Towards a Theory of the State
It would be rational for a state pursuing its own ends to escape from the treadmill where its power is used up in its own reproduction.
Did Plato's Republic "degenerate" on the way from democracy to despotism?
This is the place for drawing some of our threads closer together. Depending on the scale and perspective of the analysis, it is possible to regard the state in several ways. One is to take it as an inanimate tool, a machine. It has no ends and no will; only persons have ends. Explanation and prediction of its movements must, therefore, deal at one remove with the persons who wield the tool and shift the levers of the machine. Another is to merge the machine and the people who run it, and consider the state as a live institution which behaves as it would if it had a will of its own and a single hierarchy of ends; as if it could choose between alternatives and in doing so seemed to conform to the rudiments of rationality. We have throughout adopted the latter view, not because it is more realistic (neither is), but because it looks the most fertile in plausible deductive consequences.
Once we think of the state as having ends and a will of its own, theories and doctrines which have the state serve the interests of Hobbes's seekers after eminence, Rousseau's myopic deer-hunters or Engels's oppressor class, take on a strongly question-begging quality: for however convincing the accounts they give of how the state could or does serve such interests, they furnish no reason why it should serve them. Yet while the supposition that a will seeks the fulfilment of its ends can be taken as read (it is implicit in rationality; besides, it is hard to think of a will floating freely, not associated with any end), a supposition that it seeks to serve the ends of others needs justification, explicit support of some sort. There is, in my view, no such support for it in either the contractarian or the Marxist theory of the state. It may, in fact, be a misnomer to call either one a theory of the state, though they are both theories of the individual (or class) subject's interest in the state. Moreover, as I have contended in chapter 1, even if it had good reasons to, the state could not pursue the interests of its subjects unless they were homogenous. Its adversary relation to them is inherent in its having to take one side or the other between conflicting interests if it is to have any "policies" at all.
A successful theory of the state should not have to rely on the gratuitous assumption that the state is subservient to some interest other than its own. It should lend itself to the explanation of the state's role in political history in terms of its interest interacting, competing, conflicting with and duly adjusted to the interests of others.*35
What, however, is the proper view of the interests of the state? When do we say that it is using its power to fulfil its ends? I have from the outset reconciled the possibility of "minimalness" and rationality by laying down the "marker" that a state will choose to be minimal ("capitalist," "policyless"—alternative terms I consider to have substantially the same effect as "minimal"), if its ends lie beyond politics and cannot be attained by the use of power—if they are not the satisfactions of governing. On the other hand, all the policies a non-minimal state does adopt are, tautologically, in its interest, in the fulfilment of its ends, except when it is being foolish. Some of these policies, however, can yet be told apart from the others. Into this split, the thin end of a theory of the state might be wedged.
Certain policies, and the specific measures they call for, can at least conceptually be singled out as having a common negative feature: they appear to contribute to no plausible end, satisfy no manifest taste, augment no conceivable enjoyment of the state other than the maintenance of its tenure. They just help keep it in power. They use power in order to reproduce it. If it is right to say that Roman senators felt no altruistic love for the plebs, yet gave them bread and circuses, they "must have" done so because it seemed to them necessary for the maintenance of the existing order. If one can take it that Richelieu did not actually prefer townsmen to nobles, yet favoured the former and sought to weaken the latter, he "must have" done so in order to consolidate royal power. (The "must have" is in inverted commas to invite the reader's complicity and indulgence. So much of historical explanation is, inevitably and I think properly, no more than the elevation of the least unreasonable hypothesis to the rank of the true cause.)
Some measures, in addition to reproducing the state's power, may contribute to its other ends as well. Their nature is such that no presumption stands to the contrary. When a President Peron or a contemporary African government pampers the urban masses, we can say that it "must" be doing so because it has staked its political survival on their support (or acquiescence), but it is not absurd to allow that it likes them, too. Hence, it may be actually pleased to make workers, clerks and soldiers better off at the expense of haughty cattle barons or obtuse tribal villagers. The shape of these measures reveals their support-buying, power-maintaining function, yet it permits the supposition that some other end is being fulfilled, too. Much of the redistribution undertaken by the modern democratic state has this shape.
There is sufficient historical evidence, however, of a clear-cut class of other policies and acts of state which use state power without intelligibly, plausibly visibly contributing to its maintenance. The religious policies of James II, Charles XII of Sweden's campaigns or the profligacy of the Naples Bourbons have, if anything, weakened their hold on power. Gladstone's failed attempts to give Home Rule to Ireland, the Kulturkampf fought by the Second Reich, or American near-belligerence on Britain's side in 1940 used up some of the support enjoyed by the respective governments. Though they may have been the right thing to do, it is hard to argue that they were good politics. If such policies are nevertheless pursued, they "must" fulfil an end other than the prolongation of the tenure of power. When Peter the Great brought in Germans to run Russia, made himself odious and ruthlessly upset the old ways, he was using up power in the short run (he had a margin to spare) even if the longer-run effects strengthened the throne (which is arguable).
A parallel should make the distinction clearer still. Conceptually, we are used to the idea of "subsistence wages." Marx has built his whole unfortunate theory of value and capital on the idea of the labour-time "socially necessary" for the reproduction of labour. Only a part of the labourer's time is used up to produce the subsistence he needs to go on labouring, and subsistence is all he gets.*36 No matter that subsistence turns out to be impossible to pin down. As an idea, it is simple and powerful and it leads straight to surplus value and the class struggle. In our framework, the use of the power necessary for its own maintenance takes the place of the subsistence wage spent on the maintenance of the labourer. The surplus value that his labour time has produced in addition, accrues to capital as the pay-off to domination. In our scheme, "surplus value" would correspond to whatever satisfactions the state can afford to procure for itself over and above the maintenance of its tenure of power. Another, less "analytical" parallel is that between income and discretionary income, power and discretionary power.
Discretionary power is what the state can use to make its subjects listen to Bach and not listen to rock; to change the course of mighty rivers and transform nature; to build presidential palaces and government offices in keeping with its taste and sense of proportion; to deal out rewards and privileges to those who deserve it and to keep down those who deserve that, regardless of political expediency; to do good and aid causes its subjects care little about; to pursue national greatness; to invest in the well-being of a distant posterity and to make others adopt its values.
Our theory would not be a social theory if it had no sting in its tail, no indirect, roundabout secondary effects and no "feedback loops." Thus, it is entirely likely that once the state has made people observe the cult of Bach, and they have in due course taught themselves to like it, they will "identify" better with the state which gave them their tastes. Likewise, the splendour of the presidential palace, the achievement of national greatness and "being first on the moon" may in the end implant in the public consciousness a certain sense of the state's legitimacy, a perhaps growing willingness to obey it regardless of hope of gain and fear of loss. Hence, they may serve as a cunning and slow-acting substitute for buying consent. Like Peter the Great's administrative reform, however, they require a discretionary margin of power now even if they are certain to yield greater legitimacy or a stronger repressive apparatus or both later.
Instead of saying, tautologically, that the rational state pursues its interests and maximizes its ends, whatever they are, I propose to adopt, as a criterion of its rationality, that it seeks to maximize its discretionary power.*37, *38
Discretionary power permits the state to make its subjects do what it wants, rather than what they want. It is exercised by taking their property and liberty. The state can appropriate people's money and buy things (including their services) with it. It can also override their spontaneous intentions and order them to serve its purposes. When the state is defending its tenure in open competition, however, all the property and liberty it can take is, by the definition of competitive equilibrium, absorbed in the "reproduction" of power, i.e. in the maintenance of its tenure by redistribution. The existence of a discretionary surplus would contradict the assumption of competition, under which it is impossible so to rearrange or enrich the redistributive pattern as to obtain more support for it (cf. the earlier section of this chapter on the "profitless," break-even character of equilibrium). This condition loses some of its precision and rigidity as we move to lower levels of abstraction; we introduce fuzziness, a margin of error, but no novel set of reasons to render likely the emergence of an appreciable discretionary surplus.
At this point, the state has completed its unwitting transformation, from being the seducer freely offering utilitarian improvement, one-man-one-vote and distributive justice, to being the drudge only just coping with its self-imposed redistributive obligations. Moreover, it has entrapped itself in several predicaments at once. One is competition, being on the treadmill. Another is the changing character of society in response to its own redistributory activity, notably addiction to aid, free-rider behaviour by each interest group towards all others and progressive loss of control over redistribution. An extreme form of this predicament is to be up against an "ungovernable" society. Finally, as direct redistribution is overlaid by ever thicker layers of churning, in the ultimate democratic predicament there is no possible equilibrium: society both demands and refuses the state's redistributive role. The latter, in maintaining consent, ought both to go on expanding and to "roll itself back."
Were we to dismiss this terminal self-contradiction as mere dialectic word-play and allow equilibrium to persist, however, the latter would still not represent a proper maximum for the state, except in the tenuous sense in which the earning of the subsistence wage is a "maximum" for the labourer. With no, or negligible, discretionary power, the state is better off than in any other available posture, in each of which it would lose power altogether and be replaced by its opposition.*39 It is rational for it to cling to this position. It may well content itself with it and just soldier on. Nevertheless, if it could deliberately change some of the available alternatives, i.e. modify in its favour the social and political environment to which it adjusts when "maximizing," it could make itself better off. Recognition of some such possibility (though not necessarily any action to realize it) may in fact be regarded as a criterion of another, higher order of rationality. Making itself less dependent on its subjects' consent, and making it harder for rivals to compete, would amount to improving the environment instead of adjusting to it.
It is not, of course, actually irrational for the state not to do this. I am not arguing some historical necessity, some inexorable dynamics which must cause any state, if sound of mind, to become totalitarian. On the other hand, I would not accept that, like Plato's Republic on its way from democracy to despotism, the state "degenerates" in the process. If it has improved its ability to fulfil its ends, it has not degenerated, though it may well have become less apt to serve the ends of the observer, who would then have every reason to be alarmed by the change. I am arguing, though, that it is rational in a higher, "strategic" sense of rationality different from the "tactical" sense of optimal adjustment, for the state generally to become more rather than less totalitarian to the extent that it can get away with it, i.e. maintain majority support at the stage where it still needs it. It is also rational for a rival for power to propose, under democracy, a more totalitarian alternative if this is more attractive to the majority though more unattractive to the minority.*40 Hence, there is in competitive, democratic politics, always a latent propensity for totalitarian transformation. It manifests itself in the frequent appearance of socialist policies within non-socialist government and opposition programmes, and in socialist streaks in the liberal ideology.
Whether or to what extent this potential is realized is a matter almost of hazard, of the fundamentally unpredictable historical setting. By neat contrast, no potential the other way round, for the democratic transformation of a totalitarian state, can be logically derived from any maximization assumption that would admit of the state having the kind of ends, whatever they are specifically, whose attainment calls for the discretionary use of power.
Notes for this chapter
Historiography tends to deal more satisfactorily with states appearing in the shape of kings and emperors than with states which are faceless institutions. All too often, the latter are confused with the country, the nation; the historical driving force springing from the conflict between state and civil society is left at the edge of the field of vision. When the game is Emperor vs Senate, the king and his burghers vs the nobility, or the king vs established privileges and "ancient freedoms," historians are less apt to make us lose sight of which interests make the state do what it does.
In modern parlance, the labourer has "maximized" when accepting to work for subsistence wages. No better alternative was offered to him. A different, more "strategic" sense of maximization, however, would have him attempt to influence the available alternatives. He could try to organize a union and bargain collectively, or strike. He could seek redress in "distributive justice" through the democratic political process. He could also fall in behind the "vanguard of the working class" and join the struggle to modify the "relations of production."
If it takes the application of a fixed "amount" of power to stay in power, with the surplus (if any) available for exercise at discretion, anything which maximizes power must also maximize the discretionary surplus. The fastidious may therefore wince at "discretionary power" as the maximand; why not just plain power?
However, the convenience of a built-in separation between "being in power" and "using power to freely chosen ends" seems to me to outweigh the inelegance of the solution. If the maximand is discretionary power, we can describe competitive equilibrium in politics as the position where discretionary power is nil. This has the didactic merit of rhyming with the position of the perfectly competitive firm whose profit is nil after it has paid for all its factors of production.
Political theory, as we have seen, asks questions of a teleological nature and treats the state as an instrument: What can states do for their citizens? What ought they to do? What are the obligations and limits of civil obedience?, etc. I know of only two serious precedents of attributing a maximand to the state itself. Both do so in the context of theorizing about the production of public goods. One is Albert Breton, The Economic Theory of Representative Government, 1974. He postulates that the majority party will behave so as to maximize a function increasing in some way with the chance of re-election, power, personal gain, image in history and its view of the common good. The other is Richard Auster and Morris Silver, The State as a Firm, 1979. Here the maximand is the difference between tax revenue and the cost of the public goods produced by the state. Auster and Silver hold that unlike monarchy or oligarchy, democracy amounts to "diffuse ownership" among politicians and bureaucrats, and hence there is no residual income-recipient to profit from a surplus of taxes over the cost of public goods (leading to their over-production). I would interpret this to mean that in democracy there is no "maximizer."
Note also, as examples of an approach which proceeds, so to speak, from the "producer's" motives rather than those of the "consumer," W. A. Niskanen Jr, Bureaucracy and Representative Government, 1971, where "bureaux" seek to maximize their budgets, and B. S. Frey and F. Schneider, "A Politico-Economic Model of the United Kingdom," Economic Journal, 88, June 1978, who find that when the government is unpopular, it pursues popular policies and when it is popular, it indulges its own ideology.
Formally, discretionary power would have become negative in such postures, hence (total) power would be inadequate to ensure its own maintenance; the tenancy of the state would change hands.
Such proposals reach beyond the bounds of the simple sort of electoral competition set out earlier in this chapter. In addition to promising the majority the minority's money (equalizing incomes), they might, for instance, include the equalizing of schools (Gleichschaltung of education) or the equalizing of "economic power" (nationalization of the "means of production"), or some other property, privilege, immunity of the minority, including its creed (Huguenots, Mormons) or race (Jews). . State Capitalism
End of Notes
Return to top