The State

In the state of nature, people use their life, liberty and property for purposes adopted by themselves. A long tradition of political thought holds that this sets them at cross-purposes, leading to loss of life, insecurity of property and inability to produce the “optimal” assortment of public goods. The extreme form of this view, i.e. that in the state of nature
no public goods can be produced, is probably no longer widely held. The state of nature is coming to be viewed as capable and likely to produce
some public goods, but not as many and perhaps not as much as civil society endowed with a coercive state.
*1 The presumption is that endowed with a state, society is enabled to make the sort of choices which lead to more resources being devoted to public and less to private goods. The modern idea that the state is a device whereby society can more nearly approximate the resource allocation which it
really prefers, implies a much older belief that the “general will,” or social preference, or collective choice (or whichever species of the genus is invoked) has some ascertainable meaning.

In coercing them to realize the general will or to give effect to collective choice, the state is competing with its subjects for the use of the scarce resource that is the liberty and property of each. It restrains them in what they may or not not do and forces them to devote part of their efforts and goods to the state’s purposes rather than to their own. The same long tradition of political thought suggests that in doing this, the state is in fact forcing them to be happier (or better off) than they would otherwise be, for without at least latent coercion they could not resolve the notorious state-of-nature dilemmas of non-cooperation and free riding. At the same time, competition between the state (which successfully maintains the monopoly of force) and its subjects (whose one strong recourse is rebellion—usually risky, costly and hard to organize) is
prima facie so lopsided, so grotesquely unequal, that if the state stops anywhere short of enslaving its subjects, cogent reasons are needed to explain why.

It is hard to formulate anything more crucial to political theory than this question, which has been implicitly answered each time historians have given a satisfying account of the fall of despotism, of stalemate and accord between a king and his barons, or of how a given state has ruled by custom and law, which constrained its choices, rather than by its own discretionary reason which did not.

This chapter is mainly devoted to the largely unintended consequences of securing political consent by redistribution. The pattern of redistribution develops as a result of both the state and its subjects pursuing, “maximizing” their ends, interacting with each other to produce redistributive outcomes. These must be such that neither party can for the time being further improve his position within them. Broadly speaking, they have to reflect the balance of forces and interests concerned. Formal agreements between the state and its subjects, such as laws and constitutions under which the state is supposed to be restrained from maximizing its ends, either reflect this balance or they do not. If they do, the limits of state encroachment on the private rights of liberty and capital are naturally set by the power of the owners of these rights and a constitution or other formal agreement merely proclaims accomplished facts. If they do not, any such agreement is precarious. In abiding by it, the state is not in equilibrium. Its needs and ambitions will eventually lead it to circumvent, reinterpret, amend or simply disobey laws and constitutions. The better to clarify their role, or rather the reasons for their conspicuous absence from the subsequent argument, I start this chapter with what may seem a digression about the rule of law and constitutions, considered as binding agreements limiting the state’s discretion to dispose of its subjects’ liberty and property as and when its best interest dictates.

Montesquieu thought, oddly, that freedom could be defined as a state of affairs where man’s actions were constrained by law only. Such a definition, besides other weaknesses, seems to rest on some implicit belief in the quality, the specific content of law. Unlike rules in general, characterized by their source and enforcement (By whom? Under what sanctions?), to be consistent with freedom law must also have some particular content—for instance it could be thought of as good, benign or perhaps just. Bad law either must not be called law, or it must be agreed to have the redeeming feature that at least it replaces arbitrariness and disorder by a rule. In the political domain, law—even bad law—has from time immemorial been prized as restraint on the sovereign, as the subject’s shield from the despot’s caprice. Impartial even when unjust, general and predictable, it provides some sense of security against the random use of state power. Significantly, the distinction republicans since Titus Livius have drawn between tyranny and freedom, runs not between good and bad law, but between government by men and the government of law. Hence the much too trusting definition of freedom in the
Spirit of Laws. Subjection of the state to law,
even to law of its own devising, has strangely enough been felt to be sufficient for disarming its tyrannical potential. Not till after the Jacobin experience did political theorists of the calibre of Humboldt, Guizot,
*2 and J. S. Mill think of the possibility of the clever state creating self-serving laws which it could safely obey, while retaining its capacity to override the purposes of individuals in favour of its own.

If the rule of
mere law is not a sufficient condition for an acceptable reconciliation of conflicting claims upon the subject’s liberty and possessions and for protecting him from the powerful appetite inherent in the adversary nature of the state, one cannot aim at less than the rule of
good law. Historically, two kinds of solutions have been pursued to the problem of how to get good law. One was not only to oblige the sovereign to obey his own laws, but to constrain his law-giving powers by getting him to agree to what republican Rome called
legum leges—a super-law or constitution which can effectively make bad laws “illegal.” The other, more direct solution was to secure adequate participation by all concerned in the design of laws. Either solution, “constitutional monarchy” with the state alone making laws but only within the bounds fixed by the constitution,
*3 and democracy with the state striking
ad hoc bargains with its subjects over legislation, is designed to ensure “fair and equal” competition between conflicting public and private ends. The latter
ad hoc solution is roughly the one England stumbled into in 1688, liking it and pushing it to its logical fulfilment in 1767; since then, a majority in Parliament has been sovereign—it can make any law and govern any way it sees fit. Its sole constraint on law-making is a cultural one. This confluence of the constitutional and the democratic solution corresponds by and large to the American one, designed by the Founding Fathers with a rare combination of erudition and worldly wisdom, crowned by an astonishingly long run of success in which design must have played some part beside luck, and since copied in some of its features by many other states.

The point about having both belt and braces, i.e. a “fixed” constitution in a democratic state, where laws are in any case the outcome of negotiated bargains between it and civil society, is the relatively subtle one that the threat to people’s liberty and property can just as well come from the sovereign people as from the sovereign king. The danger, then, lies in sovereign power and not in the character of the tenant who holds it.

For obvious reasons, a sovereign assembly, a
demos or its representatives, and a sovereign monarch or dictator tend to present rather different kinds of dangers. Which is worse is at root a matter of personal taste. The view that the assembly is liable to be more unjust than the king was quite prevalent at the Philadelphia Convention disgusted by Westminster, and in the secessionist South rebelling against a Northern majority. Ordinarily, however, it is easier to conjure up the image of a personal tyrant than Pitt’s “tyranny of the majority.” Liberal thought cannot readily reconcile its faith in the benignity of popular sovereignty with approval of constitutional devices which would shackle it, hamper it in doing good and in some cases in doing anything very much at all. No wonder that in the USA, for some decades now, there has been a tendency for the separation of powers to be overcome by reciprocal swaps of functions and attributions, if not by their unilateral usurpation. Thus the executure is making a great deal of administrative law, the legislature is making foreign policy in addition to running the economy, while the judiciary shapes social policy and directs the struggles of classes and races. If the three separate branches of the American federal government were finally all merged into the Harvard Law School, much of this might be performed in a less roundabout manner. (Paradoxically, that day might conceivably mark the beginning of the end of the ascendancy of lawyers over American society.)

There is something threatening and basically “unfair” in the very notion of the sovereign state competing with its subjects for the use of their resources—”unfair” in the simple, everyday sense of an almost obscene disproportion of size and force. No single person has much of a leg to stand on, while the idea of banding together to tame the state promptly raises one of the first questions in statecraft, Why ever should the state let them band together? With the odds looking so blatantly unfavourable to anyone the least bit mistrustful, it is as plausible to predict despair and pre-emptive rebellion by people likely to find themselves in the minority as to expect them peacefully to submit, under the democratic rules, to the appetite of the prospective majority.

Agreeing to constitutional guarantees, then, is an intelligent move, a gesture to reassure the minority that nothing really harsh is going to be done to them. As disarming the mistrust of the prospective minority is, so to speak, a condition for getting everybody’s signature on the social contract, there may very well occur historical conjunctures where it is
rational for the state actually to suggest limits to its own power if its purpose is to maximize it. It has long been known that it can be rational for the wolf to put on sheep’s clothing and to refrain for a while from eating sheep. It is old wisdom that it can be rational to take one step back before taking two forward; it can also be rational to forestall an objection by stating it first, inoculate against a disease by infecting oneself with it, roll with the punches, spend to save, bend rather than be broken and take the long way round because it is quicker.

It is one thing to say that it is good for the state, or for the majority with whose consent it rules, to lull the minority into a false sense of security by offering constitutional safeguards. It is another to insinuate that states which do agree to constitutions typically have some such crafty motive in their conscious, calculating minds. The latter sort of allegation has its place only in conspiracy theories of history, and they are unlikely ever to be right. The recognition that constitutions limiting power can be positively useful for states seeking (to put it summarily) to maximize power may, however, still contribute to the proper historical appreciation of these matters. Those whose particular intellectual enterprise calls for seeing the state, not as the
locus of a single will, but as the shifting and uncertain hierarchy of diffuse and sometimes partially conflicting wills, none of which can be said knowingly to make the state’s decisions, might like to suggest that the hierarchy will tend, albeit perhaps clumsily, to grope for the choices most likely to promote its composite good made up of elements of survival, stability, security, growth, and so forth. The fact that in lurching and groping, states do not always reach worthwhile objects but occasionally fall flat on their faces, need not invalidate such a view. It may simply indicate that if there is an institutional instinct conditioning the state’s conduct, it is not an unerring one, but nor would we expect it to be.

In his brilliant exploration of some paradoxes of rationality, Jon Elster suggests that a society binding itself by a constitution (in fact, it is the state that is bound, but the distinction between state and society is not pertinent to his purpose) follows the same logic as Ulysses having himself bound to the mast to resist the sirens’ song.
*4 If Ulysses were not tempted at all by the sirens, if he were sure of his strength to resist temptation, or else if he fully intended to yield to it, he would not want to be bound. Equipping himself with a “constitution” which forbids him what he does not want to do, is rational in terms of his wish for an assurance against his own changing states of mind, his own weakness of will. Whether Ulysses stands for society, or for the state, or for a generation looking ahead and trying to commit future generations, it is
his own concern that moves him. He truly fears the sirens. Admittedly, he has shipmates but it is not to satisfy
their concerns that he has himself bound.

My own view is different. It is that anything Ulysses-the-state volunteers to do to restrict his own freedom of choice is the result of his reading of the state of mind of his shipmates,
their fear of the sirens and
their mistrust of
his character. It is not the calculus of one interest in the face of a given contingency, but the upshot of at least two, that of the governed and that of the governor. Ulysses asks to be bound lest his crew should want to get rid of so unsafe a captain.

The analogy with states and their constitutions is distorted by the bindings. Once bound, Ulysses cannot undo his shackles. Only his shipmates can release him. A state bound by a “law of laws,” being at the same time the monopolist of all law enforcement, can always
untie itself. It would not be sovereign if it could not. The proper analogy is not with Ulysses and his shipmates approaching Scylla and Charybdis, but with the lady whose lord, reassured by her chastity belt, is safely off to the wars, while she, now mistress of herself, hangs the key of the padlock of the belt on her own bedpost.

The ultimate mastery of the state over the constitution is masked, in countries with a proper “fixed” Franco-American type of constitution, by the provision of a special guardian—the Supreme Court in the USA, the Conseil Constitutionnel in France—watching over its observance. This guardian is either part of the state, or part of civil society. It cannot be in a third place outside, “above” both. If it is part of civil society, it is subject to the state and can in the last analysis always be coerced not to denounce a breach of the constitution. Failing that, it can have its denunciation denounced by another guardian appointed to replace it. The question is obviously not whether this is feasible or whether a form of words can be found to explain that the constitution is thereby
really being respected and on a “higher plane” than hitherto but, rather, whether the stake is worth it. Nature will take its course, and the padlock of the chastity belt will be opened, no doubt in the name of
real (as opposed to
artificial) chastity, depending essentially on the balance of political support to be gained and to be lost by the move (i.e. Can the state politically afford to do it? and Can it afford
not to do it?)
and on the contribution, if any, which acting outside the constitution can make to its ends other than to sheer political survival.

On the other hand, if the guardian of the constitution is part of the state, there is a presumption that it will not have a separate, sharply divergent conception of the public good or, what is in practice indistinguishable from it, a separate and sharply divergent calculus of the balance of advantages to be reaped from interpreting the constitution one way or the other. The “separation of powers” and the independence of the judiciary are, however, designed to undermine just this presumption. Their intended function is to make it altogether possible for such a divergence to emerge. The device, prior to the Crimean War, of making officers of the British Army independent by letting (and indeed obliging) them to own their commissions, was supposed to ensure that the Army’s interest would not diverge from that of property and hence would not become a tool of royal absolutism. The device of selling French magistrates heritable and transferable title to their offices had the effect (though a totally unintended one) of ultimately allowing a divergence of interests to develop between the monarchy and the
parlements to such an extent that in 1771, finding themselves confronted by a strong-willed adversary in Maupeou, they were expropriated and the loyal and the complaisant among them became salaried officers of the state.

Evidently, when the guardian of the constitution is the creature of a previous tenant of state power, the emanation of a majority gone and past, there is quite likely to be a divergence. The American Supreme Court in the face of the New Deal, the French Conseil Constitutionnel in the face of the post-1981 socialist government of the Fifth Republic, are good cases in point. The Supreme Court obstructed or retarded some of Franklin Roosevelt’s legislation affecting the rights of property till 1937, when it backed off, sensing that even if the Administration’s bill to “reform” it was running into the salutary buffers of bicameralism, it was yet inadvisable for the Court to be seen consistently to oppose the democratic majority. (Legitimacy is obeyed if it does not command much or often.) In time and with average mortality of lifetime appointees, the Court will come to think the way the Administration thinks, though a sharp change of regime can create short-term problems. Even these problems, however, will only deter the benign sort of state which it is not desperately important to deter anyway, for it is unlikely to have unconstitutional designs of major short-term impact on the rights of its subjects. Plainly, no possible conflict with the 1958 constitution would have deterred the overwhelming socialist majority in the French Assembly from nationalizing banking and most large industrial corporations in 1981.
*5 It was perfectly understood on all sides that the Constitutional Council might well not survive if it threw out the bill.

A really radical conflict between the conception of right embodied in the constitution and that of public good proposed by the state, particularly at the “dawn of a new era” when there is a bad break in continuity, reflects a revolutionary situation, or a
coup d’état (or, as in Russia in October 1917, one on top of the other). Sweeping away an old constitution is in such moments but a minor effort in the spate of other, more portentous ones. In the face of less radical divergences, a fixed constitution can remain fixed till it is amended.

Amending the law of laws is an undertaking quite possibly different in degree, but hardly different in kind from amending a law or some other less formalized arrangement of society (and if there be a law laying down how the law of laws can be amended,
that law can be amended, for it is ultimately always possible, by proposing a particular distribution of the resulting benefits and burdens, to assemble preponderant support for the amendment). At worst it may involve a good deal more fuss and legislative time and it may require a wider margin of consent over dissent. If so, a constitution intended to protect the freedom and property of the subject against certain kinds of encroachment by the state, does provide security against
lukewarm attempts by an only marginally motivated state. This much, however, is true of
any status quo, whether constitutional or just a fact of everyday life, for every status quo represents some frictional obstacle.

The task of every state, from the most repressive discretionary dictatorship to the purest legitimate commonwealth, is the reciprocal adjustment, to its best advantage, of its policies to the balance of support and opposition they engender. Though this degree of generality almost renders the statement trivial, at least it helps dissolve the notion of the “law of laws” as some sort of ultimate rampart or “side constraint” where the state pulls up hard, and behind which the individual subject can safely relax.