Equality and Political Philosophy
Summary and Keywords
Equality is an undisputed political and moral value. But until quite recently, political philosophers have not fully explored its complexity. The literature on equality and egalitarianism is vast, complex, and multilayered—with over thirty-five years of philosophical discussion. Specifically, there are three major questions to ask about equality. First, what is equality? This question can be unpacked into two sub-questions. Distinguishing first between formal and distributive accounts of equality, we may ask what the currency of egalitarianism can be. The article goes through currencies such as welfare, resources, and capabilities, showing their respective strengths and weaknesses. A second important sub-question here is: what are the relevant scope and temporal dimensions of equality? Among whom is equality valuable, and precisely in what time frame is it valuable?
This hints at the second major question, namely concerning the value of equality. Is equality indeed valuable, or are we confusing it with some other value, be it giving priority to the worse-off, or lifting individuals above a certain threshold of deprivation? The article goes through some famous criticisms of equality’s purported lack of value (e.g. the leveling down objection), explores some potential answers, and then examines the relative strength of equality’s two main rivals, namely priority and sufficiency.
The third major question concerns what the proper account is of egalitarian justice. In particular, setting aside the question of currency, should our conception of distributive justice be informed by responsibility-sensitive accounts, or rather be focused on a responsibility-insensitive accounts that moreover place an emphasis on equality of relations rather than individuals’ holdings? We explore this in the two final sections, one devoted to understanding luck egalitarianism, and the other to its rival, relational egalitarianism.
Equality is ubiquitous in our political and moral vocabulary, yet there is scant agreement on what it entails. This is because the concept of equality is so abstract that it is almost empty unless we specify what sort of equality we are talking about. To put it differently, we cannot answer the question “does equality matter?” before we answer the question of “equality of what?” Some concepts of equality may well be subsumed by other concepts once the concept of equality is specified and qualified. For example, imagine that the concept of equality is interpreted as equality in terms of moral importance of people’s welfare, i.e., “everybody to count for one, nobody for more than one.” Such an interpretation of equality, for example, would be better captured by a different concept, such as impartiality. Therefore, it is imperative to identify a broad notion of equality at the outset.
In this chapter, we will concentrate on what we call distributive equality. Distributive equality means equality in interpersonal distribution of goods, be it non-instrumental or merely instrumental. We will focus on four central issues in the recent literature of distributive equality and devote one section to each issue. In Section 1, we will discuss what is to be distributed equally. In Section 2, we will then discuss the advantages and disadvantages of three major egalitarian principles. In Section 3, we will introduce the basic idea of luck egalitarianism and its criticisms. In Section 4, we will critically assess a rival idea to luck egalitarianism, namely that of relational egalitarianism.
Equality of What?
The Currency of Distributive Equality
What is the appropriate basis for comparing people’s relative holdings for distributive equality? There are three broadly defined approaches. The first approach is focused on people’s welfare. Welfare roughly means the overall quality of an individual’s life. There are many competing accounts of welfare: pleasure, happiness, desire-satisfaction, and so on (Griffin, 1986). Equality of welfare has provoked a number of objections. Here are just two. First, equality of welfare implies that individuals with expensive tastes (e.g., a taste for champagne) would need more resources in order to obtain the same level of welfare as individuals with less expensive tastes (e.g., a taste for beer) and hence should receive more resources, which is highly counterintuitive (Dworkin, 1981a). Second, equality of welfare is insensitive to the diversity in natural endowments, namely one’s talents and health. Imagine that an individual with severe handicap has been experiencing an awful lot of inconveniences in everyday life, but obtains a high level of happiness—as high as other individuals with no disability—because she has learned to adapt to her disability. On equality of welfare, such a person is judged to be equally well-off as other individuals with no disability, which is absurd (Dworkin, 2000).
A second major approach to the currency of equality focuses on resources. Proponents of “equality of resources” concede that resources are not valuable for their own sake, but rather stress that resources serve as a means to achieving what a person values in life. One advantage of equality of resources is that, unlike equality of welfare, it is not committed to a particular conception of the good. It also bypasses the expensive taste objection rather easily: equality of resources is not hijacked by individuals’ particular tastes. It is committed to supplying individuals with an equal basket of resources, independently of their varying tastes. There are different interpretations of the currency of resources. The most famous one is Rawls’s (1971) primary social goods, that is, the goods that any rational individual would want to have, regardless of her own particular conception of the good.
Equality of resources, however, encounters a number of objections (Sen, 1985). One major objection is that this currency is insensitive to the diversity in what a person can be and do from a given bundle of resources. Imagine that a physically disabled person and a person with no disability possess the same bundle of resources. According to equality of resources, there is no relevant inequality between these two persons. However, there is a significant difference in what these two persons can achieve from the same bundle of resources, simply because disabled individuals require more resources (e.g., a wheelchair). Thus, equality of primary social goods is unjustifiably insensitive to the diversity in people’s capability to achieve what they value. Dworkin (1981b) has offered an account of equality of resources that strives to avert this problem. His account encompasses not just primary goods, but all conceivable resources, and not only social ones but also natural ones such as talents and physical capacities. Equality of resources can thus match Sen’s criticisms and offer an account that is endowment-insensitive.
A third approach can be described as lying somewhere between welfare and resources. It focuses on people’s diverse conditions, as distinguished from their quality of life and/or material conditions. Here as well there exist a variety of accounts. One example is Sen’s (1985) idea of capability to function. A person’s capability is a combination of functionings that she can derive from a given bundle of goods and services. As the capability approach looks at what one can derive from a bundle of resources, unlike equality of resources, it is sensitive to the divergence in people’s ability to convert a bundle of resources into welfare. At the same time, it is not committed to a substantive account of welfare, thus being neutral amongst competing conceptions of the good life. Another example is the approach based on opportunity for welfare (Arneson, 1989) or access to advantage (Cohen, 1989). This approach aims to equalize people’s welfare if inequalities of welfare or advantage result from the differential effects of causes beyond a person’s control (e.g., her circumstances and natural endowment). On this approach, inequalities of welfare or advantage do not give rise to any moral concern when and because they reflect the differential effects of a person’s intentional choice and/or ambition. As we will see in section 3, the theory known as luck egalitarianism is based on this approach.
The Temporal and Spatial Dimensions of Equality
Conceptual analysis of the currency of distributive equality does not yet determine the unit of distributive equality. In thinking of the unit of equality we may focus on two dimensions: a spatial and temporal one.
Let us start with the dimension of space. Analysis of the spatial dimension concerns the question of equality amongst whom. There are three types of answer: (1) equality between individuals, (2) equality between states, and (3) equality between groups within a state. These three types of answer are not mutually exclusive. The choice of interpersonal unit really depends on what one wants to analyze. If the subject of analysis is international distributive justice, the natural choice is (2). But focusing on (2) is perfectly consistent with (1). When you look at inequality between states, you may set aside inequality within each state for practical purposes. But this does not mean you ignore inequalities within each state. You can talk about the increase or decrease of international equality in the way consistent with equality between individuals in each state. The only thing you have to bear in mind is the subgroup consistency: (1) given the inequality within each nation being constant, reducing interstates inequality reduces the overall inequality in the world, and (2) given the inequality between states being constant, reducing inequality within a state reduces the overall inequality in the world. Whether we should choose (1) or (2) really depends on your focus, and therefore, there is no reason to believe either (1) or (2) must be false.
Positions (1) and (3) may appear to be in direct conflict. To wit, consider the following simple example. Imagine that there are two racial groups, each of which consists of two individuals. Compare two possible states, X and Y, where the italicized numbers represent the welfare level of individuals in one racial group, and the non-italicized numbers the welfare level of individuals in the other racial group.
According to equality between individuals, X and Y are indifferent. But some people would disagree, as it may be argued that Y is better than X because X contains inequality between the two racial groups, whereas Y does not. They would claim that distributive equality should look at equality between groups within a state, rather than equality between individuals. However, such a claim does not undermine equality between individuals, but merely highlights the differences in the focus of analysis. What is problematic about X is not inequality of welfare between two racial groups as such, but the possible explanation of what makes inequality bad. Those who assess the two situations differently, quite reasonably, speculate that inequality is explained by the racial discrimination and/or the systematic unequal treatment of the two racial groups. What makes X more problematic is the source of the inequality. Imagine a hypothetical, and highly idealized, world, where there is no unequal treatment or discrimination amongst racial groups and where welfare is allocated randomly. In such a world, there is no morally relevant difference between X and Y. This thought experiment suggests that the argument for equality between groups is really concerned with how inequality is brought about. As we shall see in Section 3, luck egalitarianism takes the way in which inequality is caused very seriously, but is perfectly consistent with equality between individuals.
Next, let us look at the temporal dimension of equality. Should we look at inequality of a temporal part of people’s welfare at a time (the time-slice view) or inequality of people’s lifetime welfare (the lifetime view)? Although many egalitarian theorists favor the lifetime view (Dworkin, 1981b; Nagel, 1991; Rawls, 1971), it is important to understand why the lifetime view is more plausible than the time-slice view. To see the difference between the two views, compare two-period two-person alternatives in Table 1 (McKerlie, 1989). According to the time-slice view, alternative 1 gives rise to moral concern, whereas alternative 2 does not. This is because alternative 1 contains inequality of welfare at both periods, whereas alternative 2 does not. In contrast, according to the lifetime view, there is no difference between alternatives 1 and 2 because there is no inequality in terms of lifetime welfare.
Nonetheless, the time-slice view is less popular, partly because it rests on a problematic assumption, namely the separability of time. The separability of time implies that the disvalue of inequality at a time is determined independently of that at other times. But this turns out to be problematic. Compare two alternatives in Table 2. The time-slice view judges that alternatives 3 and 4 are indifferent, even though in alternative 3 person 2 is worse-off at both periods. Alternative 4 is better, not only because it contains no lifetime inequality, but also because it is fair, whereas alternative 3 is unfair to person 2. The concern for fairness, thus, tips the balance in favor of the lifetime view.
The Value of Equality
Deontic and Telic Concepts of Distributive Equality
There are two main ways to understand the moral relevance of distributive equality. The first is telic, and the other is deontic. There are different ways to understand the telic/deontic distinction, but the most influential way in recent years is due to Derek Parfit:
There are two main ways in which we can believe in equality. We may believe that inequality is bad. On such a view, when we should aim for equality, that is because we shall thereby make the outcome better. We can then be called Teleological—or, for short, Telic—Egalitarians. Our view may instead be Deontological or, for short, Deontic. We may believe we should aim for equality, not to make the outcome better, but for some other moral reason. We may believe, for example, that people have rights to equal shares.
(Parfit, 2000, p. 84, original italics)
Here is an example of a deontic version. Some would claim that everyone has a right to equal share of resources in society. According to this claim, equality is understood as a deontic concept, i.e., moral right.
There are several ways of characterizing the telic version of distributive equality. The most discussed characterization of it is again Parfit’s. According to him (Parfit, 2000), telic egalitarianism endorses at least the principle of equality. This principle holds that it is in itself bad if some people are worse-off than others. The principle of equality is a necessary and sufficient condition for characterizing telic egalitarianism, although it is often (and arguably must be) combined with another principle, the principle of utility (it is in itself better if people are better-off). It is this ideal of telic egalitarianism that has been discussed, criticized, and defended in recent years.
Telic egalitarianism: It is in itself bad if some people are worse-off than others.
It is worth mentioning, however, that some have argued that the telic/deontic distinction is not as salient as it was thought to be (Lippert-Rasmussen, 2016, ch. 5). Still, one aspect that may prove important in distinguishing the two versions of egalitarianism is this. Deontic accounts assume that “ought implies can.” If nothing could be done to improve an unequal state of affairs, then on deontic egalitarianism nothing is amiss. But telic egalitarianism is not bound in this way. An unequal state of affairs may be bad even when there is nothing we can do to alleviate it.
In criticizing telic egalitarianism by far the most ubiquitous objection has been the leveling down objection (Parfit, 2000). Imagine that the welfare of a better-off person is lowered to the level of a worse-off person without thereby benefiting any of them. According to the objection, telic egalitarianism must claim that this leveling down is better at least in one respect because it reduces inequality. However, the objection continues, the leveled down outcome is better in no respect. Thus, the objection concludes, telic egalitarianism is absurd.
Many take the leveling down objection as devastating for telic egalitarianism. But by now several attempts have been made to meet it. We will refer to three responses. The first one meets the leveling down objection by imposing a condition on the instances where equality is valuable. On this response, known as conditional egalitarianism, equality is valuable for its own sake only when it benefits at least one person (Mason, 2001). According to this response, although equality is valuable for its own sake, that value is conditional upon the equal distribution benefitting some person. This simple response enables telic egalitarianism to avoid the leveling down objection because according to it, equality is not seen as valuable for its own sake in the cases of leveling down. However, it may be seen as arbitrary and ad hoc to impose such a condition on the value of equality just for the sake of avoiding the leveling down objection (Holtug, 2010, pp. 193–198).
The second strategy in response to the leveling down objection is to challenge Parfit’s characterization of telic egalitarianism. Initially, Parfit characterizes telic egalitarianism as the view that equality makes outcomes better. He then speculates which such view can be characterized by the principle of equality. However, there might be other ways to characterize the view that equality makes the outcome better. For example, some might see equality as a feature of function that aggregates the list of people’s welfare to the overall goodness of states of affairs. According to this interpretation, the goodness of a state is given by the list of people’s welfare, not the list of people’s welfare and the value of equality (Broome, 2002; Hirose, 2015, ch. 3). If the goodness of a state is given by the list of people’s welfare, then there is no respect with regard to which the leveling down is better.
The third response to the leveling down objection is to simply bite the bullet (and admit that egalitarianism may endorse leveling down) but criticize the objection’s underlying motivation. Temkin, for example, concedes that on his view leveling down is better in at least one respect. However, he challenges a philosophical assumption behind the force of this objection, namely what is known as the person-affecting view. The person-affecting view holds that a state is better than another only if it is better for some person. In other words, the person-affecting view implies that whenever we judge a state as better than another, there must be some persons for whom it is better. It is obvious that the force of the leveling down objection rests on the person-affecting view. Leveling down is better for no one, yet telic egalitarianism judges it to be better at least in one respect. However, if the person-affecting view is rejected, the leveling down objection collapses. This is exactly what Temkin (1993, 2000) undertakes to do. In order to reject the person-affecting view, it suffices to show that there are some values that cannot be reduced to individual goods. Temkin argues that the value of equality is one such value. For those who commit to the irreducible value of equality, leveling down is better in one respect. It may well be absurd to judge that leveling down is, all things considered, better. But telic egalitarianism does not commit to such a judgment because the increased value of equality from leveling down may be outweighed by the decreased value of something else, such as the reduced value of total welfare.
Theorists who are still swayed by the leveling down objection would typically align themselves with one of its two main rivals: prioritarianism and sufficientarianism. Let us state prioritarianism first.
Prioritarianism: Benefiting people matters more the worse-off these people are.
Prioritarianism does not consider equality as valuable in itself. Rather, prioritarians claim that a greater moral weight should be given to benefits to those who are at a lower absolute level (Parfit, 2000). The key idea for prioritarianism is what has been called the law of diminishing moral goodness of welfare. That is, the moral goodness of a person’s welfare diminishes as the absolute level of her welfare increases. The moral goodness of a person’s welfare is therefore determined by the absolute level of her welfare, independently of the position of her welfare relative to others. The moral goodness of a state is given by a sum-total of the moral goodness of people’s welfare: f(w1) + f(w2) +…+ f(wn), where wi indicates the welfare level of person i and f( ) is an increasing and strictly concave function, i.e., a curve that goes upward but bends downward.
Provided that the goodness of a state is given by a weighted sum of people’s welfare, there is no respect with regard to which the leveling down is better, thus avoiding the leveling down objection. However, prioritarianism still has some egalitarian bias. According to the law of diminishing moral goodness, we can derive a greater amount of value from handing some benefit to a worse-off person than giving the same size of benefit to a better-off person. (Utilitarianism, absurdly, would be indifferent between the two). This in turn means that the transfer of some benefit from the better-off to the worse-off makes the outcome strictly better, given the total welfare being constant. (This feature is sometimes called the Pigou–Dalton principle of transfer.) The goodness of a state can be increased by equalizing the transfer of benefits until the welfare level of the better-off and worse-off becomes the same. Consequently, the goodness of a state is maximized when people’s welfare is distributed equally. Thus, prioritarianism is in a sense egalitarian, and certainly much more so than utilitarianism. Nonetheless, prioritarianism does not take equality to be valuable for its own sake. It is egalitarian in a non-relational sense.
There are several important objections to prioritarianism thus construed. One of the earliest objections is that prioritarianism, or at least its difference from egalitarianism, is not that important. According to prioritarianism, the total moral good is maximized when people’s welfare is perfectly equal. When some people’s welfare is at a higher absolute level than others, then there is some wasted good that could have been obtained under perfect equality. Thus, the goodness of a state can be represented by a difference between (a) the good that could have been obtained under perfect equality and (b) the good that has been wasted in an actual unequal distribution. But (b) can be seen as a bad caused by the actual unequal distribution. Therefore, prioritarianism is extensionally equivalent to, at least, some form of telic egalitarianism (Fleurbaey, 2015; Jensen, 2003). In other words, prioritarianism ranks all possible distributions in exactly the same way as some form of telic egalitarianism. If so, there is practically no difference between the two principles. If this point is true, then prioritarianism is not seen as a distinctive principle that is superior to telic egalitarianism.
Another objection to prioritarianism is a variant of Rawls’s objection to classical utilitarianism. According to Rawls (1971), in classical utilitarianism there is no fundamental difference between the trade-off of benefits and burdens within a person’s life and the trade-off of benefits and burdens across different persons. Rawls then points out that there are important differences between the two because there is a unifying perspective in intrapersonal trade-off whereas there is none in interpersonal ones. From this, Rawls claims that classical utilitarianism fails to take the separateness of persons seriously. Otsuka and Voorhoeve (2009) appeal to the separateness of persons in order to object to prioritarianism. They first show that prioritarianism treats intrapersonal and interpersonal trade-off in the same way, whereas telic egalitarianism does not. Yet, they continue, we commonly believe that we ought to treat the two types of trade-off differently. There is a moral shift between intrapersonal trade-offs and interpersonal ones, a shift that prioritarianism simply cannot take account of. According to Otsuka and Voorhoeve, such a shift is made possible when prioritarianism does not take the separateness of persons seriously.
This objection, however, does not necessarily trouble Parfit, who was one of the first to introduce prioritarianism. In his seminal Reasons and Persons (1984), Parfit offers a thorough analysis of the separateness of persons, including the metaphysical question of whether different people really live separate lives. We cannot go into the details of his analysis here, but one radical upshot of his analysis is that, morally and rationally speaking, the intrapersonal trade-off is no more or less important than the interpersonal trade-off. Needless to say, it is highly debatable whether the separateness of persons per se, and people’s intuition about it, is what morally and rationally matters. But it is important to note that the objection of Otsuka and Voorhoeve comes down to, ultimately, a deep metaphysical question. Others have questioned how damning Otsuka and Voorhoeve’s critique is to prioritarianism even when we do not question its underlying metaphysical assumptions. It has been shown that when focusing on a telic understanding of prioritarianism, its judgment that we ought to treat interpersonal and intrapersonal trade-offs on a par is not all that counterintuitive (Segall, 2015).
Here is another objection to prioritarianism. Imagine that some benefit (say, a bottle of wine) can be given to either a best-off person or second best-off person (both of whom are extremely well-off individuals), but, for some reason, it cannot be given to anyone else (Crisp, 2003). Prioritarianism would contend that the benefit should go to the second best-off because her welfare is at a lower absolute level than the well-being of the best-off. But the idea that the moral state of the world would be improved by handing yet another bottle of wine to an extremely well-off person seems absurd. The motivating idea behind this objection is that distributions amongst the well-off people do not give rise to moral concern. Consequently, our distributive ideal should focus solely on welfare below a certain absolute level, such as the level sufficient for living an autonomous life, living without suffering, exercising the functioning of moral agents, and so on. This objection to prioritarianism has led some theorists to endorse sufficientarianism.
Sufficientarianism: It is in itself bad if some people are below the sufficiency level.
Sufficientarianism first identifies the sufficiency threshold, below which moral concern arises. It then holds that the disvalue associated with welfare below the sufficiency level should be minimized, regardless of welfare above the sufficiency. Three distinctive features often characterize sufficientarianism. First, the distribution of welfare above the sufficiency level does not matter. Second, the gain for people below sufficiency always dominates the gain for people above sufficiency. Third, below the sufficiency level, a greater moral importance is given to benefits to the worse-off, and hence the distribution of welfare matters.
Sufficientarianism has attracted many objections, but we will refer to only three of them. The first objection is that the level of sufficiency is morally arbitrary (Arneson, 2002). Now, while it is certainly true that the level of sufficiency is arbitrary to some extent, the concept of sufficiency itself is not arbitrary. Moreover, upon reflection the sufficiency level may turn out to be vague rather than arbitrary. If so, the charge of arbitrariness does not seem to undermine the basic idea of sufficientarianism. The second objection is that the second feature of sufficientarianism (the absolute prohibition on trade-offs between the two sides of the threshold) is too extreme. This feature implies that a large aggregate loss for people above the sufficiency, no matter how large, can be justified for the sake of a small gain for someone below the sufficiency, no matter how small. Certainly, this feature appears quite extreme. However, this would not trouble proponents of sufficientarianism. For them, this prohibition on trade-offs between the different sides of the threshold merely highlights how radical and attractive sufficientarianism is (Hirose, 2016).
The third objection is that sufficientarianism turns out to be so close to prioritarianism that it is not sufficiently distinct from it. Imagine that we can transfer some benefit to a person below the sufficiency threshold from either (a) a person far above the sufficiency or (b) a person just above the sufficiency. Assume that a person just above the sufficiency threshold still remains above the sufficiency after the transfer. Sufficientarianism is committed to being indifferent between (a) and (b) because it does not care about the distribution of welfare above the sufficiency level. However, some would claim that we should choose (a), in which case benefiting the worse-off matters more than benefiting the better-off even above the sufficiency. This means that sufficientarianism gives a greater weight to benefits to the worse-off both below and above the sufficiency. Then, it turns out that sufficientarianism is a version of prioritarianism, but one which includes discontinuity of welfare between, above and below the sufficiency (Hirose, 2015).
This concludes our discussion of the value of equality. Assuming equality does have a value, and supposing it is one of the requirements of justice, how exactly ought we to pursue it? In particular, what is the role of personal responsibility in an egalitarian theory of justice? And, conversely, should we be focusing on equalizing some individual distribuenda or rather focus on the relations between individuals? We turn to that now.
Equality and Responsibility
In Section 1, we examined objections to equality of welfare. One of them concerned the issue of expensive tastes. According to this objection, equality of welfare favors transferring benefits from the person with the cheap tastes to the one with the expensive tastes, which seems counterintuitive. However, it would be hasty to conclude that resources ought to be the currency of equality. For consider an alternative case in which person A has a taste for caviar not because he went out and cultivated this rather luxurious taste but because he was brought up as a fisherman’s son in the Caspian sea. Out fishing for sturgeon for their meager living, he occasionally tasted caviar as a child, ending up hooked for life. In this sort of case, our intuitions might be very different, and a transfer of resources to such a person ceases to seem so counterintuitive. If so, the problem is not with equalizing welfare per se, but rather with equalizing welfare deficits for which the person is herself responsible.
Interestingly, the lesson that responsibility ought to be part and parcel of egalitarianism has been arrived at by proponents of equality of resources as well. In Rawls’s account, recall, resources (or, more precisely, social primary goods) were to be distributed to the maximal benefit of the worst-off group. In his initial statement of the theory, however, Rawls never inquired after how precisely individuals came to occupy that worst-off position. Compare a laid-off worker with another person who is a beach bum. For Rawls, as long as they enjoy the same low level of resources, there is no morally relevant difference between the two. In that sense his particular version of equality of resources is not sensitive enough to the individuals’ varying levels of ambition. At the same time, we said, Rawls would deem distributions of resources that would benefit the worst-off as just, even if some individuals were more talented and handicap-free compared to others within the worst-off groups. This has the unhappy result that people who through no fault of their own are bad at converting resources to welfare (the naturally untalented and the genetically ill and handicapped) would end up with lower welfare compared to the naturally gifted. This demonstrates the problem of not taking responsibility into account in one’s theory of egalitarian justice. Dworkin has captured the intuition of many by finding Rawls’s account of equality of resources guilty of a double charge. It is too sensitive (reflects) natural inequalities (inequalities in talent and health), and at the same time completely insensitive to individuals’ varying ambitions (the surfers vs. untalented hardworking individuals). What was needed, in short, was an account of equality that is both ambition-sensitive and endowment-insensitive (Dworkin, 1981b).
An important upshot of the “equality of what” debate has therefore been the introduction of the idea of personal responsibility into egalitarianism. In G. A. Cohen’s memorable words, “Dworkin has, in effect, performed for egalitarianism the considerable service of incorporating within it the most powerful idea in the arsenal of the anti-egalitarian right: the idea of choice and responsibility” (Cohen, 1989, p. 933). Although Dworkin himself later came to reject the title, this was the birth of luck egalitarianism.
In fact, the idea that responsibility might be relevant to egalitarian justice didn’t begin with Dworkin. Rawls (1971, p. 102) before him had the intuition that individuals have no special claim to the fruits of talents with which they were merely born. Early on, then, egalitarians had the important intuition that the differential impact of morally arbitrary factors such as luck on advantage should be neutralized or mitigated in the pursuit of equality. However, Dworkin is the one who gave this “luckist turn” a comprehensive place in his account. His ideal of equality of resources employs the ideal of choice to the fullest extent. To do so, Dworkin draws his by-now famous distinction between brute and option luck. Brute luck is the kind of luck the triggering of which the agent should not be held responsible for, say being struck by lightning on an otherwise sunny day. In contrast, option luck refers to luck the triggering of which the agent could potentially be held responsible for, where the obvious example that comes to mind is gambling. Many things in life are a matter of luck, but armed with this distinction, Dworkin sets out to provide an account of equality that neutralizes luck. To do so he invites us to imagine an insurance market, where we could make all incidences of luck insurable. Under this insurance market, one is supposed to have the choice of whether or not to insure against any conceivable incidence of brute luck (including the remote chance of being struck by lightning). Of course to make this an instrument of justice, this insurance market cannot simply be a real-life one. That is so because in real life individuals already know certain health risks that they face (say, due to family history) and have a good acquaintance with the talents that they happened to be born with. To make this a thought experiment that would tell us something about justice, Dworkin invites us to think of a hypothetical insurance market. In that hypothetical market, much like Rawls’s veil of ignorance, individuals are asked to make prudent and rational decisions without the knowledge of how they themselves fare with regard to risks of ill health and absence of talents. It is the distinction between brute luck and option luck that enabled Dworkin to offer a radical account of equality through this mechanism of hypothetical insurance. By imagining all conceivable incidences of luck to be insurable, Dworkin has offered an egalitarian model that is ambition-sensitive and endowment-insensitive. It allows individuals to insure against having low natural endowments. Everything, including a person’s attitude toward different illnesses and handicaps, thus becomes a matter of choice.
Luck Egalitarianism Defined and Elaborated
Ever since Dworkin put forward his ideas concerning a responsibility-sensitive account of equality, luck egalitarianism has emerged as one of the leading theories of distributive equality, if not the leading one. Luck egalitarianism is most often stated in the following way:
Luck egalitarianism: It is bad or unjust for some to be worse-off than others through no fault or choice of their own.
To better understand this view, it would be useful to highlight some key features of this conception of equality.
First, in its broadest pronouncement, luck egalitarianism says something about when distributions are unjust or bad. In other words, the view can take a deontic form or a telic one. As an account of justice, it would object to all inequalities unless they are the direct consequence of choice. Similarly, as a telic view, it takes all inequalities to be bad unless they are the direct consequence of choice. (This statement is not entirely accurate, but it will suffice for now.) Importantly, the telic form of luck egalitarianism would see some disvalue in some faring worse than others through no fault or choice of their own even when there is nothing that can be done to narrow down those inequalities.
As a theory of distributive justice, however, it is important to notice that luck egalitarianism is properly understood as a partial or incomplete such theory (Lippert-Rasmussen, 2016, ch. 1). That is the second feature of luck egalitarianism worth highlighting. Let us explain. We often think of various states of affairs as undesirable, say, because they don’t condemn exploitation, or do not reward individuals for their just deserts, or because they do not guarantee some equal standing. Moreover, being concerned with relative disadvantages, it remains silent with regard to individuals’ absolute levels of welfare, as perhaps a sufficientarian account, or a Rawlsian concern with raising the absolute level of the worst-off would. It is important to notice that luck egalitarianism does not presume to say something about all these social ills. It restricts its focus to distributions of whatever currency is deemed to be of ultimate interest to individuals. Some may view this incompleteness as a shortcoming of the theory. Whatever we might think about that feature (we shall revisit it below when discussing relational egalitarianism), it is important to notice that it is a deliberate one, rather than a result of oversight on the part of luck egalitarians.
Here is a third feature worth emphasizing. Given its objection to arbitrary inequalities, luck egalitarianism is often thought to be heavily reliant on the complaint lodged by anyone who is worse-off than another. Some have objected to this reliance on comparisons between better-offs and worse-offs as manifesting an “ethics of envy.” On this understanding, luck egalitarianism is perceived as focusing on how individuals rank relative to others. That is, individuals compare themselves to other, luckier, individuals and lodge an envy-based complaint with some imaginary ministry of equality. For some people, this seems an unattractive way of cashing out a lofty ideal such as equality. Still, one thing to notice is that luck egalitarianism does not object to all manifestations of luck. It is a common misconception, for example, to attribute to luck egalitarianism the view that all differential luck is bad. But notice that the formula above objects not to people being better-off than others, but only to them faring worse. This might seem like a superficial difference, and in practice it might not amount to much. Yet, it belies a real difference of emphasis because it shows that the luck egalitarian concern does not lie with depriving individuals of their good luck, but rather in ensuring that no one fares worse due to bad luck.
The luck egalitarian credo, we said, is silent on good luck. Here is another theme it is silent about, namely equalities. This is the fourth aspect we want to highlight. Luck egalitarianism speaks of the badness of some faring worse than others. But the view does not mention the goodness or badness (or unfairness) of individuals faring equally compared to others. Some luck egalitarians, when pressed on the point, have committed themselves to the view that equalities, and not just inequalities, can also be unfair according to luck egalitarianism, predictably when they don’t reflect equivalent levels of prudence (Cohen, 2006; Temkin, 2015). Others have insisted that luck egalitarianism properly understood must keep silent on any potential injustice of equalities (Segall, 2012). One by-product of this is keeping luck egalitarianism distinct from the concept of desert.
Here is a fifth and final point. Luck egalitarianism is often understood as objecting to arbitrary inequalities, such as those brought about by luck. But there is some ambiguity here. Consider a case where two equally situated individuals place identical bets at blackjack. One wins the other loses and an inequality ensues. This inequality, observe, is owed to nothing but luck. Now, some luck egalitarians bite the bullet on cases such as these and insist that even such inequalities are unjust. This has earned them the title of all-luck egalitarianism, for they object to inequalities not only in brute luck but also in option luck (such as gambling). Yet many people, including many luck egalitarians, want to resist treating inequalities between gamblers as unjust or bad. What the counterintuitive example attests to, for them, is an ambiguity with regard to the clause “through no fault of their own” in the luck egalitarian credo. The gamblers’ case indicates that this “luckist” clause should be attached not to the fact of inequality itself, but rather to the intrapersonal shortfall that befalls the worse-off person. In other words, it is unfair for one to be worse-off compared to another when she is worse-off than she could have otherwise been, and through no fault or choice of her own. Luck egalitarianism in short incorporates both an interpersonal (worse-off than another) and an intrapersonal (worse-off than she could have otherwise been) dimensions.
Luck egalitarianism has emerged in recent years as the leading account of egalitarian justice. But it is certainly not uncontested. And, by far the most important criticism comes from what has become known as relational egalitarians. Relational egalitarianism is the idea that what matters for equality, indeed the “point of equality” is not so much an equal distribution of some currency. Rather, the point is to bring about a society characterized by equal relations amongst citizens.
Relational Egalitarianism: The point of equality is not so much an equal distribution of some currency but to end oppression, guarantee equal democratic capacities, and in general bring about a society marked by equal relations and status.
The egalitarian ideal, then, ought to concern eradicating differences of rank and class, and all manner of obstacles to social equality. In recent years much has been made of this ideal of “social” or “relational” (sometimes also dubbed “democratic”) equality, and of the fundamental difference between it and luck egalitarianism (Anderson, 2010; Fourie, Schuppert, & Wallimann-Helmer, 2015). In this final section we want to explore both the ideal of relational egalitarianism, and how it differs from luck egalitarianism. This will help us gauge how deep the divide is between these two accounts.
In charting the differences between luck and relational egalitarianism, we may notice seven major points of departure. Perhaps the most obvious difference is this. Luck egalitarianism, relational critics have observed, is obsessed with how each and every individual fares. There is, of course, a comparative element in luck egalitarianism, indeed too much so on the relational reading (recall “the politics of envy” charge). But the egalitarian endeavor of luck egalitarianism begins and ends with calculating the size of individuals’ bundles, be it of welfare or resources or what have you. Relational egalitarianism as its name denotes, has a diametrically opposite concern. Instead of inquiring after what bundles individuals hold, relational egalitarianism is concerned with the quality of the relation between individuals. Our distribuenda (or currency), as it were, is not those bundles held by individuals but, in a sense, the space occupied between them. And here of course the ideal is to have relations that are marked by equality and equal respect.
Now to say that the focus of relational egalitarianism is on relations is not to say that their ideal is oblivious to what material resources individuals get. This is so for the obvious reason that social planners are rather restricted in the way that they may shape social relations. The major, if not the only, way in which we may remodel social relations is by redistributing some material resources. Notice, even policies that facilitate social equality by increasing the self-respect of certain individuals (say, those whose social standing is low for historical reasons, say African-Americans, or professional groups whose social esteem is in need of boosting, like teachers) would require doing so by means of redistribution of material goods (e.g., Rawls’ famous social bases of self-respect).
A second, rather obvious dimension along which luck—and relational—egalitarians diverge is with respect to personal responsibility. Reverting back to Rawls’s choice-insensitivity, relational egalitarians advocate an egalitarian ideal that is unaffected by personal choice and responsibility. This fundamental difference between the two ideals—responsibility-sensitivity—also accounts for some of the fiercest critiques by relational egalitarianism on luck egalitarianism. This is of course the much-discussed “abandonment objection.” The objection is this. Luck egalitarians, we said, object only to inequalities and disadvantages for which the individual is not responsible. It follows that luck egalitarians would allegedly leave reckless drivers to die by the side of the road, for their disadvantage (being injured) is the result of their own fault or choice. Whether or not that latter charge is correct, it is interesting to note that it is not so obvious that relational egalitarians themselves avert this objection, nor, more generally, that their account is responsibility-free. Take, for example, the case of convicted criminals. Some relational egalitarians (e.g., Anderson, 1999, p. 319, n. 76) say that these individuals deserve lower priority in the allocation of resources. In other words, their claim for distributive shares is weaker than that of non-criminals. It follows that relational egalitarianism is also sensitive to considerations of personal responsibility. It is just that the role they assign it is different (or weaker) than that assigned to it by luck egalitarians.
Here is a third difference between the two egalitarian ideals. We have said that relational egalitarians, despite their emphasis on relations, are not uninterested in the distribution of material goods. Rather, we said, on their view, an equal distribution of those goods is not an end in itself but, if anything, a means to securing a society of equals. But while relational egalitarians are potentially concerned with the distribution of material, social goods, they deem natural inequalities as not in themselves unjust. That is not to say that relational egalitarians care not whether someone was born with some natural handicap (say blindness). Far from it. But relational egalitarians do hold that natural inequalities, such as between the healthy and the handicapped, or the talented and untalented, are not unjust in themselves. It follows that the redistribution of those natural resources is not part of the relational egalitarian agenda. This is yet another feature that sets them apart from luck egalitarianism. As we said earlier, Dworkin’s luck egalitarian ideal of equality of resources famously utilizes the mechanism of hypothetical insurance in order to redistribute not only external, but also internal resources such as one’s health and native talents. That has drawn the criticism and indeed ire of relational egalitarianism. The endeavor seems to them not only futile, but also disrespectful. It is disrespectful in the sense that to qualify for some distributive share one has to present oneself as somehow naturally deficient compared to others. The plight of the deaf, to take an obvious example, is not a social one for luck egalitarians, but a natural one, so goes the accusation. According to democratic egalitarians, that is deeply disrespectful. Luck egalitarians simply miss the point of what is bad about the predicament of the deaf, namely the social fact of living in a society designed by and for the hearing.
It is not obvious, however, that taking on board natural inequalities is either disrespectful or problematic. There are at least two things to say here. First, it is useful to recall that, on any plausible egalitarian theory and certainly on luck egalitarianism, benefits are never foisted on individuals against their will. If a deaf person does not consider her hearing condition to constitute a disadvantage, luck egalitarians are certainly not going to force some benefit on her. Still, the relational egalitarian might say that even if those naturally handicapped do not lodge a complaint, the very thought of their plight as calling for some compensation is disrespectful. Consider then a second observation. Suppose the deaf assert, as they should, their right to live in equality and dignity with the hearing. Still, it is not obvious that luck egalitarianism would tell us to first opt for an individual compensation as the first order of duty. Contrary to widespread perception, there is nothing in luck egalitarianism to prevent the view from first opting for some social, structural reform (say, providing subtitles and sign-language translators in the case of the deaf). Luck egalitarians are concerned with eradicating disadvantages. We have seen nothing in that theory to force them to pursue this goal through an individualized as opposed to social-structural means. To see this, think of gender-based discrimination. Luck egalitarianism would treat such incidence as an obvious case of injustice (disadvantage through no choice of one’s own). But there is nothing in luck egalitarianism to favor monetary compensation to all such affected women as opposed to a comprehensive ban on the practice. Quite the contrary, in fact. Consequently, there is nothing inherently disrespectful in the luck egalitarian registering of natural inequalities alongside social ones.
A fourth point follows, and it follows in particular from relational egalitarianism’s complex relations with distributions of material goods. We already mentioned that relational egalitarianism cannot ignore those distributions, for the simple reason that it is largely through redistribution of these material resources that we may control and shape social relations. But relational egalitarianism, or at least some of its proponents, has something to say, in fact, about material distributions that go beyond shaping social relations. And the view they take with regard to these distributions is that to be just, the pattern of material distributions must meet some threshold of what individuals truly need. Beyond these thresholds, the pattern of distribution does not matter, morally speaking. In other words, beyond equalizing social relations, relational egalitarians are in fact sufficientarians rather than egalitarians. (Recall our discussion of sufficiency and its weaknesses in Section 2.3)
This fourth point hints at a fifth one. We just saw that relational egalitarians find room within their supposedly egalitarian theory for a sufficientarian element of distributing material goods. This is consistent with their general approach to justice. If we have characterized luck egalitarians’ theory of justice as partial or restricted (and concerned with fairness, namely the rightness of wrongness of the relative distribution of goods, rather than justice more comprehensively), relational egalitarianism’s theory of justice is, in contrast, a comprehensive one. That is, following Rawls these theorists hold that a theory of justice should incorporate all sorts of intuitions we might have regarding desirable social institutions. That is, not just intuitions about comparative holdings but also about their absolute levels, concerns for incentives for work and production, concern for the requirement of publicity, and so forth. Contrast this with what we have already termed as the “leaner” luck egalitarian theory of fairness, with its concern to supply only a partial account of fairness, rather than what Cohen called “desirable rules of regulation” (Cohen, 2008).
This point of divergence may actually explain away much of the (remaining) disagreement between relational egalitarianism and luck egalitarianism. For, if the aspirations of the two theories are to begin with quite different, then perhaps it is no wonder that they reach such radically different accounts. Rawls, who is often considered the spiritual father of relational egalitarianism, writes on the very first page of A Theory of Justice, that justice is the first virtue of institutions (Rawls, 1971, p. 3). In a sense, luck egalitarians part company with him already at this early juncture. Justice, and fairness in particular, is merely one virtue of institutions. There are of course other virtues of institutions such as efficiency, publicity, freedom, and so on. When we analyze the comprehensive theories of justice, we need to take all these virtues on board and trade them off against one another.
Here is a six point of departure between the two theories. Relational egalitarians often maintain that their theory is not only comprehensive but also “political” (Waldron, 2013). This contrasts with luck egalitarianism, which is sometimes characterized as “cosmic” (Scheffler, 2003). This difference partly follows from the point raised earlier about luck egalitarians being concerned not just with social inequalities but also natural ones. It is this concern with “correcting nature” that have earned luck egalitarianism the nickname of being “cosmological” in nature, in contrast to the more solidly political conception of egalitarian justice offered by relational egalitarianism. Of course, even if this is true, to say of luck egalitarianism that it is “cosmic” (which perhaps it is) is not to deny that it is also political. It is also political in that it has something, quite a lot in fact, to say about what is bad or unfair with political communities in which inequalities in holdings do not reflect differences in choice and effort. The label “cosmic,” when used to distance relational egalitarianism from luck egalitarianism, does indeed serve to draw a difference between them, but it is less obvious that it works as a criticism.
A seventh and final difference follows from what has just been said. Because relational egalitarianism is committed to providing a political conception of equality this allows it to confine itself to quite a narrow scope. Being political, the scope for egalitarianism that it envisions is confined to political institutions, or in other words to domestic affairs. Global affairs, while not necessarily lying in a moral vacuum on the relational egalitarianism reading, are nevertheless not part of the political, and so are beyond the scope of egalitarian justice. This has some potentially interesting implications for some of the earlier points we have raised. Consider the abovementioned abandonment objection (perhaps the chief criticism in the relational egalitarianism arsenal against luck egalitarianism), say, as manifested in not offering treatment to a reckless driver. If that reckless driver happens to be a foreigner, say a guest worker, there is nothing in the relational egalitarianism principle to guarantee her medical treatment. Such treatment may be extended as a matter of courtesy or some Good Samaritan’s mercy, but it is not, strictly speaking, a matter of egalitarian justice on this view. Indeed, there is nothing on relational egalitarianism to guarantee this alien a treatment even if she was not reckless at all. In that latter case, we now see, luck egalitarianism, being global in reach, would of course extend treatment as a matter of compensation. At least on some occasions, we see, relational egalitarianism, due to its narrower political scope, can in fact be “harsher” compared to luck egalitarianism.
Equality is a key concept in political philosophy, and the more we delve into it the more complex and multilayered it reveals itself to be. Still, the past half a century has seen an explosion of writing on equality. Whatever one’s position is, this literature has certainly advanced our knowledge of what equality is, what value it might have, and how might it be pursued as an ideal of justice.
Anderson, E. (1999). What is the point of equality? Ethics, 109, 287–337.Find this resource:
Anderson, E. (2010). The fundamental disagreement between luck egalitarians and relational egalitarians. Canadian Journal of Philosophy: Supplementary Volume, 36, 1–23.Find this resource:
Arneson, R. J. (1989). Equality and equal opportunity for welfare. Philosophical Studies, 56, 77–93.Find this resource:
Arneson, R. J. (2002). Why justice requires transfers to offset income and wealth inequalities. Social Philosophy and Policy, 19, 172–200.Find this resource:
Broome, J. (2002). Respects and levelling down. Available online at http://users.ox.ac.uk/~sfop0060/pdf/respects%20and%20levelling%20down.pdf.
Cohen, G. A. (1989). On the currency of egalitarian justice. Ethics, 99, 906–944.Find this resource:
Cohen, G. A. (2006). Luck and equality: A reply to Hurley. Philosophy and Phenomenological Research, 72, 439–446.Find this resource:
Cohen, G. A. (2008). Rescuing justice and equality. Cambridge, MA: Harvard University Press.Find this resource:
Crisp, R. (2003). Equality, priority, and compassion, Ethics, 113, 745–763.Find this resource:
Dworkin, R. (1981a). What is equality? Part 1: Equality of welfare. Philosophy and Public Affairs, 10, 185–246.Find this resource:
Dworkin, R. (1981b). What is equality? Part 2: Equality of resources. Philosophy and Public Affairs, 10, 283–345.Find this resource:
Dworkin, R. (2000). Sovereign virtue. Cambridge, MA: Harvard University Press.Find this resource:
Fleurbaey, M. (2015). Equality versus priority: How relevant is the distinction? Economics and Philosophy, 31, 203–217.Find this resource:
Fourie, C., Schuppert, F., and Wallimann-Helmer, I. (Eds.) (2015). Social equality: On what it means to be equals. Oxford: Oxford University Press.Find this resource:
Griffin, J. (1986). Well-being: Its meaning, measurement, and moral importance. Oxford: Oxford University Press.Find this resource:
Hausman, D. (2015). Equality versus priority: A misleading distinction. Economics and Philosophy, 31, 229–238.Find this resource:
Hirose, I. (2015). Egalitarianism. London: Routledge.Find this resource:
Hirose, I. (2016). Axiological sufficientarianism. In C. Fourie & A. Rid (Eds.), How much is enough? Sufficiency and thresholds in health care. Oxford: Oxford University Press.Find this resource:
Holtug, Nils (2010). Persons, interests, and justice. Oxford: Oxford University Press.Find this resource:
Jensen, Karsten Klint (2003). What is the difference between (moderate) egalitarianism and prioritarianism? Economics and Philosophy, 19, 89–109.Find this resource:
Lippert-Rasmussen, Kasper (2016). Luck egalitarianism. London: Bloomsbury.Find this resource:
Marshall, T. H. (1950). Citizenship and social class. Cambridge, U.K.: Cambridge University Press.Find this resource:
Mason, A. (2001). Egalitarianism and the levelling down objection. Analysis, 61, 2456–2254.Find this resource:
McKerlie, D. (1989). Equality and time. Ethics, 99, 475–491.Find this resource:
Nagel, T. (1991). Equality and impartiality. New York: Oxford University Press.Find this resource:
Nozick, R. (1974). Anarchy, state, and utopia. Oxford: Basic Books.Find this resource:
Otsuka, M., & Alex V. (2009). Why it matters that some are worse off than others: An argument against the priority view. Philosophy and Public Affairs, 37, 171–199.Find this resource:
Parfit, D. (1984). Reasons and persons. Oxford: Clarendon.Find this resource:
Parfit, D. (2000). Equality or priority? In M. Clayton & A. Williams (Eds.), The ideal of equality (pp. 81–125). Basingstoke: Macmillan.Find this resource:
Rawls, J. (1971). A theory of justice. Cambridge, MA: Harvard University Press.Find this resource:
Scheffler, S. (2003). What is egalitarianism? Philosophy and Public Affairs, 31, 5–39.Find this resource:
Segall, S. (2012). Why egalitarians should not care about equality. Ethical Theory & Moral Practice, 15, 507–519.Find this resource:
Segall, S. (2015). In Defence of priority (and equality). Politics, Philosophy, & Economics, 14, 343–364.Find this resource:
Sen, A. K. (1985). Commodities and capabilities. Amsterdam: North-Holland.Find this resource:
Tawney, R. H. (1964 ). Equality. London: Allen & Unwin.Find this resource:
Temkin, L. S. (1993). Inequality. New York: Oxford University Press.Find this resource:
Temkin, L. S. (2000). Equality, priority, and the levelling down objection. In M. Clayton & A. Williams (Eds.), The ideal of equality (pp. 126–161). Basingstoke, U.K.: Macmillan.Find this resource:
Temkin, L. S. (2015). Equality as comparative fairness. The Journal of Applied Philosophy, online first.Find this resource:
Waldron, J. (2013). Political political theory: An inaugural lecture. The Journal of Political Philosophy, 21, 1–23.Find this resource: