The libertarian objection to redistribution policies is well-known: Insofar as there is no distinction between self-ownership and the ownership of things in which I have mixed myself in the form of my labor, any attempt to redistribute part of what belongs to me to meet the needs of third parties or to promote an equality policy amounts to forced labor or slavery [1]. This attitude is intuitively attractive because, unlike the Rawlsian approach, it takes into account the idea that individuals own themselves and that they have a primitive and exclusive right over their own person and their own abilities and talents. It therefore takes as its starting point a principle of fair acquisition, which allows us to say who owns what before entering into the cooperation process, and independently of it, and thus helps overcome a problem that seems to haunt a number of egalitarian theories of fairness today, in particular Rawlsian theory. In fact this does not seem to take into account the question of individual contributions to the process of social cooperation, and seems to consider only material and personal resources (talents and abilities) as components of the total inventory created by cooperation; so it focuses only on how we should distribute total resources so that the societal structure can be considered fair. [2] But this approach seems counterintuitive, as common sense sees social cooperation as a process in which individuals enter with the resources that they own—in particular their own person, including their personal qualities—and that it is unfair that social redistribution principles do not take account of these initial contributions.

But if, in the Rawlsian world, the structural viewpoint—the requirement for reciprocity in distributing the benefits of social cooperation—seems ready to overwhelm the personal viewpoint of libertarian theory, conversely it is the prescriptive viewpoint of the person which seems to exclude structural considerations. In other words, abstract respect for the individual’s rights trumps structural considerations; in particular it excludes fairness considerations that ask whether people are really placed in relation to each other in a way which permits cooperation by free and equal agents.

Recently, the school of thought known as “left-wing libertarianism” has looked for ways to combine the recognition of a person’s ownership rights over himself (and possibly over things, without which this right is devoid of substance) with the possibility of legitimizing a structural requirement for equality [3].

The thinkers who created this movement started from an initial intuition which is basically very simple: People are not equal, and it seems obvious that if the only purpose of the theory of social justice is to guarantee the principle of self-ownership, this principle simply transposes inequality between individuals into social inequality . This can only happen, however, if, as Nozick would want, every individual’s rights over himself are also extended to rights over the things that he acquires and is mixed in with, which together constitute his own person. If we assume that an individual’s rights over himself do not necessarily extend to rights over what he acquires by using his person, this fundamentally unequal outcome does not necessarily ensue, or not necessarily to the same extent. Of course, this means that the appropriation of things by individuals is subject to a structural condition of sharing but, after all, Nozick himself accepts that the legitimacy of the acquisition of external resources is subordinate to a structural respect for prior rights, and thus that the right over external things is essentially different from the individual’s unconditional right over his own person (and over his labor, assuming that he uses no external resource to perform his labor). Certainly Nozick interprets this conditional clause so weakly that any exclusive appropriation of a share of external resources inevitably satisfies it. Indeed, he maintains that the appropriation is legitimate if it does not result in putting third parties in a worse position than they would have found themselves in, had there never been any private appropriation [4]. Thus it is enough to establish that all members of society are better off in an economy based on private property than in a state of nature in which there is no private property, to prove that private property is legitimate and that it can have no limits. For Nozick, any egalitarian interpretation of the structural condition of the legitimacy of private appropriation (i.e., any condition which would demand that the act of exclusive appropriation not give the person who does so, an advantage in relation to those who are not able to do so to the same extent and quality) would subject the appropriation of external resources to the unanimous approval of all members of society and would destroy the reality of the principle of self-ownership, which can only be effective if there is a right to appropriate the means to give it substance which is, thus, not subordinate to the unanimous approval of third parties.

This, then, is the challenge that “left-wing” libertarians try to resolve by asserting that it is possible to give egalitarian substance to the structural legitimacy of exclusive appropriation, without voiding the principle of self-ownership. They thus try to capitalize on the benefits of the libertarian position without accepting its unequal consequences, which, in terms of justice theory, inevitably follow the idea that the rights of the individual come first before any structural precept governing the nature of relationships between individuals. For this reason they propose a form of synthesis between the individual principle and the collective principle, between an ethics based on rights and a form of structural theory that takes this structure of individual relationships into account and demands that it conform to a principle of equality in the sharing of natural resources.

This school of thought thus borrows from libertarianism the idea that each person owns himself, and allocates limits to what others can do to a person without that person’s consent. In particular, it assigns inviolable limitations on what an equality policy can do, and it emphasizes that the ethical requirement to respect self-ownership is a source of independent moral imperatives for the structural requirement for equality. Conversely, authors who claim to adhere to this view want to show that the right that we each have over our own person cannot extend to the things that we appropriate, since any appropriation of external resources transforms the conditions under which individuals can exert the right that they have over their own person. In other words, the right over oneself does not confer a right over external things and, unlike the assertion of classic libertarianism, the two rights have different foundations: Each person has an exclusive and unconditional right over himself, but the right over things is conditional and subordinate to a structural requirement for equality. The central idea of left-wing libertarianism is therefore that the two considerations are compatible, that the unconditional right over oneself is not destroyed by the conditional nature of the right over external things, and that, conversely, the egalitarian right over external resources is not annulled by the unconditional and exclusive right over oneself.

This theory therefore intends to put forward the idea that, as external resources are common, no-one can appropriate these except on condition of respecting a structural imperative which confirms that the rights of all humans are neither contravened nor annulled by the act of appropriation. We will see that the content of this structural imperative can vary, but that the idea remains the same: Although each person can freely use his person without respecting any condition linked to the rights of others, this does not extend to external resources, and thus we have a theory of justice which claims to combine a non-structural principle in the first person and a structural principle of the appropriation of material objects.

In conclusion, we will suggest that this attempt at synthesis fails and that it either moves towards an assertion of self-ownership, extending unconditionally to things (this is the theory of classic liberalism), or to an assertion of the conditional nature of the appropriation of things, which ends up voiding the principle of self-ownership in any practical sense (returning us closer to the Rawlsian theory). It is therefore essentially unstable [5].

1 – The principle of self-ownership

Let us start by trying to define the principle of self-ownership. This can be summarized in the following two characteristics:

A full right of control over the use of my own person (which includes the full right to transfer my right over myself to others, most left-wing libertarians acknowledging the shameful consequence that it legitimizes voluntary slavery).

My right over myself is unconditional, and in particular immune against any attachment or taxation: I have the right to freely use my own person without having to pay anything to anyone.

It should be emphasized that, expressed in this way, the self-ownership principle does not confer a right to external resources, or to use or appropriate them; nor does it guarantee that I can freely dispose of the results of my work since, in most cases, including intellectual work, the expression of this work requires the use of external resources [6]. Therefore, on its own, the principle of self-ownership does not provide any basis for the actual expression of freedom, as it does not involve a guarantee that I can have the use of my person; it only involves a negative guarantee that others cannot have use of my person without my permission.

One consequence of this principle of self-ownership—apart from the possibility of voluntary slavery —seems to be that I am never obligated to put my person at the disposal of others, should they need it. We must highlight this idea of a possible obligation to place my material resources at the disposal of others, as it is possible (and we see left-wing libertarians adopting a variant of this position) that I only possess these material resources subject to the implicit condition that I place them at the disposal of others when they are in extreme need of them (and I do not vitally need them to give effective substance to my own self-ownership right). These material resources are therefore owed to those who need them, because their need merely reflects their right to a share of external resources equal to that which I have myself appropriated. But, by definition, this does not apply to my person, which I possess unconditionally. This stance may seem outrageous, but although there may be many people whom I could help by placing my person at their disposal, the conclusion here is that there can be no obligation to act in this way [7].

2 – Equal distribution of external resources

As regards external resources, the assumption is, on the other hand, that these are common and that each individual has the right to an equal share. [8] This common arrangement can take two very different forms: One solution is to say that things are necessarily held in common, so that I can never use common stock without the unanimous consent of others. We immediately see that this solution is not very realistic since it makes the principle of self-ownership meaningless: having accepted that it is impossible to act without using external resources, the fact that any use of external resources requires the agreement of the community places my person at its discretion and destroys the personal element that we had thought had been protected by setting out the principle of self-ownership. It is therefore necessarily the second solution that should prevail—if we do not want the egalitarian component of the theory to overwhelm the autonomous normativity of an individual’s right over himself—i.e., that each person has the right to use external resources provided that he does not stand in the way of others’ rights to use them to the same extent. In a way this is a principle of egalitarian sharing. A good example would be a bench or a seat in a public garden: I have the right to sit there without asking anyone’s permission, but I cannot prevent others from sitting there when I am not [9].

Obviously, this usage solution is imperfect, since the right of others to sit on the bench is void if the bench is already occupied. Therefore, it is necessary, in some way or other, for the occupant of the bench to pay compensation to those who are not able to sit there in the form of a fee which reflects the ownership right that they possess over this bench, but that they cannot exercise because it is occupied. We therefore agree with the idea that each person can exclusively appropriate part of the common external resources on condition that they respect a structural rule which recognizes that each person has an equal right to carry out such an appropriation; this structural rule prescribes the payment of financial compensation to a common fund, determined by the market value of what the act of exclusive appropriation removes from the community. To the extent that the problem of future generations is always present, this fee can only take the form of income which transforms the ownership into rental, into a right to use in exchange for compensation [10]. This income is paid into a fund, and it is this fund—publicly managed—which should be used to give every individual who arrives on the scene when the world is already fully occupied, a sum of initial benefits, the value of which is equal to the equal share of natural resources to which he was entitled. The legitimacy of exclusive appropriation is therefore subordinate to the obligation to pay financial compensation, which is determined by the market value of what is removed from the community; this obligation fulfils the requirement that each person has a right to an equal share of external resources, to the extent that what I take should not nor cannot jeopardize the ability of others to take a share of the same value. Contrary to Nozick’s position, it is not enough that third parties are not made worse off by my action, and if I improve my position by becoming a private owner I cannot justify my action by claiming that I have not made anyone worse off; I also have to show that they had/have the same opportunity before and after to improve their own situation by using a share of the common resources which has the same value as that which I myself have removed from the community. The common fund, maintained by the fees paid by owners, is used for this.

As the privatization of an (equal) share of common resources is justified, it determines what we can call equality at the starting line: Each person, including members of future generations, really does have the right to an equal share of external resources (but not more than this, which certainly limits the right to pass it on and to donate) [11]. Using this common rights theory we have a range of possibilities. Strict egalitarians maintain that what I produce with privatized resources is 100% taxable, as my qualities and talents, including the results or “product” of applying my natural abilities to external resources, represent a social resource [12]. As it goes without saying that taxation at 100% (followed by an egalitarian redistribution via a common fund) would have the effect of discouraging productive effort, the tax rate is lowered for pragmatic reasons. But strict egalitarians maintain that the product is fully taxable because third parties can rightly claim that it would not be produced at all without using common resources. They also maintain that this position is compatible with the principle of self-ownership insofar as this ownership over self does not give any right over external resources without the agreement of third parties. But again, this position is not very realistic and is a good illustration of the instability of the system: If the whole of what I produce by applying my personal qualities to external resources is taxed, the principle of self-ownership is as empty as where the rule is the common ownership of external things, with the consequence that I cannot so much as use them without the agreement of others.

Even if we abandon the idea that it would be legitimate (but not sensible) to tax the revenue from the use of 100%-appropriated resources, the most egalitarian version of the left-wing libertarian theory finds it difficult to accept that equality of initial shares immediately translates into obvious inequality as a result of the unequal distribution of personal qualities and talents. It therefore maintains that, at the starting line, equal distribution of natural resources should mean real equality, and not merely a nominal equality. For that, however, we have to consider it not as equality of resources, but as equal possibility of accessing well-being or benefits. As, of course, individuals have unequal abilities to convert resources into well-being or benefits, we need to take into account differences in talents and skills and stop such differences translating into initial inequalities, and it is therefore legitimate to give more resources to those who have a lower ability to convert them into well-being or into real benefits [13].

On the other hand, there is no obstacle in principle to the idea that, having moved on from the starting point, individuals will bear the consequences of life’s vicissitudes alone and that, under the impact of differences in circumstances, considerable inequalities can develop from a starting point that was identical for all. In other words, if we accept that the requirements of egalitarianism can only be exercised under the constraint of respecting self-ownership, we also accept that equality is not the only moral imperative and that respecting individuality is also a source of legitimate demands that should be combined with the other demand, which derives from equality. From this we will conclude that there is no opposition in principle to the fact that individuals who initially benefited from equal conditions in terms of possibilities for access to well-being can consequently benefit from the advantages that they procure through the game of chance [14]. The imperative is not in fact to neutralize chance, but to equalize the initial set of possibilities for individuals to access benefits. As has been said, this can allow additional initial benefits to be given to those who have a lower ability to transform external resources into well-being (which in fact means considering people’s natural qualities as a resource which should be equalized), while at the same time opposing the idea of systematically compensating the effects of the arbitrary. The rule here will always be pragmatic: If compensating for the effects of pure chance results in reducing the value of equal initial access to well-being, we should not do this. What is contestable is not the fact of compensating the effects of pure chance, but the fact of giving this form of compensation an absolute value regardless of its consequences, because, on this hypothesis, the equality requirement would unconditionally dominate the self-ownership requirement [15]. As for the advantages of chance, each person can therefore claim to keep for himself a quantity as large as is compatible with everyone else also having equal opportunities for access to as much well-being as possible. But it is clear that if the consequence of not protecting individuals against the negative effects of chance were that society as a whole would have fewer resources (and that the equal share granted to each person would therefore be reduced), the refusal to compensate the disadvantages due to chance would become counterproductive.

The same kind of pragmatic consideration will be applied to the question of responsibility. The objective of this kind of left-wing libertarianism is not to ensure that individuals morally accept responsibility for their choice, but to maximize the value of the initial stock of possibilities for access to advantages that each person can at first enjoy equally. But if we claim that the effects of choices will never be compensated and that the effects of pure chance (brute luck) will only be compensated to the exclusion of any option luck, we are likely to completely discourage risky choices. But in fact such choices generate resources which help to compensate the effects of pure chance and to increase the value of the initial stock of opportunities to access benefits; the idea of never compensating the negative effects of some choices is therefore a bad idea, because never protecting individuals against the negative effects of their choices risks leading to a situation where there are fewer resources to protect these individuals against the effects of pure chance, and fewer resources in general to allow them to access well-being. It would therefore be unfair to claim that there should be no transfer that is motivated by individual choices; on the contrary, we see here that those who have made choices, the consequences of which are that they are more exposed to some risks, are entitled to benefit from some transfers from those who chose comfortable inaction and who refused to take the least risk. The choice not to take risks is taxed, and it is fair that the person who has not taken any risk should contribute towards funding the insurance which organizes transfers to some people who have taken risks and failed. The fact that there are people who take risks is in fact an advantage for everyone because, in this instance, if everyone chose not to take any risk, all would be penalized, compared with a society in which some agree to take risks and thus generate additional resources which increase the equal share of opportunities to access well-being that all can enjoy at the starting line. Therefore there is no reason to claim that as soon as individuals have been equalized, in a relevant way, each person has to agree to bear the consequences of his choices alone, because it is very possible that by diverging from this rule, we can better guarantee the equality of individuals in a pertinent way, because risky activities would have been encouraged and this would therefore have released more resources to raise the level at which individuals are guaranteed equality, from this pertinent viewpoint. However, the refusal to allow each individual to bear the consequences of these choices alone is only motivated by pragmatic considerations and not by a position of principle on the question of individual responsibility. The issue is simply that if each person has to bear the negative consequences of all his choices alone, some choices that are socially very advantageous will be discouraged.

3 – Is self-ownership compatible with equal sharing of external resources?

The assertion that society owes each person an equally beneficial share of external resources (i.e., a share of external resources which gives each person, with his mix of external and internal resources, an equal possibility of achieving the same level of benefits or well-being) seems to be the most attractive version of the left-wing libertarian position, insofar as it subjects the distribution of personal qualities to the egalitarian requirement. Is it compatible with an assertion of self-ownership that is sufficiently substantial for it not to be drained of its content?

First we should recall that in the initial assumption itself there is the idea that self-ownership does not confer rights over any share of external resources. Generally speaking, self-ownership is compatible with the absence of the possibility of using anything, even if, in this kind of case, the right of self-ownership becomes meaningless.

Then we observe that the right of self-ownership guarantees that if I have produced something solely by the use of my person (assuming that this is possible) I have the right to the product of my work; but it says nothing about the things that I have produced with the essential support of external resources and it does not automatically give me ownership except to establish that I had the right to those resources and that I paid others the necessary compensation for making them unavailable for use by anyone else.

Therefore the only question is not whether the egalitarian distribution of external resources is compatible with the right of self-ownership, (as this follows from the assumptions), but what form of egalitarian distribution is compatible with this right in practice. It is therefore a question of determining the point at which an egalitarian policy ceases to be legitimate because it challenges the actual reality of the right of self-ownership. It should be noted that it is not a question of redistribution since, a priori, individuals do not possess anything; contrary to appearances it is therefore not a question of taking from some to give to others, but of distributing equally what belongs to everyone (to which everyone has an equal right). The difficulty of course is to understand how this is possible over time, and when things have already been appropriated. The solution involves showing, as we have seen, that appropriations are only conditional and are subordinate to the right of each individual to have a suitable share of external resources, which involves each appropriator paying into a common fund a fee for his use of common things, and that it is from this common fund that the equal shares of those who are not direct appropriators and the members of subsequent generations are formed.

It should not be forgotten, however, that assigning to each person a quantity of resources that allows everyone equal possibilities to access advantages is an egalitarian objective which cannot be tempered or balanced by the self-ownership requirement. To better understand the consequences of this idea, let us imagine a desert island inhabited by two individuals, one of whom (Incapable) is severely disabled and has a very low ability to convert material resources into well-being, whereas the other (Capable) has on the contrary a very strong ability to carry out this kind of conversion. The theory of left-wing libertarianism demands that Incapable is assigned many more resources than Capable, but it recognizes a limit: If Capable cannot survive with the resources assigned to him in his own right and if he is forced to work for Incapable on the latter’s terms, his self-ownership remains in principle but it loses its reality. The limit of equal sharing (and not of redistribution) is therefore the possibility for each person to reproduce his existence in an independent way [16]. And yet, in a real society, especially in a rich society, the equalization process (the fees owed by appropriators to the common fund that is intended to provide equal shares for non-appropriators and members of new generations) never goes so far that the appropriators who pay the fee become slaves [17].

The idea here, then, is that we can allow a right of self-ownership to remain while at the same time cancelling any unequal consequences it has or may have. This right is not questioned as long as each person can avoid forced labor and has sufficient resources to continue to live by voluntary exchanges with others. It is therefore not true that if one allows the principle of self-ownership to remain, its inevitable consequence is an inequality and finally a dependency on each other; the reason for this is that these consequences can be avoided without affecting the right itself, which remains real as long as no-one is forced to work for others. Conversely, neither is it true that introducing a structural requirement inevitably results in the person himself being subjected to the collective viewpoint of equality.

So it appears possible, with a group of individuals who vary greatly in terms of talents and ability to convert resources into well-being, to annul the tendency of the principle of self-ownership to produce inequality coupled with dependence. This annulment could occur by distributing the ownership of external resources, which would be in line with the egalitarian premise. In a society in which resources are distributed according to this principle, those least able to convert resources into well-being would be very generously equipped with resources, and they would have the possibility to achieve a level of well-being equal to those who have fewer resources but more ability to convert them into well-being. But that does not mean that the most able would be obligated to assist the least able or forced to work for others, thereby calling the principle of self-ownership into question. We would be satisfied with organizing a resource distribution which compensates the lack of ability to convert these resources into well-being through an increase in resources. Therefore for the less able it is not a question of demanding that the more able should give them part of their work (which would indeed contradict the self-ownership principle). Quite the contrary, those who are less able to convert resources into well-being do not say that they have the right to be helped by others, but that they have the right to an equal amount of possibilities to access well-being, and that they therefore have the right to a share of resources (given their ability to convert these into well-being) which they need to be able to access an equal quantity of well-being compared with those whose ability, from this point of view, is greater. Those less talented (in converting resources into well-being) are not parasites; they simply claim the share of external resources that they are entitled to.

Nozick claims to legitimize considerable inequalities based on each person’s right to own himself and not to be forced to work for others (i.e., the separate and inviolable nature of the individual), but the left-wing libertarian theory also seems to satisfy this principle of self-ownership, and even in a much better way, since it guarantees everyone a right of self-ownership which is more than theoretical. In fact, with Nozick the principle of self-ownership is only partially satisfied since those who have nothing are owners of their own person, but they cannot exercise this right because they are forced to work for others. Nozick’s theory gives them no guarantee against this possibility. On the other hand, the version of the left-wing libertarian theory that we have mentioned includes such a guarantee for all: The best equipped will not be forced to work for others because the egalitarian pressure is contained by the personal imperative, which does not want anyone to ever be deprived of the resources needed to reproduce his existence in an independent way, but those who are less well equipped have the assurance that they will not be refused access to an equal share of external resources.

The objection is of course that this is all impractical because the world is not a stock of unowned resources. But this is irrelevant, because the members of each generation have an enduring right (and we do not see on what grounds they could be deprived of it) to own a quantity of external resources which allows them to achieve equal well-being. The egalitarian premise as it is understood here should therefore be extended to members of all generations: each person should retain sufficient resources to procure an equal quantity of well-being. Each generation should therefore make sure that, when it dies and the following generation arrives, the same quantity of unowned resources is accessible to the new generation. Each generation would therefore find the same quantity of unowned resources in the world that the previous generation had found there.

Left-wing libertarians note that to uphold such a solution it is necessary to prohibit significant gifts from one person to another within a given generation, when those gifts change the level of well-being which individuals are able to achieve, to such a degree that it offends equality, unless this disruption can be justified by the self-ownership requirement and the independence resulting from it. But we do not see why this ban on giving what we have produced by interacting with external things would violate our right of ownership over ourselves; it is at most a restriction on our use of external things (we have the right to use them during our lifetime but not to bequeath them to our descendants, or to pass them on for free to a third party).

So the conclusion is indeed that, whatever the practical problems, there is no contradiction in principle between asserting the validity of the principle of self-ownership and asserting a principle of equality in distributing external resources. But this equality does not result from redistribution, since the fees paid by the owners are not a payment for what they own by right, but a payment required to respect the condition of the legitimacy of their appropriation. Why, then, should it be surprising that those who have gained more than their share of external resources should pay a substantial tax to allow others to appropriate a suitable share of those same resources [18]?

4 — Cohen and the fundamental instability of left-wing libertarianism

The analyses proposed by Gerry Cohen in his book Self Ownership, Freedom and Equality show that any attempt to combine the principle of self-ownership with equality ownership over external things is doomed to fail [19]. The main objection to the left-wing libertarian theory is identical to that made against any “non-structural” theory which settles for defining fairness through a theory of initial acquisition, without agreeing to submit the product of individual interactions to a structural criterion of fairness; it defines a fair situation (each person owns his own self and each person has an equal quantity of possibilities to access benefits), and it supposes that the fairness of this situation is preserved for as long as the parties act fairly. But this premise is not sustainable because it contains a circular argument that Cohen explains clearly: We cannot define a fair situation as a situation which only contains fair actions, because the very definition of a fair action implies the concept of fairness that it serves to define. We therefore need a concept of fairness which allows us to define a fair society other than as a society which only contains fair actions; in other words, we need a concept of fairness which is not historical but structural.

The famous example of Wilt Chamberlain helps us to understand this [20]: Chamberlain is an exceptional basketball player and, every year, there are a million people who agree to voluntarily pay a quarter of a dollar to see him play. Whatever the initial distribution of resources that is defined at the beginning of the year and is considered to be fair, it has been significantly changed at the end of the year, because now Chamberlain has a quarter of a million dollars more than at the start of the year. We have therefore moved from distribution D1 to distribution D2, and the libertarians’ argument consists of saying that if D1 is fair and if no-one, when moving from D1 to D2, behaved in an unfair way, D2 cannot be unfair; neither basketball fans who each gave 25 cents, nor Chamberlain, behaved unfairly because they only did what they had the right to do; therefore D2 cannot be unfair.

Cohen shows that this conclusion is not valid, thus agreeing with the Rawlsian position, which states that the structural requirement is essential for the consideration of fairness. The main reason is the circularity of the argument that defines a fair action as an action which does not constrain anyone, and a non-constraining action as an action which does not prevent others from doing what they have the right to do. The idea of constraint is used in defining the idea of entitlement and, conversely, the idea of entitlement is used in defining the idea of constraint. If we apply this comment to the Chamberlain case, it produces the following result: It is not possible to say that an equalizing tax which takes a share of Chamberlain’s profits is an infringement of his freedom, without introducing the idea that Wilt Chamberlain and the spectators have a perfect right to act as they do. Indeed, without this idea of entitlement, it is the simple fact of preventing Chamberlain from acting as he intends which represents a restriction on freedom, but this assertion would mean, for example, that the fact of preventing the poor from moving onto the land or into the gardens of the rich is also a restriction on their freedom. And yet if we want to prevent this absurd consequence and highlight the difference between Chamberlain’s behavior and that of the squatters who move onto your lawn without your permission, we have to introduce the idea that Chamberlain has the right to act as he does and that the squatters, on the contrary, do not have the right to act like they do. Taxation cannot limit X’s freedom unless X has the right to act as he does. And yet, when Nozick begins to explain what is meant by the fact that Chamberlain has the right to act as he does (and that therefore we do not have the right to prevent him from doing so or to tax his profits more than is needed to maintain the minimal State), his only response is to say that he is not forcing anyone and that he is not harming anyone by acting as he does. The circularity of the argument is therefore obvious. We cannot define a fair situation simply as one that only contains fair actions, and fair actions as actions that do not constrain anyone, because the very idea of an action that does not constrain anyone must necessarily contain the idea of entitlement and, consequently, of fairness.

The construction proposed by left-wing libertarianism cannot avoid this criticism insofar as it remains a “historical” theory of the fairness of initial acquisitions. It defines a fair situation as one in which each person owns himself (criterion A) and each person owns an equal share of external resources (criterion B). It then supposes that if, in this kind of situation, only actions which are themselves fair occur, i.e., actions which do not directly harm self-ownership and equality in sharing external resources, the result is necessarily fair. But this is not true because the fairness of the initial situation can disappear without anyone behaving in a way that is deliberately unfair, i.e., without anyone behaving in a way that questions each person’s self-ownership, and without anyone trying to appropriate more than an equal share of external resources. And we cannot reply that this is impossible on the grounds that a fair situation is defined as one which only contains fair actions, because this is a circular argument. The definition of the fairness of a society cannot lie in the fact that it only historically contains actions which people had the right to carry out.

To understand the importance of this point let us imagine a form of left-wing libertarianism which would not allow the quantities of resources allocated to each person to be compensated by their ability to convert those resources into well-being and benefits; such a theory would be open to serious objection because differences in talent or ability will inevitably lead to inequalities which destroy the reality of self-ownership for some members of society by depriving them of the means to exercise this independently. Similarly, if we imagine a version of the theory which allows the quantities of resources to be compensated by the ability to convert them into benefits, it is still more open to objection here, because the effects of pure chance will again create inequalities incompatible with the reality of the principle of self-ownership for all. But we have seen that the left-wing libertarians who are committed to initial compensation for differences in talents reject the principle as being a form of permanent regulation and they impose a condition on the neutralization of chance, which is that this neutralization must maximize the initial equal stock of possibilities for access to benefits that each person receives at the start. But, obviously, maximizing the initial stock with which each person starts, protects no-one against the possibility of becoming dependent on others and being deprived of the means of independence. The left-wing libertarian option therefore allows a series of actions which comply both with the principle of self-ownership and the principle of an equal share of external resources (including the principle of compensation for initial inequality of talent), while allowing chance to distort the relationships of equality and non-dependence, resulting in non-respect of the principle of self-ownership. In this approach, nothing allows us to identify the result of interactions as unfair if the result is the product of a series of unfair actions. The dependency into which some members of society can fall (either due to inability or bad luck) cannot be considered unfair because it is the result of a series of actions that individuals have the right to carry out. But as the definition of fair actions is tainted by an obvious circularity, the conclusion cannot be maintained. Actions which unintendedly cause some members of society to fall into dependency, are defined as fair because they do not directly constrain anyone and they do not constrain anyone because they do not prevent any individual from doing what he has the right to do. But if an action is defined as fair because it does not prevent anyone from doing what he has the right to do, it is not possible to define a fair society as a society which only contains fair actions. Again we need an independent criterion of social justice and we should say that if the effect of a series of actions is that some members of society fall into dependency and can no longer give concrete meaning to the principle of self-ownership, it is unfair and the actions which lead to this are in turn potentially unfair and likely to be curbed or controlled [21].

Again, the most important question is not who has acquired what and how, but whether all individuals have the means to be free and to function as free and equal citizens. A structural consideration clearly helps to escape the circularity which defines legitimate actions by the lack of constraint and defines non-constraining actions by their legitimacy. The structural consideration claims that an action is fair if it does not have the effect of depriving any individual of the means to be free and to function as a citizen of equal value. If the principle of self-ownership and the principle of fairness in acquiring external things prohibit or do not guarantee access for all to autonomy (as is inevitably possible), we are faced with a dilemma: We must either accept unequal freedom, so an unequal valuation of individuals, or we must abandon the unconditional nature of self-ownership which, combined with an initial egalitarian distribution of external resources, does not allow us to guarantee everyone the reality of freedom by actually possessing the means of autonomy [22]. The left-wing libertarian theory either moves towards a classic libertarianism (if it accepts that the initial principles can legitimately result in a situation where the reality of self-ownership is no longer guaranteed for some), or towards an unconditional egalitarianism (if it accepts the idea that considering the reality of the principle of self-ownership for all is an unconditional value). Cohen himself opts for this second solution: Fairness cannot recognize unconditional privilege focused on the individual because in circumstances where respecting this privilege is an obstacle to the freedom of all, privilege must give way, and we come back to the idea that the individual’s right over his own person and his talents is subordinate to the legitimate (therefore mutually advantageous and reciprocal) nature of the structure of relations between people.

The solution proposed by Cohen is of course based on the idea that work is pointless and uneconomic unless it is combined with external resources or, at least, with productive elements—such as training—whereby individuals cannot say that they are the only authors. But these resources are common and they imply that the person who appropriates them, and which are the basis of his freedom and autonomy, is answerable to others who have not been able to appropriate those resources under the same conditions, which they need in order to exercise their own freedom and autonomy. It is not a question of whether and how the initial equal sharing of external resources is compatible with the exercise of freedom, but a matter of showing that those who achieve autonomy by using external resources are constantly operating under equivalent condition (in terms of value) as those who are not able to achieve it. This is merely another way of saying that freedom cannot be legitimately effective for some if it is not effective for all, or that freedom implies a form of equality.

The existence of this accountability and the contribution which consequently weighs on those who are the best equipped is by no means similar to forced labor, or to some people being placed at the disposal of others. It is true that these obligations do not have to be contractual to be legal, but they cannot be reduced to a form of slavery, not only because they are compatible with freedom, defined as having the legal and material conditions to exercise autonomy, but because they are indeed implied by freedom as, without them, some members of society would be deprived of these conditions and the equality of status which allows them to function as full members of this society.

It is precisely this position that libertarians of the right, like Nozick, contest, by arguing that we actually face an all-or-nothing logic: Either we own ourselves, our own person, and we do what we want without ever being obligated to give others the help that we have not contractually agreed to give them; or the reverse is true, that we can be legally obligated to give help to others that we have not contractually agreed to give them, in which case we cannot really be autonomous. But, Cohen says, it is wrong that, in a society, autonomy (which presumes access to material resources, something which is recognized by right-wing libertarians) is necessarily always maximized by the principle of unconditional self-ownership (without any redistribution). There are very good reasons to believe that in a world of chance, where individuals are equipped with very different talents, the principle of unconditional self-ownership will lead to situations where some will be deprived of access to the means of production and that, consequently, they will not have the conditions needed to exert the form of control over their own existence that we associate with the idea of autonomy, and this remains true even if we combine the self-ownership principle with the principle of equal division of external resources. From this we conclude that if we want all members of society to benefit equally from a certain degree of autonomy, the self-ownership principle must be limited, because implementing this unconditionally or conditionally may very well not have the effect of maximizing autonomy; in any case, it may in fact not succeed in maximizing the autonomy of those who have the least. It should also be noted that an individual’s autonomy does of course vary depending on the rights that he can exert over himself, but also on the rights that others have over themselves; thus if another person has full rights over his own talents, he may manage to reduce me to proletarian status, thereby reducing my autonomy. Therefore it is quite simply not true that autonomy is maximized when each person has an absolute right over himself. The theory may be paradoxical but it is sustainable and we are therefore entitled to put forward the following idea: Restricting the ownership right that each person has over himself could indeed have the effect of creating autonomy, and it is not self-ownership as such that creates autonomy but a certain restrictive use of this self-ownership. If we have to choose between the free exercise of self-ownership (which would harm autonomy) and the imposition of restrictions on this self-ownership (which would favor autonomy), we should choose the second solution.

[1] Robert Nozick, Anarchie, État, Utopie, trans. from English by Evelyne d’Auzac de Lamartine, Paris, PUF, 1988

[2] Peter Vallentyne, Hillel Steiner et Michael Otsuka, “Why Left-libertarianism Is Not Incoherent, Indeterminate, or Irrelevant: A Reply to Fried”, Philosophy and Public Affairs, 33, n° 2, 2005, 201–215.

[3] Peter Vallentyne, “Left-Libertarianism, a Primer”, in P. Vallentyne and H. Steiner (dirs.), Left-Libertarianism and its Critics, New York, Palgrave Macmillan, 2000, 1–20.

[4] Michael Otsuka, Libertarianism without Inequality, Oxford, Oxford University Press, 2003, p. 23; Peter Vallentyne, “Robert Nozick”, in John Shand (dir.), Central Works of Philosophy, vol. 5, The XXth century, Quine and after, London, Acumen, 2006.

[5] Mathias Risse, “Can there be ‘Libertarianism without inequality’? Some worries about the coherence of left libertarianism”, Kennedy School of Government, Harvard University, working paper RWP 03–044, November 2003, available on the Social Science Research Network (SSRN): ssrn.com.

[6] M. Otsuka, Libertarianism without Inequality, op. cit., 31.

[7] Peter Vallentyne, “Left-Libertarianism, a Primer”, op. cit.

[8] M. Otsuka, Libertarianism without Inequality, op. cit., 24; Hillel Steiner, “How Equality Matters”, Social philosophy and policy, n° 19, 2002, 342–356.

[9] P. Vallentyne, “Left-Libertarianism, a Primer”, opt. cit, 10–11

[10] Henry George, “The Injustice of Private property in Land”, in P. Vallentyne and H. Steiner, The Origins of Left-Libertarianism: An Anthology of Historical Writings, New York, Palgrave, 2000, 193–216.

[11] P. Vallentyne, “Left-Libertarianism, a Primer”, op. cit., p. 19; M. Otsuka, Libertarianism without Inequality, op. cit., p. 37–38; François Huet, Le règne social du christianisme, Paris, F. Didot, 1853, 266–275.

[12] P. Vallentyne, “Self Ownership and Equality: Brute Luck, Gifts, Universal Domination and Leximin”, Ethics, 107, n° 2, 1997, 321–343; Hillel Steiner, “How Equality Matters”, op. cit.

[13] M. Otsuka, Libertarianism without Inequality, op. cit., p. 25; Richard J. Arneson, “Equality and Equality of Opportunity for Welfare”, Philosophical Studies, vol. 56, n° 1, 1989, p. 77- 93.

[14] P. Vallentyne, “Brute Luck, Option Luck, and Equality of Initial Opportunities”, Ethics, vol. 112, 2002, p. 529–557; P. Vallentyne, “Brute Luck, Equality and Desert”, in Serena Olsaretti (dir.), Desert and Justice, Oxford, Oxford University Press, 2003.

[15] Peter Vallentyne, “Self Ownership and Equality…”, op. cit.

[16] M. Otsuka, Libertarianism without Inequality, op. cit., p. 32.

[17] Ibid.

[18] Ibid., p. 35.

[19] Gerald A. Cohen, Self-Ownership, Freedom and Equality, Cambridge, Cambridge University Press, 1995, p. 102–111.

[20] Ibid., chap. 1.

[21] Ibid., chap. 9–10.

[22] Ibid., chap. 9–10.