Conflicting Visions on Freedom?
Real-Libertarianism and the Capability Approach
Freedom, Well-Being and Justice: The Capability Approach
Master’s Thesis Political Science, Specialisation Political Theory & Behaviour Supervisor: Dr. E. Rossi
Second Reader: Dr. M.J.M. Maussen Name: Hugo van Dam
Student Number: 0521701 Date: 30 September 2014
Table of contents 1. Introduction
1.1 Research Subject: The Capability Approach 4
1.2 Research Motivation and Goal 5
1.3 Research Question 6
1.4 Research Outline 7
2. The Liberal Basis of Sen and Van Parijs
2.1 The Capability Approach and Justice as Fairness 8 2.2 Real-Libertarianism and Rawlsian Liberalism 10
2.3 Principles of Redistribution 10
3. The Concept of Freedom in the Capability Approach
3.1 Capabilities and Realised Functionings 12
3.2 The Value of Choice and the Independent Value of Freedom 14 3.3 Problems with the Preference-Based Approach 16
3.4 Well-Being Freedom and Agency Freedom 19
3.5 Ranking Capabilities: Deliberation and Lists 22
3.6 The Need for Positive Freedom 24
4. Real Freedom for All
4.1 Real-Libertarianism 27
4.2 Formal Freedom and Real Freedom 29
4.4 Undominated Diversity 34
4.5 A Basic Income for All 37
4.6 Money as Negative Freedom 40
5. Parsimonious Freedom
5.1 Sen versus Van Parijs 42
5.2 Poverty as Capability Deprivation 43
5.3 Real Freedom as Negative Freedom 43
5.4 Conceptual Parsimony and Real-World Implications 44
1. Introduction
1.1 Research Subject: The Capability Approach
The capability approach as developed by Amartya Sen is an attempt at an integral theory on welfare economics. The approach commences from the theoretical standpoint that “[...] first, the claim that the freedom to achieve well-being is of primary moral importance, and second, that freedom to achieve well-being is to be understood in terms of people's capabilities, that is, their real opportunities to do and be what they have reason to value.” (Robeyns 2011). The various beings and doings people strive to achieve are called functionings. A capability is then a bundle of various functionings of a single person (Sen 1993: 31-53). Capabilities can be distinguished from functionings by setting apart the real opportunity to achieve something from what is actually realised in the end (Robeyns 2011).
As noted above, freedom is cardinal in Sen’s work. Of course, it is important to define what freedom actually stands for in his theory, since freedom is undoubtedly an essentially contested concept (Swanton 1992: 1). This is all the more true for the concept of freedom in Sen’s philosophical work, since the nature of his theory makes it hard to distinguish between capabilities and associated functionings, and freedom in any other given sense.
The concept of freedom inherent in Sen’s capability theory can, be summarised by the term 'preference-based freedom'. It relies on "[...] the view that options a person prefers more contribute more to her freedom (when they are options she in fact has) than options she prefers less." (Carter 1996: 109). Sen's conception of freedom thus centres on the notion that options a person has only matter to his or her freedom if the person actually values these options. This view sets Sen apart from libertarians purely oriented at negative freedom, where negative freedom is understood as the absence of constraints to carry out one’s will (Ibidem). At the same time Sen distinguishes unfreedom seen purely as absence of monetary means from unfreedom as absence of capabilities to achieve certain functionings. More clearly stated: poverty can be seen as deprivation of capabilities (Sen 2002: 87).
1.2 Research Motivation and Goal
Next to Sen's argument on the relation between poverty and (lack of) capabilities, there is another strand in political philosophy that takes this line of thinking a step further. This school argues that there is no conceptual difference between absolute poverty and absence of negative freedom.
A prominent exponent of this school is the late Jerry Cohen, who shows that poverty is in itself a threat to a person’s negative freedom (Cohen 2011). This argument is further expounded by Philippe Van Parijs (1995), who proposes the highest sustainable basic income as a remedy to said lack of negative freedom. As explained by Van Parijs and Robert-Jan Van der Veen (1986: 636), economists often argue that a basic income is impossible in a strictly economic sense. Philosophers on the other hand often claim that the notion of a basic income is ethically unacceptable. Both these reasonings would then combine to form the conclusion that no further research into the idea of a basic income is warranted. Van Parijs and Van der Veen disagree by stating that research into the economic feasibility and the ethical acceptability of this idea can alter its political chances. This thesis follows that reasoning by comparing the concept of freedom that can be found in the works of Amartya Sen with notions of freedom formulated by Philippe Van Parijs in relation to the idea of a basic income as a universal grant.
The argument of this thesis is that Philippe Van Parijs’ approach of real freedom and the associated idea of the highest sustainable basic income is conceptually more parsimonious than Amartya Sen’s concept of preference-based freedom that lies in the capability approach. The goal of promoting freedom in a society must be to infringe as little as possible on citizens’ private spheres and autonomy. The line of reasoning presented here is that capabilities and their equalisation essentially entail a positive notion of freedom, and are thus conceptually an unnecessary infringement on citizens’ comprehensive views of the good life.
The figure below shows the graphic representation of this argument:
Sen Van Parijs
Basic liberal-democratic institutions
+
Basic liberal-democratic institutions
+ Traditional negative freedom
+ Capabilities (=positive freedom)
Traditional negative freedom
entails basic income
=Egalitarian liberal democracy =Egalitarian liberal democracy
1.3 Research question
The main question of this thesis is:
Does real-libertarianism offer a more parsimonious account of freedom in an egalitarian liberal democracy than the capability approach?
The argument presented here is that both Sen’s and Van Parijs’ theories can be described as liberal-egalitarian, but that their approaches differ mostly on the question what the metric of freedom should be once a liberal-democratic society is presented as a given. For Sen, it is capabilities, for Van Parijs this is access to the highest sustainable basic income.
The freedom a person enjoys, can be conceptually split in two: 1. The freedom he or she has to achieve certain goals (ex ante) 2. The actual achievement that he or she realises (ex post)
(Sen 1992: 31). For Sen, 1 would be the capability set that person has access to. For Van Parijs 1 would be the availability of the highest sustainable basic income. In Van Parijs’ theory, 2 is not as important as it is to Sen. The outcomes of the choices people make are only bound by constraints of formal freedom (see § 4.2) and undominated diversity (see § 4.4). Even though Sen’s theory strives for equality of capabilities and not of functionings, the outcomes do matter and are an integral part of capability theory and Sen’s normative position on freedom (see chapter 3).
1.4 Research Outline
This thesis attempts to answer the question posed above by comparing notions of what it is that the state should hand out in order to achieve the greatest freedom for its citizens: the highest sustainable basic income or equal access to capabilities.
Firstly, the basis of Van Parijs and Sen’s theories is traced back to John Rawls’ theory of justice as fairness (chapter 2). It is shown that their theories both build upon Rawls's framework to come to their own conclusions on what an egalitarian liberal democracy should look like.
Then the concepts of freedom as used in both Sen’s and Van Parijs’s accounts are fleshed out to analyse what they actually entail. A reading of first Sen and then Van Parijs shows how they both end up at their respective conclusions, and what steps they take along the way. A number of issues will be looked at in order to assess what the elements of both conceptions are:
-the difference between opportunity and outcome and the effects this difference has on the functioning of the concept of freedom used;
-the value of individual choice in assessing degrees of freedom;
-the difference between collective and individual freedom as worked out in both theories; -the relationship between freedom and justice will be touched upon shortly for illustrative purposes;
-and lastly the respective use that is made of concepts of negative and positive freedom. The analysis of these theories shows that Van Parijs needs less intrusion into citizens’ private conceptions of the good life and thereby less paternalism to arrive at the same place as Sen, namely a free society. This means that Van Parijs offers a more parsimonious account to reach the same outcome and that his theory may thus legitimately be preferred over Sen's when promoting an egalitarian liberal concept of freedom.
2. The Liberal Basis of Sen and Van Parijs
2.1 The Capability Approach and Justice as Fairness
Sen set out to develop a theory that would constitute an improvement on John Rawls’ theory of justice as fairness. Rawls’s theory, as laid out in his seminal work A Theory of Justice, has several well-documented problems. The goal of the capability approach is to avoid these problems in coming to a framework that can be used for normative assessment of well-being and freedom (Robeyns 2011).
Justice as fairness is founded on the idea that there are two guiding principles of justice: “[...] (a) Each person has the same indefeasible claim to a fully adequate scheme of equal basic liberties, which scheme is compatible with the same scheme of liberties for all; and (b) Social and economic inequalities are to satisfy two conditions: first, they are to be attached to offices and positions open to all under conditions of fair equality of opportunity; and second, they are to be to the greatest benefit of the least-advantaged members of society (the difference principle) (Rawls 2001: 42-43). The basic or fundamental liberties Rawls speaks of in the first principle include, for example, the right to freedom of conscience and freedom of association. The second principle is closely related to the economic structures that are prevalent in a given society (Wenar 2013).
According to Rawls, the basic structure of society (the governing arrangement of the political, legal, familial and other related systems) distributes so-called primary goods. They can be described as things that every rational man is supposed to want. Rawls distinguishes between social and natural primary goods. The social primary goods are rights, liberties, opportunities, income, wealth and (the social bases of) self-respect. Natural primary goods include health, imagination and intelligence. Natural primary goods are of less importance in Rawls’ framework, since they cannot be handed out by society’s basic structure in the literal sense (Rawls 1999: 54).
As noted, Rawls’ theory is problematic in a number of aspects. Firstly, Rawls almost completely ignores the diversity of human beings and their circumstances in his theory. The primary goods metric he develops takes little to no note of the fact that so-called ‘conversion factors’ play an important role in our daily lives. Our natural endowments (general health, genetic predisposition to certain disabling conditions etc.) and the physical surroundings we
are born into (climate, availability of natural resources etc.) play a very large role in determining whether we are able to actually put to use the primary goods Rawls speaks of. Simply holding a primary good is not enough: for example, a disabled person may hold distinctly more primary goods than an able-bodied person, and would still not be able to pursue her objectives to the same degree as the latter person (Sen 1999: 70-71, 74). To quote Sen: “Primary goods suffers from fetishist handicap in being concerned with goods, and even though the list of goods is specified in a broad and inclusive way, encompassing rights, liberties, opportunities, income, wealth, and the social basis of self-respect, it still is concerned with good things rather than with what these good things do to human beings.” (Sen 2010: 218). Sen says this in a poetic manner: “Judging advantage purely in terms of primary goods leads to a partially blind morality.” (Ibidem: 216).
A second point is that Rawls’ theory applies only to liberal democratic states, whereas the capability approach may well be summarised by the motto ‘justice applies everywhere’. This is a major disadvantage of Rawls’ framework in comparison with the capability approach. It seems odd that what is meant to be a comprehensive view of justice encounters its limits at the borders of (mostly western) liberal democracies. Sen, in contrast, would have his theory apply to lesser developed countries the same way it does to the western world (Robeyns 2011).
In sum: Sen believes that the notion of primary goods cannot accurately measure or capture the full account of human well-being and freedom. This is where he introduces the notion of ‘capabilities’ (Carter 2012). Capabilities essentially entail a positive notion of liberty (see § 3.6), as they conceptually embody the degree in which a person can actually make use of the freedom she enjoys. This conceptual use of Rawls's framework of primary goods is what makes it possible to compare Sen’s theory to that of Van Parijs. (Ibidem, see also Van Parijs 1995: 240).
2.2 Real-Libertarianism and Rawlsian Liberalism
Like Sen, Van Parijs owes a debt to Rawls (Van Parijs 1995: 227-233). This becomes clear when Van Parijs attempts to modify parts of Rawls’ theory so that he can use it as a fertilising ground for real-libertarianism and the associated idea of the highest sustainable basic income (Ibidem: 131-32). Van Parijs admits that the two theories are distinctly different, but he nonetheless feels that “[...] the formal-freedom-constrained leximinning of opportunities is sufficiently close to the fundamental-liberties-constrained maximinning of socio-economic advantages for it to raise strictly analogous worries.” (Van Parijs 1995: 227).
As briefly mentioned above, Van Parijs introduces the concept of 'real freedom', which entails that traditional negative freedom (understood as freedom from constraints to carry out one's will) is not enough; in order to achieve real freedom, an individual must possess the resources or capacities to actually carry out their will (Van Parijs 1995).
Van Parijs suggests that a society that can guarantee only the first tenet of this dyad is formally free. To achieve real freedom though, the second tenet is indispensable (Van Parijs 1995: 4). Van Parijs indicates that the highest sustainable basic income is a practical implication of his concept of real freedom (Ibidem: 232).
Here we stumble upon a contradiction: the concept of a basic income as a path to freedom (of any kind) as worked out by Van Parijs, seems to go directly against Sen’s suggestion that freedom does not have to do with money alone (Sen 2000: 87-88). How then does Van Parijs’ idea of real freedom relate to poverty seen as capability deprivation? This important question will be further addressed in chapter 5.
2.3 Principles of Redistribution
It would theoretically be possible to have a North-Korean style regime where all capabilities related to personal welfare (for example education, food, shelter) are guaranteed. In Van Parijs’ approach it would likewise be possible to have such a regime where every citizen would enjoy a guaranteed basic income. This example makes it abundantly clear that freedom understood solely as the amount of resources of some kind or another citizens possess, is nonsense. It seems logical then, that the intention of Van Parijs’s and Sen’s approach is the same, namely to make sure that people can actually use their freedom as well as the resources
they attain. The differences lie mainly in what these resources are and according to what principles they are to be divided.
When it comes to resource division Sen strives for so-called ‘basic capability equality’ (Sen 2010: 216). He argues that the primary goods approach Rawls uses for the division of scarce resources has some important flaws. It is based on the presupposition that goods are the measure of all things. Sen argues that it is not goods themselves that are of importance but the ‘relationship between persons and goods’ (ibidem), that is, what a person can actually do with her goods. It may be so that striving for enlargement of total happiness or total desire-fulfilment does not paint a complete picture in most cases, but in Rawls’ approach the relationship between person and good and all that it stands for is completely ignored.
The difference between Rawls and Sen can be easily illustrated by comparing the endowment of a severely crippled person with that of a ‘fit’ person. Say this person would be completely bed-ridden, for example due to a crippling disease such as progressive ALS, but would own a dazzling amount of resources and/or primary goods compared to simple me, still I would not want to trade with the other person given the chance. Instinctively, this seems very logical but it is nonetheless something that Rawls has not taken into account in the primary goods approach (Sen 2010: 216).
Sen uses a version of this example to show that utilitarianism cannot in itself render a just outcome when deciding how scarce resources should be divided. In a purely utilitarian system of welfare and goods division, a crippled person, for example would get a substantially smaller piece of the total pie compared to fit citizens (Sen 2010: 217) The reason for this would be that she simply cannot do as much with the resources allocated to her as a fit citizen could and would thus need significantly more resources to achieve the same outcome. Utilitarianism does not seem to lead us to a just outcome then.
Van Parijs uses a leximin distribution principle: this is a distribution wherein the greatest possible real opportunities must be offered to those who possess the least opportunities (Van Parijs 1995: 5). (Van Parijs adds a further constraint of respect for all citizens’ formal freedom, see § 4.2). It is for now only important to note that it would by its nature grant a crippled person more resources than any of the other distribution principles thus far discussed (Sen 2010: 215). This is further elaborated upon in chapter 4.
Rawls’s difference principle completely ignores the fact that the person we are discussing is crippled (Sen 2010: 217). Rawls justifies this by stating that a crippled person is one of many possible complicated cases that only serve to distract from the creation of a more general theory of justice, that is a theory that would apply to the majority of human beings (Ibidem: 215). This seems a little harsh, as Sen does not hesitate to point out: “[…] hard cases do exist, and to take disabilities, or special health needs, or physical or mental defects as morally irrelevant, or to leave them out for fear of making a mistake, may guarantee that the opposite mistake will be made.” (Ibidem, italics original).
3. The Concept of Freedom in the Capability Approach 3.1 Capabilities and Realised Functionings
The distinction between freedom to achieve and actual achievement is a centrepiece of social evaluation (Sen 1992: 31). As noted earlier, in Sen’s work this distinction corresponds to a person’s capability set and the functioning vector she succeeds in realising. The focus in Sen’s theory is on equalisation of capabilities, even though these are very hard if not impossible to measure (Rawls 2001: 171). In any place, capabilities may be defined as substantive freedoms to achieve functionings (Sen 1999: 75).
In order to assess how well a person is doing in the framework of the capability approach one need only look to what a person is able to do (what kind of activities or meaningful work) and be (for example well-nourished or happy). The various doings and beings a person can achieve are called functionings. All functionings a person performs taken together are called her ‘functionings vector’. A capability stands one level higher, theoretically speaking. It is the combination of several functionings vectors that the person is able to achieve (Kaufman 2006: 290). A set of capabilities consists of a number of combinations of functionings and as such reflects “[…] the person’s freedom to lead one type of life or another.” (Sen 1992: 40).
The capability approach by its nature prioritises some functionings over others and is as such a normative theory. Examples of functionings that, generally speaking, are given priority over random ones include education and the ability to move around freely (Robeyns 2011). This will be dealt with in great detail later (§ 3.5 and 3.6), because it is of key importance to the question here discussed that the capability approach is necessarily committed to a positive
notion of freedom and as such infringes on citizens private conceptions of the good life. The reason that Sen strives for equalisation of capabilities and not of functionings is that having a capability to do something implies having a further choice in what manner you would like to achieve the associated functioning. Even if two people possess widely different capability sets it is still possible for them to achieve the same functionings. In case this happens Sen argues that viewing their starting positions in terms of capabilities still makes it possible to measure differences in their respective amount of freedom (Kaufman 2006: 290). This is where the idea of ‘conversion factors’ comes in. As has been shown in chapter 2 many theories of justice (including Rawls’s) do not take notice of the fact that physical goods or even some immaterial goods (e.g. status, love) can only be enjoyed in a meaningful way in a certain relation to the person they belong to. A simple example would be that if I own a Ferrari but do not have driver’s license, it would be impossible for me to actually enjoy ownership of the Ferrari to the fullest extent (casu quo driving it). To clarify the same point on immaterial goods: if I would be the most loved person on earth but would be in a constant coma that I would never awake from all this love would in a manner be lost on me, that is I would not be able to actually enjoy being loved. This goes to show that one must look at the complete picture when doing any kind of social evaluation.
The capability approach tries to achieve this by taking a person’s situation into account to the fullest possible extent when analysing her capabilities to do and be. This is done by distinguishing three different kinds of conversion factors that can be of importance in a person’s relationship to a good:
-Personal conversion factors are traits that are purely internal to a person, examples include sex and physical condition;
-Social conversion factors stem from the society a person lives in, for example if women are discriminated against or if there is a very rigid class system that hinders upwards social mobility;
-Environmental conversion factors have to do with the physical environment a person lives in. Examples include climate, proneness to drought or floods, et cetera (Robeyns 2011).
Conversion factors in social evaluation are some of the key elements that the capability approach uses to achieve the full picture of a person’s well-being and freedom: “Sen uses “capability” not to refer exclusively to a person’s abilities or other internal powers but to refer
to an opportunity made feasible, and constrained by, both internal (personal) and external (social and environmental) conversion factors” (Robeyns 2011).
All beings and doings combined are that which makes a human life valuable. In order to compare achieved well-being between persons the concept of functionings can be used. When trying to compare the freedom to achieve well-being one may use the term capability. The distinction between capabilities and realised functionings is the distinction between that what is deemed effectively possible and that which is actually realised in the end (Robeyns 2011). An example of this could be that if I work a hundred hours a week and earn large amounts of money by doing so, I would have the financial means to undertake a long journey. If however I were to actually do so, I would lose my job because my boss cannot miss me for more than a day. As such it can be contended that it would be effectively possible for me to undertake a long journey but I choose not to since my job is also important to me. Translated into the capability theory this means that although I have the capability to work a hundred hours per week and to undertake a long journey, the functioning I end up in realising is only the former. This illustrates that choice is also a very important concept in the capability approach, as is discussed in more detail in the next section.
3.2 The Value of Choice and the Independent Value of Freedom
What distinguishes Sen’s conception of freedom from that of other writers is the fact that he leaves a large amount of room for individual choice in assessing individual well-being and freedom (these two concepts and the relationship between them will be further clarified in § 3.4). An example to illustrate why choice is important in measuring the real degree of freedom a person enjoys: You wear a sweater that your mother has selected for you that morning and it is the same sweater that you would have picked yourself. Are you less free than when you would have actually chosen to wear that sweater yourself? At first glance this seems a quite trivial question. However, if we use a similar example in a bit more dramatic fashion it becomes clear that choice is central to personal freedom.
Consider this situation: you run the family-owned business, have more than enough money and are an influential and well-respected member of your community. This seems like a nice outlook, but first we must stop and consider how you have gotten in this situation. For the sake of argument, say there are two options:
1. This is the life that has been presented to you as your only option, by your parents or familial social pressure. You would have rather gone surfing all over the world and deep in your heart you could not care less about the family business.
2. This is the life that you have wholeheartedly chosen, after having spent your twenties travelling the world and surfing where the best waves are. Finally you have realised that surfing is not your true calling, you need to replace your father as head of the family business, and also take up his role as a great contributor to the life of the community you were born into.
These examples relate to the perspective where the freedom to choose is valued in itself. It is called the “independent value” of freedom (Carter 1996: 110). They serve to clarify how the capability perspective differs from other approaches to the concept of freedom and at the same time it shows that there is a real difference in focusing on either outcomes or opportunities. This is so because, as the examples above have illustrated, the same outcomes are possible with a widely different set of opportunities, choices and considerations on the part of the agent. By focusing on equalisation of capabilities, and thereby on the opportunities available to people, the capability perspective allows for a pluralist conception of the good life and thereby of freedom (Robeyns 2011). This, in essence, is what makes Sen’s theory liberal: the recognition that humans may have many goals and values and there is not one correct way of living that is to be universally preferred over all others (Gaus 2011).
Carter (1996) criticises Sen’s account of freedom by stating that the capability theory does not give a convincing account of freedom without a certain degree of control over one’s own destiny. Carter argues that if freedom is intrinsically valuable, as Sen says it is, all opportunity sets available should be ranked first and foremost on the basis of the amount of freedom they provide. Here we come back to the examples 1 and 2 above: if freedom is indeed intrinsically valuable then a capability set that would rank higher than another in the sense of achievement alone would still be ranked lower if this achievement was not attained by a choice on the part of the agent. For the examples above this would mean that if you would be a happy small-time entrepreneur by taking over the family’s laundry business purely by your own choice this would be an outcome of greater value than if you would be an unhappy billionaire who has been forced by her parents to take over the family-owned chain of fast-food stores simply because someone has to do it. This sounds logical at first glance, but leads to a number of
dilemmas.
If choice is taken as the leading criterion for evaluating capability sets because it bestows a certain freedom upon the person who gets to make the choice, who gets to decide what is a valuable choice and what is not? If it has never even occurred to me that I could be a fast-food billionaire and still I am perfectly happy being a small-time laundry owner does that mean I am missing out on something? This feels extremely counterintuitive as Carter also notes: “[…] it is generally true that to deprive people of options they would not choose is nevertheless to inflict a disadvantage on them”. (Carter 1996: 110). Even if taking away many options a person would not choose may not change anything to his or her concrete situation, it can still be argued that her extent of freedom would be diminished (Sen 1988: 290).
The way out would seem to conclude that the preferences that people express by their choices cannot ex aequo be seen as valuable or as providing freedom. There is an “[…] inescapable requirement of valuation involved in the assessment of freedom […]” (Sen 1992: 64). Sen’s approach tries to measure the value of the options available to a person by looking at the preferences that she has. These preferences are then assumed to show themselves in the concrete choices people make. Unfortunately this is not as simple as it seems.
The choices people tend to make may reflect their preferences, but what is to keep their preferences from being in some way distorted? If preferences are to be taken as leading in assessing degrees of freedom how can we tell that people’s preferences are fully their own? And to what extent does satisfaction of preferences lead to more freedom? These questions will now be addressed.
3.3 Problems with the Preference-Based Approach
The problem of ‘adaptive preferences’ also called ‘the contented slave dilemma’ is that the preferences people have tend to adapt to their specific situation. Imagine a person was born into slavery and has never seen that there is a life outside of it. That is to say she has known no life where she would get to make her own decisions and choices independently and without a lot of regard for what other people think. It would not be a stretch to imagine that this person would adapt her preferences to this situation in order not to strive for a great deal of things that are without a doubt unattainable. It could be postulated that to adapt one’s preferences to a bad outlook is a self-defence mechanism to prevent oneself from being
unhappy 24/7 from knowing all the time that the things one strives for are never ever going to happen (Sen 1985a: 21-22).
Adaptive preferences pose a significant difficulty for the capability approach and the related notion of preference-based freedom. If, as Sen says, preferences are expressed by the choices people make, and the satisfaction of these preferences (or desires!) in turn leads to a greater amount of freedom on their behalf “[…] then one way of increasing our extent of freedom is by ridding ourselves of the desire to do those things we are constrained from doing.” (Carter 1996: 115). This essentially Stoic or Buddhist thought (Carter 2012) sounds very promising at the surface but if we stop to see where it leads us the initial enthusiasm should quickly cool. The ridding of desires can be used as a self-defence mechanism against almost any form of oppression since if you do not desire the things that are unavailable or bluntly denied to you, you will keep from feeling bad all the time. This situation serves to mask great injustices of which slavery itself is of course an important example. Thus someone who has foregone all desire can theoretically be said to be free (Ibidem). It can also be said that she would probably look most like a lifeless plant (hereby of course disregarding Buddhist sages and the like who actively choose to strive for a life with as little desires as possible).
If one’s preferences indeed affect one’s degree of freedom, there may be an even bigger problem for Sen’s theory. According to Sen (and many other theorists of freedom) the ideal of freedom is valuable in itself, or as stated in Carter (1996: 110) “intrinsically valuable”. This means that freedom can never only have instrumental value, more simply said: when measuring freedom it is never enough to measure only outcomes. This is so because as we have seen earlier (§ 3.1) in the capability approach and in social evaluation as a whole it is possible for an infinite range of outcomes to be achieved in an infinite number of ways with the use of an infinite number of opportunity sets. Outcomes can thus never in themselves say anything decisive about the degree of freedom under which they were achieved.
The preference-based approach to freedom is also problematic in another respect. According to Sen (1985: 210) environmental factors that lower one’s overall availability of functionings must be categorised as options. This does not seem very logical since many features of the environment a person lives in are outside the control of that person alone. The example most used in this context is that if there is a severe danger of contracting malaria where I live this is not something that I can manage to eradicate completely on my own. My freedom is affected
not only by what I can and cannot do, but also what others do to or for me. Thus, living in a malaria-free environment (without moving) cannot be seen as an option that I alone can choose. It follows then that debilitating environmental factors must be seen as constraints to freedom and that their ‘not-being’ is not an option as such. In contrast, their ‘being’ is a constraint to the functionings that I can manage to achieve. This leads us to the conclusion that freedom cannot be seen only as a relation between an agent and the options available to that agent but instead as “a triadic relation between an agent, constraints and possible functionings.” (Carter 1996: 121).
Sen is of the opinion that in evaluating the goodness of states of affairs, it is impossible to disregard the individual choices people make, “[…] no matter what motivates them towards that choice. (Sen 1991: 19, italics original). There may however be an empirical difference between observed choices and considered preferences. This can happen for a myriad number of reasons such as shortage of information, haste, greed or influence by any other situation or emotion that may distort our actual best choices vis-à-vis ourselves (Ibidem: 20). If however we accept that freedom is of the highest importance in assessing the goodness of a certain state of affairs (as Sen argues in almost all of his works, see § 3.2 for an analysis of the place of freedom in the capability approach) then it would seem strange to conclude that the preferences that a person exercises by making certain choices compete with his overall freedom for importance. To clarify, you can’t have it both ways: one option is that freedom is of the highest importance and option sets that provide us with more freedom must then always be ranked higher than option sets that offer less freedom. The other option is that individual preferences are of the highest importance and that they must always be respected even if they lead to a lower degree of freedom (Carter 1996: 113-116). Acceptance of both these axioms would lead us directly into circularity: if there are two option sets of which the second provides us with more freedom than the first “[…] we would be saying that we prefer the second set because it provides us with more freedom, and that it provides us with more freedom because we prefer it.” (Ibidem: 113).
Sen’s concept of freedom is divided up even further. This division will be the next point of inquiry.
3.4 Well-Being Freedom and Agency Freedom
The capability approach is concerned with substantive freedom. Substantive freedom is more specific than freedom to perform any random action in the sense that it only covers actions that people actually value instead of being concerned with any action they might want to perform (Carter 1996: 109). Sen further distinguishes between two types of freedom: well-being freedom and agency freedom (Sen 1992: 56).
Agency freedom has to do with the agency objectives and goals a person pursues in life. Agency freedom need not have anything to do with achievement of individual well-being of the person herself. Say for example someone strives to become a very successful businessperson even though she knows she will be unhappy from stress and long weeks away from her family. Agency freedom in this case refers to the freedom to make a choice that can be motivated by factors other than personal well-being such as, in this example, the ambition to become a successful businessperson. All a person’s goals and objectives taken together reflect her agency achievement (Sen 1992: 56).
Well-being freedom, in contrast, primarily has to do with achieving personal well-being. Examples include being well-fed or well-educated, being healthy et cetera. Of course the achievement of well-being itself can also be seen as an agency objective. Sen however decides to make this split since achieving personal well-being is, up to a bare minimum, a matter of survival and its assessment is thus of crucial importance (Sen 1985b: 206).
The conclusions that can be drawn from establishing the amount of a person’s agency freedom and well-being freedom differ in a distinct way: well-being freedom has mostly to do with the assessment of advantage where agency freedom symbolises how far a person can go in achieving her own conception of the good (Sen 1985b: 207). Logically, achieving one’s conception of the good (as in the example above, satisfying one’s ambition to become a successful businessperson) need not coincide with one’s own advantage (being able to be with one’s family and not being stressed all the time). It is even possible that achieving a greater agency freedom is detrimental to your own well-being freedom (Ibidem).
Sen himself contends that well-being freedom is only partial and that agency freedom gives the most complete picture of individual freedom (Sen 1985b: 205). Carter (1996: 116-17) criticises this distinction by stating that both well-being freedom and agency freedom are only partial concepts. In contrast, if freedom is intrinsically or independently valuable (as Sen
himself argues, see § 3.2 and 3.3) then we can only assume that Sen is interested in a total and undivided concept of human freedom. A division of the concept of human freedom in the manner explained above would automatically lead to the conclusion that freedom does indeed have important instrumental value. If a person’s freedom may be derived from the achievement of either well-being or certain agency goals this means that freedom is no longer valued above all else for its intrinsic and independent value. Neither the extent of our well-being freedom nor our agency freedom says decisively how much freedom we actually have (ibidem).
Sen makes a further distinction between instrumental agency success and realized agency success (Sen 1992: 57-58). This distinction mostly has to do with whether it is important to actually influence the bringing about of one’s own agency goals. If something that you desire happens without you playing any part in this we may call this instrumental agency success. If in contrast you achieve this feat yourself or play a decisive role in its coming about this may be called realised agency success (ibidem). This is not merely a semantic distinction: imagine that all goals you have in life come about while you just sit in your rocking chair looking out of the window for years on end. This does not seem like a very tempting position to be in, but still it would not be possible to say that your instrumental agency success is absent if we have not looked at whether you are sitting in this rocking chair by your own choice (If you are in no way prevented of getting out of the rocking chair and pursuing any agency goal you desire we might also say that you are just lazy or lack initiative). Leaving this question aside for now, we must in this context turn to Sen’s notions of “effective freedom” and “control freedom” (Sen 1992: 64) to understand what is actually happening here.
Effective freedom can be defined as the realisation of an agent’s goals without her active participation therein. If all your agency goals are realised while you just sit back voluntarily in your rocking chair your effective freedom could be deemed very high. If someone has bound you to the chair and gagged you so that you can never escape your outlook becomes quite different and a lot more problematic. In this situation you lack control freedom, which can be defined as the realisation of your agency goals by your own doing (thus the achievement of instrumental agency success) (Carter 1996: 117, Sen 1992: 65). Imagine that in the outside world your agency goals are being fully realised in the exact same manner you would have done this yourself, only then without your participation, while you are inside gagged and
bound to your rocking chair. This situation does not look like anything that can be called freedom at all (cf. Carter 1996: 117-120).
As a final example, say you do get out of the chair to go outside and produce some great work of performance art (this is your agency goal of choice) but you die while doing this because where you live the air is laden with toxic fumes from the recent explosion of a nearby biochemical waste plant, can you still be called free in realising your agency objectives? It would seem not. Here we again stumble upon the restriction of freedom by environmental constraints (See § 3.3). We must again conclude that also in the context of well-being freedom and agency freedom here discussed environmental factors have nothing to do with personal choice but are still able to constrain individual freedom. From this we can derive that it is not only important what I can do or cannot do but it is also of the utmost importance what others do to or for me (i.e. executing a terrorist attack on said biochemical waste plant) (cf. Carter 1996: 121).
Consequently, it would seem logical that when living under a totalitarian and tyrannical regime (which I contend can also be called an environmental factor) the chances to bring about any agency goal (let alone a big one, for example overthrowing this tyrannical government) will be severely constrained. It is also possible that in this same situation well-being of all citizens is guaranteed up to a more than acceptable point (cf. § 2.3. One could even argue that this is a situation that can actually be found in historical or current governments, but concrete examples need not concern us here). In this situation the importance of control over one’s own destiny comes into play: if you would be well cared for in all respects by servants of this tyrannical government but would be unable to realise your deepest ambitions (say, surfing where the best waves are) could we say that you can in any important manner be called free? It follows that if we take freedom seriously we must accept that true freedom is not possible without control over one’s own destiny (Carter 1996: 119, Cohen 2011: 73-80), and that Sen’s split of the concept of freedom in well-being freedom and agency freedom does not leave enough room for that conclusion. Lastly, the notion of effective freedom denies the agent all opportunity by only looking at that which actually gets realised in the end, which is exactly what Sen was trying to prevent when developing the capability approach. This conclusion is also drawn by Martha Nussbaum (2011: 197-202). Consequently, she will figure prominently next to Sen in the next section, which draws
attention to the ranking of capabilities and to related notions of positive freedom that lie in the capability approach.
3.5 Ranking Capabilities: Deliberation and Lists
For this section, it is important to remember that freedom as described by Sen has everything to do with possessing certain capabilities that can be described as substantive freedoms to achieve valuable functionings. Up until now we have only discussed the theoretical background of the capability approach and its notions of preference-based freedom. The next step will be to assess the manner that Sen proposes to find the ‘right’ capabilities to achieve human flourishing.
Sen contends that individual freedom is a social product and there is a two-way relationship between on the one hand the social arrangements necessary to expand individual freedoms and on the other hand the use of these freedoms to make the social arrangements even more effective thereby creating a positive feedback mechanism that will ultimately lead to greater freedom (Sen 1999: 31). Unfreedom can be witnessed in the process aspect of freedom (in the violation of certain key civil and political rights) and in the opportunity aspect (absence of opportunities for achievement of personal well-being or a minimum of agency objectives) (Ibidem: 17).
Having concluded all this, what then are the concrete capabilities that Sen would promote would he become an enlightened dictator tomorrow?
It is harder to answer this question than it would seem. If we look for a classification of important capabilities, it is first important to look at terminology. Sen himself seems to make use of the term “basic capability” (Sen 2010: 218) to point to the most important capabilities every person would need, the ability to do “certain basic things” (Ibidem). This is not so, in Sen’s (later) work the term ‘basic capability’ refers to a minimum threshold of these important human needs, measurement of which is not necessary for ranking of living standards in different places but for deciding on a cut-off point in the assessment of poverty and deprivation (Sen 1992: 45).
If the term ‘basic capability’ only refers to a certain amount of some unspecified, albeit important, capabilities, the question which capabilities these are still has to be answered. Sen does not attempt to answer this important question in his work. The decision of what is
important in life and what is not should, in his opinion, be left to the social group in question to be decided by processes of deliberation (Robeyns 2011). In any case he has always refused to endorse a concrete list of capabilities of which scholars would decide that they are universally important for the simple reason that this would exclude rational reasoning and democratic processes by the people who are concerned (Sen 2005: 158). Nevertheless, Sen has in his own work stated on multiple accounts that the centrality and importance of certain capabilities are not in question. Examples of these capabilities include health and education (Nussbaum 2011: 20)
A myriad of scholars has attempted to make Sen’s somewhat vague selection process more concrete (Robeyns 2011). The most extensive capability theory of justice is presented by Martha Nussbaum (2000, 2006, 2011). Nussbaum uses the term ‘basic capability’ in such a way as to refer to the “central human capabilities” that are necessary and valuable above all else in every human life (Nussbaum 2011: 17-45). The paradox is that Nussbaum integrates political liberalism as an important part of her theoretical framework in order to avoid paternalism or the forced prescription of one or another account of the good life (Ibidem: 19). At the same time, however, she does promote a list of the ten central human capabilities and thereby attempts to answer the question: “Which capabilities are the most important?” (Ibidem: 27, 33-34). Her defence for this double standard is that these ten central capabilities are essential for leading a life worthy of human dignity (Ibidem: 32).
Furthermore she is of the opinion that the ten central capabilities she has selected are very elemental yet in her formulation very vague. This would enable any political community to further specify the contents of the list in any manner so desired through a self-chosen political process. Nussbaum’s ten central capabilities are:
1. Life;
2. Bodily health; 3. Bodily integrity;
4. Senses, imagination and thought; 5. Emotions;
6. Practical reason; 7. Affiliation;
ii. Having the social bases of self-respect and non-humiliation; 8. Other species;
9. Play;
10. Control over one’ environment i. Political
ii. Material
(Nussbaum 2011: 33-34). Nussbaum states that these central freedoms should in any case be promoted up to a minimal threshold by a government that “[…] has the job of making people able to pursue a dignified and minimally flourishing life […]” (Ibidem: 33). Other freedoms or abilities that citizens might pursue are in her view categorically less important than the capabilities included in this list and the question whether they will be promoted or not can thus be left to the standard workings of the political process (Ibidem: 32).
Nussbaum’s arguments in favour of her list notwithstanding it is easy to see a large degree of paternalism in ‘forced’ promotion of the capabilities that she has included. Even if we concur that these ten central capabilities as proposed by Nussbaum are important we must conclude that the step she takes of defining them herself and defending their importance for every human being is exactly what Sen meant when he spoke of his refusal to defend a canonical or pre-determined list presented by philosophers (Sen 2005: 158). Other writers take issue with the fact that the capabilities on the list seem to lead us straight into political liberalism and are in that sense not as universal as they seem (Robeyns 2011). Every pre-determining of any aspect of the good life minimises the scope that a democratic or deliberative process would have to formulate a social group’s view of what the good life should include.
Now that this has been clearly ascertained the next step is to look at the role of positive freedom in the capability approach.
3.6 The Need for Positive Freedom
Up until now several key qualities of the capability approach, their upsides and downsides, have been discussed. We have seen that the capability approach consists of the important distinction between capabilities and functionings; that freedom of choice is a central concept; that freedom itself is of central importance and has intrinsic as well as independent value; that Sen
distinguishes between well-being freedom and agency freedom; that Sen does not state a concrete method for determining which capabilities are more important than others; that other authors have tried to specify Sen’s suggestion for a deliberative-democratic process; and lastly that Nussbaum has created a list of ten central human capabilities that must be promoted by governments above all others. The next step is to show that the capability approach functions by promoting a positive concept of freedom.
The distinction between positive and negative liberty1 as first posed by Kant was popularised by Isaiah Berlin (2002). Through the years the sharp contrast between negative and positive liberty as it was analysed by Berlin has come into some disuse. In general, however, it still stands: in current debates many authors attempt to achieve a certain degree of synthesis between the two concepts or try to show that one cannot exist without the other (Carter 2012).
The distinction between positive and negative liberty mostly has to do with the importance of control on the part of the agent. Simply said: negative liberty refers to the absence of obstacles that are external to the agent. Positive liberty then refers to the presence of control on the part of the agent (Carter 2012). Negative liberty is traditionally promoted by political liberals. Since it proscribes the absence of intervention by the state in the lives of its citizens as a positive thing this is altogether logical. In contrast a positive concept of liberty is generally used by a variety of perfectionist ideologies (Ibidem).
Questions that are crucial in this debate and at the same time relevant to the line of inquiry presented here have to do with the role of the state in promoting this or that kind of freedom. It would seem that the removal of barriers by the state for its citizens would be enough to make them free to an acceptable degree. Active intervention by the state in the lives of the citizens immediately carries with it the suspicion of authoritarianism or at the least paternalism (Carter 2012). If however we strive for a truly egalitarian liberal democracy we need to ask ourselves if the absence of barriers is enough in itself. If we accept that no human is born the same and we also accept that humans are morally equal, we can follow Sen’s conclusion that in a liberal egalitarian democracy we must strive for basic capability equality to assure that no individual or group is forgotten or marginalised owing to factors outside of their own control. Notice that the reasoning in the last sentence leads us comfortably into a concept of positive freedom: it would be the state that would take responsibility for all individuals to create the conditions for them to
lead a minimally flourishing life.
When interpreting this perspective in a cynical way, it would mean that the state would be taking away responsibility for citizens to create the conditions necessary for their own flourishing. If, on the other hand we accept that the promotion of basic capability equality implies the promotion of freedom by the state we would end up with Rousseau’s famous phrase: “[…] each individual will be forced to be free.” (Rousseau 1999: 58). What Rousseau meant by this was that freedom is only possible within the structure of the state, the collective. Every individual must subject to every and all of the state’s laws and if he does not subject himself he must be forced by the entire body politic to do so.
Of course this is only a thought experiment: going from Sen’s basic capability equality to Rousseau’s subjection of individuals to the general will may be going a bit too fast. The line of reasoning that they follow with regard to the fact that it is the state that should take responsibility for its citizens and that it by doing so proscribes a certain account of what the good life entails is still inescapable. For now it is enough to conclude that the capability perspective in itself implies the use of a positive concept of freedom.
4. Real Freedom for All 4.1 Real-Libertarianism
Like Sen, Van Parijs is also of the opinion that freedom is of the greatest importance. He observes that capitalism as we know it is replete with a myriad of inequalities. Van Parijs’ theory is formulated as an answer to those who believe that these inequalities must be accepted if freedom is cherished above all else (Van Parijs 1995: 1).
Van Parijs, as we have seen earlier (§ 2.2) states that freedom cannot coexist with the existence of structural inequality. If in a capitalist society only negative freedom, understood as the absence of barriers external to the agent (see § 3.6), is guaranteed those without money would be unable to actually carry out their will and can thus not be called free. Van Parijs strives to combine a negatively free society (Formal freedom, see § 4.2) with a basic income for all (see § 4.5). He calls this perspective real-libertarianism or real-freedom-for-all (Van Parijs 1995: 1).
In Van Parijs’s theory freedom has everything to do with the division of property in a society (Van Parijs 1995: 3-5). The reason for this is simple and can be illustrated by a thought experiment: say that formally speaking, according to the laws of our land, we are all free and equal citizens. However, there is one man who owns everything: the land we walk on, the air we breathe, the food we need for our sustenance. He agrees to share his possessions with us and thereby assure our survival. In order to convince him that we actually deserve a fair share of the goods that we need to maintain a decent standard of living he imposes some conditions on us. Specifically, he demands from us that we show him that we can run a mile under five minutes and that we must come by his house every other day for a performance of ballroom dancing to be judged worthy only by him. Those who are in some way physically handicapped so that running is hard or impossible and those who have no sense of rhythm whatsoever cannot perform these tasks adequately. Thus, the ultimate consequence will be that they will be left to starve. It begs no further explanation that the situation described here does not deserve the label ‘freedom’.
Widerquist (2013: 1-8) argues that capitalism as well as socialism show traits of the situation above as of the fact that they both enforce a system of property rights: one private, the other public. The situation sketched above could occur in a system of private ownership as well as in a system of public or collective ownership. The man who asks us to dance and run simply
symbolises who (or what, a company for example) actually owns a large majority of the property in a given society (or all in the case of collective ownership). Widerquist contends that it is not possible to be truly free and independent if one does not own any property (Ibidem: 7). If all property in a society already belongs to someone then the only way to acquire enough goods for subsistence is to sell one’s labour power. If the only other option available is to starve this situation cannot realistically be called freedom. A system that functions by the sale of labour power must provide its subjects with the option of non-participation, otherwise it is no different from a system of forced or slave labour (Ibidem: 4). Hayek (1960: 137) disagrees by stating that even if the threat of starvation to myself or my family compels me to accept an extremely low-wage job and even if there is only one man who could provide me with this job, I would still not be coerced by this man, anyone or anything to actually accept this job. Hayek refuses to equate liberty with the ability or the opportunity to do certain things, even if these things require money. Intuition shows that this statement is false. For if my only choices are between starvation and a choice that is in some other way horrendous to me or my loved ones (For example, becoming a sex slave) this is not an actual choice and I am not free in any meaningful sense. Van Parijs (1995: 22) takes this further by stating that in a society with a system of private property negative freedom alone can never be enough. Opportunity must be part of a concept of freedom that has any substance. It is therefore also the key concept that distinguishes real-freedom for-all from formal freedom (see § 4.2).
Admittedly, any concept of freedom always embodies ideal theory (Van Parijs 1995: 8). This may be so, but the striving for any ideal of freedom can have profound effects on the organisation of a society. If the ideal is the maximisation of the aggregate of individual freedom, this is not to say that every individual will be maximally free. It could even be that the exact opposite is the outcome. An example: in a situation of maximal aggregate freedom, even voluntary slavery could be tolerated since every individual would privately own his own body to rent out or even to sell. Additionally, if we say society itself should be maximally free, the implications for individual freedom can be grave. If a society is threatened by some external enemy and its collective freedom must be defended, conscription may be superinduced. Understandably, this would constitute a large infringement on individual freedom (ibidem: 16).
As a first step in discussing real-libertarianism it is good to conclude that the freedom that Van Parijs finds important is individual freedom. He contends that both formal freedom (see § 4.2) and real freedom are “[…] aspects of individual freedom: collective freedom […] is only instrumentally relevant to both of them […]” (Van Parijs 1995: 4). In Van Parijs’s theoretical framework, individual freedom does not mean that you are not prevented from doing whatever you want to do, it means that you are unprevented from doing whatever you might want to do (Ibidem: 19). Say what you desire is to not participate in the capitalist system. Considering the example above, how is this possible in a system with private property rights? This is where basic income comes in. Striving for this ideal of freedom means that the highest possible basic income should be made available to all citizens. Before going on to the concept of a basic income, we must make sure that with its implementation, no freedoms are violated. That is where the demand of respect for formal freedom comes in.
4.2 Formal Freedom and Real Freedom
As discussed in § 4.1, there is a distinction between formal freedom and real freedom. This section serves to further elucidate this distinction.
Van Parijs calls the negative freedom that ‘traditional’ libertarians promote ‘formal’ freedom because he feels that is does not say anything decisive about any real-existing situation. What is meant by this is that even if I had an entire catalogue of rights apply to me on paper, this would still say nothing about my actual predicament. Of course that is not to say that formal freedom is not important, it is generally understood to include important rights such as freedom of speech, freedom of religion and freedom of conscience. It only means that it is possible to have a bundle of rights apply to you while you are dying in the gutter. As was clearly stated in § 4.1 dying in the gutter (or comparable states) cannot be described as situations wherein you are actually free. Formal freedom can, in principle, only be restricted by some form of coercion. “[…] real freedom can be further restricted by any limit to what a person is permitted or enabled to do. Van Parijs (1995: 32-33) formulates the distinction as such: “One is really free, as opposed to just formally free, to the extent that one possesses the means, not just the right, to do whatever one might want to do.”
So, of what elements does real freedom consist?
According to Van Parijs, formal freedom consists of these two conditions: “1. There is some well enforced structure of rights (security).
2. This structure is such that each person owns herself (self-ownership).” To achieve real freedom, a third condition must be satisfied:
“3. This structure is such that each person has the greatest possible opportunity to do whatever she might want to do (leximin opportunity).”
(Van Parijs 1995: 25).
The first condition has to do with the fact that a society without any form of ownership cannot be said to be free. If no property rights would exist, there would be nothing but force to decide who owns what. It would not be a stretch to imagine that the result would be chaos and anarchy. Libertarians have therefore always made it one of their base claims that “[…] there can be no liberty without property” (Van Parijs 1995: 12). A second advantage of having some system of property rights is that the security it provides makes that citizens are able to plan for the future and produce goods and services in cooperation and in the mutual trust that their ownership rights will in some way (to be determined in detail) be respected (Widerquist 2013: 4).
Self-ownership is the second part of Van Parijs’s concept of real freedom. As the term already suggests, self-ownership means that you are the full and total owner of yourself. It is important to state immediately that this is also an ideal and that it can and will never be fully realised in any society. Furthermore, in reality self-ownership is a matter of degree and the question “do I own myself?” can accordingly never be answered with a straightforward ‘yes’ or ‘no’ (Van Parijs 1995: 6-7). In Van Parijs’s conception of real freedom there are many limits to the concept of self-ownership. Slavery, for example, is not compatible with a free society (Ibidem: 9). Widerquist (2013: 15) incorporates the concept of self-ownership in his definition of what a free society actually is. It should be possible to distinguish between those who are free and those who are unfree: freedom is then defined in terms of rights (Van Parijs’s formal freedom) and independence (the power to say no, that is to refuse cooperation with the economic system in place or other individuals in it). The concept of self-ownership will be further analysed in § 4.3.
The third condition is what distinguishes Van Parijs’s concept of real freedom from traditional libertarian approaches: the economic structure of society must be constructed so that the ‘leximin opportunity’ principle can be successfully realised. This ‘lexicographic maximin’ principle entails that those in society who have the least opportunities have access to an opportunity set that is not smaller than the one enjoyed by any other person under any other feasible economic regime. In case there is a feasible economic regime that does provide evenly good for the person with the least opportunities then the principle must also apply to the second worst-off person in society, ad infinitum (Van Parijs 1995: 25). There is an important difference between this principle and other much-used distribution principles: it strives to make every member of society as free as she can possibly be. The same cannot be said for a formula that tries to equalise opportunities enjoyed by all citizens, or one that takes the opportunity set of the average member of society as a starting point and then attempts bring everyone’s opportunities as close to the average as possible (Ibidem).
Of course it would also be necessary to make a ranking of the three principles in case there would be conflicts. The effective protection and enforcement of property rights may demand that those who do not respect them will be put in prison or punished in some other way that would constitute a violation of their self-ownership. It is also important to note that some violations of self-ownership will be deemed paternalistic yet serve to increase the opportunity set of all citizens. Examples may include prohibition of alcohol for those below a certain age, prohibition of certain drugs or obligatory education up to a certain age. In these examples the infringements that relate to health serve to make sure that young people do not sustain irreversible damage to their physical well-being. The infringement of self-ownership that comes with obliging all citizens up to a certain age to be educated has to do with the fact that the opportunity set of those who are not may be unacceptably smaller compared to citizens who have been educated (cf. Van Parijs 1995: 26). For the analysis of the concept of real freedom, it is important to note that all these infringements are options and are thus strictly speaking not required when striving for a really free society.
When comparing the concept of real freedom to other possible accounts of freedom it is important to understand that Van Parijs considers a just society to be a free society: “A free society is one in which people’s opportunities are being leximinned subject to the protection of their formal freedom, that is, the respect of a structure of rights that incorporates
