. . . no country can be well governed unless its citizens as a body keep religiously before their minds that they are the guardians of the law and that the law officers are only the machinery for its execution, nothing more.
— Mark Twain
The questions that I will attempt to answer in this essay are prior to any question we can ask about the shape that our political, economic, and social institutions should take in order to establish a just society. First are questions about government itself — why it is necessary and whether it is intrinsically good or a necessary evil. Then there are questions about the state — civil society, the political community, or body politic. Again our concern is with whether it is necessary and, if necessary, whether it is intrinsically good or a necessary evil.
The order in which I have placed these questions is based on the fact that the state or civil society is not the only community or association of men in which the role of government must be considered. As we shall see, the question about the necessity of government applies to any association of men living and acting together for a common purpose or a common good — a family, a village or tribe, or a private corporation of any kind, as well as to a state. Civil government is only one of the many types of government, the type that is appropriate to a civil society or state. The appropriateness of different types of government to different types of communities is a consideration posterior to the problem of understanding why government, of one type of another, is necessary for the existence of any community. We will subsequently see that the state — the political community — comes into existence only with the institution of a certain type of government, but we must first understand why the existence of any community depends upon the institution of government.
I said earlier that the propositions I was going to set forth comprised the controlling principles of political philosophy conceived as a purely normative discipline. That remark calls for a word of further comment before I proceed with the exposition of the principles.
Because of the dependence of political thought upon political history, there is an inveterate tendency on the part of political philosophers to intertwine descriptive or historical statements with their normative judgments. They often pass insensibly from describing the way things are or have been to judgments about how they ought or ought not to be set up. In many cases, normative judgments or evaluations are implicit in statements that, on the surface, have the character of statements of fact; and they are left implicit, masked or concealed by appeals to historical evidence, rather than expressed explicitly in normative terms and defended as such.
I am going to try, in what follows, to concentrate on propositions that are clearly and plainly normative in their intent and that have the universality proper to controlling principles. This does not mean that I will abstain from references to historical fact or to current experience, but, where the discussion of such matters is required or helpful, I will try to treat them in a manner that is appropriate to questions of fact and not as if they were subjects about which a political philosopher has or can have special wisdom. I will try to exercise the same kind of restraint with regard to political problems that call for normative judgments which fall below the level of universality appropriate to principles. For purposes of illustration or amplifications I will from time to time deal with such problems and comment on alternative solutions to them; but I will reserve philosophical judgment about such alternatives at the level of policy, except in those rare instances in which the controlling principles require their endorsement or rejection.
– 2 –Proceeding now to the question about the necessity of government, we must begin by distinguishing two senses of the term “necessity” — practical or pragmatic necessity, on the one hand, and logical and natural necessity, on the other.
We say that something happens necessarily in the very nature of the case when, given the operation of a cause, its effect cannot not occur. The causal laws discovered and formulated by natural sciences are, in this sense, statements of the necessary connections between one event and another. In the sphere of logic, we say that a valid inference is one in which the conclusion necessarily follows from the premises. If the premises are affirmed, the conclusion cannot be denied without contradicting one’s self. In contrast to these two related senses of necessity, we speak of a thing’s being necessary in the order of human action when it is indispensable to the end that we have in view. If it is impossible to achieve the end we are aiming at without employing a certain means, then that means is necessary in a practical sense.
Unlike natural necessity, practical or pragmatic necessity is compatible with the voluntary. We cannot violate or act contrary to natural necessities. If we lose our footing or our balance, we do not have the option of obeying or disobeying the law of gravitation. But in the sphere of practical necessities, it always remains possible for us to defeat our own purposes by voluntarily refusing to do what is required in order to achieve the end we have in view. The necessity still obtains; for the end cannot be achieved without employing the indispensable or necessary means. But nothing compels us to act in such a way that we succeed: we are free to fail by not doing what is practically necessary. If taking a plane is the only way to get to a certain place at a certain time, we can defeat our own desire to attend a meeting at that time and place by refusing to fly.
Government is a human institution; it is not a natural phenomenon, but a product of human action. Hence the question of its necessity is a question about its indispensability as a means to a certain end. To answer the question we must, therefore, look to the end that government is supposed to serve and attempt to define, as precisely as possible, the way in which government functions as a means.
The definition of government — not the government of a political community or civil society, but government per se — involves a number of steps. First of all, let us consider the difference between being governed and being exempt from government. An individual who is subject to government in any respect whatsoever is one who, in that respect, obeys a rule of action or carries out a decision that is not entirely or wholly of his own making. Thus, for example, when I and I alone decide the place where I shall live, the food I shall eat, or the book I shall read, I am not subject to government in the actions that I take to carry out these decisions. Or if I and I alone make the rule for my own conduct that I will not smoke cigarettes, I am not subject to government when I voluntarily obey this rule of abstinence. In matters of this kind, the young child is usually subject to government. We say that the child is governed by his parents when they decide for him the place where he shall live, the food he shall eat, or the books he shall read. Or when they lay down rules of conduct for him that he is expected to obey.
It may be supposed that the distinction between being governed and being exempt from government can also be expressed as a difference between government by another and self-government. Accordingly, it would be said that the child is governed by his parents, whereas the adult in obeying a rule of his own making is subject to self-government. For reasons that will presently become clear, I propose to reserve the term “self-government” for a certain type of government in which the decision that I act on or the rule that I obey is neither entirely of my own making nor wholly made by others. Instead of using the term “self-government” for the condition of being exempt from government by others, I will use “autonomy” to refer to those cases in which the individual acts on decisions or obeys rules entirely of his own making.
Government never completely replaces autonomy and never can. Even the young child exercises autonomy in many respects, for the strictest and most supervisory parents do not and cannot regulate every aspect of the child’s behavior, nor can they issue edicts that decide everything that the child does from moment to moment. What is true of the child is also true of the adult in almost every imaginable set of circumstances. Even the slave or the prisoner of war retains a certain degree of autonomy, for the simple reason that it is impossible to make the government of anyone total — covering every action that the individual engages in.
The distinction between government and autonomy — that is, between being subject to government and being autonomous — is related to, but is not identical with, the distinction between the social and the solitary condition. If man were capable of leading a solitary life, he could be autonomous in all respects. The solitary individual would decide everything for himself and obey only such rules of conduct as he laid down for himself. He could not help being autonomous in this case; government would be inapplicable. In contrast to the solitary life, the social life is one in which a number of individuals live together, each in some dependence on the others and each being affected by the actions of others as well as affecting others by his own actions. In is only in the case of social life that the question of limitations upon the individual’s autonomy can arise.
In order to avoid begging the question, I will refrain from assuming, without further analysis, that social life requires some limitation of the individual’s autonomy, which is just another way of saying that social life requires some degree or measure of government. I will, therefore, put the question in the most open fashion, by asking whether it is possible for the individual who lives socially — that is, in association with others — to retain the complete autonomy he would have, in fact, could not help having, in the hypothetical case of a purely solitary mode of life.
– 3 –For the purpose of answering this question, I propose to consider an extremely simple model of social life. Let me warn the reader at once that the model does not represent every variety of human association and so will not provide us with all the insights we need in order to understand all the functions of government. Nevertheless, it will help us to take a first step in that direction.
Three scientists voluntarily associate for the purpose of exploring the upper reaches of the Amazon. Before they embark on this common enterprise, they realize that, at a certain point in their exploration, they will be entirely on their own in the jungle. Each of them recognizes that he could not do alone or by himself what it may be possible for the small organized group to do, and each is willing to join the group for that reason. In other words, they are associated for a common purpose and with the realization that it is only the action of the group as a group that can achieve it. If they do not stay together and act together for their common purpose, they cannot succeed.
Before they leave civilization behind and enter the jungle as an isolated group entirely on its own, the three scientists face the question of how rules or decisions will be made for the action of the group as a group, as well as for the conduct of its individual members in so far as such conduct affects the success of the enterprise. The qualification just stated leaves them autonomous in matters that do not affect the concerted action of the group or the success of the enterprise. But why can they not be completely autonomous, each regulating his own conduct and deciding everything for himself?
A moment’s reflection will serve to discover that complete autonomy is impractical and will not work. Understanding why this is so will throw light on the function that government is needed to perform.
Though the scientists associate as equals, each needs the cooperation of the other two in order to succeed in their common enterprise. They must agree, therefore, upon some method of regulating their own conduct and of reaching decisions in a manner that will preserve their concerted efforts to achieve a common goal. There are only three alternative procedures available to them.
One is for the scientists to require unanimity as the basis for any rule or decision that they will acknowledge as having authority for them. One dissenting voice on their part would then have the effect of a nullifying veto. And that, in turn, would mean that each scientist is committed to obeying himself alone, since no rule or decision to which he does not assent has authority for him. This would leave each of the scientists completely autonomous.
A second procedure would be for the three scientists to elect one of their number the leader of the expedition and confer upon him the authority to regulate the conduct of the party and decide all matters affecting the success of the enterprise.
The third alternative — and the only one that remains — consists in an agreement on the part of the scientists to have all rules adopted and all decisions made by a majority vote of two against one.
Only the first alternative leaves the scientists completely autonomous. The second and third institute a mode of government to which they themselves submit — two of them to the personal authority of an elected leader in one case, and all three of them to the impersonal or collective authority of a majority vote in the other case.
To show that government is not merely preferable to complete autonomy on the part of the scientists, but indispensable or necessary, we must have some reason for thinking that the requirement of unanimity on the part of the scientists will not work. Only if that is the case, must one or the other of the two remaining procedures be resorted to for the sake of carrying the expedition out successfully.
In the strictest use of “impossible,” it cannot be said that unanimity must be rejected as an absolutely impossible method of adopting rules or making decisions. It is entirely conceivable that the three scientists might concur in their solution of every practical problem that called for the making of a rule or a decision. Reaching his decision independently, each might, nevertheless, find himself in agreement with the other two; or even if the matter were fully discussed, the discussion might eventuate in a unanimous conclusion.
Hence unanimity cannot be rejected in principle as impossible. But that does not mean that it should not be rejected on the grounds of its probable consequences in practice. The practical problems that must be solved by our exploring scientists are not like mathematical problems or even experimental ones — problems the solutions to which can be demonstrated or for which decisive evidence can be offered. On the contrary, they are problems about which reasonable men can disagree as to their solution. The likelihood of such disagreement, even among three scientists engaged in a practical enterprise, is sufficiently great to make the requirement of unanimity impractical. In fact, it need only fail to be satisfied at one crucial turn of affairs to prevent the expedition from succeeding. Since the probability of one such failure is extremely high, that is sufficient reason to reject unanimity, together with the complete autonomy it preserves, in favor of government.
– 4 –We have now discovered one reason for the necessity of government. It is necessary as an indispensable means of getting rules adopted and decisions made about matters concerning which equals engaged in a common enterprise can reasonably disagree. Stated in another way, the three scientists must set up the personal authority of a leader or the collective authority of a majority vote in order to be sure that at every crucial turn of events their expedition will be directed by a rule or a decision the authority of which each of them acknowledges even though he may disagree with it, i.e., even though he would have adopted a different rule or made a different decision if he were acting autonomously instead of submitting to government.
While unanimity will not work as a way of getting a number of equals engaged in a common enterprise to work harmoniously together for a common goal, it is the only way in which equals can institute an authority that they acknowledge or a government to which they willingly submit. Once again we must realize that a unanimous decision on the part of all, the decision of an elected leader, or the decision reached by a majority vote exhaust the alternatives; for since we are considering voluntary action on the part of the scientists who join hands as equals, we must exclude the forceful imposition of a decision by someone outside the group itself.
That being the case, we can see at once that government itself cannot be instituted by a majority vote or by the decision of a leader, since the authority of a leader or of a majority is the very thing being instituted. Hence the institution of government itself, together with the delegation of authority to an elected leader or to a majority, must be accomplished by the unanimous consent of the parties involved — in this case, the three scientists as equals.
Since the government whose authority they acknowledge is established by their unanimous consent, the scientists form a self-governing community even though each has surrendered his autonomy with regard to all matters affecting the success of their common enterprise. Each of the scientists is a constituent of the government that is established with his consent. If the government established takes the form of a dictatorship (i.e., decisions by a leader), self-government is minimal, being limited to the selection of the leader, whether by lottery or by a majority vote. If, however, the government established confers authority upon a majority vote, then self-government is maximal, for each of the scientists exercises a voice in the adoption of every rule and the making of every decision.
In either the minimal or the maximal case, the individual remains self-governing when the rule adopted or the decision made is contrary to one that he himself would have chosen were he autonomous. The fact that he is obliged to obey a rule or to act on a decision that is not of his own choosing must be combined with the fact that his consent was involved in setting up the authority to which he owes obedience and, in the case of maximal self-government, with the additional fact that he participated in the voting that eventuated in a decision other than his own. For him to refuse obedience in those cases in which he disagrees with an authorized rule or decision is tantamount to his insisting upon his autonomy instead of acknowledging the authority of government.
– 5 –We have learned three things from our limited model. (1) Acknowledging and submitting to an authority for making rules and decisions concerning the actions to be taken by a group of men associated for a common purpose is the only alternative to retaining and exercising complete autonomy. (2) Since the retention of complete autonomy is tantamount to making unanimity the condition for adopting any rule or decision, its retention will probably frustrate concerted action for a common purpose, because the matters about which rules or decisions must be made are matters about which reasonable men can disagree. Their disagreement about such matters being highly probable, individuals associated for a common purpose must surrender their complete autonomy and substitute for it an authority that they themselves set up and acknowledge. They must do this if they wish to succeed in acting together harmoniously and effectively for whatever is their common purpose. (3) Government is necessary only as a means — a means of achieving concerted action for the good commonly aimed at by a group of associated men. The necessity of government answers to the need for a commonly acknowledged authority to make rules or decisions concerning actions that affect the achievement of a common purpose.
The points just made all relate to one function of government — one reason why it is necessary as a means. But that is not the only function of government, or the only reason for its necessity. Another is the indispensability of government for the maintenance of peace. To understand this, we must go beyond the simple model we have been considering, consisting of three men alike in character and purpose. Let us now contemplate a much larger community of equals, involving individuals differing in a multitude of respects. This type of community, much larger than a single family, we usually call a civil society — a community of men living together under civil government.
The reason why our three scientists found it necessary to institute government, the authority of which they freely acknowledged by their unanimous consent, will apply without qualification and, perhaps, with even greater force in the case of civil society: the common good for which men associate in the larger community cannot be achieved if each of them insists upon retaining his complete autonomy. Some portion of it must be surrendered to establish an authority for making rules and reaching decisions binding on all by their free consent. But in the case of the more populous and humanly heterogeneous community of a civil society, there is an additional reason for government, namely, its indispensability as a means to civil peace.
In any populous community comprising men of divergent interests, conflicts or disagreements will probably arise about matters of either private or public concern. The parties to such conflicts may either be private individuals or they may be private individuals arraigned against public officials. Confronted with the probability of such conflicts or disputes, what alternatives are available for settling them? Only two: one is whatever power is at the disposal of the parties in conflict; the other is the authority of government to adjudicate disputes and to enforce its judgments. Let us consider each in turn.
In the absence of government, each of the parties to a dispute, being autonomous, must operate as judge in his own case and, in defense of his ex parte judgment, must try to persuade his opponent or, failing in that, exercise such de facto force as he can bring to bear. Disputes can, of course, be settled in this way, but not peacefully, since persuasion is likely to fail and recourse to violence will be necessitated. Hence if men who live together and interact in all the affairs of their daily lives retain their complete autonomy, there is no way of excluding recourse to violence as a way of settling the disputes that are likely to arise. It follows, therefore, that government with the authority to adjudicate disputes and with authorized force to implement the judgment of its tribunals is indispensable to the peace of a civil society, in which men are associated for their common good.
Peace is essential to the very existence of a community as a community; and so, if communal life is a means to the common good of those who are associated in order to live well as human beings, then the maintenance of peace among men living together is indispensable to their achieving good lives for themselves.
The peace of a community may also be breached or marred by acts of criminal violence, as well as by the violence that would arise if the parties to a dispute could not submit their differences to an authorized tribunal for adjudication or arbitration. In the absence of government, each individual would have to defend himself against aggression by others with whatever power is at his disposal. The probability is high that wholesale violence would ensue. For this reason, as well as for the reason that, without authorized tribunals to settle disputes, the settlement of them would probably involve recourse to violence, the absence of government is a state of war rather than one of peace.
The preceding discussion of government as necessary for the maintenance of civil peace has introduced the notion of authorized force and the distinction between such force and violence. In our simple model of the three scientists in the jungle, the institution of an acknowledged authority sufficed for the operation of government; but in the more complicated case of civil society, naked authority is not enough. It must be clothed with and implemented by authorized force. The reason for this rests on the probability of disobedience together with the probability that persuasion will not always succeed in winning compliance from those who tend to be recalcitrant.
The probability of disobedience is generated by the fact of human freedom. Even those who freely acknowledge the authority of government always remain free to obey or disobey its rules of law, its administrative edicts, and its judicial decisions. In a populous community, comprising men of every stripe, good and bad, the probability of disobedience is not negligible. The disobedience may or may not be justified in the particular case. Let us for the moment table the problem of justified disobedience; I will return later to the conditions under which there is a right to disobey. For the present let us consider only the likelihood of disobedience that is not justified.
One way in which the constituted authority of civil government can deal with such disobedience is to attempt to persuade the recalcitrants. Since persuasion can fail and since, furthermore, it is likely to fail in a certain number of cases, some other device must be available if the authority of government is not to be rendered ineffective for the purpose for which it is instituted.
The only other device is the employment of coercive force. It is strictly an emergency measure in the sense that its use is justified only by the failure of efforts at persuasion. Nevertheless, the high probability that persuasion will not be effective to overcome unjustified disobedience in every case makes it necessary to implement the authority of government with the right to employ coercive force.
The force that is used to compel obedience or compliance may either be authorized or unauthorized. It is authorized only if it is instituted to implement the authority of civil government. Only such force as is thus instituted and employed to implement the regulations and decisions that government itself has the authority to make is, strictly speaking, authorized force or force exercised de jure — rightfully or by right. All other force is purely de facto or without right, and all such unauthorized force is violence. Violence may be committed by a government as well as by the members of a society. A government commits violence when it exceeds its authorization to use force — when it uses force that it is not authorized to use, or uses it to enforce compliance with rules or decisions that it is not authorized to make.
Since authorized force belongs by right to civil government and to civil government alone, government should have a monopoly of authorized force. This does not mean that it necessarily does have a preponderance of the force available in the community. The de facto force that can be marshaled by a revolutionary party or movement may surpass and overpower the authorized force of government. Again I am postponing for the time being the question of the right of revolution, which is related to the question about the conditions under which revolutionary violence is justified.
The only point that I wish to stress here is that authorized force — force used to implement authority — is, by its very nature, the exclusive property of a civil government that is duly constituted; in other words, a government that is itself de jure and not de facto. The force exercised by a de facto government is as unauthorized as that government itself is, and so, being de facto force, is a manifestation of violence.
– 6 –I can sum up what we have learned so far in a single, though somewhat complex proposition: government, with the authority to make laws, to adjudicate disputes, and to issue administrative decisions, and with a monopoly of authorized force to coerce where it fails to persuade, is an indispensable means, proximately, to the peace of communal life; and, ultimately, to the happiness of its individual members, to whatever extent a good human life for each of them depends on their being able to live together, work cooperatively for their common good, and interact peacefully with one another. Those who concede that government is necessary for the reasons just indicated may still wish to ask whether, being necessary, it is also a necessary evil. Or, in addition to being necessary, is it intrinsically good?
What is being asked is not whether there can be bad government. Government can obviously be bad in a variety of ways: through exceeding its authority or its right to use coercive force, through the imperfection of its institutions, through the injustice of its acts, and so on. Since no one can deny the abuses, imperfections, or injustices that everyone knows can afflict government, the question should be not whether government can be bad, but whether it must be. For if it must be bad and, in spite of that, is necessary, then and only then would it be correct to regard it as a necessary evil.
I submit that there is nothing about the nature of government that makes it impossible for it to be free from abuses, imperfections, or injustices. This remains true even if one were to add that, government being what it is and men being what they are, any government instituted and carried on by men will always reflect to a certain extent the weaknesses and imperfections of its human constituents and agents. Nevertheless, the institutions of government can be so perfected and its operations so safeguarded that they can be rendered innocuous, in spite of the ever present human proclivities to the contrary. Government is, therefore, not necessarily or intrinsically evil.
The only reason that might be given for thinking the contrary would be the view that complete autonomy on the part of every individual is an absolute good; for if this were the case, then government, by taking autonomy away from the individual in certain matters, while leaving it intact in others, would necessarily be evil. This line of argument can readily be shown to be self-defeating. To be a necessary evil, government has to be necessary, in the first place. But why is government necessary? Because, as we have seen, complete autonomy on the part of individuals is incompatible with their effective cooperation for a common purpose and with their peaceful interaction in communal life. Hence if the effectiveness and peace of communal life is itself something good — good as a means to the good life of human beings — then complete autonomy, not government, is to be judged intrinsically evil.
In short, the goodness of government as well as its necessity rests on the fact that human beings, in order to engage effectively in the pursuit of happiness, must associate and cooperate with their fellowmen to obtain the goods of communal life, among which peace is a principal component, and they cannot do so unless the authority and authorized force of government replace autonomy with regard to all matters affecting communal and common goods.
[Great Books of the Western World GBotWW=”1″]