Andrew Robinson - Theory Blog

Monday, November 15, 2004

RAWLS - NOUMENAL SELF (notes - work in progress)


‘The philosopher’s task… is… to bestow the nature of man to man by thinking him as noumenon and the subject of philosophy’ – Gayatri Chakravorty Spivak, Critique of Postcolonial Reason p. 34.

Rawls’s conception of the person is central to his theory. For instance, William A. Galston has argued that moral personality is at the root both of the original position and of the well-ordered society as concepts in Rawls’s theory (Moral Personality 494-6). It can only have the role Rawls assigns to it if it expresses something primary. His formulations of this idea vary between different works. I have chosen to refer to his model of the person as “noumenal”, partly because in some works Rawls pursues a Kantian justification of the model and links his conception of the person explicitly to it (eg. CW 304), and partly because it shares with Kant’s model a disdain for the complexities and specificities of actual needs, desires and social relations. Rawls has, in his own terms, ‘an ideal of the person that provides an Archimedean point for judging the basic structure of society’ (TJ 511), and it is this conception I shall examine in this section. It is important to realise that this conception is to all intents and purposes a Barthesian myth: it is constructed intensionally within Rawls’s own theory, projected onto (rather than drawn from) actual people, and involves an image of “the” person rather than a series of different people. Richard Bellamy argues that Rawls is only able to assume a drive to reduce conflict because he essentialises his model of the person; otherwise, problems related to social exclusion would be far more visible (Lib and Modern Society 239). Similarly, Milton Fisk argues that Rawls has ‘to appeal to the distinction between the humanly essential and the humanly accidental’, and that he cannot escape the problem that, unless he naturalises historical circumstances, he has too little for a substantive theory (in Daniels ed., 53-4). Iris Marion Young also portrays Rawls’s model as essentialist because it involves abstraction from particularities to construct a singular point of view (JPD 100-1). Indeed, Rawls sometimes uses an explicitly essentialist vocabulary. For instance, the two moral powers are ‘part of the essential nature of citizens’ (CW 467). All those who are to be included must, Rawls insist, ‘appreciate’ his ‘political ideals of person and society’ (JAFAR 157), and he also admits that the validity of his theory depends on his elaborate theory of the person (PL 93-4). (Larmore thinks the idea of persons as free and equal citizens, although it cannot actually be validated, is the basic truth-claim which founds political liberalism - Larmore 356-7). However, Rawls also admits the idea of moral personality is ‘vague’ (TJ 445). The conception of the person is absolutely central to Rawls’s theory. As E.A. Goerner argues, key Rawlsian concepts such as equality and freedom are nothing more than slogans unless some degree of Kantian (i.e. comprehensive-doctrinal) context is implied by them (Rawls’s Apolitical Political Turn 716). (The implication for the comprehensive/political distinction was earlier expressed by Nagel: Rawls’s “thin” theory of the good really assumes a contentious liberal/individualist conception of the good. See Nagel in Daniels ed., 9-10, 15).

This vagueness has attracted criticism. For instance, Norman Daniels says that the status of Rawls’s generalisations about the person is unclear. If they are empirical, ‘they are probably false’; if they are idealised abstractions from distorting contexts, they may not be viable. Alternatively, they may involve a definition based on exclusions (one may or may not be a person), or an ideal. If they are an ideal, they tend to render Rawls’s theory, derived as it is from this conception, circular (Reflective Equilibrium and Archimedean Points 94-5). Robert Paul Wolff suggests that Rawls is trying to perform ‘the impossible feat of deducing substantive conclusions from purely formal premises’. In order to do this, he needs an idea of fidelity to principle. ‘Fidelity to principle is not, after all, deducible from bare formal rationality, at least without some rather powerful metaphysical assumptions about the timeless character of the moral agent (qua noumenon, in Kant’s language’ (UR 20). The molar self at the centre of Rawls’s model is constructed theoretically via his conception of the person. It is, Honig argues, the role of the original position to consolidate the self into a stable, socialised subject (PTDP 132). One should keep in mind that the construction of a conception of the person goes hand in hand with the attempted subjectification of people into such a model by means such as education and punishment. The issue is not of freedom, but of social control. As Michael Baur suggests, ‘free beings – precisely because they are free – cannot be imagined in advance as all agreeing to any particular thing at all’ (Reversing Rawls 251).

Rawls specifically differentiates his conception of the person from a conception of a human being, and emphasises that it is derived from philosophical traditions, thereby demonstrating its mythical character. The idea of a ‘person’ as ‘someone who can take part in, or play a role in, social life, and hence who can exercise and respect its various rights and duties’ (and who is ‘normal and fully cooperating’) is, he says, something derived from traditions of moral, legal and political philosophy dating back to Ancient Greece (JAFAR 24, PL 18). This leaves the possibility that there are actually-existing people who are not part of Rawls’s conception of the “person”, as well as the possibility that actual persons are misrepresented by his conception through the exclusive emphasis on what is otherwise one of a series of characteristics. A “person”, in Rawls’s sense, is to be the basic unit of ‘thought, deliberation, and responsibility’ (PL 18), and ‘the powers that equip us [sic] to be normally cooperating members of society over a complete life’ are ‘the sole relevant characteristics for justice’ (CW 317). Justice is, therefore, a mythical construct, reconstructing people by means of standards imposed from the standpoint of the social system. A single characteristic is conceived as somehow revealing the authentic being of someone who is, in actuality, far more complex: the sense of justice, according to Rawls, ‘reveals what the person is’ (TJ 583). ‘It is not our aims that primarily reveal our nature but rather the principles that we would acknowledge to govern the background conditions’ (orig. TJ 560). Further, it is something which is to be asserted a priori, not a contestable claim. ‘We start by assuming that citizens are free and equal moral persons who can contribute to, and honor the constraints of, social cooperation for the mutual benefit of all’ (CW 365). His conception of the person is by his own admission ‘given in advance’, not empirical (CW 367). His theory is to be ‘strictly deductive’, from premises which restrict participants in the original position to ‘a certain psychology’ (TJ 104). Further, even in his later work, Rawls concedes that his conception of the person is metaphysical. It uses ‘certain metaphysical theses’ based on what he terms ‘our ordinary conception of persons as the basic units of deliberation and responsibility’, even though Rawls thinks it is neutral between mainstream metaphysical doctrines (PL 29). For all his pretensions to neutrality, Rawls still, in the words of Peter Berkowitz, ‘takes a strong stand on the essence of morality, presupposing that what is morally worthy in human beings is their elemental freedom and equality’ (John Rawls and the Liberal Faith 64). Thomas Scanlon also interprets Rawls as basing his theory on the idea of a rational agent able to change life-plans, and defends this on the grounds that ‘each of us does in fact look at himself in this way’ (in Daniels ed., 171, 178).

Rawls explicitly declares that he believes his conception to serve something within actual people which is “higher” than other forms of desire. For instance, he states that, from the point of view of the parties in the original position, ‘our fundamental interests connected with the exercise of citizens’ two moral powers take priority over other interests’ (JAFAR 110). Rawls declares that his approach is sufficient as long as it gives people the ability to exercise their ‘highest-order interests’, the ones connected with his conception of the person (CW 367). ‘Both the agreements and preferences of citizens in society are counted as hierarchically subordinate to these interests, and… this is the ground for the priority of liberty’ (CW 372). People are also assumed to have an ‘antecedent moral structure’ prior to particular goals (CW 384) and to want above all a right to revise and alter their ends (TJ 160). Justice is to orient to a person’s ‘nature as a free and equal rational being’, as opposed to ‘the specific things he happens to want’ (TJ 220). This nature is, furthermore, collective, as if its realisation by different people expresses a single kind of selfhood (see TJ 226). In some passages, Rawls insists that his theory must establish itself solely on the basis of autonomous (i.e. noumenal) characteristics, avoiding any reference to ‘lower-order impulses, say for food and drink’ (CW 315) (NOTE: elsewhere, eg. CW 319, Rawls qualifies this claim a little). ‘Given the veil of ignorance, the parties can be prompted only by… highest-order interests’ (CW 315). Therefore, the account of primary goods derived from the original position (which Rawls uses to determine what people are and are not justly entitled to) is to operate independently from what people actually want (CW 367). Further, what counts as an issue of justice is to be derived solely from the account of the original position (CW 351), so there is no room for forgotten actualities to return elsewhere. To ‘compromise’ on a sense of justice, readmitting worldly concerns, ‘is not to achieve for the self free reign but to give way to the contingencies and accidents of the world’ (TJ 503). Justice is to express solely a basic nature and exclude anything which ‘one need not have as a condition of being a rational human individual’ (TJ 222-3). In other words, in his very framing of the problem, Rawls already constructs “justice” in a binary with everyday life and with actual people. The essential self is to stand above actual people and those aspects of human life which Rawls reduces to mere “accidents”. As is clear from Rawls’s words, this does not simply mean factoring morally-arbitrary advantages or alignments out of considerations of justice. It means ignoring all needs, abilities and desires, including the most basic, so as to construct an image of a bare person stripped of contingent locality and able to rise above specific alignments in the interests of “justice”. Indeed, at one point he asserts that persons have a fixed ‘nature as reasonable and responsible beings’, whereas everything else is only a possibility and a contingent social construct (PL 277). He also refers to moral action as a way ‘for persons to express their nature as free and equal rational beings’ (TJ 390). His principles involve ‘the desire to express most fully what we are or can be’, and any breach of them therefore ‘strikes at our self-respect, our sense of our own worth’. Breaching them means acting like a ‘lower order’ creature, on the basis of instinct (TJ 225). Therefore, ‘to be held accountable to the principles of justice… does not stunt our nature’, but rather, realises this at the expense of the ‘narrower impulses’ it ‘enables us to control’ (TJ 403). Because it expresses the terms people would recognise in the original position, Rawls’s theory, he claims, ‘exercises a natural attraction upon our affections’ (TJ 418-19). It also expresses our nature (TJ 222). (One might ask why something which expresses “our” nature needs to be imposed via violence, e.g. punishment). (c.f. Larmore p. 348-9: the conception of the person expresses the idea that people should only be valued to the extent that they are reasoning beings). Possession of moral personality, although one might argue it is as arbitrary as any other “natural” trait, is a necessity for admission to the Rawlsian moral world. As Galston puts it, ‘you can only be a party to the [original position] agreement if you are a moral person’ (Moral Personality 507).

The particularity of Rawls’s conception is part of its appeal to some critics. Stephen Macedo’s attack on Iris Marion Young, which involves a blatant endorsement of oppression as long as the targets are unpopular figures, is linked to Macedo’s valuation of the goal-orientation provided in Rawls’s theory. ‘The indiscriminate embrace of difference and diversity should be resisted… Assimilation is an inescapable and legitimate object of liberal politics: it all depends on the justifiability of the values toward which institutions assimilate and the reasonableness of the means’ (469-70). It is only by insisting on the validity of imposing some values that political theory can be ‘tough-minded’ and have ‘spine’ (470). This is a structural similarity to arguments which could be provided by a Grand Inquisitor, a Stalinist or a Nazi; all arrange the relationship between values and coercion in the same way, differing mainly in the values they advocate.

Despite being conceptually distinct from the “human”, Rawls’s “person” nevertheless involves a large part of his claim to universality, and he uses words like “everyone” as if they are coterminous with “all persons”. However, like other versions of universality, this model is constructed around an exclusion, and in fact involves a particular set of particularities turned into an image of universality. ‘The universal’, says Butler, ‘can be the universal only to the extent that it remains untainted by what is particular, concrete, and individual’ (CHU 23). In Rawls, there is what amounts to a pretence at universality, concealed beneath phrases such as “persons are assumed to be…(e.g. free and equal)”, “persons as…(e.g. free and equal)” and “in justice as fairness, persons are regarded as… (e.g. free and equal)”. This means “regarding”, and treating, people as if they are something which in fact they may or may not be, although Rawls also incorporates two other assumptions which strictly contradict the gesture of “regarding as…”: normalising those who do not conform to the model, and using nonconformity as a basis for invalidation and punishment. Larmore similarly states that those who do not share the model of the person are to be treated as if they do, and that it is no objection to the de facto theoretical universality-claims of political liberalism if it turns out not in fact to be universal (Larmore 353). Stephen L. Darwall emphasises the importance of rationality, planning and responsibility in Rawls’s work, stressing that these elements are drawn from a Kantian conception of the person (Is there a Kantian… 324). The parties in the original position are not neutral regarding possible conceptions of the good outside this position. They have a distinct preference for conceptions which share Rawls’s emphasis on planning and responsibility (326). Since the essence is modelled around ideas such as rationality, Darwall suggests Rawls is actually taking rationality, not persons, as an end in itself (327). Of course, one has to add that the rational is supplemented by the reasonable, but Darwall’s basis point is sound. Michael Neumann similarly suggests that the idea of a Kantian respect for persons is misleading. Kant wanted to respect, not diverse people, but ‘rational selves in all their sameness’ (Did Kant Respect Persons? 285). ‘No messy consideration of what you want as a flesh-and-blood human is required; instead it is positively excluded… I [can] have done the right thing by you, yet given no weight to… anything that might differentiate you from other rational beings. I respect you only to the extent that you think and act just like everyone else, and no further’ (Did Kant Respect Persons? 291-2). The noumenal self is an essence of conformity, not diversity (294). ‘Each person is self-legislating, but the self-legislated rules are always the same’, because ‘a decision is arrived at with all the mechanical sameness of an intelligent robot’ (Did Kant Respect Persons? 295). Neumann uses this reading of Kant partly as a criticism of Rawls’s reading, but clearly important elements of Kant’s homogenising drive find their way into Rawls’s project via his conception of the noumenal self. Rawls’s is, however, a contentious reading. Michael Gorr asks why Rawls assumes that one can express one’s essential nature only through the principles of justice. A Kantian noumenal self could in principle express itself via any principles which could be formulated in an original position, including anti-Rawlsian ones such as Nozick’s (On Natural Inequality 26).

In any case, the self-determining persons are not in fact ourselves, but the imagined shadow-people of the original position, as Baur convincingly argues. Rawls appeals to the ideal of the person partly because it allows him to avoid metaphysical problems by deducing outcomes from free choice: ‘what counts for the purpose of such self-organisation is determined by the very personhood that is organized and doing the organizing’ (Reversing Rawls 258-9). However, actual people are not self-determining, because actual people are placed in a submissive role: one is to do what the parties in the original position command. The parties cannot be identical with actual people, “us”, since if they were, the original position would be unnecessary. However, the parties also are not self-determining, because they are modelled in a way determined by the moral beliefs of existing people. Hence, ‘our personhood is supposed to follow a criterion or standard that is given to it from a source that is essentially other than it, namely [the original position parties’] personhood’, and the parties’ personhood ‘is given a narrow set of interests and concerns…by a source that is essentially other than it (i.e…. by us)’ (292). Rawls conceals this problem because he equivocates on whether the original position parties are actually “us”. However, the difference between “us” and the parties is structurally necessary, and its result is that Rawls cannot escape metaphysical controversy. The parties cannot tell us why we (as opposed to they) should obey the principles they accept. Rawls can tell us this only by invoking a contentious “comprehensive” philosophical claim that it is in everyone’s good to act as a free and rational being (276). Baur thinks this problem expresses a deeper difficulty: one cannot measure oneself against a standard, because if one wishes to, one is fallible, and if one is fallible, so are the standards one uses to measure oneself (280). The relationship between Rawls’s conception of the person and his model of the original position tends to undermine the position taken by Kymlicka that the latter does not require an epistemologically transcendent position (Contemp Pol Thought 290).

Commentators on Rawls have linked his conception of the person to the absolute primacy he assigns to justice over other moral virtues and personal goals. Clark and Gintis state that ‘no acceptable theory of moral behavior could claim the lexical priority of social obligations and personal concerns’, and claim that Rawls’s Kantianism has misled him in this regard (Rawlsian Justice… 318-19). Similarly Roberto Alejandro argues that Rawls’s assumption of the primacy of justice is a contentious metaphysical belief and is not widespread in democratic societies (Rawls’s Communitarianism 91-2). He adds that ‘there is an element of dogmatism in the claim that one virtue, justice, is the only one that best expresses the individual’s nature’ (Rawls’s Communit. 85). This primacy means that the theory of justice monopolises and dominates both individuals and communities (Rawls’s Communit. 77, 88). In other words, an impositional drive arises as a result of Rawls’s adoption of an essentialist conception of the core or noumenal person. Rawls’s ‘wager’, argues Honig, ‘is that the institutions of justice as fairness fit and express the (core) self without remainder’ (PTDP 138). His failure in this wager is the subject of her essay.

Ronald Dworkin (in Daniels ed.) is wrong to suggest that Rawls’s theory is rights-based, because he ignores the rigging of the original position so as to only include a certain kind of person. This rigging expresses a goal – the primacy of this kind of person – which is prior to any rights Rawls embraces. In fact, many of the rights Rawls specifies have in any case always-already been exercised on one’s behalf, and are conceived as rights of the essence, not of the actual self.

The essence of the self is posited in an almost irrationalist way. Galston is right when he claims that Rawls determines the boundary between noumenon and phenomenon arbitrarily (Moral Personality 514). For instance, Rawls specifically states that this essence, the “moral person”, does not have to justify itself (CW 334). It is constructed in the role of a master-signifier, regulating and even suppressing specificities. ‘The aims and conduct of citizens in society are… subordinate to the conception of citizens as free and equal persons’ (PL 366). As so often with master-signifiers, Rawls’s conception of the person operates as an exception to his broader commitment to rational-choice modes of thought and action, based on an “irrational” choice to elevate one particularity to a position where it transcends all others. Its role is contradictory, because, while Rawls admits that it is based on a model of human nature - ‘a conception of justice for a democratic society presupposes a theory of human nature’ (PL 346) - it is also something he views as contingent. It is a result of particular institutional arrangements (i.e. territorialisation and “education” through systematised social relations), and not something people would accept or desire if ‘left to their own reflections’ (JAFAR 56). It is also contradictory because, on the one hand, it embodies the right as that which exceeds every conception of the good, but, on the other, it embodies a particular partial conception of the good, ‘a conception of the good on the basis of which the parties rank alternatives’ (TJ 138). Such ideas are to be the benchmark for other aspects of Rawls’s theory, including his principles of justice. He compares different ideas on the basis of which idea ‘is more appropriate to the conception of citizens as free and equal, and of society as a fair system of cooperation between citizens so viewed’ (JAFAR 122). At one point, Rawls specifies his conception of society as his starting-point. ‘The procedure is as before: we start with the basic idea of society as a fair system of cooperation’ (PL 34). Nevertheless, its result is supposed to be that basic institutions ‘are set up to advance certain fundamental interests which everyone has in common’, ensuring that society is ‘a cooperative venture for mutual advantage’ (TJ 71, 74).

Rawls’s conception of the person has a number of components, each of which I shall examine in more detail later. The main ones (in their schematic form) are that a “person” (or “citizen”, which basically means a person who is part of a society) is “free”, “equal”, “reasonable” and “rational” and has the “two moral powers”, which are the “sense of justice” and the “capacity for a conception of the good”. For instance, ‘part of the essential nature of citizens… is their having the two moral powers’ (JAFAR 200), and every citizen is assumed to have a sense of justice, a capacity to form, follow and revise a conception of the good and a capacity to exhibit the political virtues necessary ‘to cooperate in maintaining a just political system’ (PL xlvi-xlvii). Also, Rawls’s theory ‘joins the content of justice with a certain conception of the person, and this conception regards persons as both free and equal, as capable of acting both reasonably and rationally, and therefore as capable of taking part in social cooperation among persons so conceived’ (CW 306).

The two moral powers ‘are taken as defining “moral persons” and “moral personality” (JAFAR 19), and Rawls ‘takes as fundamental’ persons’ ‘capacity for social cooperation’ (PL 306). Rawls claims to be building his entire theory from ‘the fundamental intuitive idea of citizens as persons with [the two moral] powers and with a higher-order interest in their development and exercise’ (JAFAR 175). In addition, society as a “fair system of cooperation” requires that their relations with each other are “reciprocal” in the sense of ‘mutual advantage’ (JAFAR 122). Though he is leaning on the positive connotations of terms such as “free” and “equal”, the meaning he gives them is actually distinct from wider uses. Hence, ‘in virtue of… their moral powers… we say that persons are free. And in virtue of their having these powers to the requisite degree to be fully cooperating members of society, we say that persons are equal’ (CW 397). Further, if one has ‘normal capacities’ then, according to Rawls, one desires to be a citizen (PL 85). At one point, Rawls defines reasonable persons as people who are free and equal (in Rawls’s sense) and ‘who have an enduring desire to be fully cooperating members of society over a complete life’, provided they also ‘share a common human reason’ (CW 476). For Rawls, ‘part of the essential nature of citizens… is their having the two moral powers which root their capacity to participate in fair social cooperation’ (PL 203). He also wants to give a ‘special place’ to the sense of justice (CW 126). ‘[E]very person may be assumed to have the concept of justice’ (CW 71).

Rawls also makes various other assumptions about the psychological nature of “persons”, but these are not part of the noumenal conception and so I shall handle them later. The basic conception of the person seems to be very limited in what it demands, but it is important to realise that each of the terms and phrases Rawls uses have a technical meaning beyond their ordinary use, and that what he actually infers about “persons” is quite extensive. For instance, persons are supposed to rank their exercise of the sense of justice above all else, to take responsibility for their ends, to be able to view themselves separately from every specific belief and desire, to be a “normal and fully cooperating member of society”, and so on. For instance, he defines the sense of justice as ‘a settled disposition to adapt and want to act from the moral point of view’ (TJ 430). Rawls is actually demanding conformity to a very specific and exacting ideal, an ideal which many people will fall short of by various degrees. (This has led some authors, such as William Galston and Steve Sheppard, to accuse him of perfectionism. See Sheppard 411-13). For instance, he demands that people have higher-order interests in their ability to alter conceptions of the good and in their sense of justice, which requires that one subordinate all specific desires and needs to these two “capacities” and that, for instance, one be prepared to sacrifice any particular desire if it conflicts with the demands of a just state. Citizens are also assumed to have a ‘capacity to accept’ and a ‘desire to act on’ reasonable political principles, as well as to ‘do their part’ provided others do, and also to trust others who do their part (PL 163). They are assumed to be ‘normal’, to have the two moral powers, and to ‘assume’ their ‘role’ (PL 203). Also, Rawls ‘regards each person as someone who can and who desires to take part in social cooperation for mutual advantage’. ‘[M]oral persons are said to have both the capacity and the desire to cooperate on fair terms with others for mutual advantage’, and also to have ‘a regulative desire to conform the pursuit of one’s good, as well as the demands one makes on others, to public principles of justice which all can reasonably be expected to accept’ (CW 365). As Doppelt, in an article defending Rawls, suggests, ‘[t]he references to highest-order interests… [are] tantamount to a conception of the supreme good or value at stake in the Rawlsian political order’. These references specify a ‘public or political identity which Rawlsian citizens must possess’ (Is Rawls’s Kantian Liberalism… 821). Doppelt explicitly uses such discourse to pick and choose who deserves freedom and respect (826-7).

To take an example of a restrictive definition, Rawls’s idea of “cooperation” is very specific. ‘Cooperation is distinct from merely socially coordinated activity… Cooperation is guided by publicly recognized rules and procedures which those who are cooperating accept and regard as properly regulating their conduct’. It also implies an idea of fair terms of cooperation and an idea of mutual advantage (CW 396). Therefore, an actual person does not qualify as a noumenal person simply because she or he can cooperate in the mundane sense of engaging in a mutual project with others. It is necessary that one submit to alienated procedures and rules, and that one’s cooperative endeavours be guided by a rational sense of one’s good and by relations of reciprocity. If society is seen as a system of cooperation, Rawls presumably means that it is seen as embodying all these specific characteristics. To take another example, ‘free persons have a regulative and effective desire to be a certain kind of person’ (CW 335). Of course, by the standards of my theory of oppressive discourse, Rawls’s conception is mythical regardless of what it specifies. However, it is also significant that the myth he constructs is very extensive, and that it extends far beyond any initial descriptive plausibility it may seem to have.

Rawls repeatedly emphasises that his theory aims to appeal to, to be fair to and to include those who fall within his conception of the person, without ever declaring this to be a gesture of exclusion. For instance, he states that he is trying to be fair, not to doctrines, but to ‘free and equal citizens’, so that the meaning of fairness can be referred back directly to the essence of such citizens (JAFAR 189; cf. CW 472). His aim is to appeal to ‘all citizens viewed as reasonable and rational’ (CW 488), and he wants an ideal of justice which appeals to citizens who view themselves in the way he suggests (CW 332). Citizens, and only citizens, are to specify the terms of social cooperation (PL 73). Similarly, ‘the liberal ideal of political legitimacy’ is that ‘social cooperation’, at least around constitutional essentials, ‘is to be conducted so far as possible on terms both intelligible and acceptable to all citizens as reasonable and rational’ (CW 490). Hence, for instance, Rawls’s critique of utilitarianism rests on the latter’s supposed failure to guarantee the ‘highest-order interests’ in the exercise of moral powers (TJ 152).

It is crucial to emphasise the exclusionary character of such assumptions: the two moral capacities are a precondition according to Rawls for anyone being ‘a full and equal member of society’ (PL 302). As a result, the equality Rawls posits is an exclusionary equality. Persons are “equal” in having the moral and cooperative abilities Rawls values, and ‘[a]ll who can be fully cooperating members of political society count as equals’ (JAFAR 21). Presumably, therefore, those who cannot conform to this model are not equals. Rawls clarifies this by stating that there is a ‘minimum’ of moral capacity one must have to be a citizen, even though those with more capacity are not to have greater degrees of citizenship (TJ 443). At one point, involving an appeal to “common sense”, Rawls claims that justice cannot apply to animals and that it only applies to (some?) humans because of what might be termed their character-structure. It is, he claims, ‘natural’ to think that only ‘moral persons’, who have a sense of justice, conception of the good and a rational plan, have a right to equal justice (though he also extends his conception to cover those who will develop ‘moral personality’ in due course) (TJ 441-2). In fact, Rawls does not specify whether a moral personality is necessary for inclusion - he thinks the issue is of no importance because the overwhelming majority have such a personality, and he does not want people’s lack of a moral personality to be used to withhold basic liberties because of the effect this would have on just institutions (TJ 442-3). On another occasion, he specifies that it is ‘[r]easonable and rational agents’ who are to be the ‘units of responsibility in political and social life’ (PL 50). However, it is clear on closer inspection that he does indeed allow liberties to be withheld, through practices such as punishment and the discourse of psychiatry. It is simply not possible to have an oppressive theory and then to avoid generating the possibility of oppressions from it.

As a result, the “original position”, Rawls’s main reasoning device, is rigged so as only to include people who conform to the essence he posits (eg. PL 309). ‘Citizens are represented solely as free and equal persons: as those who have to the minimum sufficient degree the two moral powers and other capacities enabling them to be normal cooperating members of society over a complete life… [T]hose similar in relevant respects are to be treated similarly’ (JAFAR 87). Persons are ‘represented solely as’ free and equal moral persons (CW 305, 310, 317). In the original position, ‘it is postulated… that the parties are moral persons, rational individuals with a coherent system of ends and a capacity for a sense of justice’ (TJ 289), and it is also assumed that everyone has capacities in the ‘normal range’ (PL 25), as well as that ‘each man has the requisite ability to understand and act upon whatever principles are adopted’ (TJ 17). For Rawls, ‘the motivation of the parties is appropriate to the representation of moral persons’ (CW 334). So people in the original position assume that those they represent have ‘the two moral powers and a certain psychological nature’ (PL 370). Of course, this also means that those who do not conform to this model of the person - who fall outside the boundaries of sameness - are not represented in the original position, and therefore, that Rawls’s idea of what “justice” entails is constructed without reference (or relevance) to them. For instance, he is prepared to calculate what people require on the assumption that they are normal and fully cooperative (CW 452). The idea of “fully cooperative” suggests that the people Rawls values are to some degree submissive, as well as that they believe in reciprocity; people who do not conform to this model are not taken into account. Further, the parties in the original position ‘are not to be influenced by any particular information that is not part of their representation as free and equal moral persons… unless this information is necessary for a rational agreement to be reached’. This leads to a ‘thick’, rather than a ‘thin’, veil of ignorance which excludes far more than is necessary to simply eliminate threat-advantage from the parties’ considerations. The more knowledge is excluded, the better the theory, since it is resultantly ‘connected more closely to the conception of free and equal moral persons’. Any information beyond this conception should if possible be excluded ‘in order to have a lucid representation of the notion of freedom’ (CW 336). (NOTE: This position is taken from the most Kantian of Rawls’s essays, “Kantian Constructivism in Moral Theory”. He contradicts this position a little in the later essay “Justice as Fairness: Political Not Metaphysical” - CW 401 - returning to the “Humean” idea of using the veil of ignorance to exclude arbitrary reasons for privileging ethical statements). It is important to realise that it is only because of the posited essence of the person that Rawls is able to deduce any conclusions from the original position, since otherwise, the parties would not know enough to reach any conclusions whatsoever.

Another thing to realise is that Rawls simply asserts that people are what he says we are. He demands that people be “assumed” to fit his model of the person or “viewed” or defined in this way. The language he uses is frequently descriptive. Hence, citizens are ‘normal and fully cooperating members of society’, and further, ‘want to be, and to be recognised as, such members’ (PL 81). Citizens are ‘described as’ having the two moral powers (CW 428), and ‘regarded’ or ‘conceive[d]’ as normal and fully cooperating members of society (PL 178, 301). He simply “assumes” that people are reasonable (CW 445), that people can honour the fair terms of cooperation (CW 317) and that everyone fits the various elements of his conception of the person (PL 106). Further, ‘[w]e assume that persons as citizens have all the capacities that enable them to be cooperating members of society’ (PL 20), and persons ‘are regarded as capable’ of changing their conceptions of the good on reasonable and rational grounds (CW 404). To take another example, he asserts that ‘justice as fairness regards citizens as engaged in social cooperation, and hence as fully capable of doing so’ (JAFAR 18; cf. PL 3, CW 357); all citizens want, or could want, to cooperate (JAFAR 202). And also, ‘starting from a political conception of society, political liberalism describes both citizens and peoples by political conceptions that specify their nature’ (LN 23). Further, ‘citizens are said to have’ the two moral powers and the higher-order interests (PL 417). In a passage already discussed above, Rawls uses three phrases of this kind: ‘start by assuming’, ‘regards’ and ‘are said to have’ (CW 365). The meaning of such claims (especially in terms of ideas such as to “regard”, “assume” and “view as”) is not specified, and I shall examine it in more detail below.

The assumption that people are “normal” and both able and willing to engage in what Rawls terms “social cooperation” is disturbingly common in Rawls’s work. He assumes this type of person to be not only the most widespread, but also to be a citizen’s ‘proper role’ (PL 186). Further, ‘citizens are equal in virtue of possessing, to the requisite minimum degree, the two more powers and the other minimum capacities that enable us to be normal and fully cooperating members of society’ (PL 79), and are included politically as a result of these capacities. Citizens are equal only because they ‘comply with just institutions’, fulfil duties and obligations and have a so-called sense of justice (PL 280).

Rawls’s conception of society limits it to the role of being an association of people who fit his model of “the person”. For instance, ‘the state should be understood as an association consisting of equal citizens’ (TJ 186). He also refers to ‘the idea of society as a fair system of cooperation between citizens as free and equal’ (JAFAR 56). Society, or ‘the life of a people’, is similarly given an essentialist definition by Rawls, as a ‘scheme of cooperation’ over time (e.g. TJ 257) and a ‘cooperative venture for mutual advantage’ (TJ 74)..

At times, Rawls admits the incompleteness of his conception of the person as regards actual people. Even on these occasions, however, he continues to insist on its primacy, due to the “fundamental” questions it ostensibly addresses. For instance, he thinks it is legitimate to start from the case of the normal adult citizen ‘and proceed from there’ (PL 245); other groups should be satisfied with their status as an afterthought. ‘Our focus… is on persons as capable of being normal and fully cooperating members of society over a complete life. The capacity for social cooperation is taken as fundamental’ (PL 301). This is one instance of a privileging of the “normal” which runs through Rawls’s construction of the noumenal self. It is also worth noting that Rawls sometimes suspends his hostility to admitting contingent needs and abilities into his theory, so as to ensure that it does not exclude the normal. For instance, he admits the ‘normal range’ of contingent variables to be considered in the original position (PL 272), and he also admits “general facts” about people and society, about human motivations and about the value of primary goods to ensure that an agreement is possible (CW 265), as well as at times admitting the historical specificity of such elements (eg. CW 303). There is also some slippage when a characteristic is accepted because it is normal and because it is therefore unrealistic to deny it, and yet it is then incorporated into ethical ideals (see below on rational-choice psychologies). In other words, the noumenal self as demand for conformity bears equally on all people, but on some more equally than on others.

One crucial idea which results from Rawls’s conception of the noumenal self is the idea of “primary goods”. This is a list of things people are presumed to want, and according to Rawls, all legitimate claims on grounds of justice must take the form of demands for primary goods. Primary goods have a crucial role in Rawls’s theory. On one occasion, he remarks, ‘given human nature, wanting them is part of being rational’ (TJ 223). However, they do not include anything necessary to satisfy basic needs, let alone any specific objects or social relations which might be desired. This incompleteness has been a source of criticism. For instance, Adina Schwartz suggests that, even if most people in fact value primary goods, the parties in the original position could not assume a bias in favour of such majority preferences (Moral Neutrality and Primary Goods 301-2). However, the parties do not value primary goods solely as means to the conceptions of the good they may turn out to have beyond the veil of ignorance. The list of primary goods is constructed with reference to the noumenal self, about which the parties are not neutral. Hence, ‘the idea of restricting appropriate claims to claims to primary goods is analogous to taking certain needs alone as relevant in questions of justice. The explanation is that primary goods are things generally required, or needed, by citizens as free and equal moral persons [with conceptions of the good]… It is the conception of citizens as such persons, and as normal cooperating members of society over a complete life, which determines what they require’ (CW 373). Sometimes, he even suggests that these goods can be referred back to the two higher-order interests (CW 366). Citizens have ‘a certain role or status’, and only things which are necessary for this role or status are counted in questions of justice. In this way, Rawls’s theory expresses the interests of those who have such a role, and is therefore ‘objective’ (CW 373-4). Rawls assumes that primary goods reflect the fact that plans of life ‘normally have a certain structure and depend upon the primary goods for their formation, revision and execution’. However, establishing this ‘does not rest on… historical or social facts’ , arising ‘only in the light of a conception of the person given in advance’, along with so-called ‘general circumstances and requirements of social life’ (PL 308). In other words, Rawls does not recognise autonomous needs at all. He assumes people to be cogs in a social machine, and then grants them only what is required for their continued functioning as such a cog. His theory serves only those who are prepared to become cogs, and it may be intended to pressure others into becoming cogs also. As Darwall suggests, ‘[t]he real basis for taking primary goods as the index of well-being is the conception of the person as a rational chooser of ends’ (Is there a Kantian… 325). Rawls’s theory depends on the implicit but undeclared assumption that primary goods are sufficient for the noumenal self to express its essence, since otherwise, it would be irrational for original position parties to exclusively orient to these (Wolff UR 150, 162). However, he also takes ‘persons’ needs as citizens’ as equivalent to what is ‘advantageous for all’ (PL 179), keeping up the appearance that his model is inclusive.

Primary goods are deduced from shared characteristics of ‘permissible conceptions of the good’, i.e. those conceptions compatible with the conception of the person Rawls specifies. They are specified as ‘things citizens need as free and equal persons’. The specific list of primary goods (which is less significant from my perspective than how it is derived) consists mainly of abstract goods, such as rights and liberties (defined legalistically), income and wealth (conceived as operating in a broadly capitalist economy) and ‘the same social bases of self-respect’ which can be assumed to be valued by all persons (PL 180). Rawls does not provide a case for the primacy of these goods, which he assumes to be ‘evident enough’ (TJ 381), until challenged by other theorists. The list varies over time (with goods such as leisure-time and even absence of pain suggested as additions [CW 454-5]), its basic role in Rawls’s theory remains the same. It is to provide a list of desired objects derived directly from the essence of the person, which can be used to trump any actual claims about what is valuable which might arise from particular people. ‘While the list of primary goods rests in part on the general facts and requirements of social life, it does so only together with a political conception of the person as free and equal, endowed with the moral powers, and capable of being a fully cooperating member of society. This normative conception is necessary to specify the appropriate list of primary goods’ (JAFAR 58). The list is restricted to items open to public view, and ‘[p]rimary goods… are what free and equal persons need as citizens’ (JAFAR 59-60). The specific contents of the list are significant in other ways, notably in that their structure encodes a legalistic state apparatus and a system of commodity fetishism indirectly into the assumptions held in the original position, and in that they allow the acquisitive individualism Rawls often disavows a backdoor into his theory.

Primary goods, Rawls states, are clearly not anyone’s idea of the basic values of human life. Nevertheless, they are to ‘specify what [citizens’] needs are… when questions of justice arise’. The reason for this is that, even though they ignore citizens’ needs and desires, they are ‘mutually acceptable to citizens generally’ as a standard of comparison. This ‘objective’ conception of citizens’ needs is separate from specific needs, ‘desires, wishes, and longings’. It is linked specifically to the citizen as a ‘role or status’. Anything which is ‘not a requirement’ for the role of citizen does not qualify as an objective need. ‘Needs in any other sense, along with aspirations and desires, play no role’ (PL 188-9; c.f. JAFAR 141). ‘[P]rimary goods… are things persons need as citizens, rather than as human beings apart from any normative conception’ (JAFAR 88); they are what people need ‘as normal and fully cooperating members of society’ (TJ xiii), and nothing counts as a primary good unless it is necessary for people’s ‘highest-order interests’ (CW 314). Primary goods are directly related to the ‘Archimedean point’ (TJ 232) of the myth of the person; the list of primary goods ‘depends on a certain conception of the person’ which determines how interpersonal comparisons are to be carried out (CW 359). The parties in the original position ‘suppose that their conceptions of the good have a certain structure, and this is sufficient to enable them to choose principles on a rational basis’ (TJ 349). ‘The characterisation of primary goods does not rest on… historical or social facts’ (CW 367). Because defined to suit their role or essence, primary goods are fair to citizens (JAFAR 61). Further, the worst-off group as defined under the difference principle is also to be taken as the worst-off among persons, not among people in general, and defined ‘within the normal range’ relative to primary goods (CW 258-9). The assumption of the validity of primary goods as a measure of assessment seems to depend on two claims: that they are necessary to all social plans, and that people in the original position can be assumed not to care about difference (TJ 125). He later weakens the first claim.

In his reply to Amartya Sen, who criticises him for oversimplifying human needs and desires and for ignoring the enabling characteristics of some “goods”, Rawls says that he simply assumes people to already be enabled, i.e. to have the two moral power and to be normal and fully cooperating members of society. Therefore, what Rawls is offering is not a minimum necessary to achieve social exclusion, but the things people need in order to remain citizens, assuming they qualify to avoid exclusion. Therefore, he says that primary goods are adequate to guarantee ‘the kind of needs and requirements that political justice should take into account’ (JAFAR 169-70). This reply only holds up if one accepts that political justice should be constructed around an essentialist conception of the person, ignoring all other needs and characteristics, and if one accepts Rawls’s assumption that all morally-valued people can be “assumed” to fit this conception.

While the role and derivation of primary goods does not change, their exact empirical status does. In A Theory of Justice, Rawls assumes there are some things ‘everyone would prefer more of rather than less’, at least if they were rational, and which have a transcendent character owing to their necessity in advancing other ends (TJ 79). ‘Rational individuals, whatever else they want, desire certain things as prerequisites for carrying out their plans of life’, and for this reason, they could consent to primary goods being the standard of comparison without the risk that, beyond the veil of ignorance, ‘they may find their plans utterly ruined by the principles they adopt’ (TJ 348-9). In the same text, however, he adds that primary goods are things ‘men are presumed to want’ (TJ 230), a somewhat weaker claim. In Political Liberalism, primary goods are things people only require ‘normally’ to realise the myth of the self, and he claims to introduce the idea only to ensure that there is an outcome to the deliberations in the original position (PL 75-6). He concedes that there may be rational systems of ends which do not require primary goods, but this does not stop him from reaffirming that they would be viewed as universal in the original position (PL 278). In a late essay, “The Priority of Right and Ideas of the Good”, he adds that he hopes that people will accept them as a standard of comparison so as to ensure the public possibility of an overlapping consensus, even though they do not approximate to what anyone actually values (CW 456-7). This change reflects changes in how Rawls conceives and presents his idea of the essence of the person, but across these changes the essence and its relationship to the idea of primary goods remains intact. The idea of primary goods tends to construct social relations in an exclusionary way by insisting that all claims must take a predetermined form. Many forms of life may lead to requirements which fall outside the definition of “primary goods”. While this may include physical objects, it is perhaps more important that Rawls leaves no space for the discursive margin of tolerance and the specific rights and liberties which may be necessary to include those who fall outside “normality” in various ways. If social insiders can claim, from a position of self-sufficient smugness, to be able to dictate in advance what form a legitimate claim may take, they implicitly cut off any possibility of actual people seeking anything concrete through the space they control. It is not simply that Rawls’s “principles of justice” limit the extent to which desire can be satisfied; it is also that it dictates the form desire must take to be satisfied at all, or even to be taken into account. The result is a situation where the in-group operate a generalised exclusion through a monopolisation of space and resources to meet their own “highest-order interests”, at the expense of denying any space whatsoever to the excluded.

It is to be noted that the “primary goods” tend to be things which Rawls thinks the mere existence of a liberal state can deliver. They are, Rawls stresses, ‘mainly features of institutions’ and ‘objective features of citizens’ circumstances’ (PL 181). They tend not to be things which could be demanded from the state, or which would put the state’s existence in question. They include some things, such as “self-respect”, which have a contentious definition elsewhere in Rawls’s theory, and others, such as “personal integrity”, which it seems to be beyond the basic structure’s capacity to provide. Some, such as the ‘diverse opportunities’ necessary for effective free movement and choice of occupation and the ‘social bases of self-respect’ (CW 362-3), are extremely vague. Their theoretical role is often a little weaselly, since Rawls allows the vaguer parts of his theory to operate as a let-out from difficulties. (Something which could not otherwise be valued in terms of primary goods may nevertheless be seen as necessary for one’s self-respect, for instance). Rawls believes that all claims on grounds of justice must take the form of claims for primary goods, even if the claim is based on special needs (PL 190, 326). This would seem to suggest that one could not, for instance, demand a wheelchair as a “primary good”; one could only claim (for instance) financial assistance. Similarly, one could not claim to be unfairly treated because actual outcomes fall short of reasonable expectations, because the relevant measure is the primary goods received by the average representative man. (This could lead to situations where, for instance, a general belief that the free market increases wellbeing is used as a pretext to ignore its actual effects in a specific context, as is common practice in the IMF, World Bank and WTO). The idea of primary goods would also preclude any inclusion of irreducible goods, and this has crucial effects in terms of the practical implications of Rawls’s theory. In general, the idea of primary goods reinforces the tendency for Rawls’s theory to assess the system on its own terms (its procedures and standards), rather than in terms of its effects on others. Rawls is of the view that the use of primary goods would ‘eliminate… socially divisive and irreconcilable conflicts’ which arise form particular conceptions of the good (PL 330). However, this can only occur if the parties to the conflict put the necessity of unity, or the supposed shared essence of the person, above all else. It may simply displace conflicts from the symmetrical relation between two conceptions of the good to the asymmetrical relation between an included and an excluded group.

Not surprisingly, the idea of “primary goods” has come under attack from critics such as Brian Barry, who suggests that the veil of ignorance would not rule out the possibility of the parties endorsing utility functions by postponing specific provisions for later stages when the veil is lifted (The Liberal Theory of Justice Chapter 9), and Adina Schwartz, who suggests that Rawls’s model is not in fact neutral. Some ways of life, such as those of an austere ascetic, a deep ecologist and a revolutionary socialist, do not require greater rather than lesser amounts of primary goods, so that Rawls is implicitly biased towards some ways of life (****). Similarly, Grundman and Mantziaris argue that the idea of primary goods contradicts the idea of autonomy, that it rests on a confusion of philosophical and social conceptions of incommensurability and that it renders it ‘structurally impossible’ to appeal for specific goods excluded form the list, thereby making it difficult to interact with opponents of the scheme (Fundamentalist Intolerance… 585-7). Similarly, David Lyons says that the idea of primary goods ‘is based on a kind of statistical norm’ and ‘favors those closest to the norm’, in contradiction with Rawls’s claim to eliminate bias through the veil of ignorance (in Daniels ed., 163). Will Kymlicka has attempted to defend Rawls against such criticisms, suggesting that even these ways of life require particular arrangements of resources (Lib. Individualism… 888). This is true, but misses the point. It proves neither that such people benefit personally from having the greatest possible amount of primary goods, nor that Rawls treats such people fairly. Kymlicka assumes they can be treated fairly in a society which is run so that all get the greatest possible access to primary goods, and that the goods required by such dissident ways of life can be reduced to the primary-goods index, with the implication that any demands which exceed their place in liberal society are excessive and unreasonable, just as if they were demanding that economic inequalities be arranged to their advantage. Kymlicka suggests that it is ‘selfish to demand’ more than one receives in a liberal society (890), and this characterological rhetoric covers the way in which he relies on the naturalisation of capitalism. The implication that deep ecologists and revolutionary socialists have demands which can be encompassed in a basically distributive and consumer-driven model is a variety of form-imposition. It goes hand-in-hand with the endorsement of a system which is not simply distributive, but is also a system of repressive territorialisation and seizure of “resources”. Again, liberals’ perverse logic dovetails with that of the capitalist state, as when animal liberationists are charged with burglary for liberating animals from vivisectionists and fur farmers. In such cases, liberals do not so much resolve the conflict as peremptorily adopt the rhetoric of one of the conflicting forces. Kymlicka also ignores Rawls’s work-cultism as regards the possibility of ascetic monks, and the fact that the resources demanded by the groups he discusses may not be on the list of primary goods in any case (for instance, untamed nature in the case of some deep ecologists). It should also be noted that Rawls’s assumption of the use of a universal equivalent in economics and therefore the operation of a general law of value are undeclared implications of his concepts of “income and wealth”, which are derived in the original position. In other words, original position parties are implicitly assumed either to desire or to naturalise capitalist relations between objects and people, presumably complete with commodity fetishism and the constant process of capitalist reterritorialisation and subsumption.

The idea of an exclusionary essence of the person is the main device for constructing an in-group in Rawls’s theory, but an early essay, “Outline of a Decision Procedure for Ethics”, reveals another such conception. In this essay, only a limited class of people, “competent moral judges”, are considered relevant to ethical theory. To qualify for this group, one must have a string of characteristics which identify one as, among other things, psychologically normal. For instance, a competent judge should know whatever ‘it is reasonable to expect the average intelligent man to know’, to have normal intelligence and to feel empathy (CW 2-3). The role of empathy expresses the theme of sameness in Rawls’s thought, expressing an ‘imaginative experience’ which implies psychological similarity between the judge and the judged (and which in fact could only lead to the exclusionary practice of misleadingly assuming others to be acting on motives familiar to oneself). Rawls claims that he accepts principles because competent judges endorse them, rather than vice-versa. He introduces the idea of a competent judge because it is ‘believed to be necessary’, and is derived from common sense (or in Rawls’s terms, what ‘we’ recognise in everyday life) (CW 4). Despite clear-cut psychological normalism, including the direct construction of a binary between competent judges and the “mentally ill” (which would mean that the normal are more competent than the psychologically different even when judging the actions of the latter), Rawls nevertheless claims that his conception of competent judges avoids criteria which are the ‘privileged possession of any race, class, or group’, offering instead criteria ‘which belong to men everywhere’. When assessing principles, what matters is only their acceptability to competent moral judges, not to anyone else (CW 10-11). ‘A man who rejects the conditions imposed on a considered judgment of a competent person could no longer profess to judge at all’ (CW 72). Though he later drops the terminology, the exclusionary logic of this analysis recurs throughout Rawls’s work. For instance, in a later essay, he stresses that only the reflective equilibrium of rational persons is relevant to ethical theory (CW 301). He also admits that, while ‘no race or recognized group’ lacks moral personality, it is possible for ‘scattered individuals’ to “lack” it through what he terms a ‘defect or deprivation’. Such people are victims of fortuitous circumstances or poverty (TJ 443). Justice is limited to those who have the appropriate powers. ‘Those who can give justice are owed justice’, and ‘We are not required to give strict justice… to creatures lacking this capacity’ (TJ 446, 448). One is only owed justice if one has certain ‘natural attributes’ (TJ 397). (Despite Rawls’s denials, one could imagine the implications if a specific group was labelled as lacking, or less likely to have, basic moral capacities. For instance, a group could be classified as genetically inferior because its members are deemed more prone to commit crime, which for Rawls shows an insufficient sense of justice. Since this shows the group to be prone to lack the characteristics of persons, they could be labelled almost as morally subhuman).

I also feel that the basic assumptions of Rawls’s theory do not change between his early and late work. The main alteration is that, whereas in the early work he claims to be constructing a general ethical theory, in the later work he claims to be constructing a more limited “political” conception which, however, is to have primacy within all the ethical conceptions included in a sphere of legitimate dissent. The role of the conception of the person is the same in both cases. As regards his apparent change of heart in “Justice As Fairness: Political Not Metaphysical” and subsequent works, it is clear that his defence against the charge that he has a metaphysical conception of the person is mainly a defence against the accusation that his portrayal of the parties in the original position reflects assumptions about what people are actually like. On closer inspection, he admits that there may be a metaphysical doctrine involved. He writes: ‘no particular metaphysical doctrine of the person, distinctive and opposed to other metaphysical doctrines, appears among [justice as fairness’s] premises, or seems required by its argument. If metaphysical presuppositions are involved, perhaps they are so general that they would not distinguish between the distinctive metaphysical views… with which philosophy has traditionally been concerned’ (CW 404, my emphasis). This leaves open the possibility that it nevertheless reflects a metaphysical view which is sufficiently widespread in mainstream theory to seem uncontroversial to Rawls, but which may nevertheless be oppressive, false and/or controversial when compared with less mainstream views (e.g. those of Deleuze, Foucault and psychoanalysis).

In any case, I think Rawls is misrepresenting his theory if he claims it does not involve a conception of an essence of the self. If the political is to be taken as primary over other fields, and a particular view of the person is to dominate the political, then logically, the view of the person used in the political sphere must be in some sense essential and fundamental in relation to all other views. There is still a noumenal self, even if its legitimation is different to that used by Kant and in Rawls’s earlier work. Larmore’s claim that only the citizen is an individualist model in Rawls’s theory (Larmore 346) is a spurious defence, because the role of the citizen has a primacy Rawls does not explicitly accord it. It is allowed to found coercive action against those acting on other identities, to limit claims based on other identities and to rule out absolutely anything which exceeds it. It is therefore socially primary and not one identity among many, so that anything which “only” applies to citizens and not to other identities is in fact socially primary. The effect is exactly the same as if Rawls had claimed directly that the “citizen” is the essence of the self (as his idea of a highest-order interest in justice in any case implies).

The mythical character of Rawls’s conception of the person should already be apparent from what he actually says. It becomes clearer when one starts to consider what exact gesture is involved when Rawls writes of “describing” or “regarding” people as or “assuming” people to be the various elements of his conception. On the surface, the rhetoric used is positivistic, and appears to reflect a claim that people in fact are like this. This is the reading of Rawls which is common among social scientists such as Clements and Hauptmann (****), who use Rawlsian language to account for the outcomes of experiments in economics and to modify rational choice theory by supplementing the rational with the reasonable (although their results would also be compatible with any theory which affirms an ethical element in choice). It is also a reading Rawls sometimes invites. Occasionally, he suggests that he is engaged in a descriptive project; for instance, he says that moral actions are a way ‘for persons to express their nature as free and equal rational beings’ (TJ 390) and that seeing them as free and equal meets people’s ‘fundamental needs’ (JAFAR 200), which clearly suggests that persons have such a nature and needs prior to Rawls’s theoretical manoeuvres. He also states that the existence of the so-called moral capacities as something ‘certain’ and as an ‘aspect of human psychology’ (CW 288). This is also suggested on those occasions when Rawls refers to his conception as a simplification, since a characteristic isolated for analytical purposes must nevertheless be a “real” characteristic and not an imagined or inferred one. This is, however, an unlikely reading. After all, the conception of the person is derived from traditions of political thought, not from analysis of actual people, and Rawls goes out of his way to differentiate it from psychological theories of the human mind. In other words, Rawls is prepared to consider people to be something they in fact are not, which clearly suggests that he is using a discourse involving mythical and impositional elements. Indeed, Rawls directly admits that he is treating people as something they are not, but thinks this is alright as long as it is accurate enough for political purposes (whatever that means) (CW 422). The reference to “accuracy” still suggests that the view must be approximately true, restoring the suggestion that it has some kind of empirical referent. He also attaches no value to truth in what he terms the “thin” theory of the good (TJ 368), which would seem to imply that he is unconcerned whether the assumptions he makes are true or not.

Even if the conception is intended as empirical, however, it is still an oppressive discourse, since the primacy of the noumenal self over the rest of the psyche is established through naturalisation and since the basis for the entire conception is firmly intensional. (In addition, of course, if the claim is empirical then it is falsifiable: “persons are assumed to be free and equal” would have to be treated in the same way as, for instance, “dogs are assumed to bark”. It would have to be tested and instantiated, and, if necessary, altered, qualified or even dropped, depending on evidence about actual “persons”. My suspicion, however, is that there is an internal or definitional relation between “person” and “free and equal” in Rawls’s work, which makes phrases of this kind structurally different from such descriptive phrases as “dogs bark”. Empirical assumptions only occur when Rawls begins to examine the “stability” of his conception and whether this conception is “realistic”. Hence, he admits that his theory depends for implementation on citizens sharing his conception of the person and actually requiring the primary goods it offers [CW 454}. I shall return to the “general facts” of psychology and the discussion of “stability” later).

I would suggest that the empirical validity of Rawls’s theory is significant for far more than merely its stability. After all, Rawls claims that ‘the arrangements of a just society are so suited to us that anything which is obviously necessary for it is accepted much like a physical necessity’ (TJ 403), that people really do want justice in his sense (PL 208) and that it delivers on the basis of ‘what our fundamental interests really are when it comes to the design of the basic structure’ (JAFAR 99). Further, people can act from Rawls’s conception of the person because they are ‘adequately represented’ in it and so can express ‘full autonomy’ through it (CW 320), and everyone, provided social structures take a particular form, can be an equal citizen (TJ 82). Also, ‘acting justly is something we want to do as free and equal rational beings’ (TJ 501). It surely cannot be so well-suited unless it in fact has some inherent relevance, although all of these phrases may be more than they seem if Rawls’s extra-empirical use of positivistic language has extended into them. Indeed, the importance of having a “just basic structure” would seem to depend on this structure having some relevance to actual people. Therefore, it is not clear what useful function is performed by separating, as Rawls does, the assumptions of the theory from the issues of “implementation”. I suspect the function this performs is to conceal the role of the systematising drive beneath a concern for “liberties” and actual distributions. Meanwhile, it constructs a situation where powerful institutions can “legitimately” make assumptions about what people actually are, which may be thoroughly mistaken and even incomprehensible to those on whom they are imposed. Unless it is descriptively accurate, Rawls’s conception of the person is nothing more than a violent, Kafkaesque interpellation (as for instance if someone who has an unconditional drive is told in court that she or he is capable of being responsible for her or his ends, when this is untrue). It is no more valid than saying, “a person is taken to be a dog”, or “a person is able to fly”, or “citizens are able to agree on a single comprehensive doctrine”. (If the claim is to avoid rendering some groups voiceless, it must, furthermore, apply to all people and not simply to the majority). If, as I suspect, the desire for a conception of the person derives from the desire for a conception of justice, this means that the systematising drive is being given sufficient priority to override actuality - people are to be assumed to be what it is convenient for social unity to conceive them to be, rather than what they are - and this gesture is retrospectively concealed beneath the claim that Rawls’s theory is “normative” not “descriptive”.

One problem with Rawls’s theory, assuming that it is not descriptive, is the problem of who exactly is doing the “describing” and “assuming”. Clearly there is a standpoint in his theory which involves some sitting in judgement over others on the basis of this assumption - at its least elitist, of citizens judging each other, but plausibly, of an elite sitting in judgement over citizens. Clearly, therefore, the assumption carries an implicit asymmetry between “assumer” and object, which hints at a possible relation of panoptical power.

Another problem is that Rawls clearly derives from his conception of the person a duty to normalise people into conforming to his model, or to give them what they need in order to attain this status and to ‘play our part in society’, for instance, through medical treatment (JAFAR 175). If people are in fact the way Rawls claims, this could not possibly lead to this kind of transformative duty. The implication seems to be that it is desirable for all people to conform to this model, and that people should be allowed, or even forced, to become this kind of person if at all possible. This is a potential threat to any liberties of those outside the model, and it involves a general entrapment of desire in particular modes of action.

Rawls repeatedly insists that his conception of the person is “normative” - for instance, it is ‘a normative political conception’, not a metaphysical or psychological one (CW 480), and even that it is normative rather than descriptive (CW 397) - although this is misleading given the vocabulary he uses. A normative claim would more usually be expressed along the lines of, for instance, the term “people should be…”, rather than in a modality which is clearly referential. Therefore, Rawls’s language is misleading, giving an appearance of inclusiveness. He does admit that his conception is selective, however. For instance, he claims that ‘the political virtues are specified as those qualities of citizens’ moral character important in securing a just basic structure over time’ (JAFAR 142). This claim is itself somewhat confusing. One possible reading is that Rawls judges people by a standard external to them, i.e. he values people depending on how well they enable the emergence of a just basic structure. This would locate Rawls as a political authoritarian using a directly impositional and invalidatory discourse. The way Rawls avoids this is by assuming that a “just” basic structure is valuable precisely because of what it offers to “moral persons”. However, if read in this way, the argument is clearly circular, and also suggests that what he supports is of value because it benefits those it happens to include. Somewhere along the line, a value superior to “persons”, which establishes them as morally significant or not, has found its way into Rawls’s account. If Rawls’s conception is “normative” only in that it selects particular elements from psychology, it is nevertheless dependent on the claim that these characteristics are present in psychology (alongside whichever other aspects are bracketed).

One suggestion which sometimes arises is that the conceptions of the person and society are ideals, which Rawls introduces in an attempt to bring them into being. This is the reading put forward by Ed Wingenbach, who suggests that the conception of the person is ‘about the political implications of the democratic tradition’ and is ‘not a statement concerning actual facts about individuals’ (Unjust Context 215). For instance, he refers to his conceptions of the person and society as ‘ideals’ latent in public culture (CW 401), and to the idea of the good citizen as an ideal, or ‘(partial) conception of moral worth’ (JAFAR 142). His conceptions of the person and of society are ‘ideals’ he wants to ‘realise’ (JAFAR 178). He also says at one point that he takes people to be what he wants them to be (PL 213). His goal is to establish a fair basic structure and eliminate injustice, and for this reason, the ideal is an aim (PL 285). He seems to see these ideals operating by directly modifying action, so the division between “is” and “ought” is blurred: ‘citizens who count as part of their good being reasonable and rational and being seen by others as such are moved by reasons of their good to do as justice requires’ (JAFAR 202). However, he does not say that people “should” be this way (which would not necessarily be oppressive), but rather, that people should be viewed as being this way. Unless Rawls assumes that pretending that someone is an X will in fact make her or him an X (which, would seem to contradict Rawls’s “general facts of psychology”), there is a great difference between these two claims, and there is no reason whatsoever why a desire that something “should” be the case should entail an attempt to assert that it is the case. Rawls does, however, seem to have this strange assumption. ‘Now let us suppose that certain principles and their justification do, in fact, articulate just such a conception of the person. Then, given our dependence on society, we could not be this sort of person unless institutions developed and encouraged our capacity so to act… People’s attaining this conception of the person would be the achievement of social cooperation’ (CW 294). The institutionalisation of the image of citizens as free and equal will, Rawls assumes, actually make people conform to the image. ‘[W]hen citizens publicly recognize that the basic structure satisfies the two principles, this public recognition itself not only encourages mutual trust… but also nurtures the development of attitudes and habits of mind necessary for willing and fruitful social cooperation’ (JAFAR 117). He assumes that the application of his conception would actually lead to the ‘virtues’ he desires (CW 444). In other words, because Rawls assumes people to passively receive whatever conception is transmitted by society, he sees nothing wrong with relying on territorialising mechanisms to force people to view themselves in the way he advocates. The use of an ideal as the basis of ethics is implicitly impositional. Iris Marion Young criticises Rawls on the grounds that his ideal is a variety of mastery and therefore does not open to others (JPD 5). It should be noted that Rawls intends to use “institutions”, i.e. mechanisms of social control, to bring about the society he wishes; he is not relying on the internal attractiveness of the ideal itself. Furthermore, his reliance on ideals is inconsistent. Why, asks Galston, is he prepared to idealise about one’s sense of justice, but not about one’s level of intelligence (Moral Personality 513-14)? Furthermore, adds Patrick Hayden, if freedom and equality are to be treated as general ideals, why are the contingencies of geography and history to determine who is to be treated equally, through the nation-state system (Rawls, Human Rights and Cultural Pluralism 51)?

The idea that people simplistically act on the basis of consciously-affirmed ideals is an incorrect assumption, and anyway, does not require the gesture of pretending that people already “are” what the ideal says they are. Indeed, such a gesture would seem to preclude the whole purpose of having ideals, i.e. to actually change the world so as to improve it in some way. If, for instance society is a fair system of cooperation, and if the supposedly commonsensical conception to this effect is to be taken as correct, this assumption is a barrier to any possible attempt to turn society into a fair system of cooperation. One cannot, after all, make something into what it already is. Yet such an assumption is precisely what is so often implicit in essentialism, and which renders it oppressive. By positing the ideal as something which is already an essence, one imposes voicelessness on the actually-existing object which one intends to change. One insists that everyone play a game of “let’s pretend” regarding the status of the object (in this case, “the person” and “society”), and to conceal or repress other aspects of it, including those which contradict the assumption that it fits the essentialist category at all. Rawls both gentrifies and exaggerates the effectiveness and “spiritual” significance of institutions of social control. Indeed, his own terms suggest that he is relying on reactive motivations, such as a desire to conform to what others expect and to be seen by others as fitting a particular preconceived role, to create the bridge between “is” and “ought”. Declaring something to be a “moral virtue” is not enough to make it come into being, and punishing or discriminating against those who lack the “virtue” is also not demonstrably useful; it could easily lead to labelling and “deviance amplification”. It is also worth noting that, if it is an ideal, it must nonetheless be justified vis-à-vis alternative ideals, especially those which offer a rather less petty-minded image of the self than is involved in Rawls’s theory. Referring back to existing beliefs and self-conceptions is not enough. If one accepts both the claim that ideals, if realised in institutions, are realised and the claim that Rawls’s conception of the person reflects how people in constitutional democracies in fact view themselves, his theory becomes a complete “repressive reduction of thought to the present”, and can never be anything else. (Rawls also implicitly admits limits to what can be constructed through ideals, but he does this as an afterthought, in the sections on “stability”. For instance, he says of the ‘ideal’ that ‘it must be possible for people to honor it sufficiently closely’ for it to be feasible, and that this requires it to be compatible with human nature and general social facts [CW 321]).

In his later work particularly, Rawls is inclined to present his conception of the person as an analytical simplification, used to gain a clearer view of a question he sees as fundamental. This is sometimes related to his desire to increase social cohesion through the construction of an overlapping consensus around a political conception of justice. His advocacy of an overtly Kantian model of a noumenal self reached its peak in “Kantian Constructivism in Moral Theory”, and went into decline soon afterwards. In “Justice As Fairness: Political Not Metaphysical”, Rawls steps back from his earlier claims about his concept of the person. ‘The description of the parties [in the original position] may seem to presuppose some metaphysical conception of the person, for example, that the essential nature of persons is independent of and prior to their contingent attributes… But this is an illusion caused by not seeing the original position as a device of representation’ (CW 402). Rawls justifies his emphasis on citizens with the characteristics he ascribes to them on the grounds that this enables him ‘to achieve a clear and uncluttered view of… the most fundamental question of political justice’ (PL 20). Therefore, the assumption that everyone has a conception of justice is ‘purely formal’, relating to the original position as an exercise in modelling, and the assumption must in practice be limited by common sense and knowledge of human nature (PL 315). Rawls’s theory is, he claims, abstract in the sense used in economic theory: he singles out one aspect of society as especially significant and brackets other aspects (CW 437), selecting relevant characteristics in order to solve a specific problem (PL 300). Abstractions are idealisations used to ‘gain a clear and uncluttered view of a question seen as fundamental’ (JAFAR 8), for instance, via ‘simplifying assumptions about citizens’ capabilities’ (LN 13). ‘The first problem of justice concerns the relations among those who in the everyday course of things are full and active participants in society and directly or indirectly associated together’, and who are ‘engaged in social cooperation’ (TJ 84). So people are assumed, for instance, to be cooperative because this is ‘fundamental’ and so this case ‘should be examined first’ (JAFAR 179).

This version reduces those who do not fit his model to a secondary status, outside the core of his theory, and this itself is significant (notwithstanding the fact that he has implicitly metaphorised such groups as “clutter”). For instance, it is of no small importance if the psychologically “normal” see relations within their group as more “fundamental” than relations with the psychologically different. In this version of Rawls’s theory, the oppressive role of the myth of the person operates via form-imposition. A structure is implicitly constructed which places particular people at the core, and reduces others to a secondary status, as a theoretical afterthought unable in practice to affect the core of the structures which nevertheless are imposed on them (even if in theory the resolution of the “bracketed” questions is assumed to undermine the validity of the core perspective). Also, this model restores the claim that the theory is basically empirical. The analytical simplification can only operate if it is actually one of the characteristics of those it describes, since otherwise it is not a simplification but a misrepresentation. Even if, for instance, the question of social cooperation is fundamental, it is misleading to assume that this means one can assume people to be cooperative. It may simply be that it is fundamental to find out when and whether people cooperate.

There is a paradox in the transition from the early to the later Rawls, because the abandonment of explicitly essentialist and universalist formulations goes hand-in-hand with a movement of the Kantian conception of the person towards the centre of Rawls’s theory. In the first edition of A Theory of Justice, Rawls adopts a strongly essentialist position of a ‘standpoint of eternity’ and ‘[p]urity of heart’, extrahistorical and atemporal (orig. TJ 587). This position later disappears, but the revisions of Rawls’s principles incorporate equally essentialist elements. In particular, his reliance on the “two moral capacities” to explain the choices in the original position – for instance, the primary goods and the choice of an “adequate” (no longer maximal) regime of liberty – moves Kantianism from a peripheral “interpretation” to a central theoretical element.

Another, more obviously oppressive, possibility is that Rawls uses the model of the person as a standard of assessment which regulates access to the rights and benefits he attaches to being a member of society. For instance, the term ‘conception of moral worth’ (JAFAR 142) could be taken to imply that only those who conform to Rawls’s model of the person deserve to be considered in any way in terms of the distribution of material goods and in terms of liberty. This implication also arises from the assumption that people are equal only because (and, by unstated implication, if) they have the two moral powers. ‘Their having these powers to the requisite minimum degree to be fully cooperating members of society makes persons equal’ (PL 19). ‘The task is to articulate a public conception of justice that all can live with who regard their person and their relation to society in a certain way’ (CW 306) (NOTE: Rawls also calls this a ‘social point of view that all can accept’ in the same passage [CW 307], suggesting he does not see any exclusion as involved). In this case, the discourse would be impositional, and operate to legitimate the invalidation of those who fall outside the conception (especially since the “sense of justice” is a reified term, referring in practice to a set of contestable beliefs and not to a “sense”). It is possible to construct any group as primary by attaching moral value to what this group shares (and others do not), but this does not explain why one should accept such a valuation, especially when it is carried out by the in-group itself. This is, indeed, how I suspect the conception operates in practice, and this confirms my broader suspicion that mythical discourses usually cover impositional ones. However, Rawls usually denies that this is the case, and he rarely discusses the “outside” which would be involved very obviously in such a conception. One finds him using a language which elides the excluded group so as to maintain the image of an inclusive theory. For instance, he uses the phrase ‘persons as capable…’ instead of the phrase ‘those people who are capable…’ when discussing his assumption that people can cooperate fully in society. In this way, the exclusionary operation of the concept is disguised.

Rawls admits that the basis for his conception is derived from the particular sub-section of “common sense” which he elevates to a position of primacy in his approach (NOTE: this reflects a wider tendency in Rawls’s theory to assume, mistakenly, that “common sense” ideas are in fact “knowledge”, and therefore to put them beyond criticism. This is a separate issue which I shall not examine here). Hence, he states that his conception of the person is drawn from how people are seen by courts, politicians and their ilk (JAFAR 19-20), and on another occasion, he adds that his conception is ‘how citizens actually think of themselves’ and each other (CW 406; cf. LN 33, PL 300). He also claims later that he is trying to ‘single out the essential aspects of our conception of ourselves’ (CW 308) and that his view is justified because ‘it is rooted in human thought and feeling, and tied in with our ends and aspirations’ (TJ 343) and ‘begins from our everyday conception of persons as basic units of thought, deliberation, and responsibility’ (CW 397). Conditions set on the original position are ways of modelling existing convictions about what people are like (JAFAR 18). It is unclear why this should be an excuse for taking the views as dogmas, and the gesture involved here is basically impositional. The fact that elites, or even the people being so viewed, view people in this way is not in any sense evidence that they are like this. Politicians and judges may simply be misguided, or biased in what they assume. (It is, for instance, convenient for courts to assume that people are capable of “responsibility for ends”, because this precludes their having to take into account the actual complexities of the flows of desire and enables them to pretend that their impositional practices reflect a basically fair and “reasonable” set of assumptions. This no more proves the validity of this claim than if, for instance, the Catholic Church were to claim, just as conveniently, to be the sole repository of truth). As regards people’s self-conception, this could be influenced by a variety of psychological factors of the kind commonly discussed in psychoanalysis; for instance, they may have internalised parental assumptions. Since Rawls specifically says it is how “citizens” view themselves, the claim might even be circular (i.e. only those who view themselves as citizens are citizens, so that the fact that this group see themselves in this way is established tautologically, and its use to confirm Rawls’s theory is mistaken). The same problem arises with the unspecified “we” of the other quotes. Another problem is that Rawls never establishes that people actually do have the self-conception he infers. He also does not establish why he gives moral primacy to the group which happens to self-identify in this way.

It should be added that, in “Justice as Fairness: Political Not Metaphysical”, Rawls wishes to deny that his theory is based on ‘claims about the essential nature and identity of persons’ (CW 388). It is not clear how he can deny this, since the model discussed above is clearly a model of an essence. The distinction seems to rely on the idea that this model of the person, although “normative”, is also “political”, and only applies to people’s deliberations in their role as citizens. He may simply be hinting at what he later states in Political Liberalism: that his position is not strictly-speaking a form of transcendental idealism because of the contract component (PL 285), or perhaps to his (valid) claim that his model of rational persons in the original position does not reflect his model of moral psychology (PL 27-8). However, since the role of citizen is taken to be primary over other roles whenever these conflict, this would seem to make it also a general principle. As Galston has noted, the political conception of the person cannot cohere with other conceptions held within comprehensive doctrines if, as Rawls assumes, it is to have primacy whenever the two conceptions collide (Moral Personality 503). A number of authors have also observed the contradictions apparent in any embracing of a political ideal by those who do not share it in their comprehensive doctrines. For example, Richard Bellamy says that if Rawls’s conception of the person only applies in the public sphere, ‘it requires all but the liberal to be endowed with an impossibly schizophrenic [by which he means split and self-contradictory] personality’ (Lib and Modern Society 240). Also, Rawls’s language appears to suggest that people are in fact this way, and such a claim is neither normative nor political (in Rawls’s sense), but rather, is empirical. If people are not in fact the way Rawls describes us/them, his model of the person is empirically inaccurate and is, in his own terms, a ‘delusion’ and ‘false consciousness’. For Rawls, a conception is a delusion if it involves ‘false and unreasonable beliefs’ or ‘irrational and inhuman values’, and he makes a claim for his theory that it tends to eliminate such ‘false beliefs’. He demands, for instance, that the general facts he uses be neither illusions nor delusions, although for some strange reason he thinks his reliance on common sense gives a good chance of this (JAFAR 121-2). Similarly, the idea of society as a system of cooperation between citizens is delusional and/or false consciousness on this definition. He also tries to say that the urge to treat people as “moral persons” is specific to justice as fairness as a “Kantian” view, although this does not seem to make it a part of a comprehensive doctrine; it remains part of the public political conception (PL 273).

Setting up a theory based on a decision that “people are regarded as X” is oppressive regardless of which of the various senses is meant, albeit in various different ways. The only way the view can escape such criticism is if it is empirical and also is actually true, although even here, the question arises of how its factual status is transferred into politics. To show this, one could construct parallel forms. For instance, one might say, “people are assumed to have two legs and therefore to be able to climb stairs”. This is, and the politics resulting from it are, exclusionary, regardless of its exact meaning (the only conceivable exception being if the claim were empirical and were true, which it is not).

Perhaps the biggest advantage for Rawls of his formulation is that it allows him to evade, elide or fudge issues regarding whether his conception is a (simplified) description, an ideal, a standard of judgement or a “common sense” dogma. In this way, the oppressive logic embedded in the assumption can be hidden beneath a constant process of shifting signifiers. In this sense, the many legitimations and specifications which surround the unitary phrase “people are regarded as…” is itself indicative. On closer inspection, the oppressive logic still emerges through a close analysis of Rawls’s discourse.

For purposes of completeness, I should add one final use of the concept of the person which clarifies its use a little. In a well-ordered society - the “realistic utopia” which is the end-point of Rawls’s theory - people are, want to be, see themselves as and see each other as the kind of person Rawls posits, and are to share a fundamental aim of social justice (TJ 475, CW 309, 324, 326, 431, PL 306). However, he also assumes that a well-ordered society is an honest society where the political principles are publicly accepted and taken in good faith. It is not a society which depends either on institutionalised delusions (false beliefs) or false consciousness in a Marxist sense. It is not to be guided by ‘useful illusions’ (e.g. CW 326, JAFAR 121, TJ 368). However, the role of the concepts of society, the person and the citizen in his own theory seem to belie this idea. Assuming people to be something they only aspire to be, or to be something they are not, is clearly either “delusion” or “false consciousness” in Rawls’s sense. Because its basic logic involves entering into self-alterity, Rawls’s theory could not conceivably lead to the transparent immediacy of a well-ordered society.

One thing that is absolutely clear about Rawls’s theory, and that overrides his denials regarding the essentialist role of his conception of the person, is that the political ideas derived from this conception are primary over everything else. This assumption is necessary for Rawls’s theory, since otherwise, he could not justify coercive state apparatuses on the basis of it, and it demonstrates that the assumptions are not merely analytical. They construct a potential system of power-relations which would impose the assumptions on everyone. Thus, Rawls refers to ‘the liberal idea that persons are citizens first’ (LN 66), later adding that the ‘adult members of families and other associations are equal citizens first. That is their basic position’ (JAFAR 166). This primacy is assumed to be based on a characteristic of “persons”: what Rawls terms a “highest-order interest” in how other interests are ‘regulated’ and ‘shaped’ by social institutions (CW 299). Lest this drive be assumed to be a drive to pursue freedom from institutional control, it turns out simply to be a drive to retain the “right” to change one’s interests, which incorporates a duty to do so if this is required by the state. People have highest-order interests in realizing and exercising the moral powers, and these interests are decisive over all others, ‘supremely regulative and effective’ and ‘govern[ing] the deliberation and conduct’ of moral persons, and are ‘moved by these interests to secure the development and exercise of the moral powers’ (CW 312). Any person’s conception of the good is ‘in essential respects subordinate to the highest-order interests’ (CW 313). The desire for justice is ‘regulative of other desires’ (TJ 462), including the other higher-order moral desire for the right to alter one’s conception of the good, and the idea of justice, along with the ideal of the person it embodies, is assumed to be shared by everyone, at least in a well-ordered society (JAFAR 199-200).

The so-called “thin” aspects of morality are also to be accepted unanimously, since otherwise it would be impossible to construct a single basis for comparisons (TJ 392-3). Justice is to operate as a meta-morality, floating above the different conceptions of the good and comprehensive doctrines and able to adjudicate between them (TJ 194), and regulative of all further agreements (TJ 10). In short, justice is not one ethic among many; it is a master-signifier, justifying a state which is prepared to override all specificities in the interests of this imagined universal. It is no coincidence that “right” is singular whereas “goods” are plural, since the singular master-signifier is to be ‘a condition of good’ (PL 334). ‘[T]he desire to express our nature as a free and equal rational being can be fulfilled only by acting on the principles of right and justice as having full priority. This is a consequence of the condition of finality: since these principles are regulative, the desire to act upon them is satisfied only to the extent that it is likewise regulative with respect to other desires. It is acting from this precedence that expresses our freedom from contingency and happenstance’ (TJ 503). There is no alternative, Rawls claims, to the sense of justice ‘governing our other aims’; it cannot be ‘compromised and balanced against other ends as one desire among the rest’, because it is ‘a desire to conduct oneself in a certain way above all else, a striving that contains within itself its own priority’ (TJ 503). In the original position people ‘acknowledge’ that the two principles ‘are to override [their deepest moral] beliefs if there is a conflict’ (TJ 194). It is a ‘good’ for everyone to act justly (TJ 350), and Rawls’s theory expresses a ‘kernel’ of ethics which should be universal (TJ 194). Further, Rawls assumes people will experience justice of his kind as a good (PL 202-3). Justice is to be a generalised system of coordination and control, a set of ‘mutual ties which bind the entire society, each and every member of it’, around ‘support for the basic structure’ (TJ 438).

Social discussions are to be constrained by statist codings to remain within definite limits. ‘It is from the position of equal citizenship that persons join the various religious associations, and it is from this position that they should conduct their discussions with one another’ (TJ 192). One is also required according to Rawls to share ‘final ends in common’ with others in the political sphere (PL 202). Further, the idea of primary goods clearly involves treating people in line with their supposed essence, regardless of all else. Such elements impose voicelessness on actual people by insisting that people enter into a discourse of self-alterity in order to be heard “in public”. This discourse of self-alterity is regulated by the fictive image of the “free and equal citizen”, and its primacy is also established via an impositional gesture. Indeed, Rawls admits that it has a repressive role, since ‘the principles of any reasonable political conception must impose restrictions on permissible comprehensive views, and the basic institutions those principles require inevitably encourage some ways of life and discourage others, or even exclude them altogether’ (JAFAR 153). There is no structural requirement in Rawls’s theory that only oppressive or otherwise specified conceptions be excluded (though he would clearly like the reader to believe this is the case, given his constant references to slavery and religious intolerance as examples). Any conception which exceeds the pre-constructed ethical “kernel” can be excluded and repressed. Just institutions are only to help citizens further their ends if these ends happen to fall within the principles of justice the institutions impose (CW 361).

Clearly this is impositional. It is not at all clear that people feel themselves to have such “higher-order interests”, and the implications of assuming they do for the structure of social relations are massive. What Rawls is insisting people have in order to qualify as moral agents includes regulative desires which suppress all other desires. This clearly restricts who can be included in his theory, and how. It gives an absolute primacy to any agent who can claim to express everyone’s regulative interests, even over and against their actual desires and needs. Further, only agents who not only have this desire but put it above all else qualify as persons in Rawls’s theory. It appears that “justice” often embodies itself in Rawls’s theory as the interests of the state (see below on the concept of “liberties”), so that Rawls is in fact advocating that people submit, even as regards their deepest needs and desires, to state control. The excuse that Rawls is simplifying by choosing the aspects of the person relevant to ethical theory is, therefore, not enough, because the elements which are bracketed out are not simply ignored theoretically but are also to be repressed psychologically and socially. I also suspect that, in spite of Rawls’s insistence that his theory involves inner emotional states, it would be impossible in practice to distinguish between reluctant public conformity and avid acceptance. Therefore, in practice, Rawls’s theory is an agenda for conformity.

The idea of an “overlapping consensus” does not significantly alter the situation, since the kernel remains the same. It is simply that, like any master-signifier, the political principles permit some leeway between different beliefs within their ambit. Hence, freedom of choice is to operate, but only ‘within the framework of just institutions’ (TJ 393). Many comprehensive doctrines are to endorse (or one might better say, to fill out) the principles, each ‘from its own point of view’, but they are to remain constrained by the principles of justice derived from ‘citizens’ essential interests’ (PL 134). ‘In a phrase: justice draws the limit, the good shows the point’, and conceptions of the good, including justifications for the political principles themselves, ‘must fall within the limits drawn - the space allowed - by that political conception itself’ (CW 450). Thus, people are free to rationalise the master-signifier however they choose, but it remains in the last instance an arbitrary imposition.

Another instance of the implications of Rawls’s conception of the person is his theory of representation. In this case, the implications in terms of subordination to the state are overtly stated. For Rawls, representatives are not to be ‘mere agents for their constituents’. Indeed, he gives them extensive autonomy. However, he claims they are nevertheless ‘representative in the substantive sense’ because they act on the citizen’s ‘first interest’ in ‘just and effective legislation’. They are only to further other interests within the limits of this first interest (TJ 199-200). In Rawls’s well-ordered society, few people would be active in politics. Rather, people would delegate political decisions to an active minority who ‘look after their concerns’ (TJ 200). This discourse is clearly substitutionist, allowing an elite to dominate through its supposed expression of an “interest”. Substitutionism also occurs in various passages regarding the use of state coercion, which Rawls mystifies as a direct expression of the coercive power of citizens, rather than as the actions of a special apparatus of repression.

The myth of the noumenal self operates in Rawls’s theory as a hurdle which every claim must cross in order to be heard. Its role is therefore to render a whole range of claims voiceless. All the things Rawls offers are derived from the essence and not from actualities. Crucially, since the essence is conceived as normal and conformist (“fully cooperating”), this usually means that everything is referred back to the good of the social system. Even something needed for personal wellbeing or survival is justified only by reference to the essence. ‘By regarding the strength of claims to medical care as tied to maintaining our capacity to be a normal member of society and restoring that capacity once it falls below the minimum required, a guideline is provided… for balancing the costs of such care’ with other concerns (JAFAR 174). Implicitly, this means that medical care which is necessary to be a citizen (e.g. to restore one’s efficiency as a worker) is to be funded, but anything necessary for personal wellbeing, but not for the role of citizen, is not. The resultant inflexibility is something which Rawls directly celebrates (PL 184). One only has a right to food and drink as a means to pursue the higher-order goods specified by the essence (PL 76). Similarly, Rawls affirms the “basic liberties” only as a way to enable the development of one’s moral powers, which is one reason why they end up limited to the so-called “fundamental cases” (e.g. JAFAR 45, 200, PL 417). These liberties are specified so as to enable the exercise of the two moral powers. ‘What distinguishes the two fundamental cases is, first, their connection with the realization of the fundamental interests of citizens regarded as free and equal as well as reasonable and rational’, and the existence of a liberty is considered more important the closer it gets to the two fundamental cases (JAFAR 113). These liberties are seen as realising people’s normal capacities (PL 337), and thus are implicitly conditional. They are affirmed because they express people’s supposed essence. ‘To exercise our [moral] powers [in the two fundamental cases] is essential to us as free and equal citizens’, in order to express our ‘fundamental interests’ (JAFAR 45-6).

Elsewhere, the conformist implications are more obvious. What Rawls gives, he gives in order to make people cooperative members of society (TJ xv). Mutual aid is a good thing, not because of its effects for people, but because it has a ‘pervasive effect on the quality of everyday life’ (TJ 298). Similarly, funding for the arts, science, mathematics and philosophy is to be justified, but only by reference to its beneficial effect on public culture. As long as it has this (reactive) basis, it is justified, but something which ‘realizes certain great excellences of thought, imagination and feeling’ - which is justified, therefore, for active reasons - is not (JAFAR 152). Education is justified only because it enables people to participate in social life and have a sense of self-worth, and (in a slightly sinister phrase) because it enables people to enjoy their society’s culture (TJ 87). Other examples of practices justified by reference to benefits to the public culture include such classic reactionary practices as so-called ‘national defence’ (JAFAR 152), i.e. the discourse of militarist apparatuses and the military-industrial complex.

The image involved in these justifications is invariably a conservative one. The “high-quality” social life Rawls desires does not require that people drop their petty-mindedness, but it does require an absence of overt conflict. It often looks very like an idealised suburbia where no-one feels happy unless “integrated” with everyone else. In other words, it is a society where self-alterity has triumphed over the actual. Also, a society where politicians and “citizens” are to decide to impose whatever is most beneficial for “public life” is one where freedom is always conditional and where the state is paramount. The system gives people only what it must give them to make them what it wants them to be. That such a society happens to justify some “liberties” is secondary to this. These liberties are, after all, themselves justified only because of their beneficial effects for society as defined by its rulers. Indeed, Rawls is inclined to define ideas convenient for constructing a system of social control to be ideas which can found political unity.

In general, it seems that Rawls permits the state to pick and choose which needs and desires it recognises, based on its own drive to maintain its own structure. It is crucial to realise that Rawls does not only apply his principles to those who fall within the model he constructs (as he should, if he follows through the claim that his theory only applies to people of this type, i.e. to “people as” free and equal [eg. CW 311]); a schema constructed only to handle ‘moral persons’ ends up imposed on those he might label immoral persons also. A society constructed as a system of cooperation among the included must somewhere relate to the excluded, and “justice as fairness” seems to have little to offer except Kafkaesque interpellations and hostile territorial regimes. Rawls generates from his principles a total territorialisation of the world, thereby implying the total dominance of everything by those people who conform to Rawls’s image. Actual people find themselves trumped by a series of essences which may have little or no reference to their beliefs, identities or character-structures. At worst, this could lead to a division of the world into two castes - those who conform closely to the essence and those who do not - with the former totally dominating the latter. In any case, as I shall show later, Rawls is not shy in his advocacy of normalisation and repression, when he slips out of his usual habit of pretending the excluded do not exist. The rights he guarantees are offered only to citizens, not to human beings in general. Therefore, they amount in effect to privileges resulting from inclusion.

Rawls faces criticism from a number of authors who suggest that his conception of the person reflects a particularity wrongly projected into the position of the universal. The problem is that he assumes the possibility of a particular “unmarked” person who can be differentiated at least conceptually from all the “marked” variants. This assumption has a long history in western philosophy, but is usually associated with the particularities of dominant groups assigned an unmarked status within particular binaries, such as men, white people and Europeans. It is little wonder, then, that Rawls has come under attack on these grounds. Seyla Benhabib accuses him of substitutionism of this kind, suggesting that his non-dialogical conception of universal personality is in fact an expression of the experiences of white upper- and middle-class men, imagined to be universal (The Generalised and the Concrete Other 81). Peter Handley (Theorising Disability, p. 115) suggests that Rawls provides a good example of the assumption Margaret Thornton terms “Benchmark Man”. This leads him to take a certain, limited type of person as the core of his theory, which leaves him unresponsive to issues around disability. Similarly, Christoph Fehige accuses Rawls of formulating the “higher-order” preferences in terms of his own particular preferences, adding that, by doing this, he ‘violates the neutrality condition’ (Justice Beyond Desires? 265-6; c.f. 275). For instance, Rawls’s principles discriminate against people who do not wish to change their beliefs, because the principles are designed for people who do wish to do so (280). Jacob Segal suggests that the self valued primarily for its capacity to choose, which he terms the disenchanted self, is a product of capitalism (Marx, Time, and the Deontological Project 249-50). Also, Steven Lukes terms Rawls’s theory ‘a theoretical attempt to isolate what cannot be isolated’, and suggests that the parties in the original position only represent ‘a subclass of rather cautious, modern, Western, liberal, democratic, individualistic men’, constructed as such via a very specific set of interests, incentives and “general facts” (Relativism… 183-4).

Rawls treats the world as something which exists solely for the private use of one particular group of people, who are permitted a total territorialisation (which may even extend to the bodies of the excluded), and he does not even seem to consider the possibility that this might be “unfair” or “unjust” to anyone else. The exclusion which results from his system is not only exclusion from the so-called “benefits of social cooperation”, but also from any kind of social or physical space. This includes things, such as natural resources and opportunities for association and action, which are not in any sense produced through “social cooperation”. To simplify a little, one might term Rawls’s theory “special pleading” by “free and equal moral persons”. If one does not accept Rawls’s unstated assumption that normal, submissive conformists are the only valuable type of human being and the sole focus of ethical theory, one also cannot accept the rest of his theory. Rawls nowhere explains why the “normal” form of life is to be given such an extensive primacy, and it is unclear if he is even aware that this is what his theory supposes. Since his theory appeals to “common sense” and to a systematising drive, both of which are characteristics of the in-group, it appears to allow this group to simply define itself as the sole ruler of the world. Indeed, Rawls’s model is so closed that at times it approaches tautology and circularity. (One might question whether “reasonableness” or a “sense of justice” exist independently from the specific principles they are used to ground).

The idea of society as a system of cooperation between persons of a particular type seems to imply that anyone who is not of this type is excluded, and that those who are included have no duty to those who are not (since society is a system of reciprocal cooperation between insiders). If Rawls’s problem is how to construct social unity among people with diverse desires, his problem is to replace the competition of many desires with a simple division between “us and them”, based on a single desire treated as primary: the desire for social unity and state control above all else. It is no surprise that he ends up with this kind of model, since he works “backwards”, from intensionally-conceived conceptions to actual people rather than from actual people to theory. It is because of this manoeuvre that Rawls constructs an essence which floats above actual situations and bears down on them like a lead weight. The liberty Rawls seeks quickly becomes a “right” to repress, on the basis that insiders already control “everything that matters” and are equivalent to “everyone”. Even the “normal” are in a sense hemmed-in by Rawls’s system, because in public they are reduced to “citizen X”. This could be repressive towards other aspects of their identities (cf. Chatterjee). An empty self, who is owed nothing prior to social rules and norms, is both demeaned and likely to demean others. Furthermore, the “backwards” method is the main reason why Rawls is so tolerant towards the system’s “injustices”. The “backwards” method means that Rawls tends to repress difference beneath identity.

To conclude this section, Rawls’s conception of the person is essentialist and mythical, as are the parts of his theory which follow from it. This also leads to an impositional and invalidatory approach to those who fall outside the closed group of smug conformists which Rawls seems to construct as the only morally valuable beings. As a result, his theory is incapable of providing either the liberty or the social unity he desires. By constructing “the person” in a mythical way, it operates as an oppressive force on actual “persons”, and even more so on those who do not qualify as “persons”. Max Stirner could almost have had Rawls in mind when he wrote, ‘[t]he kernel of the State is simply “man”, this unreality; it itself is only a “society of men”’. ‘So 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’ (The Ego and its Own, trans. S. Byington, London:Rebel Press 1993 p. 180, 179). Saul Newman adds that ‘the state… demands that the individual be man, be human, so that he can be part of the state society and thus dominated’ (From Bakunin to Lacan, Lanham, Maryland: Lexington Books 2001 pp. 63-4). To support the self-in-alterity is in effect to be a misanthrope towards actually-existing people. Rawls constructs an image of the self which is such as to leave the self at the mercy of the state, or of other forces which can claim to express its essence by representing its higher-order interests. This is an agenda for subordination, not liberty. People “would have” agreed, and so must submit. As Rolando Perez puts it, ‘[a]ny claim to a universal structure of the mind stems from the reactive desire to make life mathematical, calculable, and simple, as in a story with a necessary beginning and a necessary ending. However, relationships between human beings are a lot richer than that’ (On An[Archy] and Schizoanalysis, p. 112).


Post a Comment

<< Home