The Big Myth about Liberty
There is a widespread myth about the word ‘liberty,’ a myth propagated by philosophers from the “left” and “right,” by conservatives, radicals, modern liberals, and classical liberals.
The myth begins with a distinction: Liberty takes two basic forms: negative and positive. Negative liberty concerns the absence of constraints, impediments, or interference. For instance, a person has freedom of property—understood as a negative liberty—if others may not take her property or interfere with her use of it. In contrast, positive liberty concerns the power or capacity to do as one chooses, or the power to act autonomously. A person has freedom of property—understood as a positive liberty—if she actually owns and controls some property. Bill Gates and I both have the negative liberty to own a yacht, but only Gates can afford a yacht. He has the power to do something I cannot, and in that respect, he is more free.
Now for the myth about liberty. The myth holds that negative liberty is the special concern of classical liberals and libertarians, while positive liberty is the special concern of Marxists, socialists, and modern liberals. Many classical liberals reject positive liberty, saying that the power to do as one pleases or to make autonomous decisions, however valuable, are not really forms of liberty. If it were, this would automatically license government to do whatever it wants in an attempt to force us to be free. In contrast, many Marxists complain that if liberty is just about the absence of interference, then liberty will not be valuable or meaningful. Negative freedom, they say, is the freedom to be poor, to be unemployed, and to sleep on public sidewalks. Marxists might say, no one interferes with the homeless person, but he is not free in any meaningful sense. On these issues, the classical liberals in question are wrong, as are the Marxists.
Isaiah Berlin is a principal exponent of the myth. He might even be its originator. After distinguishing between negative and positive liberty, Berlin argued that the two concepts of liberty are not merely different conceptual categories, but rival political ideals, with conflicting implications about the proper role and scope of government. The assumption that different conceptions of liberty entail different political regimes recasts the semantic issue (“What is liberty?”) as a political one, where the debate is not merely about how to use the language but about how to use the police.
The myth is perpetuated because both sides to this debate make a common assumption: Liberty—whatever that is—is to be promoted by government in a direct way. Both sides agree that government has the job of promoting liberty (whatever liberty turns out to be), and both sides agree that the government should promote it in a direct manner.
When we say government should promote some value, it could do so directly or indirectly. The distinction is best illustrated by an analogy. Suppose you think government’s job is to promote commerce. A government might attempt to promote commerce directly, by creating new corporations, offering subsidies and grants to businesses, providing tariff protections, and buying products, or indirectly, by providing a basic institutional framework (such as the rule of law, constitutional representative democracy, courts, and a well-functioning property rights regime).
On the assumption that government should promote liberty directly, it’s vital to decide which kind of liberty is the real one, or the more important one. On this assumption, determining that ‘liberty’ is best analyzed as “the capacity to choose autonomously for oneself” really would license government to force us to be free.
Yet, this assumption is a bad one. Identifying a role for government—or any other institution—as a protector or promoter of liberty requires a real argument. It requires that we examine historical, economic, legal, and sociological evidence to determine how well (and it what way) government can do the job. Government gets the job of promoting a particular kind of liberty a particular way only if government is best qualified for that job. We discuss this point further below.
Positive and Negative Liberty
Berlin coined the positive/negative liberty distinction. (In fact, his use of the terms was somewhat different from the way we characterized them above. However, the terms have come to be used in ways slightly different from Berlin’s original meanings.) The negative/positive distinction naturally suggests that the categories of negative and positive liberty are supposed, jointly, to exhaust the possibilities. Not so. Berlin says historians documented two hundred ways of using the term, and he is writing only about two central ones. Below, we list some of the ways we have seen people use the term ‘liberty.’
(a) Hobbes describes liberty as an “absence of external impediments.” On Hobbes’s view, any obstacle whatever is an impediment to liberty.
(b) More specifically, we can define ‘liberty’ as an absence of impediments imposed by other people.
(c) Even more specifically, we can define ‘liberty’ as an absence of obstacles deliberately imposed by other people.
(d) Or we can define ‘liberty’ as an absence of obstacles wrongfully imposed by other people.
On a negative conception of liberty, it will be a matter of historical contingency whether a given liberty makes for happier or healthier or wealthier lives. Negative liberties are not guaranteed to make us better off, but neither is vitamin C, or exercise — so guarantees can be beside the point. The point of negative liberty has less to do with what liberty guarantees and more to do with what liberty gives people the chance to do for themselves.
Despite the lack of guarantees, the social sciences and history may well reveal that respecting negative liberties has a long, successful, non-accidental track record of making for better lives. It may turn out that the most effective way to promote positive liberty is to protect negative liberty. In any case, we can’t settle a debate about what negative liberty does for people by conceptual analysis alone. We need to investigate what happens to people when negative liberties are reasonably secure, and what happens when they are not. That is, if you want to know how valuable negative liberty is, you need to get off the armchair and go check.
(e) In a more positive vein, we can define freedom as a power to do what we want. Berlin would reject this as an analysis of political freedom (whether positive or negative), except in cases where one’s inabilities were caused by other people.  Still, even if such inabilities have no bearing on political freedom, they remain a part of the conceptual landscape of positive freedom.
Even on this positive, capacity-oriented, view of freedom, though, it remains a contingent matter whether increasing freedom makes for better lives. As children, and even as adults, some of our wants are self-destructive, and the power to satisfy our wants needn’t be good for us. Whether it is good for us will depend on the nature of our wants and on our level of maturity. (Note: We should avoid the heroic leap of inferring that if the power of satisfying certain of my wants is bad for me, then it would be better for me if bureaucrats and police had the power to stop me from satisfying these wants.)
We can’t settle any debate about what positive liberty does for people by conceptual analysis alone. We need to investigate what happens to people when positive liberties are present, and what happens when they are not. In other words, if you want to know how valuable positive liberty is, you need to get off the armchair and go check. We can easily imagine a world in which people have lots of positive liberty but remain miserable. However, as a matter of fact, it may be that when people have positive liberty, they tend to be happy. At any rate, philosophy cannot settle this debate.
(f) Moralizing the previous definition, we can think of freedom as a power to do what is right.
(g) Or we can define ‘freedom’ as a power to do what is right, free from all temptation to do otherwise.
(h) We note the possibility of a whole family of related conceptions according to which liberty is a power to do what we want, without self-imposed baggage (in other words being free of commitments or, more generally, free of plans, promises, hang-ups, and self-conceptions that no longer fit the person one has become).
An awkward, and by no means merely theoretical, puzzle is highlighted by this conception of positive freedom. Namely, we could have real choices and still be thwarted by factors like insufficient motivation, insufficiently independent judgment or insufficient information about our options, about our capacities, or – for that matter – about the authenticity of our goals. We might want too much, or too little, to be able to deal with our world of options as it is. In sum, not all the impediments to freedom are external. Psychological freedom, and how it relates to other forms of freedom, is the subject of the final chapter of our new book, A Brief History of Liberty.
We continue to speak of positive and negative liberty, but we remain aware that, as Berlin and as his critics stressed, positive versus negative liberty is a false dichotomy. Negative and positive liberty can themselves be viewed as clusters of related concepts, and there are conceptions of freedom such as Philip Pettit’s conception of freedom as non-domination that straddle the categories. 
Picking a Conception
People use the term ‘liberty’ to refer to a range of different things. Time-honored conceptions of liberty tend to be time-honored for a reason. They play different, often complementary roles in common-sense thinking. There is value in identifying the essence that these various conceptions all share, but there is also much to gain from being aware of the differences. Each of these freedoms is something people have for good reason struggled to secure. One is concerned with liberty in all such contexts, but the concerns one aims to mark by using the word are only related, not identical.
Some theorists think a minimal set of protected negative liberties is all we need in order to launch a society that, over generations, produces explosive gains in positive liberty. Other theorists seek guarantees and do not find them in a system of mere negative liberty. I might be free from government interference, free from oppression by a rigid caste system, and so on, yet I might remain unable to do much because I am poor.
Such issues are real, so it would be a shame to let debate about negative freedom’s real effects degenerate into a terminological dispute. Perhaps, as a matter of fact, negative freedom often leads to poverty. How would we know? Manipulating definitions would not tell us much. The point of defining terms is not to cut off but rather to facilitate debate: not to stipulate that negative liberty leads by definition to prosperity, but to be precise enough to make a question answerable. For example, where there are fewer obstacles to seeking employment of one’s choice (fewer migration restrictions, fewer licensing or union membership requirements), are there fewer unemployed people? If so, then we can infer (not in the way a logician deduces but rather in the way a scientist guardedly infers causal connections from empirical regularities) that negative freedom is in that respect positively liberating. We can ask well-defined questions about the consequences of specific forms of negative freedom, such as the freedom from trade restrictions or from state-mandated religion. If we can document trends, making the debate less about whether a trend is real and more about why the world sometimes departs from it, we have made progress in lowering barriers to understanding — which is what we realistically hope for from philosophy.
The Role of Government
Simply acknowledging positive liberty as a valuable species of the genus liberty does not commit us to any particular view about what regime promotes it best. We share Berlin’s concern about giving governments a license to do whatever it takes in order to promote positive liberty. (In the real world, to give government officials the power to do x is to hope that officials will use it to do x, knowing that, no matter who actually ends up holding such office, the person in question will duly pay lip service to doing x, then will use the power for purposes of his or her own.) None of the conceptions of freedom discussed earlier entails that it is a government’s job to secure that kind of freedom. Defining terms cannot settle a government’s proper role as protector or promoter of particular liberties.
Whatever Constant and Berlin meant to be doing by warning that different conceptions of liberty are also different political ideals, the takeaway value is not that different conceptions of liberty automatically translate into different government mandates, but rather that naïve conceptions of what is possible or what is at stake can leave people vastly overconfident about the likelihood of achieving a result by creating the political power to obtain that result by force.
Many people think that government’s job is to guarantee that we achieve a satisfactory level of liberty. However, there is a difference between ‘guaranteeing’ as rendering inevitable (as when an economist says doubling the minimum wage would guarantee rising unemployment) versus ‘guaranteeing’ as when the government expresses a firm intention. Clearly, guaranteeing something in the latter sense is no guarantee in the first sense. Imagine a world where every time a government guarantees that people will achieve a given level of welfare, an evil demon makes sure (guarantees in the other sense) that people do not. In that world, if you wanted people to be well off, you wouldn’t want to issue guarantees. You’d permit people to be badly off, because that would be their only chance to prosper in that demon-plagued world.
Of course, we don’t live in a world of evil demons, so perhaps that is irrelevant. Yet many real factors can and do disrupt, corrupt, or pervert our best-laid plans and guarantees. Therefore, imagining a world devoid of corruption and of unintended consequences—as some theorists like to do— is no more interesting than imagining a world of evil demons. We have to check how well legal guarantees work in our world. We want governments to issue guarantees only to the extent that such guarantees work.
Only so much of practical consequence hangs on how we define our terms. By contrast, a lot depends on what we want from our lives and from our communities — whether we want to be free to stand or fall by our own merit, or whether we want to be free from the risks and costs that go with personal responsibility. A lot depends on how powerful and unconstrained we want our government to be, which in part turns on how confident we are that it will be used by, rather than against, our children.
We think both negative and positive liberty matter. Negative liberty matters in part because it is a highly effective, if imperfect, way of promoting positive liberty.The result of freedom of thought, of freedom of association, of the division of labor within firms and of the specialization of roles that evolves between firms is that society becomes an unimaginably complex web of cooperation, moving ever further away from individual self-sufficiency. Although it may sound somewhat paradoxical, this is actually a contribution to positive freedom, because, as particular roles within society become redundant, a given individual grows less dependent on particular providers of a given service. Freedom in the positive sense can and sometimes does burgeon along with the increasing complexity of this web of interdependence. A Brief History of Liberty is a story of those preconditions of real choice slowly coming together.
 Thomas Hobbes, Leviathan. Indianapolis: Hackett (1994): 79. The context is that Hobbes is describing the impediments to the Right of Nature as the liberty of people to do whatever they judge to be most conducive to self-preservation.
 We acknowledge the possibility that conceptual analysis alone might establish a given liberty’s intrinsic value.
 Berlin says this in a discussion of negative liberty, but he never suggests he would categorize such inability as a lack of liberty of any kind.
 Pettit, Philip (1997) Republicanism: A Theory of Freedom and Government. New York: Oxford University Press.