a state of nature, when this point had been reached, the deliberation would be at an end. Assuming that some way could be found to quantify the various "weights," a sum would be struck and the decision taken either to cooperate with my friend's lie or not. But according to Rei-man, in certain political communities one more "morally relevant" factor must be added into the calculation before it is complete. If the system of laws as a whole is what I have for the sake of brevity called consequentially maximal, then that fact will create a new additional prima facie obligation of obedience to the law, and I am morally bound to add its weight to that of the other arguments against complying with my friend's request.

And here, I contend, we see the precise nature of Rei-man's confused double counting. Some philosophers use the notion of a prima facie obligation merely as a short-hand way of summarizing the aggregate weights of those factors which I have called "general." To say that there is a prima facie obligation to obey the laws of a generally beneficent state is for them just a compendious way of reporting that the general factors weigh out on the side of compliance, allowing of course for shifts in probabilities from case to case. Since the general factors remain relatively constant over a span of time in which many particular decisions must be made, this reminder to myself in the form of a prima facie obligation claim will shorten my calculations. Presumably, I would then stand under a general necessity of periodically reviewing my estimate of this portions of my deliberations, revising and altering it as changes in the facts, the system of laws, and the solidity of predictions of the future required.

If this summary sense of a prima facie duty is all that Reiman has in mind, then he has no right to add in a new weight for the prima facie duty after he has already added in the weights of the individual general factors. Either he must simply add the prima facie obligation to the estimate of the particular factors, or else he must carry out the entire calculation of both particular and general factors. To carry out the entire calculation and then add the prima facie duty is to count the weight of the general factors twice. But if Reiman thinks that there is more to prima facie duties than a mere summary of the aggregate weight of certain sorts of factors, then he has said absolutely nothing to justify such a belief. Indeed, his general adoption of consequentialism in moral deliberations should rule out any appeals to emergent or supervenient prima facie duties.

The net effect of Reiman's mistake is to load the scales in favor of compliance with the law. It is easy to see this in the imaginary case we have been considering. The violation of tax laws by two private citizens will, in most cases, have so little effect on the general operation of the legal system and the social level of law-obedience that an honest calculation of effects will assign a vanishingly small weight to this factor even when one judges the system as a whole to be highly beneficial. A prima facie obligation, unquantified but vaguely felt to be of some significance, is likely to be assigned a considerably greater weight. Hence, individuals who are misled by Reiman into confusing the summary and supervenient sense of prima facie obligation are likely to conclude for obedience in many cases where a straightforward consequentialist calculation would lead to disobedience.

Leaving to one side such estimates of actual deliberative outcomes, however, we arrive at the clear theoretical conclusion that the notion of a prima facie obligation to obey the laws of a consequentially maximal state offers no way of distinguishing the moral condition of the state of nature from that of civil society. Even in the state which Reiman chooses to call legitimate, an individual will carry out exactly the same moral calculation that he would perform in a state of nature (or, what is presumably the same thing, in an illegitimate state). No new morally relevant consideration will be introduced into the calculation in the legitimate state. Which means, of course, that in any ordinary sense of the word, the state is not "legitimate." In short, I am not wrong. Political philosophy, as the study of that legitimate political authority which distinguishes civil society from the state of nature, is dead.

Thus far, I have confined myself to the particular attack which Reiman mounts against my thesis, but other critics might not wish to limit themselves to his line of reasoning. I have argued that only a contract could bring into existence a new moral situation, but could I not resolve the conflict that I perceive between authority and autonomy by putting the notion of prima facie obligation to my own uses? In my original essay, I made much of the totality of the forfeiture of autonomy which the classical social contract demands of each individual. Suppose that a watered-down contract were proposed, calling only for such forfeiture as was involved in granting to the state a prima facie right to command and laying upon the citizen only a prima facie obligation to obey. Does not such a contract leave room for autonomy?⁵

My answer is, of course, no. But before explaining why, I want once again to emphasize the absoluteness of the typical state claim to authority. In the United States, for example, the law permits what is called conscientious objection in cases involving religious or quasi-religious scruples. But the state claims the right to decide what shall be accepted as a conscientious objection, and in which contexts the appeal to conscience shall be allowed. The law also permits private citizens to use force in self-defense, but the state claims the right to decide when force may be used, against whom, for what purposes, and within what limits. The state claims the right to change its decisions on these matters at any time, so long only as it does so in accordance with its own rules. No state anywhere, to my knowledge, relinquishes in the slightest its claim to be the supreme authority within its territory. If it were to relinguish that claim even so little as to make room for the most marginal exercise of authority by private citizens, then when the state and some citizen came into conflict within that margin, the two together would have to resolve their differences either as equals through force or negotiation or else by submission of the dispute to a higher authority. In the first case they would be back in a state of nature and in the second the state would thereby have acknowledged that it was not sovereign.

Let us consider, then, a social contract calling only for a transfer to the state of prima facie rather than overriding authority. Wherein lies the forfeiture of autonomy in such a contract? In my original essay, I pointed out that autonomy could be forfeited totally or by degrees, across the board or with regard only to particular areas of decision, and so forth. Obviously a contract of the sort we are now considering would involve a diminished forfeiture of autonomy, and I suppose it would therefore be preferable to a total forfeiture of autonomy. But the signatories of this contract would still have bound themselves to obey commands whose rationale they either denied or could not comprehend. To revert to the example of my friend and his tax troubles, I would be bound under such a contract to refuse to help him, even though the outcome of my calculation of pros and cons was a clear conclusion to comply with his request. To produce that result, it would only be necessary that the net weight in favor of lying for him be less than the standing weight of my prima facie obligation to obey the law. Since the good consequences of law-obedience would already have been added into my calculation, I would be reduced to saying, "I shall obey the law solely because it is the law, irrespective of any reasons I can give myself concerning the goodness or badness of the law and the goodness or badness of my complying with it in this case." That is heteronomous obedience just as surely as is the absolute obedience to state commands demanded by classical social contracts.

The belief in state authority comes naturally to men, it would appear. A band of robbers ride into town with guns drawn and demand all the gold in the bank. They are called criminals. They return the next year on the same day and repeat their demand. Again they are called criminals. They put on uniforms and return each year on the same day Eventually, they are called tax collectors. Finally, the smallest and least offensive of the bandits rides into town un armed and the townspeople give him their gold wxthout struggle. The state has arrived.

1. Strictly speaking, I don't agree with Reiman about this now, although I did when I wrote In Defense of Anarchism. My present views are rather different, though I am not now able to articulate them clearly or defend them adequately. But this is a reply to the last book I wrote, not a preview of the next I hope to write, and since the disagreement between Reiman and myself does not turn on those deeper issues of moral philosophy, I have adopted the simplifying device of speaking from my earlier point of view.

2. I say "in principle" because in some special set of circumstances it might actually be true that one and only one person was morally right to use force to accomplish his ends. That mere contingency, however, would not give him adequate grounds for claiming a monopoly of the legitimate use of force.

3. Lest it seem that I am somehow sliding around Reiman's definition by speaking of enforcing wishes rather than commands, let me say that Reiman's argument rests in part on the institutional or systematic character of law, as opposed to the noninstitutional character of the actions and decisions of private individuals. I shall deal with that part of his theory of political authority a bit later on, when I turn to his criterion of the legitimacy of political theory.

4. In the discussion of prima facie obligation which follows, I am drawing on and in part reproducing remarks which I made in a reply to two other critics, Professors Malcolm B. E. Smith and Michael S. Pritchard. See The Journal of Value Inquiry (Winter 1973) for the entire exchange.

5. It is clear, I hope, that Reiman takes no such line. For him, the validity of the prima facie obligation rests on the goodness of the systematic effects of the laws. For even this sort of diminished contracctualist position, the validity of the prima facie obligation rests on the bindingness of the original contract.