Next to his just-war theory, Walzer is most famous for arguing, in Spheres of Justice and related works,1 for a social-democratic approach to distributive justice that he calls “complex equality” (Walzer 1983: 3–30). This argument is of the utmost importance for understanding Walzer’s contribution, as it showcases the distinctiveness of both his situated approach to political theory and his vision of an egalitarian society. Put briefly, Walzer uses an interpretive method, focused on social meaning as the foundation of a theory of justice, to argue for a political conception of equality that overcomes relationships of domination and oppression. Complex equality is the subject of the next three chapters. In this one, I focus on the philosophical underpinnings of Walzer’s method. The next chapter focuses on the political vision of Spheres and Walzer’s conception of social democracy. Chapter 5 considers the application of complex equality to a society marked by cultural diversity and contested traditions of meaning.
In Walzer’s theory, goods must be distributed in accordance with their socially specific meaning (Walzer 1983: 6–8). As goods have different meanings from each other, this requires separation between different spheres of social life such that possession of one good does not automatically bring possession of other goods. For example, holders of political power should not be able to guarantee their children access to higher-quality education. In Walzer’s favorite example, people will receive healthcare if they need it – that is, if they are sick – not because they are wealthy (88–91, Walzer 1986c, 1994a: 28–31). This challenges Rawls’s “justice as fairness” – then, as now, central to much political theorizing about justice – both in suggesting that the attempt to develop principles of distribution must incorporate a greater diversity of distributive principles and by arguing that equality is not primarily about distribution of goods but about decision-making power.2 Walzer argues that his approach will prevent people dominating others by virtue of their possession of particular goods that are dominant in the society in question, most likely either money or political power (Walzer 1983: 10–13). It will thus lead to a type of equality in which, because “[d]ifferent goods … are distributed to different people for different reasons” (Walzer 1973a: 242, 1993b: 86–90), nobody oppresses or dominates others. As it relies on a diversity of principles, Walzer’s argument is also pluralistic, so the subtitle of Spheres is A Defense of Pluralism and Equality.
Walzer’s argument also requires pluralism across societies, or what he calls “radical particularism” (Walzer 1983: xiv), because he insists that goods have social or cultural meanings, rather than natural or universal ones. This is because the creation of goods is, according to Walzer, a cultural process that turns otherwise indeterminate objects into goods that we have reason to value (6–8; 1993a; for discussion, Reiner 2016: 369–73). This means that distributing goods in accordance with their meaning will lead to different distributive principles in different societies. Even food can have multiple meanings: bread can be “the staff of life, the body of Christ, the symbol of the Sabbath, the means of hospitality” (Walzer 1983: 8). Walzer thus argues that the first task of a theorist of justice is to interpret the social meaning of goods. I call this the social-meaning thesis. As I argue in Chapter 8, it is the analogue of Walzer’s commitment to the common life in just-war theory.
Walzer’s theory thus produces a two-part method for theorizing about justice (Reiner 2016): theorists must first interpret the meaning of the good in question, and only then make an argument for distributive principles that accord with that meaning. Both aspects of this method showcase Walzer’s important contributions to theorizing about justice. First, he provides an important challenge to universalist theories that seek to deduce a set of distributive principles that are valid everywhere by asking how that is possible when the goods available for distribution are spatially and temporally particular. Second, he insists that egalitarian theories must understand equality as a political relationship that avoids domination and oppression, and so that they must not stipulate principles of justice too finely so as to leave more room for democratic decision-making.
I start by situating Spheres in the context of the lively debate in political theory that Rawls’s A Theory of Justice (1971) spurred. I then turn to Walzer’s project, explicating the twin ideas that are at its heart: the social-meaning thesis and the notion of distribution for right reasons (the meaning-dependence thesis). I explain why they are integral to Walzer’s social-democratic equality. I illustrate with discussion of his healthcare example, before concluding with an account of Walzer’s contribution.
There was a significant difference in intellectual context between Wars and Spheres. When Walzer wrote about just wars, there had been little recent work on the subject by philosophers. That was why he had to “recapture the just war for political and moral theory” (Walzer 2015a: xxvi). When Walzer was at graduate school in the 1950s, this was true more generally: most research on political theory was historical. Walzer’s own PhD thesis (Walzer 1965) was a study of Puritan attempts to produce a holy commonwealth in the seventeenth century. His adviser, Samuel Beer, encouraged Walzer to make that history comparative (for Beer’s own comparative history, see Beer 1965, 1970), so in some of his early academic work, Walzer compared different revolutionary moments (see especially Walzer 1970c, 1974). However, Walzer’s intellectual interests always tended toward the normative. One reason he found his early work for Dissent so rewarding was that it enabled him to make the sort of engaged, contemporary argument that he could not make in his academic work (on this, see Walzer 2013a, 2013b, and discussion in Isserman 1987: 88–90, 115–16, Sorin 2002: 96–9). For a variety of reasons,3 little work in political theory in the 1950s and 1960s engaged with vexed moral questions of contemporary concern. Indeed, in 1956, Peter Laslett declared that, “For the moment, anyway, political philosophy is dead” (cited in Laslett and Runciman 1962: vii).
However, starting in the 1960s, a group of political theorists set up the monthly discussion group SELF (the Society for Ethical and Legal Philosophy), whose members used the techniques of analytic philosophy to develop normative theories (Walzer 2007: 307–8).4 Shortly after Walzer took up his first academic position at Princeton in 1962, he joined SELF, which provided him with his “philosophical education” (307). Members of SELF included the main figures in the next generation of political theorists: Robert Nozick, Judith Thomson, Ronald Dworkin, Thomas Nagel, and John Rawls. Rawls is especially important because A Theory of Justice (1971, but circulating in manuscript form for most of the previous decade) was the major statement of SELF’s approach and prompted a new wave of theorizing about distributive justice.5
Rawls developed an updated notion of the social contract based on the idea that justice requires fairness or impartiality. To imagine what this might mean, he deploys a hypothetical device that he calls the “original position,” in which participants are stripped of knowledge of particulars about themselves that differentiate them from fellow citizens by a “veil of ignorance” (Rawls 1971: 17–22, 118–92). The original position is a fair form of the state of nature of Locke, Rousseau, and Kant. Reasoning as if one were in the original position will ensure fair outcomes because nobody will be able to tailor suggested principles of justice to benefit themselves: after all, participants do not know who they are. This means that principles have to be fair to everyone and consider the points of view of all citizens (see Okin 1989: 101–9). Rawls argues that the resulting conception of justice will not be utilitarian, because utilitarianism allows some to be sacrificed to the general welfare, but egalitarian, as participants will choose an equal distribution of resources unless inequality benefits everyone (Rawls 1971: 60–5). Rawls argues for two principles of justice, which in their final form read:
(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 … for all; and
(b) Social and economic inequalities are … to be attached to offices and positions open to all under conditions of fair equality of opportunity; and … to the greatest benefit of the least-advantaged members of society.
(Rawls 2001: 42–3; see also the first rendering in Rawls 1971: 60–1, with the revised second principle at 83)
Rawls thus reconciles liberalism with equality. While the revised version of the second principle does permit some inequality, Rawls insists that there will be a “tendency toward equality” (Rawls 1971: 100–8). One critic suggests that Rawls “goes farther in the direction of equality than even Marx proposed” (Schaar 1981: 157).
In 1974, Walzer’s close friend Robert Nozick wrote an influential libertarian critique of Rawls, holding that egalitarian theories do not consider people’s actions in the production of goods. Nozick argues that justice requires considering how people stand in relation to particular goods. He holds that this requires what he calls an “historical” theory against the “patterned” principles that, he claimed, characterize egalitarianism (Nozick 1974: 153–60), and that producing goods creates legitimate entitlements to them. Nozick rejected egalitarian theories on the grounds that, to maintain a distribution of goods in accordance with a pattern mandated by egalitarian principles, governments would constantly have to interfere with citizens’ freedom. For example, it would have to prevent them from paying talented artists to perform for them and prevent the stars from earning a living in this way (Nozick 1974: 149–50, 161–4). If Wilt Chamberlain – the prominent basketballer of the era – were to receive money from fellow citizens in order to play basketball for their entertainment, the government would have to step in immediately to return the money to the other citizens. If it did not do this, the distributive pattern mandated by the egalitarian principles would no longer hold. This means that equality is incompatible with a free society, and that egalitarian theories cannot account for individual merit,6 or indeed personal differences of any kind.
The political context is also important. Rawls had the ascendancy within political philosophy. From the 1970s onwards, liberal theory took a marked egalitarian turn,7 so much so that Ronald Dworkin and Will Kymlicka have argued that no political theory can be taken seriously unless it is in some sense a cashing out of the idea of equality (Dworkin 1977: 179–83, Dworkin 1983b, Kymlicka 2002: 4–5). By contrast, the 1970s was also the era in which the post-war consensus, when conservatives accepted the welfare state, began to disintegrate, with neoconservatives arguing increasingly vociferously for freer markets and the inevitability of inequality (Kristol 1972). In the USA, this movement reached its apex when Ronald Reagan won the Presidential election of 1980. For Walzer and Dissent, rumblings had begun long before (Walzer 1980b: 78–91), thanks to neoconservatives arguing with increasing vigor in the early 1970s that the social-democratic project of ensuring greater equality was doomed to failure because some people are innately more gifted than others. Of particular importance to Dissent was Philip Green’s critique of neoconservatism for latching onto the arguments of some social psychologists that IQ is heritable and that there are statistically significant differences in IQ across racial groups (Green 1976a, 1976b, Coser and Howe 1974). Walzer’s first statement of his theory is a response to the prominent neoconservative Irving Kristol – a former colleague of Walzer’s mentor Irving Howe from their undergraduate days at City College but by the 1970s prominent in Dissent’s right-wing rival Commentary (for discussion, Dorman 2000: 41–56) – that inequality is inevitable because the distribution of talents is unequal (Walzer 1973a). Like Nozick, Kristol argues that this means that attempts to redistribute income will both cause substantial coercion and inevitably fail to produce an equal society.
While Walzer is broadly sympathetic to Rawls’s egalitarian conclusions, he notes at the start of Spheres that he “mostly disagrees” with Rawls’s approach, and in particular with its foundation in analytic philosophy and appeal to the universalizing social-scientific disciplines of economics and psychology (Walzer 1983: xviii). Walzer believes that Rawls appeals to the wrong sort of legitimating foundation in invoking the original position, because appeal to fairness misses the reasons for egalitarian concern: domination and subordination (xiii). Walzer’s objection to the original position is that our endowments are part of our identity and so should and must have consequences in the social world (Walzer 2007: 304). That is why Rawls’s theory did not win the political battle (308). Walzer was a good friend of Nozick (they acknowledge each other at Nozick 1974: xv and Walzer 2015a: xxx), and engaged with his arguments in their co-taught Harvard class in 1970–1971 (Walzer 1983: xvii). Walzer accepts Nozick’s critique that Rawls minimizes the importance of action and history. Moreover, Walzer suggests, egalitarian liberalism misunderstands and depoliticizes the distributive process, describing it “as if it had this form: People distribute goods to other people,” when in fact goods cannot be defined in abstraction and “don’t just appear in the hands of distributive agents” (6–7). However, Walzer denies that all egalitarian theories are susceptible to the critique of inequality advanced by Nozick and the neoconservatives. In Spheres, Walzer attempts to show how considering the creation of goods helps illuminate an egalitarian, social-democratic theory that evades Nozick’s criticisms.
Walzer’s task in Spheres, then, is to advance a theory of equality that recognizes that people’s standing in relation to particular goods is relevant to the distribution of those goods. Complex equality argues that Nozick’s individualism misses “the social character of production” (Walzer 1983: 323, n. 2). Considering people’s roles in creating goods will lead to a society of equals (Miller 1995a, Gutmann 1995, Van der Veen 1999), divided into different “companies,” each of which enjoys preeminence in its own domain (Walzer 1973a: 243). For example, those worthy of public honors for particular accomplishments will receive them (Walzer 1983: 259–68), but will not for that reason also gain greater access to divine grace (243–8). Walzer understands this as allowing room for political and social action without leading to an unequal society. In defending equality, Walzer argues that there is no reason to believe that all talents are distributed along the same bell curve. In the different companies, different talents are nurtured and different relationships created. Thus, while inequality will indeed exist within each sphere and some may end up with very little of the good in question in that sphere, these inequalities will not mount up across the spheres, because “there is no single talent or combination of talents that necessarily must win the available goods of a free society” (Walzer 1973a: 241; see also Walzer 1993b, 1995a). Charm, intelligence, strength, entrepreneurial talent, and so on, are different skills: while each is indeed unequally possessed, they are so in different ways, because in the course of social life, we necessarily develop particular relationships with particular goods and groups of people. If this is so, then distributing goods in accordance with their meaning will be egalitarian because it requires that “we pay equal attention to the ‘different qualities,’ and to the ‘individuality’ of every” citizen (Walzer 1973a: 245; for critique, Arneson 1995, Barry 1984, 1995, Bounds 1994, Den Hartogh 1999, Kateb 2014: 196–7).
Walzer’s argument in Spheres rests on two major theses that produce a two-part method for theorizing about justice (Reiner 2016: 364; for discussion, Scanlon 2014: 183–4):
1. The meaning-dependence thesis: following the British social democrat Bernard Williams (1967),8 Walzer argues that the relevant distributive principle for any good must depend on the meaning of that good. Thus, love should be returned for love, political power to those able to persuade people to vote for them, etc.
2. The social-meaning thesis: following the anthropologist Clifford Geertz (1973), Walzer argues that goods are not just out there in the world, waiting to be distributed, but must first be conceived of and created by human beings. Thus, rather than taking goods to be natural, Walzer argues that, “All the goods with which distributive justice is concerned are social goods” (Walzer 1983: 7).
The first thesis is the egalitarian one: Walzer argues that complex equality will arise when goods are distributed for relevant reasons. The second is the pluralist one: discerning what the relevant reasons are requires cultural inquiry into the meaning of goods. On Walzer’s account, theories of distributive justice: (1) properly start with cultural inquiry into meaning, and (2) must then construct a case for distributive principles that accord with social meanings. That case will, always, be both socially situated and politically contestable. It is crucial to Walzer’s theory that political knowledge is both “particular” and “limited” (Walzer 1981: 14–15) and that “philosophers have no special rights in the political community” (19, Walzer 1984: 65–6). Rather than seeking to stipulate distributive principles, philosophers should seek to provide persuasive interventions in public debates and help guide citizens and politicians through their cultural morass of meanings (Walzer 1980d: 175). If distributive principles must depend on meaning, then they must not only be culturally but also temporally specific, rather than universal and timeless (Walzer 1981).9
This makes Walzer’s brand of equality differ markedly from the liberal variant Rawls espouses in ways that, he thinks, evade both the neoconservative critique of Kristol and the libertarian one of Nozick. By pointing to the numerous human talents, Walzer seeks to allay the claim that human beings are inevitably unequal by invoking a “democratic wager” that, in a diverse society, the diversity of skills will help avoid the emergence of a class of “flatly incompetent people” who lose out in every sphere (Walzer 1993b: 92). By allowing for shifts in meaning and hence of distribution, he seeks to avoid excessive reliance on the state (although see Walzer 1995a), and holds that, rather than being incompatible, “liberty and equality … stand best when they stand together” (Walzer 1973a: 256). Walzer does not define equality as involving equal possession of the same set of goods. Indeed, he denies the notion of “primary goods” that is at the heart of Rawls’s theory (Rawls 1971: 54–5). Primary goods are those things that all people want, regardless of what else they want; in other words, “general-purpose means for the pursuit of one’s comprehensive goals” (Sen 2009: 64, Walzer 1983: 8). On Walzer’s account, there are no goods that everyone wants, so it is a mistake to measure equality on such a metric. Rather, we achieve equality by abolishing relationships of domination so that there is “no more bowing and scraping, fawning and toadying … no more masters, no more slaves” (Walzer 1983: xii–xiii). Walzerian equality is close to equal citizenship (Downing and Thigpen 1986, Gill 1987, Miller 1995b, Van der Veen 1999, Den Hartogh 1999; for critique, Trappenburg 2000, Keat 1997). However, even citizenship is on Walzer’s account a cultural good whose meaning varies over time and space (Walzer 1989), while even a looser term such as equality of status is not an “objective condition” (Walzer 1995a: 283–4). For example, Walzer insists that, in an equal society, some will be engaged citizens, while others will be kibitzers who mostly disavow the political process (Walzer 1968b; see also Walzer 1970a: 203–28). The best we can say is that equality is a political relationship that overcomes oppression, subordination, and exclusion, and replaces them with inclusion, democracy, and spherical separation.
Walzer’s theory, then, allows for inequality within spheres; indeed, it requires such inequality. It is even possible that, in a particular sphere, some would be left with very little of the goods in question. The lack of an objective metric by virtue of which citizens would be one another’s equals leads some critics to deny that Walzer’s theory is egalitarian (see Arneson 1995, Rosenblum 1984, Kateb 2014). Why, then, does Walzer hold that distributing goods in accordance with their meaning would lead to complex equality?
Walzer borrows the meaning-dependence thesis from Bernard Williams, who introduces it as a way of achieving equality in unequal circumstances. Williams suggests that differential treatment does not interfere with equality, provided that reasons are given for the difference that are both relevant and socially operative (Williams 1967: 120–5). Walzer turns this argument into the claim that, “If we understand what [a good] is, what it means to those for whom it is a good, we understand how, by whom, and for what reasons it ought to be distributed,” with each sphere governed by different or “autonomous” reasons (Walzer 1983: 9–10, propositions 4 through 6 of the theory of goods). Reasons that might be entirely appropriate in one sphere will be inappropriate in others, and this means that inequalities in each sphere need not worry us, so long as they do not carry over into others. For example, while love is in contemporary society the appropriate reason for having a relationship with someone – so much so that we often think that love given for other reasons is not love at all – it is inappropriate in public office, where it amounts to nepotism (Walzer 1983: 234–9, 147–8). The plurality of social meanings ensures that inequalities will not mount up if distributions are for relevant reasons. Walzer concludes that trying to ensure equality in each sphere is a false or “simple” equality (13–17), and that the egalitarian task is not to prevent people from monopolizing certain goods but rather to prevent those goods from becoming socially dominant (10–13).
As equality is a political relationship, it does not mean possessing identical amounts of things and does not require that goods be equally distributed (or that philosophers stipulate how goods be distributed according to a pattern). Rather, it means preventing people’s legitimate claims to certain goods from enabling them to possess others. Walzer calls this threat to equality “dominance”: it occurs when “the individuals who have it, because they have it” can get other goods for which they do not stand in the appropriate relationship (10). This is tyrannical for it enables the holders of the good to “invade the sphere where another company of men and women properly rules” (19). There is, for example, nothing wrong with some people having political power. Indeed, Walzer starts Spheres by noting that the “simple” egalitarian belief that allocating political power unequally makes equality chimerical is the false claim that motivated his writing (xi). What does make for inequality, Walzer holds, is when apparatchiks use their political position to gain access to money, healthcare, recognition, free time, love, or any social good that is not within the sphere of political power. When that happens, political leaders are the masters, not the servants, of the people. So long as office holders gain no advantage in other spheres, citizens are not “unequal generally,” just within the sphere of politics (19). Thus, Walzer reformulates the doctrine of right reasons in negative terms, as what we might call a doctrine of not-wrong reasons: “No social good x should be distributed to men and women who possess some other good y merely because they possess y and without regard to the meaning of x” (19–20). This doctrine restrains the sort of distributive argument that should be made in such a way that it impedes dominance. In particular, it prevents the politically powerful and the wealthy from using their power or wealth as justifications for gaining goods that belong to spheres in which other reasons are operative.
On Walzer’s argument, we prevent either market or state becoming dominant by “blocking” certain exchanges or uses of power (Walzer 1983: 97–103, 281–4; for discussion, Andre 1995) on the grounds that they involve things that should not be for sale or should not be obtained politically. For example, Walzer notes, the contemporary United States prevents the purchase of human beings, political power or influence, criminal justice, exemptions from military service, love or friendship, and criminal objects. The point of blocking these exchanges is to prevent the conversion of every available good into a commodity (Andre 1995: 172).10 As a social democrat, Walzer seeks to resist market imperialism, a type of universalism in which everything is for sale that, by converting citizens into consumers, makes us all the servants of capital (Walzer 1984; for discussion, Bevir and Reiner 2012: 194–200). At the same time, Walzer, as a Dissentnik wary of state power after the experience of socialism in the twentieth century, insists that the state is not a panacea to the problem of inequality but potentially the site of an equally deadly form of power. Walzer suggests that political power serves to sustain the community (Walzer 2012a), so the blocked uses of power prevent those uses of it that would be of merely personal benefit. For example, he notes, state officials may not use their power to corrupt criminal justice, confiscate property, control religious or family life, or stipulate the meaning of social goods (Walzer 1983: 282–4). When they violate this list, officials fail to act as holders of political power. They do not really exercise political power at all, because the meaning of that is communal life. Corrupt officials threaten equality by subordinating “all the companies of men and women to the one company that possesses or exercises state power” (284). One important way in which Walzer tries to limit government is by localizing the provision of welfare services and using voluntary, often religious, organizations wherever possible (Walzer 1986d, 1988d; see discussion of the “welfare society” in Chapter 4). This will make for a more participatory democracy and thus enhance the agency of citizens.
It is important to note that, on Walzer’s account, neither market logic nor political power threatens equality once suitably constrained. He even makes the striking argument that the distribution of commodities is not a matter of justice, holding that “we can’t redistribute commodities directly, not if we are to allow individual men and women to choose for themselves the things they find useful or pleasing” (Walzer 1983: 106–7). Walzer implies that commodities are those things that may be useful or pleasing to us privately but are not necessary for effective membership in the community. The list of blocked exchanges prevents the undermining of communal provision. By contrast, commodities differentiate us from other people because of our different (albeit culturally inflected) views about what is pleasant. Commodities can be exchanged freely, because free exchange is the appropriate principle for those goods whose value varies according to taste. In other words, each particular commodity – Walzer’s list includes hi-fis, yachts, and rugs – has merely “use value and individualized symbolic value” (108). While there should, Walzer holds, be some redistribution of money, it is more important to strengthen “autonomous distributions” in other spheres (107, but cf. Walzer 1973a: 248, which accepts that until the convertibility of money is diminished, it should be “more or less equally held”). In a world of complex equality, we would obtain relatively few goods using money. In Walzer’s account, the problem of free exchange is that it is used with regard to goods that should not be seen as commodities. We tame the power of money by blocking its exchange, not by ensuring its equal possession because, as Nozick pointed out, equal possession of money is a condition that cannot last and that ignores people’s relationship with particular goods (Nozick 1974: 61–4).
Walzer’s favorite example of distribution for right reasons is healthcare. Like Williams, he holds that “the proper ground of distribution of medical care is ill health” (Williams 1967: 121; Walzer 1983: 86–91). Here, however, Walzer also parts company with Williams, which is an important example of Walzer’s pluralism leading him to disagree with analytic philosophers with whom he is otherwise politically sympathetic. For while Williams insists that distributing healthcare on grounds of ill health is a “necessary truth,” (Williams 1967: 121), Walzer suggests that it is socially contingent on the meaning of healthcare. That meaning varies across societies because, like every good, healthcare has a social, not an “essential” meaning (Walzer 1983: 9; for discussion, Reiner 2016).
The source of this thesis is Clifford Geertz’s anthropology. Geertz was Walzer’s colleague at the Institute for Advanced Study from 1980, which explains why the social-meaning thesis was added to Walzer’s theory in between “In Defense of Equality” (1973) and Spheres (1983).11 Geertz bequeathed to Walzer the claim that meaning does not inhere in nature but is a collective human product. Geertz argues that, because recent developments in neurobiology suggest that human behavioral patterns are not dictated by our physical structure, it would be impossible to act meaningfully unless directed by “culture patterns – organized systems of significant symbols” (Geertz 1973: 45–6, 75–6). This makes us “unfinished animals who complete … ourselves through culture” (49). Human behavior is remarkably plastic because our understanding of any situation is susceptible to a great variety of interpretations. The task of culture is to direct us toward particular paths of action by endowing the world with meaning. This means that we are “suspended in webs of significance” spun by our cultures, which makes the analysis of culture as an interpretive quest for meaning crucial to social understanding (5). As there is no reliable source of meaning outside these cultural webs, there is no way to understand or evaluate them from a neutral standpoint: they must be understood on their own terms. For example, in Geertz’s famous analysis of a Balinese cockfight, he is at pains to deny that the betting practices surrounding cockfights – in which participants run a serious risk of financial ruin – are irrational. In Balinese society, “money is less a measure of utility … [than] a symbol of moral import,” so being able to risk a great deal is crucial to participants’ public self (432–7).
The thesis, then, is that meaning is a human product. Walzer takes the cultural variability of goods that he discovers to be powerful support for Geertz’s claim (Walzer 1994a: 26). The premise of his social-meaning thesis is that goods acquire whatever meanings they have via social construction, which turns indeterminate objects into goods with social meanings (Walzer 1993a is the classic statement). Meaning is cultural, not natural: whatever meanings exist, they do so only by virtue of social construction. Again, we can best understand the thesis if we reformulate it in negative terms: meanings, and therefore the goods they create, are neither universal, nor idiosyncratic, but intersubjective. That is why Walzer starts Spheres with a “theory of goods” based on the claim that, before we can distribute goods, we must conceive of and create them (Walzer 1983: 6–10). A social “good” is never merely an object, as objects have no “essential nature” (Walzer 1994a: 26). Before an object becomes a good, it must be constructed by sets of subjects in a particular way (Walzer 1993a: 39–40). That is why the first proposition in the theory of goods is that every good is a social good (Walzer 1983: 7). Following Geertz, Walzer concludes Spheres by holding that, “We are (all of us) culture-producing creatures; we make and inhabit meaningful worlds” (314; for discussion, Galston 1989: 124). This means that – like the world of war – the societies in which we live are moral worlds that contain the resources necessary for their improvement. I discuss this further in Chapter 8. For our immediate purposes, the key point is that without the cultural production of goods, there would be nothing to distribute.
That is why Walzer rejects Rawls’s notion of primary goods: as meanings are variable, there are no goods that all people need if they are to live a good life. At best, each society will have its own set of primary goods needed by all members, but not necessarily by outsiders. Furthermore, attempts to stipulate a set of universally necessary goods by deduction from a general account of the process of meaning creation cannot work. Although social construction is a universal human activity, it has virtually no universal content: as soon as we assign “use and value” to objects, they become social goods that can be understood only through interpretation (Walzer 1993a: 39, 51–2, n. 1, Agnafors 2010: 167). Walzer concludes that the first task of a theory of justice is to interpret the meaning of goods. For example, we cannot take equality of opportunity to be universally desirable because it depends on the idea of the career open to talents, and that in turn on the notion of life being lived according to a plan. Those ideas simply did not exist before the advent of liberal capitalism (Walzer 1984: 54–5, 1993a: 43–4, 1994a: 22–4, 1989/90: 33, 1986b: 74–6).
Walzer often puts the thesis in terms of “shared understandings” of particular societies (see, for example, the important discussion of the relativity of justice in Walzer 1983: 312–14). I think this has led to considerable misunderstanding and is the cause of the thesis being “subject to more criticism than any other facet of the theory” (Mayer 2002: 344, Armstrong 2000: 87, Downing and Thigpen 1986). It overemphasizes the degree of agreement to which Walzer appeals and draws attention away from the theory of meaning-creation that underlies Walzer’s thesis. It has led critics to argue that, if shared understandings exist, they likely result from domination and lack normative force (Walzer anticipated this criticism at 1983: 9). In particular, critics tend to take Walzer to suggest that there is a shared understanding of how goods should be distributed. Ronald Dworkin’s important early critique of Spheres puts this emphatically, claiming that Walzer’s theory is that, “it is part of social meaning … that medicine and other necessities of a decent life should be distributed according to need, punishment and honors according to what people deserve” (Dworkin 1983a: 4). Dworkin insists that, “Our political arguments almost never begin in some shared understanding of the principles of distribution” (4, emphasis added; for similar critiques, see Den Hartogh 1999: 500, Van der Veen 1999: 240, Trappenburg 2000: 345, Miller 1995a: 5, Barry 1995: 70, Daniels 1985, Cohen 1986).
However, Walzer’s theory does not rely on the claim that societies agree about how goods should be distributed and he acknowledges that such a position would be implausible (Walzer 2014a: 170). Rather, his argument is that distributive disagreement is nested within cultural traditions that create and then define what goods are. Walzer’s use of the term “shared understandings” causes confusion because it has led critics to view his theory as presupposing fixity of meaning and a degree of consensus within society that does not exist (for versions of this critique, see Warnke 1993, Erskine 2007, Bounds 1994, Orlie 1999, Balfour 1994). That this is a misreading is shown by Walzer’s argument that meanings are “not just there, agreed on once and for all … they are always subject to dispute” (Walzer 1993a: 42). Social meanings change over time and at no point can be defined such that competing accounts are implausible. What Walzer means when he says that there is a shared understanding of the meaning of goods is not that all members of communities have an identical sense of what a good is, but that whatever meaning the good has is intersubjective. This means merely that there is sufficient comprehensibility to serve as the basis of normative debate within a community and that there are no universal goods.
Healthcare is Walzer’s favorite in part because it illustrates the particularity of goods. Dworkin is right that Walzer argues that healthcare should be distributed on the basis of need. Echoing Dworkin, Govert den Hartogh takes Walzer to collapse the meaning of healthcare into its principle of distribution, and comments that, “if we are only told that the use of medical care is to cure the ill, it doesn’t seem at all plausible that we know enough to decide how to distribute it” (Den Hartogh 1999: 505). In fact, however, Walzer does not define healthcare as a need, and he tells us more about it than that its use is to cure the ill. He argues that it currently has two meanings: health and longevity (Walzer 1983: 87, 1994a: 30). By health, Walzer seems to mean a sense of wellbeing and the absence of pain. This meaning might be universal, although its value varies across societies. However, longevity is a culturally specific meaning, yet one that is, Walzer holds, necessary to support the conclusion that healthcare should be distributed on the basis of need. The ubiquity with which these meanings now hold is such that “the gross structure of justice-in-cures is given in advance of “arguments about distribution” (Walzer 1994a: 31). In other words, his claim is that no serious argument can be given against the social provision of at least basic healthcare. However, Walzer’s argument is specific to contemporary societies. What makes it so striking an example is that it shows that what most take to be a universal good is actually culturally specific.
Seen as a social good, the meaning of healthcare differs across societies. We miss this only because we take healthcare to be a natural good, and believe that people everywhere feel that curing the ill occupies the same place in a list of social priorities. For example, Martha Nussbaum argues that health and longevity are universal goods, or what she calls “human capabilities” (Nussbaum 2000: 78; on the human capabilities approach, Nussbaum and Sen 1993, Sen 2009: 225–68, Putnam 1993). A human capability is something that everyone needs, wherever and whenever they live, if they are to have a good life. Nussbaum thus takes human capabilities to be similar to Rawlsian primary goods (Nussbaum 2000: 88–9; but cf. Sen 2009: 260–62, Sen 1993: 47). The difference is that Nussbaum’s list is arrived at via cultural analysis and seeks to combine universal and particular (Nussbaum and Glover 1995: 1–7). The first two items on Nussbaum’s list of human capabilities are “life” and “bodily health,” with the former defined as the ability “to live to the end of a human life of normal length” (Nussbaum 2000: 78) and so resembling Walzer’s longevity. Bearing in mind the extensive empirical research undertaken by Nussbaum and her collaborators, it may be that every contemporary society values longevity. However, this might be due to cultural diffusion in a globalizing era. To consider whether longevity is a human good, we should consider variation over time, as well as space.
Walzer contends that past societies did not always value longevity. In medieval Europe, for example, doctors could do little to help their patients, so the “common attitude in the face of disease … was a stoical fatalism” with medicine marginal (Walzer 1983: 86). Not having much chance of a long life, people did not value it so highly. Rather, spiritual care – “the cure of souls,” the social meaning of which was “eternal life” – took the place of longevity (Walzer 1994a: 28–9). That means that there was a socially felt deep desire to achieve salvation, such that “eternity was both real and important” (29). As it was believed that salvation is achieved by repentance and atonement, Walzer argues that from the agreement about salvation’s importance, there followed “the moral necessity of socialized distribution”; that is, the guaranteed availability of a priest in every parish (29). As longevity was less important, the distribution of healthcare did not need to be socialized.
So, Walzer shows that the meaning of healthcare varies across societies: in medieval Europe, it meant only mitigation of pain and suffering, while today it also contributes to longevity and is more important. In Nussbaum’s terms, “bodily health,” may have been a central capability in medieval Europe, but “life” was not. Walzer’s account thus challenges the idea that the meaning of goods is universal by showing that even those that seem most obviously central to a good human life are in fact social: it is not a necessary trait to prize longevity. In other cases, goods that are highly valued in some societies do not exist elsewhere. Not only the life according to a plan but even the love match did not exist in medieval Europe. Walzer’s healthcare argument illustrates the importance of changes in meaning over time, not just cultural space.
The structure of this argument is independent of its content. Walzer may have overstated the importance of longevity, but proving that would require cultural analysis. The logic of the theory is that we cannot know how to distribute a good until we know what it means to the people to whom it is to be distributed, and we cannot know that until we know how they understand and value the good. His interpretation of the meaning of particular goods is independent of the claim that goods have social meanings. Yet the theory poses a daunting challenge to universalist accounts by showing the variability in meanings and the difficulty of making a positive demonstration that a good exists in all possible societies.
While healthcare illustrates the social-meaning thesis, it also demonstrates how the meaning-dependence thesis constrains distributive arguments. One might ask why medieval European values did not require socialized healthcare, given that the relevant reason for receiving treatment there was being sick (Nozick 1974: 233–5; for discussion, Wolff 1991: 123–6). Walzer’s response is that healthcare’s “internal goal” is not its “social meaning,” because the latter includes a good’s value (Walzer 1983: 88). The increasing importance that we attach to health, as well as the addition of longevity as healthcare’s second meaning, thus form part of his argument for socialized distribution (Walzer 1994a: 30). So, the reason that healthcare did not have to be socialized in medieval Europe is that it was not sufficiently important. In arguing for distributive principles, the meaning-dependence thesis requires that we compare goods to establish their overall place in social life. This puts restraints on distributive arguments in two ways. First, it requires rejecting claims that a good should have a particular meaning that it does not actually have, which means rejecting claims of egalitarians such as Nussbaum that healthcare should be distributed as a need in societies in which it is insufficiently valued. Second, it rebuts principles that distribute a good by the standard appropriate to some other good. This rebuts libertarians such as Nozick, who recognize healthcare’s value but insist it should be distributed on the basis of ability to pay (Walzer 1983: 88).
Walzer notes today that, whereas most of his friends in political philosophy broadly agreed with his argument in Wars, they were critical of Spheres because of its cultural pluralism.12 When it first appeared, Spheres was often treated as part of a “communitarian” critique of liberal egalitarianism (Mulhall and Swift 1992, Avineri and De-Shalit 1992, Kymlicka 1989, 2002, Sandel 1984b, Taylor 1989, Neal and Paris 1990; see also Walzer 1990b). Although Walzer normally rejected the label, there is some truth to it. Like other communitarians, Walzer rejects liberal neutrality with regard to the meaning of the good life and insists that communities invariably form conceptions of the good in the course of their social life. That is integral to the social-meaning thesis. Like Michael Sandel, Walzer insists that liberal individualism makes a poor foundation for egalitarian argument and that Rawls in particular is vulnerable to Nozick’s critique of equality (Sandel 1984a). On Walzer’s account, Rawls’s appeal to fairness misses the importance of oppression and subordination, leaves insufficient room for political action, and relies too much on the state. However, Walzer’s pluralism fits uncomfortably with the republicanism of the other communitarians, because his cultural particularism negates the view that there is a vision of the good life and rejects Sandel and Alasdair MacIntyre’s attempts to overcome contemporary obstacles to viewing human life as unified (MacIntyre 1981: 204–25, Sandel 1996; see Walzer 1998a, 1998b, 2004a: 1–20).
If we instead think of Walzer’s theory on his terms, we see it as a social-democratic critique of liberal abstraction that also rejects the Marxist notion of justice as a bourgeois mystification. This helps us view his egalitarianism and his pluralism as aspects of an argument for socially situated, politically engaged discourse about justice and equality. To put the point in Walzerian language, as a political relationship, equality is above all about decision-making power. It is not an identity of possessions or positions. Complex equality seeks to prevent people dominating others by making decisions for them about matters of grave concern. It is for this reason that Walzer is suspicious of philosophical attempts to stipulate timeless principles of justice. He even goes so far as to compare attempts by Rawlsians to use the Supreme Court to ensure rights with Lenin’s vanguard party (Walzer 1981: 9; see also 1977, 1980c, 2015b: 83–90), a striking claim from the point of view of the Court after the Kavanaugh nomination. Walzer’s commitment to decision-making equality also leads him to advocate industrial or workplace democracy, which holds that ownership of commercial enterprises cannot be allowed to convey unlimited power within the firm, as that would lead to something like political rule over workers (Walzer 1983: 291–303, cf. Mayer 2001; see discussion in Chapter 4).
In insisting that theories of justice be socially situated, Walzer follows his mentors on Dissent, Irving Howe and Lewis Coser, who also held that social democracy must be democracy of a particular society, reflecting its norms and values (Howe 1982, Howe and Coser 1955, 1962, Coser 1954). Walzer’s insistence that a theory of justice engage with social meanings follows this critique of abstraction on the grounds that the notions of equality abstract theories advance do not feel like equality to their supposed beneficiaries. Liberals in the Rawlsian tradition tend not to wish to allow the meaning of equality to be worked out politically, because that risks allowing morally arbitrary differences to factor into that meaning, and Rawls’s original position seeks to abstract from such differences (Rawls 1971: 100–8, 1993: 22–9, 58–66). Yet Walzer’s political conception of equality objects to the notion that differences between people are morally arbitrary (Walzer 2007: 303–4), because that idea negates the value of existing commitments and fails to recognize that all meanings are socially constructed. As identities are constructed by those meanings, an egalitarianism that does not refer to social meanings does not speak to who we are.
Walzer’s social-meaning thesis, for all the criticisms it has received, marks a major contribution to distributive justice, for in demonstrating the variability in meanings, Walzer raises grave difficulties for universal approaches. Instead, he suggests that any theory must start with cultural inquiry into meaning, and so that political theory must draw on anthropology and history. It is important to note that this does not necessarily mean that moral standards are culturally relative. Rather, it suggests that without inquiry into meaning, theories of justice simply cannot provide an accurate account of what the theory is about. Principles of justice are culturally variable because the goods available for distribution are social products. When distributing different things, we must use different principles. This does not necessarily mean that the standards of evaluation are particular: using the same criteria but applying them to different goods will produce different distributive principles. Nor does the social-meaning thesis mean that Walzer is uncritical of dominant distributive standards. He continually insisted that healthcare should be socialized during Reagan’s era to attempt to show that Reaganomics was incompatible with American understandings of longevity (Walzer 1986c: 11). Rather, Walzer’s thesis suggests that social meanings should form the starting point for analysis, which should then proceed in critical dialogue with the set of meanings in the society in question. Here, too, Howe is an inspiration (“Discussion” 1976: 70). Walzer’s notion of social criticism, which I discuss in Chapter 8, explains how egalitarian theories can work within sets of social meanings.
Walzer’s theory has attracted adherents, who have pointed to the utility of his analysis with regard to universal healthcare, workplace democracy, and theorizing exploitation (for example, Mayer 2002, Armstrong 2002, Carens 2004, Agnafors 2010, Sandel 2014, Scanlon 2014). However, while Walzer’s social-meaning thesis is a major contribution, it is not of course beyond reproach. It is worth noting that, in Spheres, he is at his most particularistic, whereas part of Dissent’s thought was cosmopolitan (Howe 1976: 599–600, 1982: 276–7, Isserman 1987: 233–4, Cooney 1986: 5–7, Bloom 1986: vii, Rodden 2005: 5–9; for discussion, Reiner 2017a: 458–60). After Spheres, Walzer modified the claim that all meaning is social, suggesting that a minimal universal morality is always present alongside the culturally particular moral systems of each society (Walzer 1990a, 1994a). Yet even in Spheres, Walzer acknowledged that some meanings are individual, as we saw in discussing commodities. If this is the case, then while complex equality cannot rely on an objective metric by virtue of which we are equals, perhaps it might incorporate a default toward equality equivalent to the Rawlsian one. After all, while each commodity matters differently to different people, we all need some commodities, because we all want a pleasant life. Walzerian social democrats might, then, do well to adopt a hybrid of the distributive principles of Rawls and Williams and hold that goods should be distributed equally, unless there are relevant reasons for an unequal distribution.