[1] John Rawls, A Theory of Justice, revised edition (Cambridge: Harvard University Press, 1999), p. 62.
Justice and Anarchy: on a Contradiction in Rawls
By Crispin Sartwell
In my view, enough and more has been written about Rawls's A Theory of Justice. Indeed it has been attacked, refuted, and revived so many times from so many different angles that it would be, so to speak, rational to replace the volume on the shelf and not pull it down until some decades or perhaps millennia have passed.
However, I regard the criticism that follows as fundamental, both as exposing a contradiction or at least an extreme tension in Rawls's account and as showing something fundamental in political philosophy: namely, the extremely problematic nature of state power with regard to the matter of justice, the conceptual tendency of state power to produce injustice. The objection can be stated briefly as follows. Political power or authority is, by Rawls's definition of the good, a good for every person. Indeed, it is a fundamental good in the sense that it is a presupposition of other goods; it is necessary in order to be secure in the other goods, such as material welfare and self-respect, that Rawls - and, one would think, anyone else - regard as important. However, the political power that is constituted in Rawls's account of the decisions of the people in the original position to achieve justice also entails that they constitute an inequality of political power so severe as to compromise utterly the security of the other goods that they rationally want to achieve. Another way to put this is that there is a tension within the principles of justice concerning equality that Rawls generates: the power constituted to achieve a just degree of equality (qualified by the difference principle) is incompatible with that very equality.
Primary Goods
Rawls writes
For the present, it should be observed that the two principles [of justice] (and this holds for all formulations) are a special case of a more general conception of justice that can be expressed as follows.
All social values - liberty and opportunity, income and wealth, and the bases of self-respect - are to be distributed equally unless an unequal distribution of any, or all, of these values is to everyone's advantage.
As a first step, suppose that the basic structure of society distributes certain primary goods, that is, things that every rational man is presumed to want. These goods normally have a use whatever a person's rational plan of life. For simplicity, assume that the chief primary goods at the disposition of society are rights and liberties, powers and opportunities, income and wealth. (Later on in Part Three the primary good of self-respect has a central place.) These are the primary social goods. Other primary goods such as health and vigor, intelligence and imagination, are natural goods. . . . Imagine, then, a hypothetical initial arrangement in which all the social primary goods are equally distributed. This state of affairs provides a benchmark for judging improvements. If certain inequalities of wealth and organizational powers would make everyone better off than in this hypothetical starting situation, then they accord with the general conception.[1]
The serial or "lexical" ordering of Rawls's principles of justice, he remarks, does not allow an exchange of rights and liberties for other social goods, such as "economic or social gains" (63).
It is interesting that Rawls in this passage includes "powers" among the primary goods, a theme to which he does not return. But surely, given both the general characterization of goods he gives above, and the more specific formulation he gives later (pp. 399 ff.), powers, or at any rate certain powers, are primary goods. Indeed, like liberty and opportunity, power may be said to be a primary primary good, in the sense that it stands as an empirical precondition for the pursuit of various other goods. Specifically, political power should be seen to be a primary good. Notice, for example, that wealth, security, opportunity, and self-respect are historically connected to political power. I can think of no society, for example, wherein the group that exercises primary political power is impoverished, even in cases wherein the society as a whole is impoverished. In such circumstances, indeed, the strongman is usually stashing the country's wealth in his Swiss accounts.
And it is a familiar point, as well, that exclusion from political power is incompatible with self-respect or with educational equality, for example. This sort of criticism, for example, was fundamental to the civil rights movement, for example as regards voting rights. Martin Luther King, Jr. and many others argued that without the minimal political power exercised through equal access to the electoral process, black people in America could not make themselves secure in any other way. Exclusion from political participation, they also argued, compromised self-respect and dignity among black folk, as did the system of apartheid in schools and other social contexts. African-Americans came to achieve legal equality in such respects through the exercise of political power in a variety of ways.
The Problem
Thus it is clear that by Rawls's own account, political power is a primary social good. That is, political power is obviously something that a rational person would want, and it is subject to procedures of social distribution.
The reason that, within Rawls's system, political power is subject to social distribution is because, on his account, the persons in the original position must constitute a state. Rawls, I would say, is not perfectly clear on this; yet it is indeed perfectly clear. As Nozick argued elaborately in his libertarian or minarchist critique of Rawls, Rawls must constitute a set of institutions with the power to make an initial distribution of goods, and to oversee the future of distributions in order to keep them roughly in line with the principles of justice. This will require, at a minimum, that the institutions thus constituted have the power to take certain goods from some persons and give them to others, in order to enforce equality or liberty and opportunity, and to ensure that private exchanges or confiscations of other goods redound to the benefit of all.
In Rawls, the political institutions or institutions of the state appear to be, as we might put it, impersonal. They are portrayed as forces or procedures (in particular, structures of rules) rather than as consisting (in part) of a group of persons. However, surely the state consists in part of actual human individuals, individuals who are themselves part of the population to whom rights, liberties, wealth, opportunities, powers, and other goods are to be distributed (in this case by themselves) according to the principles of justice. If powers are indeed goods, and if political power in particular is a primary social good - which is obvious - then the assignment of political power to these persons is itself a distribution of goods which is subject to the principles of justice. But to the extent that the individuals who operate state power are necessarily empowered politically to an extent that others cannot be, the distribution of a political power is necessarily extremely unequal. Then the question of whether it is just or not would appear to turn on whether this unequal distribution redounds to the benefit of all.
The situation appears offhand to be extremely unpromising. The distribution of political power in a situation where a group of individuals constitutes the state is unequal. But the political power is itself the power to distribute other goods. To the extent that people tend to pursue their self-interest - that is, to the extent that they seek the greatest possible quantity of goods, even if this desire often observes certain limits - we should expect that the resulting distribution of all goods will favor those who exercise political power. Indeed, Rawls's own view is precisely that rationality consists in the pursuit of the good. I think that is certainly false, or at a minimum incredibly simplistic, but if it is anything like true, then we should expect that the asymmetrical or prima facie unjust distribution of political power will be followed by an unjust distribution of all other social goods.
Furthermore, the situation is fundamentally irremediable. Since the distribution of goods is made by those who have already received an unequal share of the power to make such distributions, there is no recourse short of continuous insurrection. Indeed, the power to distribute will increase, as it is put into the hands of those who distribute it, and the more unequal the distribution becomes, the more unequal it becomes, unto eternity.
This is, I believe, a conceptual incoherence at the heart of A Theory of Justice. But it is, of course, anything but an academic problem. It could not be more obvious that the ruling elite - class, race, gender, religion, caste, and so on - of any given society assigns itself an inordinate share of the menu of goods. And even if one distributed political power, not according to any of these arbitrary categories, but according to desert, or by lot, for example, the people thus assigned would quickly constitute an elite, a class, a caste, and so on. That is, according to Rawls, the persons in the original position have already constituted a profoundly hierarchical society, and one which, other things being equal, will become ever-more hierarchical in the fullness of time.
Yet another way to put this is that the original contractors cannot constitute a polity that realizes the principles of justice without constituting a polity that massively violates them. Now on the other hand, there may be no recourse. That is, the asymmetrical assignment of power and resulting hierarchical distribution of all goods - as a matter of fact typical of all state-ridden or state-addled societies (here I resort to the pejorative, obviously) - may be the best we can do. That is, it may be to the benefit of everyone. This leaves us in a position in which no social situation can be expected to produce anything but profoundly unjust distributions of goods. That may of course be the case, and perhaps we ought merely to face up. Or for that matter, give up.
Just Institutions and Coercion
It is truly remarkable that, with regard to this massive work of political philosophy, neither 'power' nor 'state' appear in the index. Indeed, any general arguments that Rawls gives for the legitimacy of state power are, we might say, superficial, and A Theory of Justice, we might say, deploys itself within the presumption of the legitimacy of state power, rather than, as in traditional social contract theory, devoting itself to the task of trying to establish it. Nevertheless, it is clear that Rawls constitutes a coercive state. This entails that among other things the original contractors are distributing political power, or at any rate setting out principles or procedures according to which it will be distributed.
Rawls defines "institutions," including state institutions, in terms of rules and behavior in accordance with these rules rather than in terms of powers of some persons over others. That is all very well, but these rules themselves concern among other things the conditions under which some people can legitimately coerce others. That is, though at least in ideal circumstances the group of people who operate political power do so according to a set of rules and procedures, they do so through coercion; they possess political power that rests on force, force to which the people not part of the institutional apparatus have no access.
It is reasonable to assume that even in a well-ordered society [wherein principles of justice are universally agreed on] the coercive powers of government are to some degree necessary for the stability of social cooperation. . . . [E]ven under reasonably ideal conditions, it is hard to imagine, for example, a successful income tax scheme on a voluntary basis. Such an arrangement is unstable. The role of an authorized public interpretation or rules supported by collective sanctions is precisely to overcome this instability. By enforcing a public system of penalties government removes the grounds for thinking others are not complying with the rules. For this reason alone, a coercive sovereignty is presumably always necessary. (211)
Rawls indeed argues that the original contractors will want to constitute a "paternalistic" power, that being rational they will want to hedge against the possibility of their becoming-irrational through mental illness or addiction, for example: "the parties adopt principles stipulating when others are authorized to act in their behalf and to override their present wishes if necessary; and this they do recognizing that sometimes their capacity to act rationally for their good may fail, or be lacking altogether" (219).
In doing these things, the parties constitute other persons as political authorities, not only to enforce rationality on themselves, but to determine whether or not they are in fact rational. They constitute others, as well, as authorities over their goods, for example their incomes. Such authority is to be exercised according to the rule of law, which makes the persons thus authorized subject to the rules they enforce. This, I submit, is impossible, in that in being constituted as the authorities who, for example, determine the extent of rationality of citizens, these authorities are by definition exempted from judgments concerning their own rationality. (Rather, there may be a hierarchy of such paternalistic authorities, each level of which answers as regards their rationality to those who operate at the higher levels; one level must be exempt from anyone's judgment as to its rationality.) And in non-deal circumstances, the parties are simply placing all their social goods at the mercy of other parties.
But my argument here does not rest on such observations, richly confirmed though they are by experience and history. What concerns us here is that, since the segment of the population that wields political authority has access to coercive power that is, as is evident, sufficient to redistribute social goods and to strip them entirely from those it judges irrational, this coercive power is sufficient to be practically irresistible; the distribution effectively distributes all political power to one segment of the population while stripping it entirely from another. In order to assure a just and rational distribution, Rawls's just institutions pre-suppose a radically unequal distribution of a particularly fundamental social good.
That is, the distribution of social goods in accordance with the principles of justice is literally impossible; the distributive mechanism or institution contradicts the principles on which it is itself supposed to make that distribution.
Anarchy
If indeed the original contractors were making a distribution of political power, they could not constitute some of their number as political authorities with access to coercive power. Thus I assert that they would have no choice, in the original position, but to choose anarchy. This is evident if we keep in mind that political power is the mechanism by which other goods are distributed. Once they have distributed political power in a radically unequal fashion - investing some with it entirely and stripping others of it completely - they would expect that the distribution of other goods would come to mirror the distribution of coercive power.
The fact that, once political power is seen clearly to be a social good, Rawls's original contractors would be constrained to choose anarchy would seem to constitute a reductio ad absurdum of his theory of justice. This would be true only if anarchism is absurd.
Crispin Sartwell
Associate Professor
Dickinson College
Carlisle, PA 17013
717 245 1474
sartwelc@verizon.net
[1] John Rawls, A Theory of Justice, revised edition (Cambridge: Harvard University Press, 1999), p. 62.