In the summer of 1996, the following question occurred to me:
If you can conceive of morality without God, why can you not conceive of society without government?
Dostoevsky wrote that "Without God, anything is possible." This epigram assumes that religion, faith, and the threat of divine punishment provide the only effective limits to human behavior -- that if individuals did not have faith in certain divinely-ordained rules, they would run amok. Is there not a similar view of government? We could paraphrase Dostoevsky by stating: "Without government, anything is possible." The assumption here is that government force is the only thing that keeps people in line -- that if individuals were not forced to abide by certain rules, they would run amok.
Beneath both of these views lies a further assumption: that human beings are inherently violent and irrational. It is an assumption we commonly associate with Thomas Hobbes, who worried that human life in the absence of a Leviathan state would be "solitary, poor, nasty, brutish, and short". Surprisingly, it seems that Ayn Rand held the same view: "if a society provided no organized protection against force, it would compel every citizen to go about armed, to turn his home into a fortress, to shoot any strangers approaching his door..." (Rand 1963a, 108). Rand offers no argument for her position, which ignores the fact of mostly peaceful relations in pre-state societies and modern societies alike. Indeed, there is little to no evidence that the threat of government action is what prevents people from running each other off the highway or pillaging the strangers who live across town. In any society, criminality occurs in only a small minority of individuals because it is a deviation from human norms, a highly risky strategy, and a waste of precious time, energy, and attention (which is almost always better directed to creation than to destruction).
Unfortunately, Rand's few writings on the foundations of political interaction are marred by similarly questionable assumptions, as has been pointed out by many thinkers in the modern libertarian movement and elsewhere.
Overall, Rand accepts the views expressed in the Declaration of Independence and the United States Constitution: human beings have certain individual rights, governments derive their legitimacy from the consent of the governed, and governments are instituted among men to protect individual rights and only to protect those rights (although in general Rand is most interested in the rights of creative innovators, not the individuals who make up the democratic masses).
Rand also accepts the traditional story that American government was founded exclusively for the sake of protecting the universal rights of man. Unfortunately, that story, while appealing, is not historically accurate. From the first, American society accepted the absolute sovereignty of "the people" in the form of the executive, legislative, and judicial branches of government (a direct borrowing from the English idea of "the King-in-Parliament"). Thus unlimited government was present from the beginning, in germ if not immediately in fact.
This is important, because much of Rand's justification for the state rests on the validity of the American experience. For example, she argues that the source of government's authority is the consent of the governed and that "any undertaking that involves more than one man, requires the voluntary consent of every participant". She even tries, via her theoretical preference for the voluntary financing of government, to make individual consent a requirement for the recognition of governmental authority. However, her attempt falls short on several counts, mainly because it is unclear what could represent "consent" on her fundamental premises. As Lysander Spooner argued in 1867, no one living today ever signed the Constitution, so we have not directly consented to that form of government. It might be argued that paying taxes is a sign of consent, but given that taxes are extracted by force, such an argument is void. Voting might be taken as a sign of consent, but in the absence of knowledge about the intent of the voter, many interpretations are possible (e.g., a vote for one party might be a vote for gridlock rather than support for that party's program). In the Randian politics, one possible means for financing government might come from a fee on contracts (whereby contracts on which the registration fees are not paid could not be enforced in government courts). Payment of the contract fee could be construed as consent, but it could just as well be construed as a safe business practice; similarly, nonpayment might arise from a simple oversight, trust for the other party to the contract, a wish to save money, or a desire to use a private arbitration service. (Strictly speaking, it is not clear if Rand would approve of such services, since they could be seen as challenging the government's monopoly on dispute resolution; however, outlawing them would establish black markets in voluntarily-chosen arbitration services, which would seem odd coming from an advocate of the free market.) Thus resting the legitimacy of a government on the consent of the governed appears to be relying on rather shaky ground.
In her essay "Man's Rights" (Rand 1963a, 92), Rand makes the point that "there is no such entity as 'society,' since society is only a number of individual men"; this is admirable from the perspective of methodological individualism, but it is not a principle to which she consistently adheres in her thinking on political matters. For example, in her essay "The Nature of Government" (Rand 1963b, 107-8), she writes that "a society that robs an individual of the product of his effort, or enslaves him, or attempts to limit the freedom of his mind, or compels him to act against his own rational judgment ... is not, strictly speaking, a society, but a mob held together by institutionalized gang rule." Here one might wonder: what is this "society" that is robbing and enslaving the individual? Since there is no such thing as society, it must be that, under certain conditions, some individuals or groups of individuals are able to rob and enslave other individuals by using or threatening to use force against them. So Rand's argument that "if a society provided no organized protection against force, it would compel every citizen to go about armed, to turn his home into a fortress, to shoot any strangers approaching his door..." is null and void, since it commits the fallacy of reification. How can a society compel individuals to do anything? What is this amorphous society that protects individuals against force? As Rand liked to say, "blank out". All we can legitimately claim is that in any collection of individuals, some of those individuals may choose to protect other individuals against the initiation of force (as, for instance, parents may protect their children, a bodyguard may protect his employer, neighbors may patrol their neighborhood, individuals may form associations for the purpose of mutual defense, one may contract with a security firm, or one may depend on the agents of a monopoly government). But "society" does not do this.
The principle of methogological individualism applies with even greater force against Rand's arguments in favor of government, specifically its ability to objectively define laws. Rand consistently ignores what have come to be known as public choice arguments, for she assumes, utterly without foundation in experience, that government is objective whereas individuals, groups of individuals, or non-monopoly institutions are arbitrary, subjective, and biased. She thinks it is possible to construct a government purely of laws. Yet who will be running this government? Individuals. And those government employees will have the same drives and motivations as other human beings. We know they will not be angels. What is to stop government employees from abusing their power? Rand says that the doctrine of individual rights is the main protection and limitation on government power. But that is a thin reed indeed.
Interestingly, Rand talks about law as being created in legislation (i.e., she is quite the opposite of a natural law theorist), but she never talks about legislatures: how they shall be elected, their powers, what kinds of laws they may make, or their role within a constitutional government. Yet legislators too will not be angels, and one can be certain that (barring fundamental changes in human nature) they will be just as interested in power as legislators always are. What is to stop legislators in a Randian state from increasing their powers, meddling in the affairs of individuals, making bad laws, and the like? Again: "blank out". Rand argues that a constitution with no loopholes would end up "leaving the government with no power to enlarge the scope of its services at its own arbitrary discretion" (indeed she says at one point that the U.S. Constitution contains "a loophole", i.e., only one!). Yet, given historical experience, that is at best wishful thinking.
Indeed, historical experience, especially in the United States, indicates that government power tends to expand indefinitely even within a society that originally was quite opposed to arbitrary and unlimited government power. Why did Rand, who received her academic training in history, not see the seemingly inevitable rise and expansion of the state in modern times? On reflection, I've concluded that although Rand did away with faith and force (which she decried as the "destroyers of the modern world"), she did not necessarily do away with religion and government, which traditionally have been the institutional manifestations of faith and force. Rand's philosophy contains an element that is supra-ethical, in that she advocates a secular, rational, man-centered ideal that is in some sense religious while dispensing with faith (see her introduction to The Fountainhead). Similarly, her concept of government is the "unknown ideal" of a minimal institution that is strictly limited to the protection of individual rights through the objective use of force only in retaliation and never for initation. Rand seeks to banish faith and the initiation of force from human affairs, without necessarily banishing institutions or forms of living (religion and government) that have historically been based on faith and force.
Thus Rand might reply to my initial question with one of her own:
If you can conceive of religion without faith, why can you not conceive of government without the power to initiate force?
Yet I have my doubts as to whether Rand has established that either a rational religion of man or a voluntary government of objective laws is feasible. If morality or law is truly objective, presumably it can be discovered by reason based on experience and can be applied outside the context of a religion (even a humanistic one) or a monopoly government. Most pointedly, there is no reason to think that because a law is made by a constitutionally limited government, therefore it will be objectively valid. Indeed, the opposite is likely true. The principles of common law -- protection of life and property, restitution for harm, and sanctity of contract -- are common for a reason, because they are objectively valid and arise naturally in pre-state societies and in mutual associations that do not involve monopoly government.
Throughout her writings on the foundations of political philosophy, Rand sets up a false alternative between individual (and irrational) action and government (and objective) action. For Rand, it seems to be "government or nothing": only government can possibly enforce law; either we have an enforced monopoly on dispute resolution or we have chaos. Unfortunately, Rand utterly neglects the role of institutions other than governments in resolving disputes -- councils of elders, arbitrators, mediators, insurance companies, and the like. She claims that "if physical force is to be barred from social relationships, men need an instititution charged with the task of protecting their rights under an objective code of rules". Notice that an institution is required -- that is, only one. And that institution is to be an enforced monopoly: a government. (Indeed, her premises could be used to argue for one world government, since otherwise rights could not be protected across borders.)
Rand was not familiar with the scholarly studies of fully voluntary legal systems (such as Bruce Benson's Enterprise of Law and Randy Barnett's The Structure of Liberty) that were published late in her life or after her death. Thus we can almost forgive Rand for her cursory dismissal of non-monopoly political theories. But the question remains open: is the necessity of government an issue of philosophical principle, or is it a matter to be proved through human experience and advancements in the "technology" of freedom? Personally, I see the question of monopoly vs. non-monopoly political systems as a scientific or experimental problem, not a philosophical one: an area for research and experimentation, not philosophical dogma.
Interestingly, both monopolists and voluntarists (statists and anarchists) have argued that what is most important in the arena of social relations is objective law. The statists say that only government can ensure the objectivity of law; the anarchists say that objectivity is impossible when there exists a coercive monopoly in legal matters. So perhaps the focus on minimal government vs. no government is misguided. Adherents to both positions claim to be in favor of the rule of objective law -- or, to mint a word, nomarchy. The question is: what is the best way to achieve that? I submit that the answer to that question will be found not through insisting on some ideological dogma, but through the hard work of attempting to discover a scientific basis for objective law, and then applying it impartially to human relations within society. Only time will tell if Objectivists will contribute productively to that program of research and action.