Andrew Robinson - Theory Blog

Monday, November 15, 2004

Why Rawls is Wrong


This paper is part of an ongoing critique I am developing of the work of the political theorist John Rawls. I will start by introducing this theorist for those of you who perhaps aren’t familiar with him because you come from political science backgrounds. John Rawls was probably the most famous living political philosopher prior to his recent death, and he was in effect the founder of the dominant tradition in Anglo-American political thought. He identifies as a liberal, and associates himself with the traditions of Kant and Mill. His first major work was A Theory of Justice, in which he developed an original normative theory based around the concept of “justice”. His methodology consists of going into a state of thought he terms “reflective equilibrium”, and reflecting on what he calls “our considered convictions”. While in this state, he claims to be able to develop a theory which, in his vocabulary, can “account for” these convictions, which in practice seems to mean to give them an overarching and systematic structure. He expresses his theory through an analytical device called the “original position”. This is an openly fictive version of the old idea of the “social contract”. By modelling people in a certain way, he suggests he can deduce what people would agree to in an ideally fair situation, and deduces his principles of justice from this situation. The principles he actually favours are, firstly, the prioritisation of what he calls a “scheme of equal liberties”, and secondly, the difference principle, which asserts that inequalities are permissible only if they benefit the worst-off group in society. In his later work, such as the book Political Liberalism, Rawls portrays his work as the development of what he terms a “public political conception”. This is to be a limited conception able to construct unity among people with diverse beliefs and ways of life, which constructs social unity around what he calls an “overlapping consensus”.

Much of the secondary literature on Rawls has concentrated on debating the specific principles he sets out and developing alternative versions of liberalism with slightly different basic principles. My own critique, however, is coming from somewhere quite different. As some of you already know, I am currently working on a general theory of oppressive discourse, and in my view, a number of the forms of discourse I consider to be oppressive arise within Rawls’s work. The operation of these oppressive forms gives an exclusionary impulse to his work, and its resultant oppressive character has been missed by most of the secondary commentators. In fact, I don’t just have one critique of Rawls, but rather, a string of accusations, including the following:

Firstly, that his idea of “our” considered convictions involves an excessive generalisation of his own personal beliefs and an uncritical attitude to common sense;
Secondly, that his work is motivated by a systematising drive, which subordinates all other drives and which turns his theory into a means of pursuing a controlled and coordinated society;
Thirdly, that there are power asymmetries built into his concept of the “basic structure”;
Fourthly, that the “original position” and other ideas in his theory rely heavily on discourses of self-alterity;
Fifthly, that his concept of the “reasonable” is badly defined, and operates mainly as a way of anathematising opponents as “unreasonable”;
And sixthly, that his methodology involves an overcoding operation which cannot in any sense be called “accounting for” beliefs.
However, I do not have time to run through all these critiques today, so I shall concentrate on the most important criticism I have of Rawls, which in my view is my critique of his conception of the person.

In my view, this conception is a form of essentialism and also a Barthesian myth. These two forms of discourse are in my view oppressive, for the following reason: they render particular people voiceless by reducing these people to a particular symbolic position. A mythical discourse occurs when actual phenomena are understood through the projection into them of a second-order signified through the double use of an already-constructed sign to connote an implied ideological meaning, and essentialism occurs when actual phenomena are reduced to a linguistically-specified essence, to the exclusion of supposedly accidental characteristics. For instance, people are only included in social relations on condition they conform to the stipulated model of the essence, which is in fact constructed outside of actual people and often over and against them. In my view, it would be quite surprising if Rawls’s methodology actually yielded an inclusive theory, because it is clearly deduced without much reference to what people actually think, need, or desire, and is postulated as what Rawls calls a “freestanding” view which interacts with actual people as if from the outside. We shall see shortly how Rawls manages to convince himself that such an externally-constructed theory is nevertheless inclusive.

The language Rawls uses in his work is typically inclusive, but on closer inspection one usually (though not always) finds that the inclusion is qualified by a reference to “persons” or “citizens”. A claim that, for instance, “all persons are to have basic liberties” seems at first sight to be inclusive, but one has to bear in mind that it is inclusive only of those who fit into the category “person”. If Rawls gives this a technical definition, the same phrase permits the denial of basic liberties to those who are not properly “persons”. Rawls’s definition of the “person” is indeed quite technical. He constructs an image of “the person” as an essence of all people, floating above the complexities and contingencies of everyday life. Hence, he says that a claim to speak for all persons is ‘too broad, unless we suppose that they are in their nature basically the same’ (CW 608). The everyday concept of the person is too ‘vague’, so Rawls introduces a precise concept to ‘crystalise’ it (CW 357). A person is not identical to a human being, and is derived from philosophy (JAFAR 24, PL 18). Thus, when Rawls discusses what people “would” agree in the original position, he is really talking about what they “must” believe in order to qualify as persons. His theory is, by his own admission, fair to persons conceived according to his model, and not to what actual people think or feel (PL 40). Further, what they “would” agree is taken to be morally binding, regardless of what anyone actually thinks, and the power to ensure that it comes about is alienated into repressive state apparatuses.

What Rawls terms his “political conception of the person” is an attempt to establish a “shared nature” without reference to empirical or theological claims. It is something akin to a Kantian noumenal self, though limited to a public political conception. Rawls terms it an ‘Archimedean point’ from which to assess society (TJ 511) as well as directly declaring it to be people’s ‘essential nature’ (CW 467). He openly admits that the validity of his entire theory depends on his theory of the person (PL 93-4). It is crucial to realise that, by Rawls’s own admission, only the characteristics of persons defined in Rawls’s way are relevant to his theory of justice (CW 317). This certainly suggests that it is operating in a mythical and essentialist way. Rawls ‘start[s] by assuming’ the essence (CW 365), which is by his own admission ‘given in advance’ and not empirical (CW 367), and then uses a ‘strictly deductive’ procedure to derive political principles from this essence (TJ 104). He also admits to having derived the conception from what people think they are, rather than what evidence suggests they actually are (e.g. CW 406, LN 33, PL 300), even though this would seem to contradict his claim to reject any reliance on false beliefs and useful illusions (e.g. CW 326, TJ 368, JAFAR 121). His procedure is to start with the conception of the essence and deduce his theory from it. This leaves no room for any other voices which exceed the limits of the essence. However, he continues throughout to use “persons” interchangeable with “all” and “everyone”. Thus, it is patently possible to be a “no-one” in relation to Rawls’s theory. Rawls has the strange idea that relying on such an essence will eliminate social conflicts and divisiveness (e.g. PL 330), but actually, it simply displaces issues from the sphere of interpersonal difference into the sphere of the definition of “the person”.

The conception of the person is constructed by exclusions. Justice is to orient to a person’s ‘nature as a free and equal rational being’, instead of ‘to the specific things he happens to want’ (TJ 220). The “veil of ignorance” which restricts the information available in the original position is a way of filtering out everything except the essence. Justice is to exclude anything one need not have in order to be a “person” in Rawls’s sense (TJ 222-3). It is according to Rawls a way ‘to express most fully what we are or can be’, as differentiated from ‘lower order’ impulses similar to those of animals (TJ 225). Therefore, Rawls states that ‘to be held accountable to the principles of justice… does not stunt our nature’, but rather, realises our nature at the expense of the ‘narrower impulses’ it ‘enables us to control’ (TJ 403). Thus, both externally-mandated ethics and mechanisms of state control are used to ensure that people conform to an essence, conceived a priori as one’s real or superior nature. It is crucial to realise that Rawls as theorist, and the coercive agents he supports, claim to know this essence from the outside, without reference to actual people, and further, are prepared to use force to ensure that people do not deviate from the essence. This essence is, by Rawls’s admission, not something people would desire unaided (JAFAR 56).

Rawls’s “person” has a whole string of specific characteristics which operate to specify who is included in his theory. For instance, a “person” has two higher-order interests in exercising the so-called moral powers. Other beliefs are taken to be ‘hierarchically subordinate’ to these interests (CW 372). He states that: ‘The aims and conduct of citizens in society are… subordinate to the conception of citizens as free and equal persons’ (PL 366). The characteristics include being “free” and “equal”, “reasonable” and “rational”, having the two so-called “moral powers” and being a “normal and fully cooperating member of society”, as well as entering into reciprocal relations with others through society as a fair system of cooperation. All these characteristics are subject to more-or-less precise definitions in Rawls’s work, and some seem to operate tautologically. For example, one of the “moral powers” is a “sense of justice”, and the way one knows if someone has a “sense of justice” is by observing whether they act on and use the ideas of the principles of justice. Therefore, if Rawls strictly speaking deduces his theory from the assumption that persons have a sense of justice, he is deducing his theory from his theory. It seems in fact that several of the criteria are displaced or reified representations of Rawls’s own views, so that one has to agree with or think like Rawls to be a citizen. Having the various characteristics of a citizen are a precondition for being included as a member of society in Rawls’s model (PL 302). Only moral persons, in Rawls’s special sense, have a right to equal justice (TJ 441-2). Only moral persons are represented in the original position (CW 305, 310, 317), and Rawls’s model of society is that it is a system of cooperation between moral persons (even though it clearly may also contain people who do not conform to the essence). Primary goods, which are the kind of goods Rawls thinks it can be just to claim, are constructed solely by reference to moral persons, and express things that moral persons, as opposed to actual people, do or would want (PL 308).

In some of his later essays, Rawls denies having a conception of the essence of the person (CW 388). This claim should not be taken seriously. Its basic meaning seems to be that Rawls’s conception of justice only applies in public, and is not binding in private on the doctrines which form part of an overlapping consensus. However, since citizens are supposed to prioritise their public identity when the two conflict, the “political” conception of the person remains in effect a general essence. Persons are ‘citizens first’, and this is their ‘basic position’ (LN 66, JAFAR 166). Further, the desire for justice and the so-called “higher-order interests” are supposed to be primary over all other desires (TJ 462, CW 313). One still finds Rawls writing of ‘citizens’ essential interests’ in Political Liberalism (PL 134).

The exclusionary character of Rawls’s assumptions is rarely stated clearly, but this is mostly because he uses obscuring formulations such as “persons are to be regarded as free and equal”. At best this means that people are to be treated as if they express the essence, regardless of what they actually are or do. Rawls does not make clear what he means by “to regard someone as something”, and I have discovered at least a half-dozen different versions of what it means in various places, ranging from it being an empirical characteristic of human beings which is isolated for analytical purposes, to it being an ideal, to it being a standard for assessing actual people. The differences between these ways of presenting his position do not alter its mythical and oppressive character. For instance, an “ideal” would not seem to require that people be treated as if they conform to it, and a simplification is only valid if it extrapolates from actual rather than imagined characteristics. His assumption seems to be that enshrining his image in institutions will in fact bring it into being, and this assumption is unfounded. That actual people are confronted by institutions which assume we are all Rawlsian moral persons would not make people in fact become such persons, any more than being confronted by institutions which assume us to be dogs would make us all dogs. It is to be emphasised that Rawls explicitly derives from his assertion that people “are” moral persons an imperative to normalise people to fit this model (e.g. JAFAR 175), so that its actual working would be impositional. Therefore, the claim that people already “are” what Rawls claims is a cover for such imposition. On occasions, he admits directly what is implicit in his approach: that his model of the person is a ‘conception of moral worth’ (JAFAR 142) regulating who is and is not entitled to public voice and resources. Being socially included is dependent on conforming to Rawls’s model of one’s essence. Hence, it is only by having the moral powers that people qualify as equal members of society (PL 19), and Rawls states that ‘[t]he task is to articulate a public political conception of justice that all can live with who regard their person and their relation to society in a certain way’ (CW 306). Thus, it constructs an exclusionary in-group assumed on a dogmatic and undeclared basis to be superior to out-groups.

The essence of the self operates as a hurdle over which actual people must jump to be heard in Rawls’s “public” discourse. Everything Rawls offers is conditional on this essence: for instance, the basic liberties are only to be valued because of their significance for the moral person, and some liberties are limited to those cases most important for this essence (JAFAR 113). Actual people find themselves trumped by a series of essentialist claims which may or may not have any relationship to their actual identity, needs or desires. The essence floats above actual situations and bears down on them like a lead weight. People who are not sufficiently close to the essence are to be excluded and normalised. In effect, Rawls permits people who conform to his model of the “person” to carry out an exclusive territorialisation of the world at the expense of all others. This is all the more disturbing for the fact that “citizens” are assumed to want the existence of a just state, above all else. Therefore, in practice, people find their claims assessed in many cases on the grounds of how useful they are to the liberal state. If one does not accept Rawls’s assumption that normal, submissive conformists are the only valuable type of human being, one also cannot accept the rest of his theory. Rawls’s conception of the person is essentialist, and reminiscent of Stirner’s critique of humanism: ‘the state betrays its enmity to me by demanding that I be a man… [I]t imposes being a man upon me as a duty’ (Stirner p. 179).

One of the characteristics of “persons” is “rationality”. This term has a specific meaning in Rawls’s theory, and an ambiguous status. It is treated as a kind of zero point of human subjectivity. One ‘quite properly’ follows one’s rational interest, he claims, and one is thereby ‘above reproach’ (TJ 371, 21), at least so long as one also acts justly. According to Rawls, ‘everyone accepts’ such self-interestedness, provided it is contained by justice (TJ 277), and it is ‘unreasonable’ to expect people to act in a directly ethical way (PL 267-8). People are to be left “free” to pursue their aims rationally (PL 268), with justice operating as a limit on, not a challenge to, this rationality.

Rawls’s account of rationality includes some elements drawn from rational choice theory, some associated with what Barthes calls “moral accounting” and some associated with his conception of the planner self. Implicitly, Rawls suggests that one must conform to all these aspects in order to be included in his theory. “Moral accounting” includes aspects such as the counting and ranking of desires, ordering, scheduling, balancing, adjusting and maximising (PL 83, LN 88, TJ 364) like a good servant of King Abacus. I am using the term “planner self” to refer to Rawls’s assumption that rational interests take the form of a “rational plan”. In his terms, ‘[w]e are to see our life as one whole, the activities of one rational subject spread out in time’, and ‘a person may be regarded as a human life lived according to a plan’ (TJ 369, 358). A plan has to fit a whole string of further criteria to qualify as a rational plan, which include not contradicting established truths, accepting the “normal conditions of human life” and repressing desires which interfere with it. Its basic model seems to be that of a packaged commodity, whether this be a job, a set of commodities or a packaged belief-system. The rational plan is to overcode and coordinate desires, and also establishes the basis for a molar conception of one’s selfhood. It reconstructs desire as “rational interests”, a process Rawls endorses. He says that ‘[p]eople must have interests - otherwise they would either be inert and passive, or likely to be swayed by unreasonable and sometimes blind passions and impulses’ (CW 44). In other words, Rawls does not simply assume that people are rational, but introduces a moral imperative that one be rational (in his sense), as part of his idea of the essence of the person. Someone who succeeds in a rational plan achieves complete happiness (TJ 482), and for Rawls, happiness can be defined objectively, in relation to one’s rational plan (TJ 481). (He also assumes that rational interests can be deduced externally and are “objective”). Rawls has to perform various manoeuvres to include in his model of “plans” the possibility that it may be irrational to devote too much time to planning and too little to acting, conjuring an image of people rationally calculating how much time to spend rationally calculating (TJ 367). Although the “just” are not supposed to be pure rational egoists, they are supposed to be structurally atomistic. Interests ‘must always be interests of a self’ (TJ 111). Even the concern for others which is expressed in the veil of ignorance is reconstructed, through the idea of the original position, as fictively egoistic. In other words, this way of imagining the self is a central tenet of Rawls’s theory, even if Rawls problematises it as a factual postulate.

One should keep in mind that, as part of the conception of the person, the concept of the rational is a condition of inclusion. Rawls hints that his model is to include rational persons and only rational persons (e.g. TJ 350, 387), and the “irrational” arises as an anathema. Justice is rigged so that it favours “rational” people over “irrational” ones. Therefore, the “rational” cannot simply be an unfortunate fact, but must be a positive ethical value. Those who are irrational are excluded. On the other hand, Rawls treats the rational as something unfortunate and as a reason for distrusting individuals. It is the reason that people need an overarching state to keep them in line, and it is the reason why one cannot demand equality of outcomes. It is the source of the various ‘tendencies to injustice’ which are the reason why Rawls leaves people so little effective autonomy (e.g. TJ 4, 110, 338, JAFAR 185). This is a self-contradiction, and requires that people flagellate themselves for being what the social system also demands them to be. Frijthof Bergmann has written an excellent critique of the role of egoism in Anglo-American political theory, suggesting that the assumption of self-interest is inaccurate and leads to a mistaken support for extended social control. Rather, people are already too prone to submit to others, and need instead to become more autonomous. Another effect of assuming people to be rational egoists is that it makes one suspicious of their claims. A good practical example of this is provided in Roland Barthes’s analysis of the Dominici trial. The ‘psychology of the masters’ allows the judge to condemn Dominici by taking him as an object. The law ‘is always prepared to lend you a spare brain’ so it can condemn you for being ‘as you should be, and not as you are’, for a psychology it has itself projected or imposed (Mythologies 43-6). Rawls’s assumption of bourgeois rationality is also a good way of avoiding issues to do with politics, psychology and everyday life.

Another characteristic of “persons” is “freedom”. Of course, the meaning of “freedom” depends on what exactly it is “freedom” from or to. In Rawls’s lexicon, it turns out to mean “freedom” for the essence of the self, from the rest of the self, including one’s desires, ends, needs, and beliefs. Rawls regards ‘individuals as free and responsible masters of their aims and desires’ (CW 266), echoing Nietzsche’s discussion of those whose desire to dominate turns inwards on themselves. One has ‘freedom’ if one can make demands to meet one’s highest-order interests (CW 330). This “freedom” includes being viewed as capable of changing aims and desires at will, whether one is capable of this or not (JAFAR 21, TJ 131-2, 331-2, CW 260). This turns out to mean that one is “free” if the social system can demand that one change one’s beliefs and desires so as to conform to the public identity imposed by the state (PL 30, TJ 131, CW 381). Therefore, as in 1984, “freedom equals slavery”. Freedom for the imagined essence of the self equals slavery for the actual self, dominated by social institutions which claim to express the essence. This is the “freedom” of the so-called “free” wage-labourer of capitalism, “freed” of any commitment which exceeds capitalism and therefore “free” to take the role assigned by this system. “Persons as free” also means that “persons” are sources of legitimate claims, although such claims may nevertheless fail, and are defined as an other in relation to ‘self-mastery’ (TJ 205). The right to be a source of legitimate claims turns out simply to mean that the laws passed by the state are justified by reference to one’s interests (JAFAR 23-4). In other words, one is not really a source of claims at all. These claims have always-already been made, registered and assessed via one’s “interests”, constructed and assessed in alterity.

Freedom also means “responsibility for ends” (PL 33). This is an even clearer case of freedom meaning slavery. Citizens are to be ‘at liberty to take charge of their lives’ - which in practice means being ‘expected… to adapt… to [one’s] expected fair share of primary goods’, as fixed by the social system (PL 189-90; c.f. CW 371). One can no more be free to adopt such a pre-fixed role than one can be free to be imprisoned. “Responsibility for ends” means that people whose desires are frustrated or whose way of life is menaced because of the state or the officially-sanctioned distribution of resources are to be silent and accept their lot. People are to ‘make out as best they can’ or to re-mould their desires to suit the social system (PL 186), or to just ‘learn to deal with’ their situation (PL 185). If someone needs or demands more than the system offers, this is automatically a lack of ‘foresight or self-discipline’ and it is unjust for others to pay up (PL 186). A good example of the social effects of this kind of acceptance is that it would endorse the right of the Mexican state to destroy the way of life of subsistence peasants in order to build an airport for a local and foreign elite. The reasonableness of such demands is established as a matter of dogma: Rawls simply asserts that people are assumed to have the appropriate characteristics to enable them to take responsibility for ends (e.g. CW 407, PL 185-6, 34). This is a logical deduction from his conception of the person, and is not based on any discussion of actual people. It functions as an impositional discourse which establishes the social superiority of those who “take responsibility for ends” and which excludes or oppresses those who do not. The concept is linked to the problematic claim that people are not ‘assailed’ by desires and that desires are under one’s conscious control (CW 369). “Taking charge of one’s life” therefore turns out to be Rawls’s term for submitting to the demands made by the social system. It is a form of violent interpellation of an impositional variety, and it operates through social exclusion. For instance, if someone is unable to take responsibility for ends and thereby exceeds her or his social place, Rawls labels this as a psychiatric problem (PL 185). The system is therefore able to decide what people are to receive, and then to pathologise them if they are unable or unwilling to submit to their lot. “Responsibility for ends” is an excuse for callousness on the part of social insiders. It is based on the impositional gesture of “holding responsible” and the correlative submissive gesture of “accepting responsibility”. It is about submission to the social system, or more precisely, to the “dull compulsion of economic relations”.

Another use of the concept of “freedom” is an adaptation of the Kantian notion of autonomy. In this version, one acts freely if one acts from the principles of justice, because these are the principles which express one’s essential self (TJ 224-5, PL 24, 77). Autonomy is identical with acceptance of the public political conception of justice (PL 98). Of course, this supposedly autonomous self is itself the result of a particular process of social subjectification, a contradiction Rawls never overcomes. His sole answer is that one can retrospectively rationalise the subjectification (TJ 452). However, this does not solve the problem because the appearance of rationality may itself be a result of the process of subjectification.

Rawls’s conception of freedom is in fact an outgrowth of a fear of freedom, a fear which is clearly expressed when he demands to be ‘unencumbered by… singularities’ and contingency, and when he demands that one not be ‘required’ to ‘choose from countless possibilities without given structure or fixed positions’ (TJ 453, 494).

While Rawls constructs his essence without reference to actual people, the realisation of his principles would have to occur through people’s actual libidinal structures. The issue of relevance to actual people returns in the form of concerns about the “realism” and the “stability” of Rawls’s ethics. Despite his denouncements of the ‘fanaticism’ and ‘dogma’ of those whose views focus on a single primary good, his own theory requires such an emphasis for its realisation. It requires that the principles of justice form a master-signifier which trumps and suppresses other desires. It requires this because Rawls repeatedly emphasises that justice is normally primary and that the intensity of desires is not a legitimate consideration in matters of justice (e.g. TJ 203, CW 372, PL 190). In Rawls’s words, ‘justice as fairness is not… at the mercy… of existing wants and desires. It sets up an Archimedean point… The long range aim of society is settled… irrespective of the particular desires and needs of its present members’ (TJ 231). ‘The constraints [of justice] do not refer to, although they limit, the substantive content of comprehensive conceptions of the good’ (PL 211). This requires in practice that people desire justice above all else, and that justice be able to override or repress other desires which exceed it. Repression - or in his terms, regulation and revision - of desires is part of what Rawls counts as ‘moral’ (PL 186). In order for this to happen, people have to develop a particular kind of character-structure involving both a submission to “society” conceived as a set of rules and a petty-minded (or in Rawls’s terms “reciprocal”) relation to others. “Justice” relates to desire mainly as a form of repression, and amounts to a desire to repress one’s desire. It operates in bulldozer-like fashion in relation to actual people.

Rawls’s ethics is strongly reactive in structure. It is constructed primarily around negative goals such as the avoidance of “strife” and social division (e.g. JAFAR 151), so that his approach focuses on making the social system function smoothly, rather than on making it responsive to actual people. Characteristics of individuals which make social cooperation easier are assumed to be virtues (PL 318). Individuals relate to “justice” in a more-or-less submissive manner, acting on principles of justice as if these were overarching externalities or an imagined authority (e.g. CW 188). Reactive ethics relies on self-suppression through an ‘ideologically imposed sense of “duty” with its attendant guilt, self-sacrifice, and self-deadening “shoulds” ‘ (Rev Pleasure 3). Rawls never gives a reason for preferring reactive ethics over an active ethics based on themes such as hope and the gift, simply assuming that ethics has to take a reactive form (e.g. PL 210, TJ 28). Reactive desire is, however, highly destructive in ways Rawls does not consider, not only because of its deadening effect on individual actions but also because the repressed desires tend to return in neurotic forms. He is implicitly distrustful of unconstrained desires, which he sees mainly in terms of selfishness, sadism and vindictiveness (e.g. TJ 110, CW 570) and sees as arbitrary and random. One should keep in mind that Rawls’s construction is exclusionary, so that social inclusion is made conditional on having the right kind of character-structure, and so that those with other structures are to be normalised.

In order to construct the kind of person he wants, Rawls has to rely on specific forms of desire. In his own words, ‘a Kantian view does not deny that we act from some desire. What is of moment is the kind of desires from which we act and how they are ordered; that is, how these desires originate within and are related to the self, and the way their structure and priority are determined by principles of justice connected with the conception of the person we affirm’. A desire for justice is supposed to operate as a ‘regulative highest-order desire’ (CW 320). This regulative role is to occur via the conception of the person, internalised as an ego-ideal, that is, a desire to be a certain kind of person, or even as a self-in-alterity. This desire is supposed to override all object-based desires (e.g. PL 82-4, CW 320). The ideal is realised through ‘self-control’ and ‘self-command’, and failure to develop or conform to the ideal shows a ‘weakness of character’ (TJ 390-1, PL 76-7). I am reminded of Nietzsche’s remarks about people whose will to dominate is so pervasive that they even try to dominate parts of themselves. This is also, however, about domination by others, since Rawls permits the state to promote those ‘forms of moral character’ (PL 194, TJ 230-1) he sees as ‘good in themselves’ (LN 111). These forms and the ideal they embody are to be ‘specified by political institutions’ (PL 194-5) and achieved by means which seem to include territorialisation and coercion. As long as it is acceptable to people with the correct character-structure, Rawls thinks there is nothing wrong with his theory (TJ 498). The impositional operation of this ideal and its invalidatory implications for people with active character-structures demonstrates that it is an oppressive form of discourse.

The emotional states through which Rawls sees his theory operating are nearly all classic reactive emotions. Guilt, shame, resentment and indignation all perform central roles in Rawls’s theory. It is clear from specific discussions that Rawls sees “justice” as actualised through negative emotions which impede certain kinds of actions (TJ 339-40). Further, the appropriate version of these is not an emotion as such, but is defined partly by its rationalisation. It is by accounting for these feelings in certain ways that they become “moral” according to Rawls (TJ 425). In other words, having “moral feelings” means having gut emotional reactions similar to Rawls’s own. The “moral feelings” are all primarily negative, and are linked to social repression. For instance, someone who feels ashamed is supposed to feel unworthy of association with others. He therefore relies on emotions which have fear of exclusion built into them. The “moral sentiments” could in practice be used as a form of emotional blackmail. Rawls admits that the only difference between “moral sentiments” and emotions which result from authoritarian situations is the kind of rationalisation given for the former (TJ 451).

Other forms of character-structure favoured by Rawls include an identity as a molar self continuing over time and with no potential to become other. He also constructs people as extremely vulnerable to others’ assumptions, so our self-respect is dependent on constant affirmation by others. The good of self-respect is a sneaky category in Rawls’s discourse which he often uses to reintroduce things he wants to include in his theory but which wouldn’t otherwise fit into his model of “primary goods”. It also leads to a case for a phatic and ritualised model of politeness. Since his concept of self-respect includes a need for justice, Rawls is again slipping very close to tautology when he uses this concept (because it suggests that people in the original position would already know they needed Rawls’s theory prior to their deliberations). Perhaps the most disturbing assumption Rawls makes is the assumption that people are basically submissive and conformist, so that our desires and interests are constructed by social institutions. People are supposed to have a ‘desire to conform’ (JAFAR 194), a desire to be part of a greater whole (PL 204) and a desire to be in sync with the feelings of others (TJ 439). He adds that ‘[o]nly in a social union is the individual complete’, because humankind has a ‘social nature’ (TJ 458-60). People are to be, and to wish to be, emotionally dependent on other people, and to feel like ‘mere fragments’ without such dependence (TJ 464). Rawls thinks that human nature and psychology are ‘permissive’ towards attempts by theorists to construct social forms (PL 87-8). This naturalisation of conformism is a good excuse for the impositional practices through which states try to force conformity by actual people. It is an advocacy of slave morality which hides its contentiousness behind a mask of obviousness. In spite of this, Rawls claims that his theory is a theory of self-affirmation (TJ 436).

With people stripped of agency, it is little surprise that abstractions such as “rules”, “institutions” and the “basic structure” gain a vampiric life-energy which gives them agency within Rawls’s theory. This agency is possible only on the basis of the repressed active desire which actually motivates them. The reactive structure of desire Rawls posits as “moral” is to be constructed through a process of entrapment and repression of active energies during childhood. Before people can learn to be liberals, they have to go through the stages of “morality of authority” and “morality of association”. During the first stage, injunction are to be arbitrary, and autonomous expressions of will are to be seen as a wrong (TJ 406-8). It is this process of subordination to another’s will which teaches the so-called ‘moral attitudes’. This argument clearly shows how Rawls’s liberalism is built on a substratum of domination and repression, especially since the morality of authority is also to be a fallback position when liberal ethics fails. It is not overcome in later life, but simply becomes subordinate to the principles of justice (TJ 409).

In addition to the “freedom” required of the person, there are also in Rawls’s theory the “freedoms”, or “rights and liberties”, which arise as a result of the first principle of justice. These are supposed to be the main political good and the reason why Rawls’s theory is so satisfactory (JAFAR 115, CW 445). According to Rawls, he guarantees the basic freedoms, allowing the most extensive liberty compatible with a like liberty for all and restricting liberty only for the sake of liberty itself (e.g. JAFAR 102, TJ 213-14, 220). In practice, however, this is not what his theory would involve. The priority of liberty is mostly only its priority over the difference principle, social welfare and perfectionist goals. It is not a priority of liberty over social order, or a conditionality of social order on liberty.

Rawls defines “rights and liberties” in such a way that they project the systematising, liberticidal logic of state and related institutions into the concepts themselves. For Rawls, ‘[l]iberty… is a complex of rights and duties defined by institutions’ (TJ 210). A liberty always takes the form of a legal duty not to interfere with something (TJ 177), and no right can exceed the institutional practices which define it (CW 49). The outcome of a particular court case can be taken as establishing a liberty (PL 345). Therefore, Rawls does not actually offer any possibility of concrete and actual positing of rights. Although he identifies liberties and rights with formal commitments by institutions (TJ 180, CW 364), he also identifies it with actual absence of constraints (TJ 177), which is not at all the same thing. In order for the “liberties” to have the pacifying function Rawls assigns them, they would have to exist in the latter sense, but he only ever constructs them in the former sense. It is possible for an effective right or liberty to exist without taking the form of a law, and even for rights to be constructed over and against the state, as “liberty against the law”. It is equally possible for a legal right to be ineffective: Stalin’s Soviet Union had one of the most liberal constitutions in the world. I remain unconvinced that legal rights are an effective way to generate actual rights, because legal rights are enforced by the state agents who are, directly and through their allies in dominant elites, the main threat to the actuality of these rights. Rawls’s analysis is a reification: it directly identifies the legal and actual versions of the concept of “rights” and fails to establish that they should be linked.

Thus, Rawls supports, not actual rights and liberties, but what he sometimes terms a ‘system of equal liberties’, i.e. a social system which claims to represent the greatest possibility of liberties through its particular system of laws. The liberticidal implications of this displacement are clearest around the issue of so-called “interference”. For Rawls, any deviation from the “proper” conduct mandated by the social system is a form of interference (TJ 210), and any act of interference can legitimately be prohibited (TJ 291). In this way, someone who rebels against a social system which claims to express “liberty” is labelled as a would-be tyrant and repressed. Defence of liberty comes to mean defence of the ‘security’ of ‘the institutions of liberty’ (TJ 193), i.e. the social system. It is not true that Rawls only restricts liberty for the sake of liberty. In fact, he restricts actual liberties for the sake of the so-called system of liberties, or in his own words, to ‘strengthen the total system of liberty’ (TJ 220). The repression which could be justified on the basis of this displacement is potentially unlimited, and Rawls explicitly notes that it can, in severe enough circumstances, include the suspension of traditional liberal rights for the good of supposedly liberal institutions (e.g. TJ 213, PL 353-4). In a classic piece of doublespeak, he justifies ‘the more or less temporary suspension of democratic political institutions… for the purposes of preserving these institutions and other basic liberties’ (PL 355). Free speech can be restricted to protect the state from serious harm, and even ‘enforced silence’ is justified in an emergency (PL 351-2). Speech can be forbidden if it leads to ‘disorder’ or is ‘disruptive of the democratic process’ (PL 336; c.f. JAFAR 114). All liberties ‘may be regulated as always by the state’s interest in order and security’, and the state is to be given an ‘enabling right’ to maintain the system, apparently by whatever means necessary. This is because the system is taken to be a necessary precondition for any and all liberties (TJ 186-7). The freedom Rawls offers is not, therefore, a freedom one can assert over and against the state: it is a freedom which the state has always-already taken into account, and which operates as a hooray-word to justify the maintenance of the state. They are rights an individual can never feel confident about demanding, because they are hedged in by an overwhelming concern for “order” and a substitutionist identification of “rights” with state power. Foucault could have had Rawls in mind when he attacked humanism for holding sovereign ‘consciousness (sovereign in the context of judgement but subjected to the necessities of truth), the individual (a titular control of personal rights subject to the laws of nature and society), [and] basic freedom (sovereignty within, but accepting the demands of an outside world and “aligned with destiny”)’ (Language, Counter-Memory, Practice, Oxford: Basil Blackwell 1977: 221). One could also add Saul Newman’s insight that liberal rights are granted in return for the relinquishment of power, and that this is inadequate because without power, rights are meaningless (From Bakunin to Lacan p. 85).

Rawls’s discussion of liberty reflects a broader tendency to reduce social relations to issues of submission to an external structure. By reducing the complexity of interpersonal relations in this way, he atomises people, leaving ordinary people vulnerable when faced with far stronger elites (c.f. Scott DAR). As a result, “consideration for others” as “ends in themselves” is a cover for a “consideration” which in fact subordinates both self and other to the social system.

He also supports a compulsive work-ethic. People are supposed, in his terms, to ‘support themselves’ and ‘do their part’ (JAFAR 179), and he subscribes to a principle of “work or starve” (see PL 182). This means that, in practice, people’s free time is treated as a resource to be distributed by the state and the market, and therefore, puts social activity under the control of elites who decide which activities it is worthwhile to fund. Hence, someone who helps the oppressed may well be engaged in an activity which is not a form of “work”, whereas a useless or harmful activity such as manufacturing B52 bombers is taken to be “work” and to be valuable. The work system leads to elite control and to pervasive social irrationalities, especially when (as is usual) it is necessary to create additional work to ensure that there are enough jobs. Rawls misrepresents this system via a mythical schema in which it really is some kind of collective good. He assumes that anyone who escapes the work system is somehow exploiting others (JAFAR 179). Actually, it is the imposition of the work-system, and of “responsibility for ends”, which is exploitative, because it subsumes all people within a singular social system expressing a particular set of power-relations. Rawls is quite prepared to create work artificially, on the grounds that work is a necessary part of a meaningful life (CW 50, PL lix). This shows how his conception of what is meaningful is tied to the capitalist system.

Rawls’s theory is pervasively exclusionary. He admits that there are people to whom he could not recommend justice as a virtue, because it is incompatible with what he calls ‘peculiarities of their nature’ (TJ 504). Someone who does not have the two moral powers cannot be a member of a Rawlsian society (PL 74, CW 338). One should keep in mind that one of these powers, the “sense of justice”, is a reified version of “agreement with Rawls”. There is a lot of selective representation which goes on around the issue of exclusion, because, while Rawls occasionally admits that the “everyone” of his theory is not really literally everyone, the exclusions he admits are nearly always directed against overtly oppressive groups, such as fascists, racists, intolerant religious sects, sadistic torturers and owners of slaves (e.g. TJ 395, PL 596-7). I think, however, that his theory excludes a lot more. My own theory of oppression also encourages resistance against these groups - without endorsing liberalism. In addition to such intolerant and oppressive groups, Rawls also excludes a series of other groups which one might term “disorderly”. These groups need not be intolerant or oppressive to be excluded: they need only exceed the liberal social and political order in some way. They may, indeed, fight for more extensive liberties than Rawls allows, or for the same liberties in a non-statist form. Rawls draws no distinction between groups which try to impose their views on others and groups which simply demand sufficient space for their own way of life. For instance, any sect whose members unreliable as members of civil society should be banned on public-order grounds according to Rawls (TJ 189-90). I have in mind groups such as peasant rebels, urban and rural insurgents, anti-capitalists, eco-warriors, workers’ councilists, anarchists, black nationalists, New Age travellers and various other kinds of radicals, as well as people who do not work, and those, such as political graffiti artists and subvertisers, whose pursuit of voice, or of art or science, brings them into conflict with the state. I also have in mind real cases such as the Mexican airport plan mentioned above, the Cochabamba uprising against water privatisation, the Zapatistas, the Stonehenge conflict, the Battle of the Beanfield, and so on. One is thus dealing with a far more pervasive series of exclusions than Rawls’s examples suggest.

For Rawls, social insiders have a right to coerce others, solely because this is something which social insiders “would” consent to in the original position. There, social insiders would agree to ‘authorize the measures necessary to maintain just institutions’, even through violence against outsiders (TJ 504). Unreasonable doctrines are to be contained and isolated, as if they were a disease (PL 64, TJ 339). Rawls openly admits that the in-group, the so-called ‘reasonable citizens’, are to be ‘dominant and controlling’ over others (CW 573). For some reason Rawls would not see this as an ‘oppressive’ use of state power, because he denies engaging in any such thing. Once one is dealing with outsiders, the issue is no longer one of justice, but one of the ‘politically attainable’ and ‘strategies of enforcement’ (?TJ 203).

Despite their exclusionary character, Rawls sees his views as having a universal ‘value’ (e.g. CW 614). Because outsiders cannot deny that insiders would agree to the principles of justice in the original position, this is supposed to mean that outsiders’ “nature” is the cause of their suffering: ‘one can only say: their nature is their misfortune’ (TJ 504). They are also guilty of a ‘moral fallacy’ because they do not share insiders’ perceptions (CW 68). Rawls does not believe in ‘laying aside’ his principles when the assumptions underlying them prove to be invalid (CW 36). Rather, they are to be imposed by means of punishment and other violent institutions. Outsiders are simply to be anathematised as unreasonable, irrational or crazy, a trick Rawls uses in his own work (TJ 26). One can only problematise Rawls’s view on his own terms by reference to the convictions of insiders (PL 65-6). There is therefore an impositional and dogmatic backdrop to Rawls’s theory, a backdrop which is altered only slightly by Rawls’s attempts to bend his own principles in a few cases. In order, for instance, to allow any space for conscientious refusal of military conscription, Rawls has to maintain that pacifism as an ideology is bound to remain tiny and that it contributes positively to the public culture (TJ 325, 335).

All the rights Rawls offer are conditional on belonging to the in-group (a conditionality linked to Rawls’s idea of reciprocity, which is basically a theorised advocacy of petty-mindedness). For instance, he believes in a right to a subsistence income, but as I have already shown, he also believes this should be restricted to those who work. In another passage, Rawls argues that intolerant people have no right to complain if others do not tolerate them. Legitimate complaint must be an appeal to shared principles (TJ 190-1). Thus, Rawls does not demand consistency of the in-group, still less prefiguration; he wishes only for an in-group which adopts certain positions with regard to insiders.

Although he sometimes says his theory is for “everyone”, it is patently possible to be a no-one in this theory. He has, for instance, a concept of character, which provides a basis for the incorporation of punishment into his theory. He mistakenly thinks that crime ‘is a mark of bad character’, ‘faults’ and a ‘weakness of character’ (TJ 276-7, PL 76-7), rather than a social phenomenon. Worse, leaders of rogue states can be typified as ‘beasts’ (LN 100). He also tends to represent intransigence against the state as evidence of character-flaws. Outsiders are also portrayed through a character-based schema of the “unjust”, the “bad” and the “evil” which misrepresents social conflicts in mythical terms (TJ 385-6). There is also a double standard here, since irrationalities of the normal are taken to be “laws of nature” or limits on political action, so that only the unusual or minoritarian can be excluded. The duty to pursue justice is taken to include a duty to normalise outsiders, who are in effect, to use a term Rawls uses elsewhere, ‘socially dead’ (see CW 407) and non-existent as persons. The strong displeasure of minorities is never a case against a law according to Rawls (TJ 317). Alongside the exclusions operating through the construction of outsiders, there are also exclusions operating through the existence of a core and a periphery in Rawls’s theory. Issues which are not part of the “fundamental” core are termed ‘clutter’ and ‘distraction’ (e.g. JAFAR 176, CW 259, 398), and are left aside to be dealt with as an afterthought. In fact, Rawls never returns to most of these issues. It is surely no coincidence that the core issues are those associated with “normal and fully cooperating members of society” whereas the excluded periphery are those associated with issues such as disability, psychological difference and medical needs.

He has a tendency to view disability and psychological difference in terms in individual ‘defects’ which should be remedied through a process of normalisation (e.g. TJ 402, 404). Children have no rights in Rawls’s theory, being taken into consideration only as potential future citizens. They are explicitly excluded from his account of the person (TJ 126) and are to be taken into account only indirectly, via discourses of paternalism (TJ 183) emphasising what they would want if they were moral persons. Neurotic forms of psychological difference seem to be permitted only on condition, firstly, that they are incurable and, secondly, that the neurosis is compatible with sublation into the work-system (TJ 379-80). As regards more extensive differences, Rawls labels these ‘disorders’ and ‘defective’ (CW 259, PL 272). It is for Rawls ‘unfortunate’ that the mad exist (see PL 144), and so-called ‘burdensome and irrational constraints’ such as compulsions would not be acknowledged in the original position (TJ 417). Individuals are pathologised if they exceed the logic of place constructed by “responsibility for ends”, and psychiatry is to be used to eliminate such excess. At one point, Rawls even defines “our considered convictions” in such a way as to explicitly differentiate them from the so-called “mentally ill” (CW 10). There is also, rather strangely, an exclusion of those with excessive abilities (such as strength, imagination or intelligence) (TJ 452, 383). People are to wish that others remain stupid and weak, so that no-one can exceed the logic of place. Rawls’s account is also ethnocentric, involving a so-called ‘duty of aid’ to put other societies on the path to justice (LN 37). Groups such as so-called ‘Arctic Eskimos’ are to be ‘handled’ in an ad hoc way so as to subsume them within justice (LN 108). His treatment of the issue of immigration (LN 8, 39) is based on myths about proprietary relations between “peoples” and territories which ignore the structure of the capitalist world system and the history of colonialism. More generally, Rawls’s model is an expression of am urban and western tradition of thought. His model of autonomy and his striation of ethics into different spheres and levels denies the uncondititonalities which are widespread in hunter-gatherer and subsistence-peasant societies. Rawls is not a neo-liberal, but his model expresses the same kind of capitalist and statist monoculture which is so characteristic of this ideology.

To conclude, Rawls is wrong because he uses oppressive forms of discourse. Beyond this, in my view, where Rawls is wrong is twofold. Firstly, he emphasises modes of action which cannot be effective in achieving liberty or wellbeing. Secondly, he emphasises the wrong problems. He is concerned about systematisation instead of emancipation. He first of all sets up a positive pole of justice which then invalidates anything outside it, instead of starting by rejecting oppression as a negative pole and by demanding freedom of and affirmation for any way of life which does not directly oppress others. In this way, Rawls is becomes an advocate of a form of state power which is bound to be oppressive.


TJ: A Theory of Justice, revised edition, Oxford: Oxford University Press, 1999
PL: Political Liberalism, New York: Columbia University Press, 1996
CW: Collected Papers, ed. Samuel Freedman, Cambridge, Mass.: Harvard University Press, 1999
LN: The Law of Peoples, Cambridge, Mass.: Harvard University Press, 1999
JAFAR: Justice as Fairness: a Restatement, ed. Erin Kelly, Cambridge, Mass.: Harvard University Press, 2001


Post a Comment

<< Home