Freedom Is Not Permission
These phrases are not mere rhetoric. They express a profound and elegant political philosophy. To understand it, we must begin with a basic presumption or default position. Logicians, lawyers, and laymen use such presumptions as the foundation for any argument. Presumptions of the “when in doubt do x” variety serve as starting points for any sort of discussion or agreement, so that we know where to go in the event that the agreement or argument fails. These sorts of “default rules” surround us every day. Many subscription services, for example, require people to opt out of annual renewal; unless they unsubscribe, the company presumes at the end of the first year that users want to pay for a second. These defaults are not set in stone, of course – the subscriber can decline the second year if he wants – but as a rule of thumb, it makes things easier by placing the burden on the subscriber to refuse a second year, rather than on the company to ask again after the first.
As this example suggests, choosing who bears the burden can have important consequences. If we start from a flawed initial position, we risk a dangerous and costly error. A subscriber who forgets to cancel might be surprised to see the second year’s charge appear on his bill.
Choosing an initial presumption is extremely important in the realm of politics or law, where the stakes are much greater. When we establish a presumption or a starting point for a political or legal argument, we are choosing what the normal rule will be, in the absence of good reason to deviate or in the event that we make a mistake. The obvious example is criminal law: courts presume a defendant to be not guilty and place the burden on prosecutors, which means that if the prosecutor fails to persuade the judge, or makes a mistake, the accused person is free to go.
In discussing politics, there are two possible candidates for an initial presumption. We might presume in favor of totalitarianism – everything is controlled by the government, and citizens must justify any desire to be free – or we can presume in favor of liberty and require anyone who proposes to restrict freedom to justify that restriction. Either everything is allowed that is not forbidden, or everything is forbidden that is not allowed. As Professor Richard Epstein observes, there is no third, middle-ground option, because there is no obvious midpoint between the two extremes: people will bicker endlessly about what qualifies as exactly halfway.14 So we must start by presuming either in favor of freedom or against it.
Yet these two candidates for starting points are not mirror images, and their differences are crucial. The differences are both procedural and substantive.
As a matter of procedure, starting with a presumption in favor of freedom is preferable because each step people take away from a state of liberty can be justified in theory by measuring whether they are better off. When two people sign a contract, they bind themselves, and in that sense are less free. But they consider themselves better off, and that is good enough, as long as they harm nobody else. It is not so easy to justify the reverse – a movement from a state of total unfreedom to one that is freer – because each step affects far more people. The totalitarian state is frozen solid, so that every action inflicts consequences on everyone else, and the slightest deviation from rigid order must therefore receive the approval of everyone affected. This means it is not always possible to determine whether people are better off at each step when they move in that direction. This, writes Epstein, “is why the restoration of even modest elements of a market system seem to pose such radical problems for Eastern European and Third World nations.”15
The point becomes clearer when we think about an individual: a free person can choose to become less free – he can sign contracts that limit his future choices, can voluntarily give up certain rights, or can surrender property he once could have used – but an unfree person cannot choose to become more free. Precisely because he starts out with no freedom, his capacity to choose alternatives for himself has vanished. He must ask his master for permission instead. This is why the road between freedom and unfreedom usually moves in only one direction. As Jefferson said, the “natural progress of things is for liberty to yield, and government to gain ground.”16
There is a deeper procedural reason why it is better to presume in favor of freedom than against it. The liberty presumption rests on a basic rule of logic: anyone who makes a claim must prove it – or, as a classic legal textbook puts it, “the issue must be proved by the party who states an affirmative, not by the party who states a negative.”17 In Latin, this is sometimes called the rule of onus probandi. Since proving a negative is technically impossible, the rule of onus probandi applies across the board to all claims: a person who asserts that the moon is made of green cheese, or that Herman Melville wrote Moby Dick, or one who claims that he may justly stop another from publishing his opinions or praying to the god of his choice, bears the burden of proving those claims.
The opposite rule – “throwing the burden of proof on the wrong side”18 – is a logical fallacy, and it too has a Latin name: probatio diabolica, or “the Devil’s proof.” Requiring someone to prove a negative is devilish because it perverts thought and leads to absurd results. It is also sometimes called “Russell’s teapot,” after an example given by the philosopher Bertrand Russell: nobody can prove that there is not a tiny teapot orbiting the sun, so small as to elude the world’s best telescopes.19 If you cannot see it, why, that just proves how tiny the teapot is! If you cannot detect it with the finest instruments, that only shows that you are not looking hard enough – and so forth. It is not possible to disprove even an absurd claim. Just as a person who says there is a teapot in space has the duty to prove it, so a person who claims the right to govern another person bears the burden of justifying that assertion. It is not possible to prove that one should not be ruled by another.
These procedural differences between the presumption of freedom and the presumption against freedom are a consequence of two substantive differences. First is the asymmetry between the past and the future. The past cannot be changed, but the future is what we make of it. That is why, offered the choice between a prize in cash or frozen assets of equal worth, a rational person chooses cash. Liquidity itself has value: cash can easily be converted into whatever the holder wants, while frozen assets may not serve one’s present needs, and selling or trading them takes time and effort. Freedom of choice is the liquidity of human action, just as property is human action in frozen form.
The second asymmetry is the asymmetry of personal consequences: someone deprived of freedom suffers a personal injury that is qualitatively different from the cost that a person suffers when he is stopped from taking away another person’s freedom. If Tom beats Joe or takes his property, Joe suffers a direct, personal injury different in kind from the harm Tom suffers when someone intervenes to stop him from robbing Joe. Our right to control our own lives and our rights (if any) to control the lives of others are simply not equivalent.
When thinking about government, therefore, it is better to presume in favor of freedom than to presume against it, not only because of the basic procedural rule of onus probandi, but also because of the substantive difference between freedom and its opposite. Simply put, the cost of having too much freedom is far smaller than the cost of having too little. At the very least, if we start with a presumption of freedom and later decide that less freedom would be preferable, we can move in that direction – whereas the reverse is not always true. And if we begin with the presumption of freedom and later conclude that that was an error, we are less likely to have hurt somebody as a result of that mistake than if we began the process by assuming away everybody’s liberty.
As