By Jan Narveson

This substantial discussion was originally composed for the Stanford On-Line Encyclopedia of Philosophy; readers may wish to compare with the alternative exposition that they have adopted.

Much political and moral philosophy over the centuries has concerned itself with human liberty (and some even with non-human liberty, as in “animal liberation”). The philosophical outlook on politics known as libertarianism, however, takes this idea to its extreme, proposing to make liberty the only interest that a state may properly have with respect to its citizens. The question even arises whether this interest is incompatible with the existence of the state. In any case, it has serious implications for the philosophy of law as well as for moral and political philosophy more generally.

1. Introduction
2. Defining the View
3. Why Libertarianism?
4. Applying Libertarianism
5. Historical Note on Sources of the Liberty Principle


1. Introduction

1.1. Two Versions

At the outset, we must distinguish these two views:
(1) that liberty is the sole value to be promoted by governments and individuals (sometimes called the “teleological” version of libertarianism) and
(2) that liberty is our sole right (sometimes called “deontological” libertarianism; this is the view that the word “libertarianism”, unqualified, is generally taken to stand for nowadays.)

It is not clear whether the first entails the second, but quite clear that the second does not entail the first. The point of making liberty a general right is to prevent governments from forcing people to do things. On that view, helping to set other people free is something we may not be forced to do. By contrast, if the theorist merely makes liberty a value to be pursued, by whatever means, then governments and individuals could feel free to impose on some people in order to promote freedom for others.
We can align these two versions with a distinction proposed by various recent theorists, between “left” libertarianism and “right” libertarianism, the two splitting over the issue of the reach of private property. The “left” libertarian holds that the right of private property for individuals is either nonexistent or very much less than absolute, and that society needs to bring about a measure of equality in the distribution of certain things, usually natural resources, perhaps providing, to that end, a social minimum. Left-Libertarianism, then, gives rise, fairly readily, to support for the welfare state. Some communalists and socialists, especially in the late 19th C., claimed to be “Libertarians”. [Example, Kropotkin (1842-1921)]; a new anthology will be of immense value for pursuing these views [see Steiner and Vallentyne].

“Right-libertarianism” takes the right of private property to be absolute, or at least very strong (see below for the exceptions). According to that view, then, the welfare state is in principle wrong, along with a great deal else that is familiar in contemporary states. It has become the standard form of Libertarianism in recent times, at least in America; it was rocketed into prominence by the publication of Robert Nozick’s Anarchy, State and Utopia, which caused the view to be taken seriously in the Anglo-American academic world. According to it, our sole fundamental right is the right to liberty; all other rights are subordinate to that — they are either special cases of that one, or derived from it, directly or indirectly. Proponents of this version tend to argue that it is the only one that makes sense. Some account of the issues between them will be made below, but we will mainly concentrate on this second type, which can claim to be the more radical and perhaps more interesting for that reason. It is also the idea that is now standardly associated with the term. In the remainder of this entry, “libertarianism” refers to “Right-Libertarianism” unless specifically otherwise indicated.

1.2 Three Questions: Definition, Grounding, and Application

The area to be sketched in this essay can be divided into three general parts.
First, considerable attention needs to be paid to defining the view, which is widely misunderstood. Just what is the libertarian idea or principle? What would a principle of liberty look like? Can we formulate one that is clear and coherent (without looking crazy)?

Second, some attention is devoted to the foundations of the view: why should we be libertarians? How important is liberty, really, and in what contexts? Why, if at all, should we seek it? Are there any good arguments on its behalf?

Third, what are the real-world, institutionalizable, implications of the liberty principle — how does it apply in practice? What would a society enshrining liberty as its fundamental political value look like? Would it, for example, have a government at all? If so, what kind? If not, how would things now familiarly done by government get done? Or would they?

The three questions cannot, of course, be sharply and definitively distinguished, but dividing the questions in this way should help to clarify the issues. We’ll pay some attention to all three questions here, but especially the first, and also, sketchily, to the history of the subject.

2. Defining the View

2.1 Rights

What is a right? That there is a close connection between liberty and rights becomes clear when we consider the general notion of a right. The by-now generally accepted analysis is that rights are duty-imposers: to say that person A has a right to do some act or set of acts, x, in relation to some other person or group of persons, B, is to say that there is something about A such that B has a duty, is required, to act in certain ways — at the least, non-adversely — in relation to A’s attempting to do x. The something that grounds these duties would have to be identified and clarified by the theorist, and precisely which duties the right entails in this case would likewise have to be clarified somehow; as would just who is in the B-position.
A very important aspect of the idea is that B’s duty is of such a kind that B may legitimately be compelled, by force if need be, to do or refrain from the things that A’s right imposes on B. This last notion, which is especially appropriate for political contexts, makes particularly clear the connection of right and liberty: one person’s right reduces, that is, it provides justification for limiting, another’s liberty. The libertarian insists on minimizing this curtailment of liberty: only the need to respect other people’s liberty, according to the libertarian, provides the needed justification for limiting anyone’s liberty.

2.2 Negative and Positive Rights — a Basic Distinction

Libertarian theory, as will be evident from our next several discussions, makes major use of a distinction between two sorts of rights, distinguished by the kinds of duties they entail: negative and positive, respectively.
(a) Negative Rights: A has a “negative” right against B if what B must do in order to respect it is to refrain from various possible acts, namely ones that would hinder, or interfere with, A’s attempts to do x, or (if this is different) damage A’s person, or, more generally, worsen A’s situation in whatever respect is in question in the context in which the right under discussion obtains or would obtain. The proponent of the right could further specify which sorts of interfering actions were forbidden – perhaps not all. For instance, B may not do what would make it outrightly impossible for A to do x, no matter how hard A tries, but perhaps B may do certain acts that would make it slightly more difficult for A to do x.

(b) Positive Rights: A has the “Positive” right to do x if B must not only refrain from hindering A, but also do things which would positively assist A to do x if A is otherwise unable to do x unaided. A further important clause would also except the case where A is assisted by the purely voluntary actions of others. A has the positive right to do or to have x only if that right requires certain people, B, at least in certain possible circumstances, to assist A in doing x, or to supply A with x, whether B acts willingly or not. Obviously, the question of how much B would have to help A, that is, how great a cost B would have to bear before his obligation ceased, is a very important question and would need to be somehow specified by the propounder.

The difference between the two can be materially tiny in some cases, but can be, and usually is, very great. Take the general right to life: in its negative version it says only that others must not kill (take the life of) the rightholder; but in its positive version, it would also require that others do something to help save the rightholder’s life if it is possible for them to do so. (How much? That becomes a crucial question.) The distinction cuts across various others that have been confused with it. For instance, “action” rights and “welfare” rights are both susceptible of negative and positive forms. We could have a positive right to perform certain actions, entailing that others are required to help us — say, by supplying us with canes; as well as negative rights which merely forbade others from getting in our way. Similarly, I could have a merely negative right to welfare forbidding others to lower my existing level of welfare, whatever it was, but not obligating them to raise it if it got too low. By contrast, a positive welfare right — what people usually mean by ‘welfare rights’ – obligates others to do what is necessary to keep the rightholders’ welfare from slipping below, or to raise it up to, some threshold, which again would have to be specified. The importance of this distinction is that the libertarian holds that people have no basic positive rights — that all positive obligations have to be in some way assumed or undertaken by the obligated individual, e.g. by promising that he will perform the indicated actions.

Some writers [e.g., Shue] claim that the negative/positive distinction is ill-formed or fake, on the ground that negative rights require police and courts for their enforcement, and yet those things are positive acts that somebody would be required to do. This is mistaken, however. The question of what the right is a right to do, and who if anybody will enforce it, are separate. If our rights are purely negative, it will also mean that no one has the duty to enforce them, as such, although everyone has the right to use whatever means he can avail himself, with the cooperation of others who also have no duty to do so, to secure his rights. The distinction between negative and positive is quite robust.

It must be noted that libertarianism is ordinarily understood to be exclusively a theory about enforceable rights. It may ely enough be held that there are other moral categories in which we may assert that people have positive duties of a weaker kind. For example, the libertarian may hold that charity is a non-enforceable, but nevertheless real, duty. We should feel badly if we do nothing for the sick and exploitred if we could easily do so, for example. And of course there is no objection to holding that it would be a good thing if various ends were brought about, and to hold that it is morally commendable to try to bring those things about, and even that those who do nothing about those things are bad persons. Ttheory’s central focus, in short, is on the morally legitimate use of force.

2.3 Liberty

The short answer to the question, What is liberty?, is that people are at liberty to do some particular thing when nothing prevents them from doing it if they want to. They are at complete liberty if nothing prevents them from doing anything they want to do. But this is clearly too general for our purposes, because political and moral philosophy are about what people ought and ought not to do, and some things that keep us from doing as we’d like are simply natural — diseases, hurricanes, the size and strength of our muscles, and so on. So we must add that for social and moral purposes, what we are interested in is the absence of impositions by other people, specifically those impositions that are caused by their intentional actions. If A ties B to a tree, then, prima facie, A has imposed on B – acted on B in such a way as to prevent B from moving, and thus from doing a great many things B might like to do. But if A hits B, yet leaves B able to continue moving, say, it is less obvious that A thereby curtails B’s liberty. Yet we can point out that A has brought it about that B is unable to continue his preferred course of experience, which he enjoyed prior to being hit, and in particular that A has upset B’s intended course of action regarding B’s own head.

It is obvious enough that, very often, events other than the actions of other humans impose on us in various ways, as when lightning strikes us dead. It might be thought that libertarians should favor the removal of these nonhuman obstacles, where this is possible, with a view to promoting people’s liberty. For discussion on this, see below, 2.4.

Berlin’s Distinction of negative and positive liberty

Sir Isaiah Berlin [Berlin, “Two Concepts of Liberty”] distinguished between negative liberty, defined as above, and “positive liberty”, which refers to an individual’s acting as he “really” wants to, rather than what he merely thinks he wants: you are free if you act in accordance with your “real nature” — freedom as “self-realization.” But that confuses two different distinctions. First: The opposite of negative liberty, the absence of external obstacles, should be the presence of conditions that actually enable people to do things — muscles, natural talents, or health, say — rather than their aspirations to self-realization. Second: Berlin’s category of “positive liberty” reflects the intrusion of a quite different political outlook — conservative as opposed to liberal. [Narveson, 2000] Liberalism has it that political institutions should cater to people’s wants, as they are, rather than some version of the good other than what people actually have. The latter would be crammed down their throats and is on that accord a violation of liberty. It is thus not a realization of the same thing that negative liberty consists in. That is done much more easily — by getting people to desist from interfering with and damaging others, or threatening to do so.
Now we can say that negative rights are rights to Berlin’s “negative liberty” — the absence of interference. Positive rights, however, need not be rights to the sorts of things he identified with positive liberty. It would even be possible to call for positive rights to negative liberty, as well as negative rights to any or all of the positive liberties that Berlin identifies.

2.4 The Liberty Principle

The “Left-Libertarian” versions

Hearkening back to our initial distinction between the view that liberty is a good that should be maximized in society, and the view that liberty is a right, not to be infringed, let us ask what the leading principle would look like on these two views. On the first kind of view, the obvious suggestion is that society should maximize liberty for all. But to say this is to get into the very intractable problem of “measuring” liberty. The terminology of “maximizing” liberty suggests that a net gain would be had by curtailing Smith’s liberty in order to increase Jones’s to a greater degree. To say such things we need a metric for measuring liberty. But it’s pretty clear that we do not have one. [See Wolff, pp. 89-92; Steiner (1983); Kymlicka, pp. 135-6.] We can say with some confidence, sometimes, that Jones has more liberty, or less, given a certain event, than he had before: if his right arm is now missing, there’s a lot less he can do. (He also feels severe pain, of course, and he doesn’t like that either.) But can we say that Smith has more liberty than Jones if Jones is in jail and Smith is not? Perhaps: but what if Jones, despite his incarceration, can do all sorts of wonderful things that he very much wants to do, while the neurotic Smith is scarcely capable of doing any of the things he’d like to do, even though he’s free to come and go? These and many other examples dispel quickly any idea that we can confidently talk of maximizing liberty across persons.

Intra-personal comparisons, to be sure, are quite another matter. A given individual may well feel that he or she is more free in one condition, less so in another. And we may certainly say that some infringements or violations of people’s liberty are greater or lesser incursions. There, the measure will be the individual’s own overall sense of good. How much a given restriction on my right to be free matters is frequently an important question. But only sometimes would the variable along which value is determined be exclusively or even primarily concerned with how “much” freedom he enjoyed in that condition. Someone might prefer life in a fairly despotic place, in return for being able to live with the woman he loved, for example.

The Standard Version

When we turn to the “rights” version of libertarianism, the now-standard one, things look more promising. In this version, each person is to be entitled to do as that person likes, or judges best, except only when his or her actions would impose on others — would interfere with the intended, desired courses of action of someone else, or (if this is different) damage that person, in the sense of doing what that person did not want to be done with or to his or her body or mind. Hobbes talks of “seeking peace”, that is, of not “making war” on others; Locke of not “harming” them in respect of life, health, liberty, or property; Kant, of acting only on maxims that “can coexist along with the freedom of the will of each and all” — a formula echoed in the contemporary American philosopher John Rawls‘ formulation of a liberty principle, that “each person participating in a practice, or affected by it, has an equal right to the most extensive basic liberty compatible with a similar liberty for others.” [Rawls, p. 60 — also echoing Hobbes’ “Second Natural Law” (Leviathan, Ch. XIV)]; and John Stuart Mill, who insists that society may use force against people only in self-defense. Interpersonal measurement would seem, in principle, not to be required on this view, for either one person has invaded another, or he has not. How much damage his invasion has done is important, for determining the compensation that might be due from the assaulter; but it is not necessary to say that one person is more or less free than another in order to determine the output of the principle. If, like Rawls, we talk of “equal freedom,” all that means is that everyone is entitled to this general liberty – not that there is something or other of which each person is entitled to get a certain “amount.”

Note that in these formulations, the emphasis is negative: people encounter each other, and in doing so, they are to refrain from actions that would cause the other person harm, danger, disease, and the like. Any other actions are permissible, whether or not they have the effect of “maximizing” something – they do not even to maximize liberty itself, nor of course anything else either.

A noted above, plenty of restrictions on human actions stem from natural causes, not just the actions of other humans. The left-libertarian may hold that society should attempt to remove these other restrictions when it is possible to do so without excessively imposing on people. This illustrates again the need in this type of theory to make interpersonal measurements of liberty. For a given natural obstacle to be moved requires someone some effort and time. The left-libertarian may argue that there would be a net gain in liberty by requiring some to expend their energies removing those things. For example, biological researchers clearly do a great deal to enable us to be more nearly free from diseases. Nevertheless, the standard version of libertarianism forbids us to conscript biological researchers, or any other kind of human effort along this line. If Albert would like some natural obstacles to his liberty to be removed, and someone else, Brenda, could do this, Albert must either persuade Brenda that it would be the thing to do, or offer her something in exchange that she would find sufficient to induce her to do this – or Abert must go without. It is this kind of case that makes libertarianism such a distincitive philosophy.

The Emphasis on Action

Libertarians are concerned with restrictions, imposition, on others’ liberty. This might be thought myopic: are there not other evils we can suffer besides impediments to our courses of action? A proposed answer to this is to be found in the section on Self-ownership below. (2.6) The point is that when we act, we use our limbs, our brains, and the other parts of ourselves, and to damage any of those things is to impede the agent’s capacity to do as he or she prefers. Whether that is too much of a “stretch” is matter for discussion and analysis of course.

Intentionality and Responsibility.

Are impositions to be regarded as wrongful only if they are intentional? Here we must distinguish two views. One is that you interfere with my liberty only if you do something which is intended to interfere with my liberty. That would be too restrictive: your actions can restrict my liberty even though you don’t intend that. The other view is that you must intend something or other — be within the reach of intentional description — when you act, for otherwise whatever you do that restricts my liberty does not violate it. No question of fault can be raised regarding bodily motions that are not intentional at all; indeed, these are not what we would normally call “actions” at all. As you tumble down the stairs, having been pushed, you may crush my leg, thus preventing me from playing tennis next day; but you are not at fault. Despite its not being an action of yours, people in your path may take steps to protect themselves. My right to protect myself doesn’t just extend to protection against other people’s intentional actions. If there is a rights-violator in this example, it’s the person who pushed you downstairs. On the other hand, you may be held responsible for possibly unforeseen side effects of other intentional actions. The issue here is whether the libertarian means to hold to a standard of strict liability. The working answer is that we do, up to the point where it was out of the question that the agent could have foreseen the result even with maximal information. But this is a debatable issue. [Nozick, ch. 4.] And in any case, we have the right to do things about the unintended, accidental actions of others, as also their sheer unwanted presence on our property.

Internal obstacles to freedom

It will certainly be said that a person might suffer a restriction of liberty at the hands not only of other persons, but of himself. The addict, for example, is reasonably described as being unfree to refrain from taking the next fix, despite the fact that nobody seems to be forcing it upon him. But two points need to be made. First, this does not require an alteration of the definition of liberty. After all, it remains that the addict, if indeed his liberty is restricted, is so because something — in this case, something internal to the addict, though arguably external to his “inner self” or his will, rather than some other person’s actions — prevents him from doing what he wants. He wants to take the next fix, yes — but he does not want to want to do so: he’d like very much not to be in the situation he’s in. But something prevents him, and it is something which, despite being inside of him, he finds very difficult to control. Second, and more to the present point, a theory of political and moral justice, justice among distinct people, can insist that the addict’s problems are his own, not other people’s. To argue that they are is to advocate a wholly different outlook, rather than to identify an inconsistency in libertarianism.

Liberty and The Good for Man

The libertarian takes it that liberty is the absence of obstacles to doing what one wants, or might want, to do. What about obstacles preventing you not only from wanting something, but preventing you from doing what would be good for you even if you don’t want it? It is unclear that we should put theses about this under the heading of “liberty”. The claim that the person in question is not at liberty to do x is the claim that she could not do x even if she wanted to; for if she could do x if she did want to, then the claim that her not wanting to is an “obstacle”, something that interferes somehow with her “liberty” is unintelligible — obstacles to liberty are obstacles to one’s will. Or if she can’t do it at all, no matter how much she wants to, then the connection with will is again affirmed, for in that case she can’t do it at will. In any case, libertarians insist that matters of this kind are for the agent to decide about, and not ones that one person may properly decide for some other (normal) person. Views that we may force people to be good whether they want to be good in the proposed way or not are rejected by libertarians. But of course individuals may change their views of what is good for them, and may try to force themselves to adhere to those views.

Actions and Inactions

An extremely important issue concerns the bearing of nonactions and omissions, by comparison with actions. Suppose that unless A does x, B will be unable to do y, which he wants to do. x is, we’ll suppose, an action that would cost A an appreciable amount of time and energy, money, or whatever. Should we say in this case that A’s inaction is an obstacle to B? This widely held view is surely wrong. It would, from the point of view of B, be very nice, certainly, if A did x. but the idea that the nondoing of x constitutes a deprivation or incursion on B’s liberty is absurd. The test here has been supplied by Gauthier. In A’s absence, B would also be unable to do y; so A’s presence cannot be what causes B’s inability. Therefore, A’s activities, including his selected inactivities, cannot be said to have caused B’s situation. We cause whatever happens in consequence of our actions that would not have happened anyway, had we not done as we do; we do not cause the lacks that we would be able to rectify if we chose. Sometimes, of course, people have specifically contracted for positive obligations, such as to control the flow of water over a certain dam. If they then neglect those obligations, they are at fault. But that is because of their prior undertaking of the obligation, and not because their inactions are, simply as such, obstacles to others. So the difference between imposing obstacles on others by our activities and failing to supply deficiencies, answer to needs, or in general make improvements in the situations of others that we possibly could make, remains absolute. Killing is conceptually different from letting-die.
This does not mean that there can be no case for requiring people to help others; it does mean that that case cannot rest sheerly on a right of liberty.


Coercion is usually thought to collide with liberty. Does the coercer impose costs on his victim? This has been denied, e.g. by Steiner [Steiner, 1994, pp. 22-32] who points out that to coerce is to threaten what would indeed cut into liberty – but if the threat is effective the violence doesn’t happen, leaving the victim free to do as he wishes.

But it is wrong to conclude that coercion does not curtail liberty. Coercion is the credible threatening, by individual A, of adverse consequences to the threatened person, B, if B does not comply with the will of A in some respect that A prefers not to comply with. The baseline condition at the point where the threat is made is preferred by B to any of the outcomes that are available to B, given A’s credible threat. B would rather continue down the street, wallet intact, than forego it, be beaten or shot, or try to counter A’s threat by force of his own. But A’s making the threat, if credible, therefore worsens B’s situation, and so his liberty is diminished: he is no longer free to select certain preferred options which were previously available to him. Coercion is therefore condemned by the libertarian, along with overt force. Liberty is the absence of obstacles, imposed costs, the coercer does impose a cost on his victim.

Much recent work on coercion has concluded that coercion is an essentially normative concept. [see Nozick] The foregoing analysis, [due to Rhodes], shows that not to be so. The victim, B, must of course evaluate his options, but the claim that B has ranked his options as he has is not an evaluative claim, but a fact about B’s values or preferences. He is coerced whatever one thinks of those preferences. So a would-be libertarian may maintain that coercion is wrong without circularity.

2.5 The Baseline Problem

If it is possible to apply the idea of encountering others in a way that does not worsen them, then there is hope of proposing as a general principle that all of us are to refrain from acting toward others in ways that impose on them, or worsen their situations. Liberty, says one recent writer, is “the absence of imposed costs.” [Lester, pp. 57-59] Will this work? Some writers have professed to doubt it, claiming that we would have to have a background of laws or rules to determine who was interfering with whom — we can’t just recognize as a fact that A was imposing on B or vice versa. [See Viminitz, for a lucid if formal exposition of the claim.] Those who say this do not press the question of whether the background of rules they claim to be needed for this purpose is arbitrary or not. If it is, then presumably there are no impartial, nonideological principles of politics at all. Libertarians, on the other hand, are committed to holding that at root, imposition is naturally discernible — that there is a nonarbitrary, natural basis for saying that certain cases exemplify imposition while others do not.

2.6 Ownership

To own something is to have the right to determine what happens to it, insofar as the owner is able to determine that, within the limit of the rights of others. To have the right to x is to have the right to do with x as one wills (within those limits), and therefore the right to allow or forbid any others to use x except with one’s own permission. Since acting simply is using one’s body and/or mind, it is clear that the right of self-ownership is precisely the same thing as the general right of liberty: (social) freedom of action is, obviously, freedom to use one’s body and mind, this being what action consists in, unobstructedly by others. [Narveson, 1995, p. 27]


The natural baseline to which libertarians appeal in the first instance is our bodies. One person, A, striking another, B, or stabbing him, shooting him, or in innumerable other ways damaging B’s body – these present a large body of clear cases to which we can appeal. Of course, the question would arise, whether the object of these incursions has done something to deserve the ill treatment in question. If we can answer that question objectively, then we can identify a very large class of paradigms for getting the initial idea off the ground. We can examine the history of person B and find out whether there is any person C, such that B has at some previous time struck, killed, maimed or otherwise damaged the body of C (who might also be A, of course). If B’s not having done so were sufficient for “innocence”, then we are free to formulate a principle to the effect that if A does this to B under that condition, then A has violated B’s liberty. It is not sufficient, of course, but it is a useful start. (See next paragraph.) The libertarian holds that we ought to make anyone’s liberty into a right: that is, that we ought to make it the case that impositions or proposed impositions against anyone’s liberty is a ground for taking action to rectify or prevent them; and that is what the Libertarian Principle does. This right is equivalent to a right of self-ownership. Each person would be regarded as “owning himself” in the same straightforward sense as that in which we can own all sorts of things such as cars and footballs: namely, being able to do what one wants to with the self in question; at the same time, one is not able to do that with others; rather, their willingness or consent must be established before one may do things with others.
There are many problems in attempting to work out this idea. Notably, there is the question of how to decide who must give way when paths cross. Do we “own” things like paths and routes when we walk about in otherwise unowned property? Whatever we say about this, it is clear that people need to, and will, devise rules to regulate traffic on these matters, and it may be very difficult to do this on the basis of the pure liberty principle if we do not make some kind of assessments of degrees of freedom after all.

Nevertheless, there is a natural basis that will be easy to apply in some cases: first occupancy or first use. First use or occupany is important because if Alice is at a certain place at a certain time, performing certain types of action that she has initiated, and Bob comes along at a later time, then in order for Bob to be able to use the paths, or natural resources, that Alice has begun to use, assuming that Alice does not wish to share these with Bob, then he would have to use force to get her to accommodate his wishes – and that is exactly what libertarianism proscribes. First-use is, of course, limited by many things, notably cases in which it would be hard to say who was “first”. Thus if there is a “commons” among some group of people, who long have felt free to use it and none has any idea who might have first established it, no member can claim, on libertarian grounds, the right to privatize some portion without negotiation with the rest: all members can be said to be equally “first.” (See next paragraphs.)

External Ownership

Everyone’s having rights over their bodies and minds limits everyone’s actions considerably; but they also leave a large area for maneuver — literally, in that we can walk, run, or make various other motions, in many circumstances, without colliding with anyone. Sometimes we can say of someone, B, who collides with A, that B has “crossed A’s path”. There is often a basis for awarding right-of-way to one person (or vehicle, say) rather than to another, and declaring those who run afoul of it as being in the wrong. When we cannot say this “naturally”, we can do something else: the several of us in the area at a given time can make agreements to the effect that this is Smith’s territory, that is Jones’s, this is common property that both can use provided they respect certain rights of way, and so on. What remains, however, is the priority of those who come first — who happen upon unused resources and undertake use of them before others come along. The first user invests energy and attention in this; his efforts can be undone and put to waste by others who do not respect those earlier efforts. The libertarian accounts this as an invasion, and so forbids it. The rule when encountering persons already on the scene is to negotiate rather than to use force. This, plus our interests in using things, is the basis of the right of property in objects outside our bodies. People begin by using things in situ; once use is undertaken, further alterations at the hands of others must be by unforced agreement between the original user and the would-be new user. This is the basis for economic exchange, as persons undertake to promote their goods by offering services, such as the transfer of rights to goods they have legitimately acquired, to others in exchange for preferred services or goods in the legitimate control of the latter.

First-use and first-occupany have several problems, as has been widely recognized. Robert Nozick [see Nozick 1974] offers a famous example: if he pours his can of tomato ketchup into the sea and stirs well, does he thereby come to own the sea? Clearly not. But it is less than clear why this is so obvious, and quite important. One obvious point is that the sea is used by many, and its shores occupied by many, who were there before the ketchup-pourer. Another is that the difference made by the ketchup to the whole sea is infinitesimal, making the claim far too thin. But establishing precise criteria for such things would be extremely difficult. Nor is it usually necessary.

It is a matter of much discussion whether there is an inherent limit, for example, to the amount of a natural resource that any individual may legitimately acquire simply be being the first user of that resource. “Left libertarians” generally assert a basic right to an equal share of natural resources, for example, and many libertarians think that some kind of limitation is inherent in the “Lockean Proviso”, as it has come to be called. [see a famous discussion in Nozick 1974, pp 175-182] Others reject the proviso in any other sense than that inherent in the general idea of libertarianism, namely that people may acquire whatever they can of natural resources, so long as they do not thereby interfere with the previously acquired property and the previously initiated activities, or of course the bodies, of others [see Narveson, 1999]. How far the libertarian can plausibly get on this matter is not easy to say. On the other hand, introducing considerations such as equality, or need, or other possible dimensions, deviates from the basic idea; a theory drawing heavily on any other source than liberty gets us into territory thought to be foreign to libertarianism.

Lbertarians point out the social advantages of the institution of private property, which are indeed many – indeed, those advantages were noted by Aristotle, who points out that individual people take care of their own things, and do not take good care of common property. But two points need to be born in mind. First, the libertarian insists that people have a perfect right to establish communal group-living if they like; the only concern of the theory is that relations among people be voluntary, and not that they be of one sort rather than another. Second, it is very unclear whether the appeal to the social advantages of private ownership, with its implication of free enterprise capitalism, is really a fundamentally different type of justification for the libertarian rule or not. Is it really logically accidental that free-enterprise societies end up materially wealthier than others? That remains an interesting subject for reflection.

2.7 Liberty and the Free-Will Question

All talk of actions ‘insofar as they are within our control’ raises the famous question about the “freedom of the will.” Aren’t human actions always ultimately caused by something or other outside themselves? We have no choice about who our parents are, what genetic profiles we inherit, what influences we will be exposed to in infancy. Do influences we have no control over ultimately determine what we want to do, and thus what we will choose? If so, how can we be “free”? The question arises, too, whether it is rational to hold people responsible for their actions, under the circumstances.
Pursuing these questions would take us far afield from the present inquiry. The best we can do here is to point out that theories about determinism, if true, must be compatible with the known facts. People somehow make choices, which make a lot of difference to what they do. We also know that people can be influenced by threats, offers, advice, and exhortations from others. No theory can deny these facts. Social philosophy must assume these familiar things without, generally speaking, trying to explain them. In this respect, as in most others in what follows, we remain within the realm of what philosophers have recently come to call “folk psychology”.

One other point should be made. Some writers appear to think that whether metaphysical determinism is true makes a significant difference to ethics. But that is very doubtful. Suppose we argue that since every action is determined, blame and praise can never be reasonable, and for this reason we oppose, say, the institution of criminal punishment. But this would make no sense — for if the general theory is true, it must also be true that we were inevitably determined to blame these people and praise those, etc. The view that we “should not” blame because determinism shows it to be “unreasonable” really presupposes that determinism is either not true or that it is irrelevant, since it cannot provide any reason justifying any particular choices as compared with any others.

2.8 Redistribution and Equalization

Perhaps the most distinctive and controversial feature of libertarianism is its opposition to “redistribution” — taking voluntarily-made earnings or products from some persons and giving them to others without the contributors’ consent, in the interests, usually, of sheer equalization or of fulfilling needs. (The view earlier distinguished as “left-libertarianism,” of course, does not make the same objection.) Libertarians sometimes say that to do this is essentially to render the involuntary contributor a slave to the persons imposing the redistribution, or perhaps their recipients. [Nozick, Ch. 7; Thomson, 49-65; Cohen (1995), chs. 9, 10] Perhaps invoking “slavery” is too colorful, but the issue is straightforward enough: are actions that impose on one person, against that person’s will, for the purpose of promoting the good of someone else, legitimate? To this question, the libertarian responds in the negative.
To have a clear view of the issue, we must distinguish between acts that, at the moment, are against the individual’s will right now, so far as we know, but would be agreed to by her if we had time to ask, and if she had time to reflect. We must also take into account that sometimes we have “package” offers, and would agree that we are sufficiently recompensed by some benefit to make us receptive to the cost being imposed without our immediate agreement to it. Nevertheless, it seems possible to formulate the idea of a “net” cost to someone, a cost not accepted in context, all things considered, by that person, who we will assume is also innocent of past or intended evils to others. The libertarian claims that imposing such costs is never justified unless, in context, that individual’s actions would in turn harm some further person. The crucial case will be where that isn’t so either, but that the person benefited is, say, in desperate need — at death’s door — while the benefactor can readily afford the incursion proposed.

Sharpness of Margins

The last point brings up an important question about the precision or sharpness with which we can draw lines around what is one person’s and what is another’s. Suppose that in A’s efforts to rescue B, A brushes lightly against C: does that render A’s efforts illegitimate in the libertarian’s view? Few would say so. Or suppose A must make some minor use of B’s property, without permission. If A needs to take one apple from B’s large orchard, thus keeping C from starving, should this deter A? It should probably be assumed that the theory is not so precise as to make those actions clearly wrong. [Thomson]

3. Why Libertarianism?

The issues of meta ethics are extensive and hundreds of thousands of pages in books and articles have been devoted to them. But foundations for views as contrary to widely accepted views as Libertarianism would seem to be a more urgent matter than foundations for ordinary views, whose very ordinariness recommends them to people. The brief discussions that follow serve to indicate the varieties of approach that have been taken to this matter.

3.1. Natural Rights Intuitions

The American Declaration of Independence begins with a famous proclamation: “We hold these truths to be self-evident…” Many libertarians do seem to regard it as self-evident that we have the general type of rights that the libertarian principle asserts. [Hospers] There are two problems with this. The first is the general problem in any appeal to intuition: what if others don’t have the same intuitions that the author does? If all we have to convince others is a statement of our position, and they don’t accept it, the proponent is in a rather weak position, dialectically speaking. [Narveson, 1988, pp. 110-121] For moral and political views especially, there is ample reason to think that we should be appealing to premises accepted by all — clearly not the case with the libertarian principle. But that brings up the second point, which is that what the libertarian can plausibly claim to be, if not exactly self-evident, yet very widely accepted, are claims such as that people do own themselves and that aggression is wrong. However, what makes libertarianism distinctive is its claim that the only human actions that we are justified in resorting to force to control are aggressive ones. A great many people think that human need, for example, or perhaps equality, also has a claim on us. We can’t refute the claim that we are justified in forcing well-to-do Peter to do what would keep down-and-out Paul from Starvation merely by pointing out that it would be using force against innocent people — that is, people innocent of the use of force and fraud against others. That, after all, is the issue.

There is also some danger of confusing definitions with substantive principles. Consider the formulation, “it is wrong to steal other people’s property.” But what can we mean by “their property” if not what belongs to them, that is, what they have a right to? That people have a right to what they have a right to is, of course, true, but it hardly shows that what this or that person claims to own really does “belong” to him in the strong sense that there can be no just reasons for depriving him of it. The libertarian purports to have a strong substantive principle. The person who makes x out of previously-unowned y, says the libertarian, owns x; the history of his dealings with y is what justifies the restriction of others’ activities in relation to x. That is not pleonastic or question-begging; the libertarian claims that it is also right.

3.2 Theological Appeals

John Locke, in support of his very libertarian-looking Law of Nature, “For Men being all the Workmanship of one Omnipotent, and infinitely wise Maker … they are his Property …” [Locke, sect. 6.] Such appeals are less frequent than they once were, but still influential in some circles. But they have too many insurmountable problems to be of any use. For one thing, many people do not accept the claim that there is a god, of any sort; and others might well deny that god is a libertarian. In general, a believer will ascribe to his god the kind of moral beliefs that believer actually has, so that appealing to theology to try to settle disagreements about it can hardly be expected to help. Meanwhile, the reach of moral and political principles is obviously such that theorists cannot just ignore the existence of billions of apparently reasonable people who don’t accept his premises. Appeals to theology to settle issues in moral and political theory are, necessarily, useless at best, and of course potentially irritating or worse — as the Thirty Years’ War illustrates.

3.3 Aristotelian Views

Some prominent libertarians — notably those who take inspiration from the writings of the novelist Ayn Rand — hold that a proper appreciation of human nature will show us that the virtuous person will be libertarian. The theory appeals to teleology, the thesis that living things are essentially engaged in a developmental process aimed at “fulfillment”; in the case of humans, this would, of course, be human fulfillment. In turn, to live successfully is above all to live rationally, and this in turn leads us to recognize a number of virtues, prominent among which is productiveness. For this we require freedom. More generally, “the right to liberty specifies a norm for social behavior that protects a necessary condition for human life.” [Rasmussen, p. 48] In order to make all this work, however, it appears that the major questions are begged, or misleading appeals are employed. Libertarian-level liberty, for example, could not be necessary for human life, since any libertarian would agree that liberty is rare and full libertarian liberty almost nonexistent — yet here we are, six billions of us, and almost none have lived in thoroughly libertarian polities. That political and moral liberty as described by libertarians is literally necessary for life is clearly false. The appeal must be, not to living at all, but to living well.

Even then, many qualifications are needed; and it is surely not at all obvious that we cannot live well at the expense of others. If “the practice of virtue requires liberty,” [Rasmussen, p. 49] then, the net effect seems only to be to proclaim that liberty is indeed a virtue. But that leaves us with the original question: why is it a virtue, and in particular, why is the singular adherence to it enjoined by libertarians a virtue — a virtue that dominates all others, practically speaking?

3.4 Rationalism

Another argument [Hoppe] has it that people effectively presuppose self-ownership even in undertaking to argue for something. Thus, one’s vocal chords and in general one’s capacity to produce speech must be thought by you to be yours in the process of argumentation. One who denied this, he urges, would be committing a sort of logical contradiction. In similar fashion, Thomas Hobbes argued that promise-breakers contradict themselves [Hobbes, Ch. XIV]. But Hoppe seems here to have confused de facto control with ownership; as we noted above, you can’t get from one to the other just like that. One who argues does, of course, employ lips and the like, but he need not claim that he has a right so to do. A similar mistake is found in Alan Gewirth’s arguments for human rights [Gewirth, 16-20; Narveson, 1997, 487-489] What he can’t do, of course, is reasonably complain if someone should decide to punch him in those lips — and that’s pretty important, to be sure. Still, rationalism of this general type seems a fairly weak reed. To make it stronger, we need to show that we all have real reasons to accept the sort of norms we are arguing for, and not, by some feat of logical prestidigitation, that we supposedly already do acknowledge them. [Kinsella] Perhaps this is what these rationalists are really getting at anyway.

3.5 The Value of Liberty

Why would people value liberty? Can we even presume that they do? Or that they should? And if it is a value, what sort is it? These are misleading questions. Liberty is not another good, like peanut butter or Rachmaninoff: it is, rather, a necessary condition of action, in the sense that if we do x, it has to have been the case, by definition, that nothing prevented us from doing x. Liberty is the condition of being able to do whatever it is, the liberty to do which is in question. And that means that it is not really up in the air whether liberty is a good thing or not. Liberty is as good as whatever can be achieved by acting: if x is good, then the liberty to achieve it is good. So insofar as the opening questions make sense, there is an obvious and ready answer to them. Suppose we want to do something, x; we see x to be valuable, either in itself or because it will produce something else we value. In that case, anyone who prevents us doing x, or for that matter just makes doing x more difficult than it would otherwise be, does what we have reason not to want to have happen, so far as it goes. That reason is that we wanted x and now either can’t get it, or can get it only at greater cost. But cost is by definition a disvalue, so its absence is a value. Thus we have reason to want to be at liberty.

To be sure, we weigh and rank our ends. Some matter a lot to us, some less. The value of liberty is also going to be greatly affected by this. We will forego even the sheer liberty to do some things we might like to, in the interests of being able, or better able, to do what matters more. Prima facie, “the value of liberty” is not a single quantity: abstract declarations on the subject are, we must admit, likely to be of modest interest.

Are they of any interest at all? Perhaps. This would be so, especially, if we could find some generic reason for ranking liberty above other ends from the social point of view. That is to say, even if I value my liberty to do x much more greatly than my liberty to do y, and especially if I value having x so much that I am willing to forego even my liberty to do y if that is the price of having x, still I may object to other people’s making that decision for me. Retaining the choice of which particular liberties I will retain and which I will forego looks to be a plausible generic value. But this is very different from valuing classical music, or archery, or Van Goghs. Liberty isn’t an intrinsic value, and indeed is only misleadingly called a “value”; rather, it is a condition on action.

A General Reflection about Liberty

This brings up a general reflection about what liberty is. Our ends are what matter to us. Precisely because they do, though, we are concerned to think about means toward those ends. The means I may employ are, in the first instance, unilaterally controllable actions of mine. If I can enlist or command someone else’s actions, those actions are means of mine only insofar as something I can do (such as issuing a command) is under my own control. In that sense, I don’t simply have the choice to achieve a given end with my own action or someone else’s: mine are all I have at my disposal, fundamentally. Whether and to what extent I will act so as to constrain my own options is a general and important question for me. But whether and to what extent I will constrain others’ actions is not a question of the same kind. My control of others’ actions depends on their choices, not just mine: so what I must do is choose among options aimed at influencing their behavior in such a way as to bring about the desired constraints.

Liberty has to do with the relation between a person’s ends and the means by which they are to be attained: it obtains when no outside interference exists to prevent the person’s own actions from carrying out his ends. The point of the “general reflection” is this: The relation between a given person’s ends and that same person’s actions directed at achieving those ends is not really a subject for deliberation. It is the way things are. As Kant puts it, “he who wills the end wills the means necessary to achieve it.” But the subjects of possibly preventing someone else from acting to achieve his ends, or alternatively of trying to assist him in achieving them, quite definitely are subjects for deliberation. But for the reason just given, the baselines of all these deliberations is going to be pursuit of A’s interests by A, if anybody. Whether I am going to try to do what I want is not in general a question for me; whether I am going to do what you want, in general, is. And of course, whether you will do what I want is a question for you, and a matter for calculation, concern, persuasion, and the like, for me.

3.6 Self-Interest

This is perhaps the place to bring up the subject of self-interest. It seems widely to be felt that the “libertarian” view on political and social matters is in some fundamental way bound up with egoism, the idea that we are all just “out for ourselves,” understanding that to mean that we are trying to promote our own good, our own interests, in a sense that excludes the good of others from consideration, or brings it in in a wholly subordinate way. The explanations above, however, should make it clear that that is incorrect. A person’s interest in liberty stems from the fact that he is interested, and not from the particular object of that interest. It is clear that people are often very interested in others. When they are, efforts to prevent their efforts on behalf of those interests will have the same unwanted status as would actions to prevent self-interested action. Interference with A’s ends is what matters to A; whether A figures prominently in A’s list of ends is another question. Libertarianism proposes a right to do whatever you wish to do, consistent with the similar liberty of others. Highly self-interested persons have the right to be so, but highly other-interested persons have the right to promote the interests of those others, so long as they promoted them with the others’ consents and not violate the rights of still others in the process.
Thus understood, however, it is promising to suppose that it is our interests that provide the foundation of libertarianism. But how do they do that? That is where our next theory comes in – contractarianism.

3.7 Contractarianism

The classic philosophers most reasonably regarded as the fathers of libertarianism, Hobbes (libertarian in his moral, not his political, philosophy) and Locke, were both contractarians, though of different kinds, and there is a natural affinity between that outlook and the Libertarian view. According to the most thoroughgoing kind of contractarianism, the principles of morals and politics represent a sort of rational agreement among everyone regarding the sort of general behavior that everyone can be expected to engage in or refrain from. Obviously this cannot be a negotiated agreement, as if by a sort of constitutional convention. It must instead be based on a sort of thought-experiment: what is it reasonable for me to demand of others, on condition that I live up to certain demands from them? There is no presupposition of a moral sense, built-in moral principles, or even moral concepts: contractarians of this type propose to derive the basic principles of morals and politics from a non-moral starting point. Each person consults his interests, whatever they may be, and his capabilities, whatever they may be; and then considers how we might expect others, with specifically differing interests, to view various proposals along this line. The principles of morals will be the one set of principles, if any, such that everyone sees it to be maximally advantageous to accept it, provided all others do too, as constraining pursuit of our various values. (By contrast, Locke assumes the libertarian principle, asserted in his “Law of Nature,” and proceeds to justify government, rather than the whole of morals, by appeal to the consent of the governed.)

From the very first mention of this idea, by the character Glaucon in Plato’s Republic, through Epicurus [Epicurus, p. 148] Hobbes, Locke, and Kant, the principal concern has always been taken to be the use of interpersonal violence or force, and fraud (lying, deception, chicanery), and the proposed principle has been a general prohibition on the use of those methods of achieving one’s goals. Why would we expect this result? The basic reason is that whatever we want, we thereby want that we not be, insofar forth, prevented from trying to satisfy those wants. More precisely and more generally, when we not only want something but, if that is different, value it, and value it enough to try to do something about it, then interference with those efforts is unwelcome. However, that is not enough to get us the liberty principle. To do that, we have to add another premise, one that was provided by Thomas Hobbes: that people are, by and large, fairly equal in their capacities to make life miserable for each other. In particular, as Hobbes pointed out, there is the fact that almost anyone has enough strength of body and enough rationality to kill anyone else, if that person takes a mind to do so. Given that all are potential enemies, and potentially lethal ones, we do well to propose general peace. The terms are: you don’t hit me, I don’t hit you; you leave me alone, I leave you alone. Specific interactions for benefit are to be agreed on by the parties concerned, which guarantees that they will be only for mutual benefit.

Why stop there? Because as soon as we start agreeing to help people, we are taking on costs, which cut into our ability to do the things we want. If those costs are justified by proffered benefits, well and good; but the case for supposing that a general imposition of positive duties on all will result in general gain for all is weak. In principle, if, say, 10% of us, or 75%, or any number short of 100%, would find it advantageous to undertake certain positive obligations to each other, they can do that without imposing these on the remainder as well. All that is needed is a principle of allowing and respecting local, specific agreements, as Hobbes’s Third Law of Nature [Hobbbes, ch. XV], which bids us to keep our agreements. It seems very plausible that we simply don’t need anything more. The contractarian argument for liberty says: people should help each other voluntarily. Any other arrangement makes no sense to the conscripted, who could easily expect to end up “paying” more for such an onerous morality than it’s worth to him.

The logic of this idea has been extensively explored in recent times by David Gauthier, who does not account himself a libertarian, however, and by Jan Narveson [Narveson, 1988], who does. The theory is not to be confused with that of John Rawls, who thinks that although one of two principles of justice is a liberty principle, and not a comprehensive one at that, the other — the “maximin” or “difference” principle, which he thinks calls upon us to redistribute considerably to the worst-off in society — is not, and undercuts the first one in to an extent difficult to determine. [Rawls, 150-161;but see also Narveson 1976] True contractarianism starts with people as they are, and asks them to consider whether their situations cannot be improved by adopting certain general constraints on their conduct. Its answer is resoundingly in the affirmative: in the absence of moral constraint, we can expect no end of evils at the hands of our fellows, and life will be worse for all. The way to make it better, argues the libertarian, is to renounce force as a way of dealing with our fellow humans, except only as required to deal with force initiated by others against the innocent.

4. Applying Libertarianism

Most philosophical theories of morals, and even of politics, don’t readily generate implications for the Real World. Libertarianism, however, has a great many adherents outside the academic world — far more than inside, actually — and they are often seriously involved in political issues. Libertarianism certainly appears to hold out the promise of practical application — though there is at least one serious reason for doubting that, which we’ll discuss in the next section. Meanwhile, the view is thought to have serious and fairly direct application in at least two political directions, which we may call, respectively, Against the Left, and Against the Right. Indeed, libertarians complain, reasonably, that there is no good political category along the “right-left” dimension for their view.

Against the Left, libertarians famously object not only to socialism, but also to the welfare state, and more generally to a great deal of the kind of political initiative that is familiar to all who live in contemporary “first world” states. Restrictions on trade, employment conditions, or on gun-ownership, tax-funded education, regulation of work-related activities, affirmative action and anti-discrimination laws, and much more, are targets for libertarian criticism. It is generally taken as given that a libertarian state would be a “minimal” state, providing basic police protection and little or nothing more.

Against the Right, meanwhile, libertarians object to restrictions on life-styles, such as on gay and lesbian relationships, on the use of recreational drugs, on religions and irreligion, on speech and expression and assembly. In short, libertarians object to restrictive legislation of any and all kinds other than on interpersonal violence and fraud. Rightist regimes generally extol the power and glory of the State; libertarians will have nothing to do with such motives, or of course with imperialism, wars for anything but self-defense, and the like. All of these objections to both left and right stem from the same source: opposition to imposing by force on individuals except, alone, for punishment of interpersonal imposition of harm.

4.1 The Past-Ownership Problem

Advocates of redistribution tend to do so either on the basis of a right to equality, say of control over natural resources, or of a right to have one’s basic needs met. Libertarians object that both claims collide with a superior right of general liberty. In so doing, however, they soon encounter a problem. According to the libertarian view of property, people legitimately own property if they have acquired it freely: they didn’t steal it, or defraud anyone in the course of acquiring it. But what if the people you buy it from stole it, or the people they got it from? Much durable property has probably, at some time in the past, been involved in an involuntary or partly-involuntary transaction. Does that infect the items in question from then on? At the political level, as the contemporary world is only too acutely aware, there are frequent claims that the people in area A have a claim to area B because many centuries ago, their ancestors were driven from B by force of arms — which is wrong according to libertarians. Do they accordingly have a rightful claim now? (Native Americans’ claims to many parts of the U.S. and Canada come quickly to mind in this connection.) Robert Nozick famously speculated that perhaps those at the bottom of society are there in part because they or their ancestors were victims of the most serious injustices, and in consequence, as a way of rectifying those injustices, some redistribution to the least well-off groups might be called for. [Nozick, p. 231]

Two responses to this problem lend perspective, however. The first is a straightforward disavowal. If A buys x from B, A having no prior reason to suspect B, and it turns out that x was stolen, then the responsibility lies with B — not with the innocent A — to fix matters up. If my great-great-great-grandparents were shysters or bandits, that’s bad news for my family tree, but it may well be too late to complain by now. Moreover, so far as ownership of useful material objects is concerned, hardly any of those even existed five centuries ago. The case for a hefty statute of limitations on these things is very strong. The second point is made by the anarchist economist Murray Rothbard, who points out that the persons to whom property would be transferred on the basis of claims from the past must not only be able to show that theft or illicit force produced the original transfer, but also that they, or some other definite person, has in fact superior claims to it now virtue of a clear and known connection to the earlier actors, before any retransferring of the supposed offender’s property is in order. And the farther back into the past we go, the more difficult it is to establish those superior claims. [Rothbard, p. 99] Few claims will survive this requirement, and so the effect of this consideration in real practice can be expected to be much more modest than some writers insist, and perhaps even, on the whole, negligible. Note that this is independent of specifically libertarian political philosophy: anyone who attaches any significant weight to ownership will be faced with this problem, and will need a solution along those general lines.

One final point about this important subject: the value of almost all property is extremely responsive to the actions of agents. Durable artefacts, and bits of unimproved real estate, may figure in arguments of the foregoing kind, but normally what a person owns now will have little relation to what a remote ancestor may have “handed down.” That is especially why the libertarian need not accept that receivers of goods stolen a hundred or a thousand years ago must simply surrender their properties to rightful claimants without further ado. If what A gives back to B is worth a hundred times what B was entitled to, simple restitution is not a fair option. (The example of trying to do justice to contemporary Native Americans serves well as an illustration of these difficulties.)

4.2 Is Libertarianism Compatible with Government?

If no one has any positive duties to others that have not been specifically incurred by promises, contracts, or similar undertakings by the agent, then there is an evident problem about government. All government proceeds by taxation, which by definition exacts things (money) from people without getting the specific consent of the person from whom it is exacted. If that is illegitimate, then government is illegitimate. That conclusion is explicitly embraced by many libertarians today. Their view tends to be known as “Market Anarchism” or “Anarcho-Capitalism” — anarchism with strong property rights, in contrast to “communist anarchist” views, popular in the late 19th century. Proponents of market anarchism claim that their idea can actually work in the real world, whereas the communist sort is impossible. [Narveson, 1996, pp. 196-7]

Market anarchism is the stateless society in which everything is done by voluntary associations. Since only some of those would be of the familiar profit-seeking type, that is, businesses as ordinarily so called, the term is misleading: other equally voluntary associations would include cooperatives, nonprofit charitable organizations, innumerable nonprofit public-service groups such as many theater troupes and churches, social clubs and the like. In many cases, no doubt, there would be competition between different types of association for the consumers of the services in question, even as at present cooperatives compete with single-owner or corporate enterprises in many areas. (For much more on anarchism, see [Caplan])

An Example: Conscription

The public-goods argument most often used on behalf of the need for government and against the market-anarchist version of libertarianism concerns national defense. May we not require people to assist in the defense of their own community? This much-discussed example is a good one for thinking about the commitments of libertarianism. In being argued that if person A refuses to help, and those who volunteer successfully defend the country, then A may owe his liberty, and perhaps his life, to those who have taken up arms. Has he not in effect inflicted a loss on his fellows, then?

4.3 Minarchy

“Minarchy” is the name coined for the view that although the state is justified, it is so only for the minimum case of it — in general, we should have as little of it as possible. Which functions would the minimal state perform? The usual answers call for police services, judicial services, and national defense. [Hospers, 1971] All three have been argued to be capable of being provided without coercion by writers of anarchist persuasion. The minarchist needs to show, first, that the state really is necessary for some such purposes, and, second, that if his argument shows those things to be legitimate, that an expansion of the state in the direction of today’s very extensive one can and should nevertheless be avoided.

4.4 Democratic Liberties

Most people nowadays tend to identify political liberty with democracy. As such, an individual’s liberties include the right to vote, to run for office, to engage in political discourse and speech, and perhaps to certain securities such as general bodily security, jury trials, and the right of habeas corpus. However, democratic rights can by no means be identified with libertarian rights, for democracy in principle permits majorities to make laws that might cut substantially into people’s liberties without justifying this via the non-harm principle. For example, if the majority can impose a tax on all to support a public school system, the individual taxpayer has no choice whether to pay the tax and thus whether to support the school; and the level of the tax might be such as to make it difficult or impossible for him to give his child the schooling of his choice. Examples may be multiplied indefinitely. [Boaz, ch. 9]

4.5 Protection

This leaves us with, at most, the protective state, which enforces the fundamental rights of its citizens — and only that. Or does it? In fact, there remains the problem identified in the previous paragraph. Government undertakes to protect people, but it does so by disallowing competitors who might offer the same — or better — protective service. But the libertarian’s principle doesn’t admit of this. No one may impose a restriction on the offering of services to anyone other than by explicit contract with the persons accepting the restriction. This point makes it difficult to see how any State could be justified in strictly libertarian terms.

Recent work by libertarians has shown how what have been regarded as services that can only be provided by a monopolizing state are not in fact so. People could sign up with protective services who would compete for customers or members, as the case would be. Much protective service is already offered by private firms in America and elsewhere as of the present (1998). [Benson, 1990, 1998] In fact, it is said that private protection far outstrips public protection, especially of property — warehouses, schools, apartment buildings, and so on.

There is serious question whether any protective services absolutely must be provided by the State. When crimes are committed, the victims or their representatives (as in the case of murder) have a motivation, and given the libertarian principle the right, to restitution from their aggressors, and this restitution may indeed be compelled. That compulsion need not, in principle, be provided by governments either. It could be provided by agencies selling their services, or by members of cooperatives acting on behalf of fellow members. Recent work on all these matters is impressive, and should be consulted by anyone entertaining serious doubts about the possibility of anarchism, at least in principle.

4.6 Law

There is an interesting question whether it is possible for law, as we know it, to be voluntary. But some theorists believe it can, and that it would be possible for differing communities within a single geographic area to have different systems of law that applied only to the members of those communities, who would be “subscribers” to the legal system in question. Work on this interesting question, which cannot be said to have been settled, is going forward today. [Friedman]

4.7 Punishment and Restitution

Libertarians tend toward the view that the practice of criminal punishment is wrong or at least problematic. In their view, crimes are essentially violations of individuals’ rights, and the proper way to respond to them is to force the criminal to rectify his misdeeds by making appropriate restitution. How much, and how is that to be determine? Those are major questions that have been extensively considered by many libertarian-leaning theorists. [Barnett; Ellin]

4.8 The Future

Many libertarians are of the view that developing technology will have much to do with the future of free institutions. They point to the internet, for example, and to innumerable ways in which individuals can be efficiently connected by voluntary means that make state intervention unnecessary or even impossible. Perhaps, they argue, the state is “obsolete” or will soon be so. [Boaz, ch. 11]


Libertarianism offers a single principle that is both remarkably elegant and apparently substantive. Its historical roots in philosophy are extensive and impressive (see Historical Appendix). It also has a strong following among many Americans, less so in other parts of the English-speaking world, and but a tiny smattering of people elsewhere. Why, one might ask, are so many people not accepting it? For this we need to turn to analyses of the workings of the modern state. The modern analysts of these matters who are sympathetic to the libertarian idea point their fingers at many things, but chiefly the fact that the modern state claims to offer benefits to its subjects, and offers ample opportunity for those subjects to enable the state to undertake to provide them — at what they suppose will be other people’s expense. [de Jasay] This brings up many issues, on both the empirical and conceptual fronts: fundamentally, of course, whether the state is indeed morally obligated to promote the good lives of its citizens on their own terms, as most of us think nowadays; and assuming that it is, then, especially, whether there are requirements, such as equality, that are both legitimate and call for redistribution from some individuals to others; and whether the state is capable of being more efficient than voluntary associations such as businesses in providing for whatever it is legitimate to provide for citizens. These major questions are the subject of enormous amounts of work in the literature not only of philosophy but also of political science and economics, especially. Very few people nowadays deny some claim to individual liberty; the question is whether there is anything else that may be legitimately pursued by coercive means. It will be long before the dust settles on that one.

5. Historical Note on Sources of the Liberty Principle

Hobbes’ “first, and Fundamental Law of Nature” is “to seek Peace, and follow it… The Second [branch], the sum of the Right of Nature; which is, By all means we can, to defend our selves.” Hobbes adds, significantly for our purposes, that “From this Fundamental Law of Nature … is derived this second Law; That a man be willing, when others are so too, as far-forth, as for Peace, and defense of himself he shall think it necessary, to lay down this right to all things; and be contented with so much liberty against other men, as he would allow other men against himself.” [Leviathan, Ch. XIV] Hobbes’ moral theory, then, is libertarian. (But in his political philosophy, Hobbes turns over all power to the state, which has a free hand — the very antithesis of what modern libertarianism holds.)

Locke’s own, more articulated version of the Law of Nature “teaches all Mankind .. that being all equal and independent, no one ought to harm another in his Life, Health, Liberty or Possessions.” [sect. 6] Significantly, Locke later (sect. 87] refers to, “his Property, that is, his Life, Liberty and Estate”; again [sect. 123] “their Lives, Liberties and Estates, which I call by the general Name, Property.”

A bit later still, Immanuel Kant proclaims as the “Universal Principle of Right” that “Every action is just that in itself or in its maxim is such that the freedom of the will of each can coexist together with the freedom of everyone in accordance with a universal law.

If, therefore, my action or my condition in general can coexist with the freedom of everyone in accordance with a universal law, then anyone who hinders me in performing the action or in maintaining the conditon does me an injustice, inasmuch as this hindrance .. cannot coexist with freedom in acordance with universal laws.” [Kant, 35]

And, of course, John Stuart Mill tells us, in his Principle of Liberty, that

“the sole end for which mankind are warranted, individually or collectively, in interfering with the liberty of action of any of their number is self-protection. That the only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others. His own good, either physical or moral, is not a sufficient warrant.” (An Essay on Liberty, Introductory, § 9)

The latest, David Gauthier’s, is also perhaps the best. Gauthier calls it “The Lockean Proviso”; it forbids attempting to better one’s own situation by worsening the situation of others. (Morals by Agreement, p. 209)


Barnett, Randy, The Structure of Liberty , New York: Oxford University Press, 1998.

Benson, Bruce, 1990, The Enterprise of Law , San Francisco: Pacific Research Insgtitute, 1990
——, 1998, To Serve and Protect ,New York: New York University Press, 1998.

Berlin, Sir Isaiah, “Two Concepts of Liberty” in Four Essays on Liberty , London, Oxford University Press, 1969: Oxford paperbacks, no. 116.

Boaz, David, Libertarianism — A Primer , New York: Free Press, 1997.

Caplan, Bryan, Anarchist Theory FAQ. He is one of the world’s most learned people regarding anarchism; his website is a goldmine for students curious about the subject.

Cohen, G. A, Self-Ownership, Freedom and Equality , Cambridge, UK: Cambridge University Press, 1995.

Ellin, Joseph, “Restitution Defended“, in Journal of Value Inquiry , Vol. 34, No. 2-3 (2000)
Epicurus, Principal Doctrines, XXXVI , Ai. I. Melden, ed., Ethical Theories (Prentice-Hall, 1967.

Friedman, David D, The Machinery of Freedom, 2nd ed.:Lasalle, Ill: Open Court, 1989. He is a very influential writer on libertarianism and related matters. See his home page.

Gauthier, David, Morals by Agreement , NY: Oxford UP 1986.

Gewirth, Alan, The Community of Rights , Chicago: University of Chicago Press, 1996.

Hayek, F.A., The Road to Serfdom , Chicago: University of Chicago Press, 1944. (an anti-socialist classic, though not explicitly libertarian.)

Hobbes, Thomas, Leviathan , 1651; many editions.

Hoppe, Hans-Hermann, A Theory of Socialism and Capitalism: Economics, Politics, and Ethics , Boston: Kluwer Academic Publishers, 1989.

Hospers, John, 1974, “What Libertarianism Is” ,in Machan (1974).
——, 1971, Libertarianism, Los Angeles: Nash, 1971.

Jasay, Anthony de, The State , Blackwell, 1985; reprinted, Liberty Fund, 1998.

Kant, Immanuel, Metaphysical Elements of Justice, Indianapolis: Bobbs-Merrill, Inc., Library of Liberal Arts, 1965.

Kinsella, Stephan, New Rationalist Directions in Libertarian Rights Theory (pdf). The Journal of Libertarian Studies, 12:2 (Fall 1996, 323-338).

Kropotkin, Peter, The Conquest of Bread, New York: Putnam’s, 1892; see the anthology by Sprading, below.

Kymlicka, Wil, Contemporary Political Philosophy, New York: Oxford University Press, 1990.

Lester, Jan, Escape from Leviathan, New York: St. Martin’s Press, 2000.

Locke, John, Second Treatise on Civil Government; many editions. In Ernest Barker, ed., Social Contract,
Oxford: Oxford University Press, 1960, pp. 3-147.

Machan, Tibor, Individuals and their rights , La Salle, Ill.: Open Court, c1989. [One of many books by one of the leading exponents of libertarianism today.]
——, (ed.), 1974, The Libertarian Alternative, Nelson-Hall, 1974.
——, (ed.), 1982, The Libertarian Reader, Totowa,NJ: Rowman and Littlefield, 1982.
Machan, Tibor, and Rasmussen, Douglas ed., Liberty for the 21st Century, Lanham, Md: Rowman & Littlefield, 1995.

Mill, John Stuart, An Essay on Liberty, Everyman Library, London: J. M. Dent & Sons, 1968).

Murray, Charles, What it Means to be a Libertarian, New York: Broadway Books, 1997.

Nagel, Thomas, “Libertarianism without Foundations“, in Paul, Jeffrey, Reading Nozick.

Narveson, Jan, “A Puzzle About Economic Justice in Rawls’ Theory“, Social Theory and Practice, Vol. 4, No. 1, Fall l976, pp. 1-28
——, 1988, The Libertarian Idea, Philadelphia:Temple University Press, 1988; reprint Broadview press, 2001.
——, 1997, “Alan Gewirth’s Foundationalism and the Well-Being State” — Journal of Value Inquiry, vol. 31.4, December 1997, pp. 485-502.
——, 1996, “The Anarchist’s Case“, in Sanders and Narveson
——, 1995, “Contracting for Liberty“, in Machan and Rasmussen.
——, 1999, “Property Rights: Original Acquisition and Lockean Provisos” Public Affairs Quarterly, vol. 13, No. 3, July 1999, pp. 205-227.
——, 2000, “Liberal-Conservative: The Real Controversy,” Journal of Value Inquiry, Vol. 34, No. 2-3 (2000).

Nozick, Robert, Anarchy, State and Utopia, New York: Basic Books, 1974.

Nozick, Robert, “Coercion,” in Philosophy, Science, and Method: Essays in Honor of Ernest Nagel. Sidney Morgenbesser, Patrick Suppes, and Morton White, editors (New York, NY: St. Martin’s Press, 1969).

Paul, Jeffrey, ed., Reading Nozick, Totawa, NJ: Rowman and Littlefield, 1981.

Pollock, Lansing, The Free Society, Boulder, Cpol., Westview Press, 1996

Rand, Ayn, The Fountainhead, Indianapolis, New York: Bobbs-Merrill, 196 ( c1943); Atlas Shrugged, New York : Random House, 1957).
Also, The Virtue of Selfishness (NY: Signet paper back, 1961), and Capitalism: The Unknown Ideal, NY: Signet, 1967. [Extremely influential novelist, widely regarded as the major “guru” of libertarianism; originator of “Objectivist” philosophy.]

Rasmussen, Douglas, “Essentialism, Values and Rights,” in Machan, The Libertarian Reader.
Rasmussen, Douglas and Den Uyl, Douglas J., Liberty and Nature: An Aristotelian Defense of Liberal Order, LaSalle, Ill: Open Court, 1991.

Rawls, John, A Theory of Justice, Cambridge, Mass.: Harvard University Press, 1971.

Rhodes, Michael, “The Nature of Coercion,” Journal of Value Inquiry, Vol. 34, No. 2 (2000).

Rothbard, Murray, Power & Market. Menlo Park, CA — Institute for Humane Studies, 1970, p. 76.

Sanders, John T. and Narveson, Jan, eds., For and Against the State, Rowman and Littlefield, 1996.

Shue, Henry, “The Bogus Distinction- “Negative” and “Positive” Rights“, in Norman E. Bowie, Making Ethical Decisions, NY: McGraw-Hill, 1985), pp. 223-231.

Sprading, Charles, Liberty and the Great Libertarians, San Francisco; Fox & Wilkes, 1995 — excerpts from a great variety of 19th Century libertarians, including Kropotkin.

Steiner, Hillel, 1983, “How Free: Computing Personal Liberty“, in A. Philips Griffiths, ed., Of Liberty, Royal Institute of Philosophy Lecture Series: 15, Cambridge: Cambridge University Press, 1983, pp. 73-90.
——, 1994, An Essay on Rights, Oxford, UK : Blackwell, 1994.
Steiner, Hillel, and Vallentyne, Peter, eds., Left-Libertarianism, Its History and its Future (2 vols.), London & New York: Macmillan Publishers Ltd., 2001.

Suits, David, “On Locke’s Argument for Government“, Journal of Libertarian Studies, Vol. 1, No. 3, 1977, pp. 195-203.

Tannehill, Morris and Linda, The Market for Liberty, NY: Laissez Faire Books, 1984.

Thomson, Judith Jarvis, “Some Ruminations on Rights“, in Rights, Restitution and Risk (Harvard University Press, 1986; also reprinted in Jeffrey Paul, Reading Nozick Totowa, N.J.: Rowman and Littlefield, 1981), pp. 130-147.

Viminitz, Paul, “A Proof that Libertarianism is either False or Banal,” Journal of Value Inquiry, Vol. 34, No. 2-3 (2000)

Wolff, Robert Paul, Understanding Rawls, Princeton, NJ: Princeton University Press, 1977.