All governments rely, at least in part, on the co-operation of the governed. The power of a simple despotism, exclusively dependent on the terror it evoked, would be restricted to activities that it could supervise in every detail. Even in rather primitive conditions, such a regime would be a feeble thing, so much so one would hesitate to say that it was governing its people; in more advanced societies, with complex divisions of labour among experts, it would be certain to collapse in days. Though hated alien conquerors have sometimes governed such societies, they have invariably had to work through an existing cultural apparatus, that is, through personnel and institutions which can secure obedience without the use of bribes or punishments. They have had to find ways of presenting their instructions as being legitimate.
A given command has legitimacy to the extent that it secures willing obedience even where it conflicts with the obvious interests of those commanded. The best known modern treatment of the concept is in the later writings of the great sociologist Max Weber (1864–1920). Weber approached legitimacy as a subcategory of ‘domination’, by which he meant ‘the probability that certain specific commands (or all commands) will be obeyed by a given group of persons’. He noted that ‘every genuine form of domination implies a minimum of voluntary compliance, that is, an interest (based on ulterior motives or genuine acceptance) in obedience’.1 The probability of obedience thus rests upon the presence of motives to obey (including purely altruistic ones). But no relationship of domination is likely to depend indefinitely on a coincidence of interest:
An order [that is, system of domination] which is adhered to from motives of pure expediency is generally much less stable than one upheld on a purely customary basis through the fact that the corresponding behaviour has become habitual. The latter is much the most common type of subjective attitude. But even this type of order is in turn much less stable than an order which enjoys the prestige of being considered binding, or, as it may be expressed, of ‘legitimacy’.2
It is no doubt for this reason that ‘domination’ never limits itself to ‘the appeal to affectual or material or ideal motives’. ‘Every such system’ also ‘endeavours to establish and to cultivate the belief in its legitimacy.’3
Weber thus used the word ‘legitimate’ to characterise one type of domination: the type in which obedience is based in part on the belief that the command is binding. Weberian legitimacy is less an objective attribute of powers, entitling some person or persons to be obeyed, than the defining quality of one particular power-situation: a relationship in which obedience is partly explained by the attitude of the subordinate to the bare issuing of the instructions. The reason that Weber addressed this situation was that he studied social practices ‘insofar as the acting individual attaches a subjective meaning to his behaviour’;4 legitimacy is the subjective meaning attached to the conscious acceptance of a relationship of domination.
Weber has long been famous for classifying legitimate domination into three categories, depending on whether obedience was based on a regard for the qualities of a leader (charismatic), for the sanctity of custom (traditional), or for the legality of a system of rules (bureaucratic). The moral basis of all three is purely arbitrary, but the eventual triumph of the last will be ensured by its efficiency. Though Weber’s treatment of this rule-bound future has an oracular obscurity, he clearly felt hostility towards it, in part, no doubt, because bureaucracy is of its nature sterile. Bureaucracy can discipline the bureaucrat himself into acceptance of a professional ethos, but it cannot on its own supply the values by which a rounded human being lives. The modern world, which witnesses the victory of bureaucratic rule, will also see what Weber calls a modern ‘polytheism’, a moral anarchy of clashing values.5
If Weber is broadly correct, legitimacy might be said to fall outside the scope of political theory, if only because it seems to have no content that is susceptible of theorisation. Charismatic and traditional domination are fundamentally non-rational, while it seems that bureaucratic domination is not in a strict sense legitimate; it may persist by virtue of its sheer effectiveness in satisfying our material wants, but it cannot induce a belief that its orders are binding. There is, however, a history of trying to explain legitimacy. It is a history that begins when the reflective individual imagines herself as devoid of political ties and sets out to account for their existence. We speak about legitimacy because we can imagine authority as non-legitimate. Capacity to do this derives from a certain conception of our selves, one that excludes the possibility that selves are constituted by their shared activities. Political legitimacy (the subject of this chapter) only becomes a necessary concept when people can imagine human beings as neither governing nor being governed, because their intellectual starting-point is an image of completely separate selves; it consists, one might say, in whatever beliefs may glue selves back to states.
The heart of a legitimating theory is thus the self-conception of the individual. The problem can only arise if there is something in that self-conception detaching his identity from the idea of his obedience; it can only be solved if acceptance of subjection (in whatever form subjection locally takes) can be shown to be consistent with the way that the self is imagined. This suggests two important distinctions. A person who thinks of his state as a part of himself, experiencing his rulers’ reverses as his own, has what might be called an ‘identity-theory’, as does a person who believes he is a natural slave, and therefore incomplete without a master; but to the extent that they construe these common situations as answering needs of pre-existent selves (in no sense constituted by subjection), they have ‘legitimacy-theories’. These legitimacy-theories are successful if they are logical and realised: logical in that whatever characteristics the theories impute to the selves imply a belief that given commands are binding; realised in that the relevant self-conception attains sufficient plausibility to be embraced by actual individuals (a theory might be ‘logical’ without achieving plausibility to people of a given cultural background).
In fact, of course, the image of the self is likely to have features shaped by the experience of social life; legitimacy becomes possible because a self is pictured that has needs that present practices can satisfy. Our image of the self was shaped by a variety of practices; if legitimate authority seems elusive, it may be that those practices imparted inconsistent expectations.
Pre-modern Europeans still took government for granted. They devoted much time to discussing the characteristics of the ideal ruler, but they showed much less interest in justifying rulership in general; the very existence of people who could secure willing obedience was not yet seen as being problematic. Though Greek and Latin writers could imagine the coalescence of communities from isolated individuals, the primary purpose of such narratives was not to explain why governments enjoyed obedience, still less why they could be seen as deserving to do so. The most elaborate attempt to rationalise these assumptions is found in Aristotle’s Politics.In this first recognisable work of political science, Aristotle of Stagira (384–322 bc) described the behaviour patterns of the inhabitants of city-states (poleis), the communities he saw as the natural endpoint of biological developments. He showed less interest in theorising the supra-political power by which non-Greeks were generally governed; although he was a Macedonian, who taught Alexander the Great (356–323 bc), he was not interested in his pupil’s unnaturally distended polity. Because he was a theorist of small communities, he was able to take it for granted that citizens would have enough in common to make it possible to speak about a ‘common good’, including the good of mutual interaction by taking it in turns to rule and be ruled. But in any case, this good of interaction was not a need of individuals so much as a fulfilment of their natures; enjoyment of the practices that we call politics was seen as constituting human beings.
Thus Aristotle presupposed that somebody should govern; the problem of legitimacy simply did not arise, because co-ordination for a collective good was something that was part of being human. The ultimate source of legitimacy-theories was probably the bias towards individualism that was introduced by Christianity. Because the objective of Christians was salvation (in practice almost invariably conceived of as the avoidance of the pains of hell), political activities were inessential to their self-conception, and it was possible to hold that earthly governments were something contingently willed by Providence. Some Christians, some of the time, especially under the influence of Augustine (354–430), were thus enabled to see government as something extrinsic to humanity. But their type of legitimacy-theory demands for its realisation a homogeneous theistic culture of a kind that is no longer possible. The earliest legitimacy-theories with elements that we could hope to borrow date from the post-Renaissance period.
The first articulation of the problem in something resembling the form it is known today was in the political works of Thomas Hobbes (1588–1679). One explanation of his novel project was that he was facing a novel situation, in which legitimacy was required by the agent that he came to call ‘the state’. During the Middle Ages, religious and cultural authority had been to some extent divorced from military and economic power, the former being located, in the main, in the international church. But from the early sixteenth century onwards, the rulers of many communities were forced to decide for themselves if their dominions should accept or repress some version of Protestant doctrine (almost all serious thinkers held that toleration was unthinkable). This meant the early modern polity was exercising a new kind of power: the power to judge the truth of moral teachings. Hobbes was the theorist of an institution that needed to decide between beliefs in order to impose them, through schools and the clergy, on every member of the population.
One reason that legitimacy was more of a problem for Hobbes than his precursors was thus his need to justify new types of government activity. But in any case a dogma about method helped to propel him in the same direction. He wanted to give an account of political life that would deserve the name of civil science, a ‘science’ being knowledge by means of which ‘when we see how any thing comes about, upon what causes, and by what manner; when the like causes come into our power, we see how to make it produce the like effects’.6 This ‘knowledge of consequence’ was ‘not absolute, but conditional’, embodied in statements of the form ‘that if This be, That is; if This has been, That has been; if This shall be, That shall be: which is to know conditionally’.7 To grasp the workings of a commonwealth was thus to grasp how its component parts, its human population, could jointly cause the whole phenomenon. Hobbes did not seriously maintain that human beings ever met as isolated equals, but his conception of a civil science dictated ‘a need, not indeed to take the commonwealth apart, but to view it as taken apart, i.e. to understand correctly what human nature is like, and in what features it is suitable and in what unsuitable to construct a commonwealth’.8
The unsuitable features were more obvious. When seen in isolation, the members of a commonwealth were nothing but material mechanisms, whose ‘life’ was constituted by an endless succession of appetites and aversions. Contrary to what Aristotle thought, they were not naturally political; indeed their only shared and natural purpose was to continue to have appetites. In the absence of a state’s authority, each of them would enjoy a liberty ‘to use his own power, as he will himself, for the preservation of his own . . . Life; and consequently, of doing any thing, which in his own Judgement, and Reason, he shall conceive to be the aptest means thereunto’.9 Under these circumstances, competition for scarce resources, mutual distrust and a competitive instinct known as ‘glory’ (the wish to do down others) were certain to drive these creatures into conflict, a conflict in which ‘there is nothing he can make use of, that may not be a help unto him, in preserving his life against his enemies; it followeth that in such a condition, every man has a Right to everything’.10 The only known means of escaping this condition was through a covenant (that is, a contractual arrangement in which one party has to trust the other). Only if human beings covenanted to lay down this unfettered liberty could they attain security and comfort, but nobody would act on such a contract unless they thought that it would be enforced upon the other parties. The solution was a covenant erecting a covenant-enforcer, the sovereign state, thus setting up conditions which made it rational for them to trust.
One possible reading of Hobbes holds that a truly sovereign state would be so terrifying that everyone would keep their covenants and follow its instructions. Hobbes certainly often maintained or implied that ‘the Passion to be reckoned upon is Fear’.11 He thought that it could override all other human motives and he explicitly maintained that ‘excepting some generous natures, it is the only thing (where there is apparence of profit, or pleasure by breaking the Laws) that makes men keep them’.12 But he was also well aware that something must be added to physical terror to make a commonwealth sustainable. The sovereign may be frightening, but he cannot be an ever-present threat to somebody not physically restrained. There was a clear distinction in Hobbesian thought between the position of someone who was temporarily at the sovereign’s mercy (like someone who is captured in a war) and someone who is permanently subject. Thus captives ‘kept in prison or bonds . . . have no obligation at all; but may break their bonds, or the prison; and kill, or carry away captive their Master, justly’.13 Subjects, by contrast, are obliged to be obedient, ‘which natural obligation, if men know not, they cannot know the Right of any Law the Sovereign maketh. And for the Punishment, they take it but for an act of Hostility, which, when they think they have strength enough, they will endeavour by acts of Hostility to avoid’.14
Thus if a Hobbesian commonwealth is to be feasible, then subjects need to know themselves to have an obligation: they need, in other words, to have legitimacy-beliefs. At times, Hobbes argues that to be obliged quite simply is to have performed a special kind of action, ‘there being no Obligation on any man, which ariseth not from some Act of his own’.15 On this view, only if one knows that one has performed such an act can one be said to know one’s obligation. This doctrine is frequently cited by theorists who hold that legitimacy is based upon consent, but it is worth remembering that Hobbes was easily convinced that someone had consented. As his conception of consent was perfectly consistent with overpowering fear, he presupposed a tacit consent to obey whenever a person with somebody else at his mercy allowed the latter corporal liberty. Thus the authority enjoyed by mothers (unless a subsequent contract intervened) derived from the fact that ‘every man is supposed to promise obedience, to him, in whose power it is to save, or to destroy him’.16
In practice, then, the act Hobbes called ‘consent’ was something he imputed to anyone reduced to a state of obedience. A prudently constructed polity would no doubt encourage ‘knowledge of obligation’ by an insistence on express consent, but this type of completed historical act is not the theory’s ultimate foundation. Consent is only binding because the institution of consenting has a tendency to favour self-preservation. Though it is true that in a state of war (the condition where we lack a common sovereign) we are inevitably driven to act ferociously, we are nonetheless bound to an inner disposition that favours sociable behaviour. This is evidently a type of obligation which cannot be deemed to arise from an act of our own, unless, that is, we are deemed to have consented to any type of action that we must consistently will, given we also will self-preservation.
The reason Hobbes found it so hard to make himself clear was probably that he could not explain how a self-interested automaton could make a binding promise. Hobbes was no doubt quite right to believe that somebody who is disposed to co-operate with others can be expected, other things being equal, to have a safer, more agreeable life than someone who is not; but it is easy to construct examples that would induce a rational calculator to see advantages in breach of contract. A person with an hour to live, a person whose behaviour could not be scrutinised, or somebody whose breach of faith would place him beyond reach of retribution might all, in Hobbesian terms, have excellent reasons for failing to fulfil their obligations.
Hobbes did show some awareness of objections of this type, and put a vaguely specified example into the mouth of someone called ‘the Fool’, who ‘questioneth, whether Injustice, taking away the fear of God. . .may not sometimes stand with that Reason, which dictateth to every man his own good; and particularly then, when it conduceth to such a benefit, as shall put a man in a condition, to neglect not only the dispraise, and revilings, but also the power of other men’.17 Hobbes’s answer to the Fool’s very pertinent question presents interpretative difficulties unlikely to be finally resolved, but the reading that appears to make most sense is that the type of person who reasons like this is likely, in due course, to be found out. The rational self thus chooses dispositions (that is to say, habitual virtues or vices) that will in general aid self-preservation.18 But this involves the unappealing claim that the purely hypothetical decision of someone who can freely choose what type of individual to be ought to affect the conduct of someone with determinate characteristics.
Hobbes was the first great thinker to face up to our political predicament. The Hobbesian detachment of a reflective self from the activities that it engaged in involved him in developing a picture of that self as holding beliefs supportive of his state’s authority. Unless there is an answer to the Fool, Hobbesian legitimation of the state is plainly unsuccessful. But even if there is a cogent answer, Hobbesian legitimation could not be realised in the modern world unless a fully Hobbesian self-conception could somehow attain plausibility. A moment’s thought reveals this is unlikely, in part because our picture of the self has several quite unHobbesian characteristics.
The question that Hobbes set himself was how the individuals that his ‘science’ abstracted from observed activities could be combined so as to form a stable polity. The fundamental basis of his answer was the belief that everybody shares a powerful motive: a common aversion to death that overrides our several and conflicting appetites. Where Aristotle believed in a shared Supreme Good – the good, among other things, of interaction – Hobbes believed in a shared Supreme Evil: violent death. It followed that the main ideological threat to the stability of Hobbesian states was the belief, promoted by the clergy, that there is something worse than violent death: the everlasting torment of the damned.
Our situation is more difficult, because our picture of the self is much more complicated. There are, of course, still social scientists attempting to answer the Fool by pursuing the essentially Hobbesian programme of trying to see political arrangements as products of individual greed and fear (they tend to forget about ‘glory’), but most of them present themselves as merely exploring a ‘model’ of human nature. Their caution is quite understandable, because the basic Hobbesian postulates were much more in tune with Stuart common sense than with more modern assumptions. The Hobbesian conception of the individual as hell- or death-avoiding was an adequate rationalisation of seventeenth-century practice; it was easy to construe the Christian religion as primarily a rational strategy for maximising pleasure (the joys of heaven), and/or avoiding pain (the fires of hell). But rationalisation of our accepted moral practices promotes a quite different conception of the individual.
To begin with, we think of the self as a bearer of rights, and in particular of the right to choose and to express its own beliefs. Hobbes thought that the state was well advised to interfere as little as possible in the behaviour of individuals, but his assertion of its right to judge precluded the more modern view that there is an intrinsically private sphere, intrusion on which is a violation of individual prerogatives. The origins of this attitude are usually and probably rightly traced to the political theory of John Locke (1632–1704), who founded his ideas upon a duty, given by God, of preserving one’s self (and secondarily others), and therefore of protecting all the rights (revealingly known by Locke as ‘Property’) by which that preservation was assured. There seem to have been two social practices shaping the Lockean self. One was the rationalist Christianity of liberal Protestantism; the other was the experience of a proto-capitalist legal order, assuring to the individual indefinitely transferable property rights. From the perspective of this self, the only purpose that was served by states was the enforcement of his private rights, using coercive powers that they borrowed from pre-political individuals. It was consistent with this view that there was still one sphere of life in which the individual’s rights could not be given away or even lent. Where Hobbes allotted to the state the right to determine the details of religion, Locke held that the right to free religious practice was too important to be handed over to the political authorities. Given the risk of hellfire, this claim was eminently rational, especially when supported by the claim that God rewards sincerity (as opposed to dogmatic correctness).
Though the decline of Christianity has taken away its original foundation, responsibility to a Creator, the secularisation experienced by most societies has actually entrenched the core assumption, for the tendency to abandon dogmatic religion has been in the name of the value of working out one’s own morality. Both Christianity’s decline and liberalism’s survival are no doubt functional with respect to market practices, but both create some rather obvious problems. If Locke’s ideas were wholly dominant, it is extremely hard to see how any government could be legitimate and stable, because legitimation based on Lockean self-conceptions encounters two important difficulties. The first is how to deal with the emergence of anti-Lockean moral theories. Locke supported coercive state action to discourage atheism, the theory that threatened the logical basis of Lockean natural law,19 but secularised Lockeans, with their attachment to sincerity, find it extremely hard to justify the suppression of sincerely held ideas. The freedom Locke defended was freedom to follow one’s reason; the freedom prized by many of his successors (including, incidentally, Max Weber) is freedom to make an arbitrary commitment to some particular values. There is, of course, no guarantee that those values will themselves be tolerant; but, in any case, post-Lockean liberalism leads to diversity, and the steady diminution in the stock of shared beliefs is bound to affect the shared belief that state commands are binding.
The second is a problem shared with Hobbes. Like Hobbes, Locke was driven by logic to the view that his abstracted individuals could only be subjected by some action of their own, but the importance he attached to the free exercise of moral judgement meant his criteria for ‘consent’ were more demanding than his predecessor’s. In consequence, his doctrine of tacit consent was even less convincing. Lockean legitimacy would be founded on the memory of an uncoerced decision to put oneself into subjection, but such events are virtually unknown. No major modern state attempts to elicit even the fiction of express consent from native members of its population; at most such states rely on the tacit consent allegedly involved in casting votes.
The self that seems to be implied by modern practices in fact owes more to Rousseau (1712–78); it is a self whose leading characteristic is horror at the notion of being subjected to another’s will. Rousseau escaped from the insoluble problems created by resort to historic consent by tracing the legitimacy of political institutions to the present will of individuals. The crucial move that made this possible was abandonment of the Hobbesian postulate that natural man desires to excel. The passion to out-do that Hobbes called glory (and Rousseau amour-propre) was said to be characteristic of human beings denatured and corrupted by their society, for natural man is indifferent to his fellows except in wanting to avoid any dependence on another’s will. Rousseau believed that he could find such selves both in contemporary savages and in the citizens of past republics, but the ready acceptance of his self-conception suggests that it was somehow already implicit in eighteenth-century culture; religious introspection (both sides of the confessional divide), respect for private ‘sentiment’, and bourgeois self-assertion were all of them certainly likely to encourage an ideal of non-dependence. At all events, in Rousseau’s theory, love of autonomy replaces glory in precluding the view that man is innately social and setting the problem legitimacy must solve.
The essence of Rousseau’s solution is quite simple. As Rousseau’s selves prize self-determination above convenience and even safety, legitimacy can never be realised among people with a healthy self-conception unless a government’s laws can be construed as an expression of a General Will directed to achieving the shared good of the whole. But there can only be a common good if people are shaped in such a way that they have goods in common: ‘if there were not some point on which all interests agree, society could not exist’.20 This was probably why Rousseau attached significance to the institution of the Censorship, a moral watchdog that enforced correct behaviour in private life, and to the existence of a civil religion whose few and simple doctrines were consistent with, and reinforced, the values of the state.21 He went out of his way to acknowledge an implication: the anti-political nature of Christianity, a religion whose adherents cared too little about conditions in the present life.22 The same objection would apply to any way of life that made peremptory demands distracting a person’s attention from the collective good. Rousseau’s society cannot afford the presence of values that undermine commitment to the common interest.
Most of us have of course imbibed both Locke’s and Rousseau’s self-conceptions. We see ourselves both as the bearers of Lockean rights and as averse to government by others. The shared priorities Rousseau guaranteed through civil religion and pressure upon manners have been assured, in modern times, by the power of national feeling. Though it is very difficult to say any more about nations than that they are groups with something important in common that leads them to aspire to govern themselves, the wish to be a part of such a unit appears to be a very powerful motive. The urge to create and defend a nation-state has brought about the kind of sacrifices, up to and including the laying down of life, no Hobbesian self could possibly envisage. Though this phenomenon can appear to rest upon identity-beliefs (the loyal subject feels his country’s triumphs as his own), it can also be construed in a more individualistic spirit. Rulers who come from the same national background find it easier to create within their subjects the impression that they share a common good, in part because they are presumed to share an interest in cultural survival; the rational explanation of the ferocity of nationalism in places like the former Yugoslavia is the entirely plausible assumption that only one culture is likely to survive in a given political unit.
The self that seems to underly much modern political practice is thus a person who attaches value to the persistence of some characteristics she feels that she shares with her neighbours. She may believe this way of being human is better than all others, but she is motivated by the accident that it happens to be hers. Though it is threatened in its turn by the emergence of sub-nationalisms (few nations are so homogeneous that they lack proto-national subdivisions) legitimacy based on this conception has plainly been quite widely realised. The obvious danger with this situation is that promotion of the nation’s values is likely to lead to subversion of Lockean ones, whether to do away with competing loyalties or to forestall such loyalties from emerging.
But even if the nation-state had ways of containing this threat without unduly transgressing the rights of its members, it would still be bound to clash, sooner or later, with the international economic order. Modern nations are in practice interdependent in ways that demand the enforcement of quite intangible entitlements. In the foreseeable future, a high proportion of the wealth of the most powerful will take the form of claims on foreign assets and so-called ‘intellectual property’, that is, of bonds, shares, licences and patents, and other entities whose very existence is constituted by a legal system. Even if we suppose, implausibly, that everybody has an interest in the existence of this legal system, it is bound to be experienced by some people (and some peoples), some of the time, as standing between them and self-determination.
This chapter has been arguing that a legitimating theory has to be based upon a self-conception that people actually find acceptable. Precisely this thought can be found in the most celebrated work of later twentieth-century liberal theory. John Rawls’s A Theory of Justice (1971) maintained that just arrangements were those that would be chosen by rational selves behind a ‘veil of ignorance’ concealing, amongst other things, their talents, social status and ‘conception of the good’.23 This remarkably ignorant self would still, however, be a ‘moral person’, that is, someone ‘with a fundamental preference for conditions that enable him to frame a mode of life that expresses his nature as a free and rational being as fully as circumstances permit’.24 Rawls thought this self-conception has actually been realised in our culture: ‘The hypothetical nature of the original position invites the question: why should we take any interest in it, moral or otherwise? Recall the answer: the conditions embodied in the description of this situation are ones that we do in fact accept.’25
The question is obviously right, but it invites a gloomier response. The postulated Rawlsian self, abstracted from its actual characteristics, seems as irrelevant to us as the virtue-choosing self to Hobbes’s Fool. Rawls’s more recent work has explained that ‘free and rational’ ought to be understood as ‘reasonable’, as ‘ready to propose principles and standards as fair terms of co-operation and to abide by them willingly, given the assurance that others will likewise do so’.26 Borrowing from T.M. Scanlon, Rawls claims that we are given by our culture ‘a basic desire to be able to justify our actions to others on grounds they could not reasonably reject’.27 This Rawlsian self-justifying self can no doubt be abstracted from some current practices (on a somewhat idealised view of the character of university life, it might be found, for instance, in the activities of academics). It is a cousin of the self implicitly imagined by Jürgen Habermas, with his principle that ‘just those action norms are valid which all possibly affected persons could agree as participants in rational discourses’.28 But nobody spends all their time as a rationally discursive individual, and there is no obvious reason why rational discourse should be seen as more essential to the self than, say, sex, sport, jokes, poetry or religion.
Even a sketchy survey of the history of legitimacy-beliefs raises a troubling possibility. Our present intuitions about a separate self owe something to Hobbesian ‘science’, something to secularised Lockeanism, and something to Rousseau’s hatred of dependence. Because there is no answer to the Fool, no Hobbesian self-conception can really support legitimate arrangements. But the more elaborate selves we can imagine are also fundamentally unsuited to the legitimation of our actual practices. The privileges attributed to the post-Lockean self will offer it the space to make non-Lockean commitments, while Rousseau’s self-governing selves will be affronted by dependence on an international order. What actually sustains our present arrangements may be no more than habit, combined with faith in their effectiveness, effectiveness being narrowly defined in terms of the delivery of economic growth. We find ourselves, in fact, in Weber’s world. If and when our economic growth should falter, the outlook will be bleak.