Andrew Robinson - Theory Blog

Monday, November 15, 2004

RAWLS AND THE REASONABLE (notes - work in progress)


“Be reasonable. Do it my way.” - Jasmine Birtles, The Little Book of Abuse, London: Boxtree 2000, p. 123.
“Soyons realistes. Demandons l’impossible! [Be realistic. Demand the impossible!]” - Situationist slogan

The “reasonable” is one the recurring themes of Rawls’s work. Although especially important in his later work from Political Liberalism onwards, it can also be found as early as his first essay, “Outline of a Decision Procedure for Ethics”, and is a crucial regulatory concept in several ways. Firstly, it is one of the assumed characteristics of his ideal of the person. It is ‘part of a political ideal of democratic citizenship’ (PL 62). Secondly, it has a special place as the aspect which is modelled into the original position as a structural condition subordinating the parties. The Reasonable, as prior to the Rational, is ‘imposed on the parties’ who are ‘subject absolutely’ to it (CW 401). Thirdly, it is the criterion by which Rawls assesses his own theory: he claims to be offering a “reasonable” political conception. His principles should be seen, not as ‘true’, but as ‘reasonable for us’ (CW 340, JAFAR 136), or ‘the most reasonable basis for social unity available to us’ (CW 430) - whoever “we” might be (and, as I shall suggest later, one major problem is that the reasonable, as an exclusionary concept, tends to construct the “we” who affirm it in a tautological way: reasonable people, as defined by their affirmation of a reasonable political conception, affirm the reasonable conception, and it is therefore reasonable). ‘[T]he aim of justice as fairness is to try to derive all duties and obligations of justice from reasonable conditions’ (TJ 111), and the original position is supposed to use simple and reasonable conditions which ‘most everyone would grant’ (TJ 457). The aim of Rawls’s “constructivism” is to derive ‘principles that no one could reasonably reject’ (PL 123). Fourthly, it regulates the borderline between legitimate and illegitimate violence (e.g. JAFAR 84), and possibly also the regulation of rights (via reciprocity limited to the reasonable in-group). Therefore, it is at the centre of the exclusionary mechanisms operative in Rawls’s theory. Indeed, at one point he goes even further, suggesting that any act which affects others’ interests must be evidenced beyond reasonable doubt (CW 17), i.e. within the field of the reasonable. Fifthly, it is central to his other ethical concepts. For instance, he states that one is ‘fully autonomous’ if one ‘freely accepts the constraints of the reasonable’ (PL 306), and that a well-ordered society is one endorsed by reasonable doctrines and in which unreasonable doctrines are too weak to subvert it (JAFAR 187). Indeed, it usually appears in a binary with the “rational”, and is given primacy over it, as the explicitly ethical pole of the pair. For instance, ‘the reasonable frames and subordinates the rational’ (PL 339; c.f. CW 319, JAFAR 82), and people in the original position are ‘constrained absolutely’ to be reasonable - though it is nevertheless in Rawls’s view a conjecture which may prove incorrect to assume that this makes his principles reasonable (PL 381). As far as I can tell from Rawls’s remarks, the “reasonable” seems to be intensional and perhaps even intuitive; it can be assessed internally by a single consciousness, and does not require any kind of dialogue. It is also assumed to be general (maybe even a character-trait), and not situation-specific. Furthermore, it is also at the very centre of what Rawls sees as essential: it defines not only justice as fairness and not only political liberalism, but the very idea of political theory. ‘No sensible view can possibly get by without the reasonable and the rational as I use them. If this argument involves Plato’s and Kant’s view of reason, then so does the simplest bit of logic or mathematics’ (LN 380-1). Any plausible theory must involve the idea of the reasonable in ‘appropriate conditions’ to which agents are ‘subject’ (PL 52). Furthermore, it is itself unconditional - it can no more be voted on than can mathematical or logical principles (PL 308). This is how fundamental Rawls sees this concept as being: it is no more avoidable than the simplest logical operation, and political theory cannot exist without it. It is, however, neither necessary for such operations nor as minimal as them; it is a complex conception which operates, often tautologically, to construct an exclusionary in-group. Its use is, furthermore, stretched over a number of different objects: “persons” can be reasonable or unreasonable, but so can comprehensive doctrines, political conceptions and specific judgements and opinions. There is often, however, a practical identity between these terms. In particular, all reasonable persons hold a reasonable comprehensive doctrine (PL 59). It seems, however, to be the final word in ethics according to Rawls. Even ‘schemes of thought’ are subject to judgement as reasonable or unreasonable (PL 56). Rawls never asks, “reasonable for whom?” (Perhaps the reasonable is reasonable for the spook “person”). It is also unclear whether the “reasonable” is an assertive or a declarative, i.e. whether Rawls describes particular doctrines as reasonable, or whether he passes judgement on them by using this label. It is only if the label is assertive that conclusions resulting from the idea of the reasonable (e.g. the possibility of articulating an overlapping consensus of reasonable doctrines) can be taken as arguments for Rawls’s theory.

As I shall show shortly, Rawls uses the word “reasonable” to mean a variety of different things, and the exact meaning of the concept is imprecise. In Political Liberalism, Rawls introduces terms such as ‘more’ and ‘most reasonable’ before he has even defined the term (PL lv, lvi). One reason for such vagueness is that Rawls defines it in terms of “practical reason”, which seems to be an excuse for imprecision. ‘The terms “reasonable” and “rational” will not be explicitly defined. We gather their meaning by how they are used and by attending to the contrast between them’ (JAFAR 82). He also claims to have derived it from everyday language (JAFAR 191), linking it to common sense and to a study by W.M. Sibley which apparently shows an idea of cooperation among equals in common sense (JAFAR 7). (The link between any such idea and Rawls’s theory is rather tenuous; one could also compare Rawls’s theory to the idea of practical reason, i.e. common judgement free from individual inclination, in Kant). There is no definition either of “reasonable” or its international correlate, “decent”, with both apparently referred back to so-called intuition. (In other words, one is supposed to receive their transmitted meaning rather than to read it, i.e. the concept operates mythically - a connotative logic which provides a handy excuse for dogmatism on the basis of a populist and irrationalist assumption that present thought must be “reason” and therefore right). ‘Just as with the idea of the reasonable in political liberalism, there is no definition of decency from which the two criteria can be deduced. Instead we say that the two criteria seem acceptable in their general statement… We give it meaning by how we use it’, and the fact that most reasonable persons will accept a view renders it reasonable (LN 67). Further, conceptions of the reasonable necessarily differ (PL 53), and reasonable persons disagree, albeit only within a range which Rawls refuses to specify (PL 56). Such later claims may be in contradiction with earlier attempts to define the reasonable, and may express a change in Rawls’s final writings from his earlier views. In Political Liberalism [p. 94] Rawls demands ‘criteria to judge’ reasonableness in persons, institutions, judgements and so on). Its characteristics ‘are not deduced, but enumerated and characterized’, and it therefore easily becomes a dogma. Rawls thinks its exact meaning will be subject to ‘differences of opinion’, but will ‘hang together’ well enough for it to be ‘affirmed by us’ (LN 87). The problem here, which recurs with most uses of the concept, is that the “us/we” who affirm the reasonable seem also to be defined by it. This is one of the characteristics of what Estlund (****) terms “the insularity of the reasonable”, and as Estlund shows, insularity cannot itself be an exclusive defining feature. There are many possible insular groups. Nevertheless, the characteristics which differentiate the “reasonable” from other possible insular groups - and in particular, the characteristics which render it legitimate for this group to exercise exclusive dominance through violence, when similar practices by (say) an insular religion would be in Rawls’s own terms “oppressive” - are ill-defined. How “we” use a concept is presumably structured, even when it is not explicitly defined; the motive which is not declared may turn out to be repressed and unconscious, or to be some kind of disavowed mythical belief (c.f. Fairclough on New Labour’s foreign policy). The various defined and operational meanings of the concept of the “reasonable” rely perpetually on oppressive assumptions, especially the impositional and typically asserted operation of the concept itself, and they are also prone to circularity and tautology. Also, if the “reasonable” is to act as a guarantee against “oppressive” uses of state power, it would seem to have to be clear and inclusive, and not be prone to the kind of slippage between “opinions” which Rawls allows. (To have practices of violence and exclusion regulated by varying “opinions” on supposedly intuitive concepts, or given meaning simply by how they are used, does not seem to justify a rigid designation of a boundary between legitimate force and oppressive violence - a rigid boundary which is necessary for Rawls’s theory to hold. The “reasonable” is defined vaguely and often tautologically, yet its social role is supposed to be specific, referential and precise, determining what is and is not permitted).

If the meaning of the “reasonable” is derived from everyday language, it is worth noting that it is often a discourse of exclusion used by the powerful. One is unreasonable if one refuses to submit to an obvious balance of force. For instance, the American state depicts the pursuit of independent economies by less-developed countries as ‘unreasonable’ (Chomsky Reader 251). The state uses the idea of the “reasonable” as a denkverbot against radicalism, a means to threaten and de-radicalise, and a way of reducing social movements to recuperable demands. It also uses it to feign necessity (e.g. the “impossibility” of the state backing down - as it ultimately did - in the case of the Pentonville Five). It should also be noted that its use tends to be asymmetrical: Rawls asks what it is reasonable for individuals to demand from the system, but rarely looks into what it is reasonable for the system to demand from individuals (perhaps because he misunderstands the system as a set of reciprocal relations and thereby deagentifies it).

The reasonable is supposed to be primary over the rational, but there is a strange relationship between the two, because the rational is also part of the ideally-required conception of the person. Therefore, the reasonable is constrained by the necessity that, while it may manipulate and engineer the rational, it may not directly demand irrationality (see PL 54). This makes it, in a sense, the worst of both worlds: it is an exclusionary ethical conception, yet it is incapable of ethical critique of “normal” modes of action naturalised as “rational”. In one passage, Rawls says that the “reasonable” means acting rationally when surrounding conditions are fair (CW 599), and elsewhere, that being reasonable means accepting fair terms of cooperation provided others do so (LN 25, JAFAR 6-7), although this simply inverts the usual relationship between the two. (c.f. also the claim - CW 459 - that Rawls’s theory is only neutral towards those conceptions which ‘respect the principles of justice). The idea that reasonable persons ‘are ready to propose, or to acknowledge… the principles needed to specify what are seen by all as fair terms of cooperation’ (JAFAR 6-7), and even more so the idea that principles of justice ‘impose restrictions on what are reasonable conceptions of one’s good’ (TJ 27; c.f. PL 210, CW 314, 317), let alone the view that a reasonable society must include some version of the two principles of justice (CW 48-9; c.f. JAFAR 191), define the reasonable in reference to the principles of justice, which rules out justifying these principles on the grounds of their reasonableness. (Normally, he sees fairness as specified by its being reasonable). Circularity also arises if “all” is taken to mean “all reasonable”, since this would mean that reasonable persons are ready to propose the principles accepted by all reasonable persons, i.e. by all persons who accept the principles.

The distinction between a public political conception and a comprehensive doctrine is central to Rawls’s later work. Despite its theoretical importance, it is ill-defined (especially since it is also possible for a partial view to be a ‘partially comprehensive doctrine’). In my view, it is primarily a means of ontological privileging which avoids symmetrical confrontations with other doctrines by hiding behind an overstated binary. For instance, Rawls’s theory would imply that someone who imposed a particular liberal doctrine - provided it is some kind of “public political conception”, not a fully comprehensive theory of everything - would not be oppressing anybody. On the other hand, someone who fights for an anarchist world without state institutions may be accused of trying to “impose” a “comprehensive doctrine” (e.g. PL lvi-ii, 138, CW 606; the examples are mine, not Rawls’s). Someone who does not accept the dictates of a liberal state is taken to be trying to impose a comprehensive doctrine (CW 606). It is hard to see much to this distinction aside from the fact that the latter conception is anti-state and not a form of liberalism. (c.f. below on “legitimate law”). The liberal conception may not represent a full doctrine of everything (nor do most religions), but it is certainly full on the level of its motivation (the systematising drive as primary drive) and in its social effects (the subsumption of specific ways of life into a single overarching system). This is perhaps the best example of the reason why Rawls would not accept the symmetry in the actions of protesters and police discussed in the introduction. (See especially the discussion below on legitimate law: if obedient people obey despite ostensibly not believing, it is impossible for an opponent to argue with an obedient person from a symmetrical position. The unbelieving conformist can concede the argument yet still use “the law” as an alibi for fetishistic conformity). (Even on Rawls’s preferred terrain of disputes between liberals and rightists, Rawls’s version does not necessarily lead to any less conflict; for instance, conflict over abortion does not disappear simply because the state weighs in on one side or the other. It is merely that the state claims ontological privilege and does not admit to being simply one of the agents in struggle. The same conflict without the state would involve two sides who both claim to be right, whereas the conflict with the state involves two sides who claim to be right and one who claims to be right on principle even if wrong). (I would also add that, say, a surfer who demanded a share of resources which are only moderately scarce would not at all be “imposing a comprehensive doctrine”, as Rawls’s binary would require. By guaranteeing the preconditions for the continuation of a basically tolerant way of life, she/he would simply not be doing the equivalent of converting by the sword).

The instance of anarchism is not simply taken as speculation. In fact, the judge in the Angry Brigade trials uses rhetoric very similar to Rawls’s, portraying the activists as if they were members of an intolerant group trying to impose its own worldview, when in fact they were struggling for greater freedom. Hence, the judge remarks as follows. “Your participation arose because you objected to the orderly way of society. One of the most precious rights is that an individual should hold his own opinions and be able to express them and be able to protest, and when one finds others who set out to dominate by exercising their opinions to the point of enforcing them with violence it undermines that precious right” - he later adds rhetoric about protecting the public which invokes his belief in the vulnerability system and resultant tendency to deindividuate (Tom Vague, Anarchy in the UK, p. 114). (c.f. also p. 118 - the press denounced them because they ‘thought the world owed them a living’, in other words, because they rejected the blackmail by the work system). It is quite clear, whatever one’s criticisms of the Brigade’s tactics, that they were not in any way threatening anyone else’s freedom to protest, that they did not attack anyone for expressing an opinion and that they were motivated, not by a desire for domination, but by a desire for freedom from domination. (The judge is, like so many of his ilk, blissfully unaware that his own complicity in the state organs of repression, which certainly “enforce” their “opinions” “with violence”, renders his entire position self-contradictory). In other words, a blatant and very dangerous confusion between opposition to liberalism and “imposition” of “one’s own opinions” is directly involved in statist ignorance and persecution.

Rawls is inclined to make varied claims about the reasonable which can easily become circular. For instance, while he claims that acceptance of particular institutions is a defining feature of the reasonable (see below on “legitimate law”), he also wants institutions designed so that ‘everyone with reason’ will accept them (PL 66). (The knots in which he gets tied seem to be the outcome of the impossible task he sets himself, i.e. to elevate one desire into a position above all other desires without affording it total ontological primacy). Take for instance the following combination of claims: political liberalism is reasonable because it is acceptable to all people who hold reasonable comprehensive doctrines; reasonable comprehensive doctrines are differentiated from unreasonable comprehensive doctrines because they can form part of a reasonable overlapping consensus; a reasonable overlapping consensus is one which constructs political liberalism. Rawls makes or implies all these claims at some point, and it is easy to show their circularity in relation to each other. The usual problem is that Rawls does not decide on priority between different parts of his argument. In particular, he does not specify whether an overlapping consensus/political conception is justified as reasonable in relation to prior reasonable persons/conceptions, or whether persons/conceptions are judged to be reasonable because and to the extent that they accept a prior overlapping consensus/political conception. Such accusations of circularity have already been made by Chantal Mouffe (Dem. Paradox p. 26). Indeed, Rawls almost admits it at one point, stating that comprehensive doctrines are unreasonable unless they confirm as true or right the views of the political conception (CW 609). Although I intend to question the logical structure of Rawls’s arguments, I also feel that his conception of the reasonable has some kind of referential content, even if he does not specify the borderlines of this content himself. I suspect it operates as a “mythical” or “concealed imperative”, with a specifiable but unspecified structure. Therefore, I shall outline one at a time the different criteria which are used to define and specify the “reasonable”.


“It is error alone which needs the support of government. Truth can stand by itself” - Thomas Jefferson, cited in Heretic’s Handbook of Quotations p. 33.

“In American there are two truths, the legal truth and the actual truth” - Mos Def and Talib Kweli, on the case of Mumia abu-Jamal, cited in SchNews Survival Handbook, 1999

The “reasonable” is crucial partly because it serves as a functional substitute for the concept of truth. Rawls’s theory is structured in a positivistic way, but he is claiming that it is “reasonable” instead of “true”. He aims to seem ‘reasonable and useful, even if not fully convincing’, across a range of ‘thoughtful political opinions and thereby express an essential part of the common core of the democratic tradition’ (TJ xi). It is therefore the criterion upon which his entire theoretical work hinges. It is, however, a lot less clear than the positivist conception of truth. It seems, indeed, to depend only on the inner reactions of Rawls himself: in “Outline of a Decision Procedure…”, he says that the reasonableness of a method can be shown without reference either to psychological effectiveness or to how much allegiance it can obtain (CW 1). Obviously the idea of “stability” involves a slight modification of this view, but it seems to express a general tendency in Rawls’s work.

For Rawls, the division between comprehensive doctrines and a public political conception permits the latter to avoid (for the most part) dealing with issues of truth (though Estlund challenges this on the grounds that the primacy of the “reasonable” would itself have to be true for his theory to hold). Rawls’s theory, he says, may or may not be ‘true’; ‘given its political aim, [political constructivism] need not go beyond’ the criterion of reasonableness. It is up to comprehensive doctrines which affirm it as ‘correct’ to decide whether it is true, whether there is such a thing as “truth” in ethics, and indeed whether the political conception is reasonable beyond the field of the political. However, it is necessary that it be more than a modus vivendi, and therefore, it is important that the comprehensive doctrines actually acknowledge it as correct in some way, ‘not simply acknowledging the political power of others’ but accepting that others whose views differ from one’s own are nevertheless also reasonable (PL 126-7). For Rawls, to refer directly to truth in politics is to ‘impose… beliefs’ (PL 61). The comprehensive doctrines are to connect reasonableness to truth, and Rawls claims to neither use nor deny the concept of truth (PL 94). It would be ‘fatal’ to Rawls’s theory were it to be in conflict with the idea of truth, as this ‘would put political philosophy in opposition to numerous comprehensive doctrines’ (PL 150), and it also must not rank individuals on any basis of truth as a criterion of inclusion, because its sole good is to give justice to citizens (PL 276). Rawls tries to avoid committing himself to philosophical scepticism, or to any particular version of truth. However, in some passages (e.g. PL 233), he clearly tends towards epistemological scepticism. There is a certain degree of scepticism implicit in acceptance of the “burdens of judgment”, several of which emphasise the fallibility of human reason and insist on a need not to demand that one’s view be taken as true. (The evasion of issues of truth reflects Rawls’s preference for a strategy of subsumption over a strategy of persuasion).

A situation where Rawls ends up endorsing Estlund’s position arises around the problem of “unreasonable” doctrines. If a doctrine affirms the truth of a set of propositions which radically negate the political conception, this would seem to force the latter to defend itself on grounds of truth. (It would, after all, seem to be intolerable for the one true doctrine, or for that matter the one virtuous way of life, to be prohibited; similarly, for the public conception to be based on untrue statements which one nevertheless must “reasonably” affirm as true, a la the emperor’s new clothes). Rawls is effectively forced to accept that this is the case, and that he needs to claim that truth (if it exists) must fall somewhere within the range of reasonable conceptions for his theory to hold (even if not all reasonable conceptions are true). (Hence, for Rawls’s political conception to be valid, either the truth must be expressed by one of the comprehensive doctrines which affirm or could affirm it, or there must be no such thing as truth). Rawls tries to avoid attacking unreasonable views as untrue, but also admits that this may sometimes be unavoidable (PL 138). When one declares some view to be unreasonable, ‘there may be no way entirely to avoid implying its lack of truth’ (CW 483). Indeed, he admits that he has to fall back on debates about truth whenever someone thinks ‘questions are so fundamental that to insure their being rightly settled justifies civil strife’ (PL 152). Though he permits this, it is to be done with ‘restraint’ and as little as necessary (CW 436). The freestanding political conception avoids engaging in philosophical disputes - unless these involve unreasonable views (PL 375). He says that his political conception is true as long as one reasonable doctrine is true, introducing via a rhetorical question the truth-claim that at least one such doctrine must be either true or morally reasonable. Such a doctrine incorporates and is strengthened by the political conception and the resulting reasonable pluralism (PL 128-9). Comprehensive doctrines, even if false, cannot be said to distort public outcomes, unless these outcomes are unreasonable (CW 585). However, he wishes to avoid saying that the presence of an idea in an overlapping consensus shows that it is, or probably is, true (CW 436) - whereas it can show an idea to be reasonable. He also wants to avoid calling the political conception true, or at least to avoid deriving its legitimacy from this. ‘Holding a political conception as true, and for that reason alone the one suitable basis of public reason, is exclusive, even sectarian, and so likely to foster political division’ (PL 129) - which for Rawls seems to be the primary evil. Hence, full reflective equilibrium and the idea of reasonableness is to be endorsed as ‘the best justification’ a political conception can have (PL 388). (In order to argue that an “unreasonable” doctrine is true, one need not necessarily confront Rawls on factual issues such as the “fact of pluralism”. One could also contradict his specification of the boundary between oppressive and legitimate violence. Alternatively, one could deny the supposed possibility and/or desirability of social unity, or the logic by which a liberal state is derived from Rawls’s “facts”. One could also claim that some other good “outweighs” the supposed “political goods”).

While in theory Rawls suggests he can argue against unreasonable groups on grounds of truth, in practice he tends not to do so. In the case of intolerant religious sects, he says it is ‘unnecessary… to argue against’ their absolutism, because while a ‘person may indeed think that others ought to recognize the same beliefs and first principles as he does’, he ‘cannot expect others to acquiesce in an inferior liberty’ or ‘to recognize us as the proper interpreter’ of religious and moral obligation (TJ 182-3). The irony here is that Rawls is himself posing as the proper interpreter of moral obligation. A similar passage arises in “The Idea of Public Reason Revisited”. ‘Of course, fundamentalist religious doctrines and autocratic and dictatorial rulers will reject the ideas of public reason and deliberative democracy. They will say that democracy leads to a culture contrary to their religion, or denies the values that only autocratic or dictatorial rule can secure. They assert that the religiously true, or the philosophically true, overrides the politically reasonable. We simply say that such a doctrine is politically unreasonable. Within political liberalism nothing more need be said’ (CW 613). He says the same on another occasion when replying to the claim that there is no salvation outside the church. It may be impossible to avoid implying that it is untrue, but it is also possible to state that it is unreasonable to enforce a claim without implying it is untrue (JAFAR 183-4). In other words, if opponents can be dismissed as unreasonable, it then becomes unnecessary to say anything further. On one occasion, Rawls even says that what is adopted for justice need not be valid for ‘meaning and truth’ or in ‘philosophy and science’ (TJ 188), implying that the state could validly suppress subversive but true doctrines. (One could add that Rawls’s theory would not be much use in actually resolving religious conflict, because it would tend simply to dismiss both sides as unreasonable. Two reasonable religious doctrines would affirm free faith and therefore not be at war to begin with).

One line of critique following from this arises from the truth or falsity of so-called “public reason” (see below). While Rawls insists that this cannot embody the whole truth, he also implies that it is not fundamentally false. His theory is not indifferent to truth in the sense of holding a view that something can be untrue and yet still the most reasonable conception, and that one therefore has a duty to endorse untruths (CW 434). “Public reason” is supposed to answer Rawls’s demand that the ‘general facts’ he uses be accepted ‘for good reasons’, and that they not be illusions, delusions or false consciousness (JAFAR 121). (An example of a “general fact” he uses in this way is the so-called ‘fact’ of reasonable pluralism - PL 153). If, in other words, “common sense” is awash with unfounded and blatant falsities, and these filter into political liberalism through “public reason”, Rawls’s entire approach is at risk. His theory is also at risk if crucial assumptions, such as the assumption that people are capable of taking responsibility for ends, are untrue. Rawls admits that his theory is dependent on the truth of the general facts it uses (CW 355). Furthermore, while Rawls affirms a need to engage “unreasonable” conceptions on matters of truth, he nowhere shows any sign of doing so, and he does not specify what form such an engagement might take. He also does not say what the implications would be if political liberalism proved unable to win such an argument - in other words, if it turned out in dialogue to be untrue. For Rawls’s theory to be valid, it is necessary that truth (and necessity) never exceed the reasonable, and never exceed the logic of sameness established by the reasonable and its companion-concepts. (At least it is necessary that truth continue to be sufficiently close that it would not necessitate a radical break with the reasonable, i.e. that it would not necessitate the equivalent within Rawls’s theory of a Žižekian Act).

In relation to the effects of Rawls’s theory, it seems likely that his commitment not actually to promote falsity is of little significance. It is clear that a view which is true could nevertheless find itself invalidated as unreasonable, whereas a false but widespread view (especially a bigoted view which has become part of “common sense”) could be treated as “reasonable” and imposed on its opponents. The “reasonable” tends to function as an operationalist attack on truth. One should also note that Rawls specifies that a reasonable person is ‘ready to accept reasonable accommodations or alterations in one’s own view’ (PL 253), that public reason will ‘contain at least some error’ (CW 329), and that a view can be reasonable and also fallacious and mistaken (CW 606). This could easily lead to a “public” obligation to believe or conform to the conclusions of patently untrue but widespread statements.


This relation of truth also requires a reliance of the supposedly “freestanding” public conception on a variety of philosophical bases which it cannot itself articulate. It is up to each citizen or association to decide ‘how the claims of political justice are to be ordered, or weighed, against nonpolitical values’: the political conception cannot state anything here, since it does not judge nonpolitical values at all (PL 386-7). It is, therefore, quite possible that the comprehensive doctrines will decide that nonpolitical values outweigh political values, or that political values do not matter at all. Rawls provides no reason why they should not do so, yet he does provide a threat: if they do so, they become “unreasonable”, and of course, this means that they are potentially subject to violence. Hence, as Rawls states, citizens may decide that their doctrines override or invalidate political values - but if they do, they cease to be reasonable (CW 609). It is via this forced choice that Rawls can accomplish the impossible goal of constructing a limiting and exclusionary conception which he can also claim includes “everyone” (a “hegemonic” gesture, in the Laclauian sense). (I am reminded of a joke Žižek likes to quote, in relation to the idea of “forced choice”: “My fiancée is never late for an appointment, because when she is late, she is no longer my fiancée”. This implies that the choice to be “reasonable” is a “forced choice” in the Žižekian sense, i.e. that it is a moment at which the system engages in doublespeak: one is under an obligation freely to choose that which is anyway compelled, like the soldier given a formal order to take a voluntary oath). In other words, the supposedly “very great values” of the political are values only for insiders, built on the basis of a system of exclusion. Rawls denies that this criterion of the “reasonable” renders political values primary (CW 609), but it is clear that they must have a structural primacy for anyone to enter the sphere of the reasonable. Rawls puts political values in the position of the master-signifier, and makes them a legitimate basis to use violence (when other values are not such a basis); this is clearly a form of primacy. At the very least, he implies that they are primary in a certain sphere of questions, and that this sphere is immune from contamination by other spheres. (In practice, someone is likely to be labelled “unreasonable” unless she/he believes (consciously or unconsciously) the political doctrine to be primary, because otherwise, there is no motive for the actions which are necessary for one to avoid being labelled “unreasonable” - e.g. unconditional obedience to “legitimate law”. In this context, the comprehensive doctrine would seem to be a “fantasmatic supplement” - it rationalises the forced choice after the event, making it seem to have been a genuine choice). The idea of “doing one’s part” is revealing here: if a reasonable person is someone who does her/his part in just institutions, one is reasonable only if one accepts a particular interpellation by pre-existing institutions, as if it were something one had always-already chosen. One thereby identifies with one’s “part” as defined by a coercively-based system.

Although it appears as an opening concept (since it enables tolerance of others who do not share one’s own comprehensive doctrine), the “reasonable” also operates so as to close political space. In particular, the “reasonable” are only to feel solidarity with others within this in-group, which is restricted by the conception of the person and other criteria. As is clear in the case of Rawls’s discussion of whether a Quaker could be a head of state, the “reasonable” tends to preclude anything better than negative reciprocity, establishing a single set of limited but absolute political claims as politically dominant. (Quakers are not supposed to run for high office because they hold supererogatory ethical views in excess over the reasonable: Quakers ‘could not in good faith… seek the highest offices in a liberal democratic regime’ in normal circumstances because of their opposition to just wars - LN 105). The concept of the “reasonable” and the limited solidarity it entails act as a barrier to anything better; in particular, terms such as “just” and “legitimate” tend to be fused with the minimality of the concept of the “reasonable”.

The “reasonable” is an expressly exclusionary concept, and a range of potential out-groups fall under the label “unreasonable”. The existence of a well-ordered society depends on “unreasonable” groups remaining too weak to compromise basic institutions (CW 489), but Rawls also affirms that unreasonable doctrines ‘will always exist’ - even his belief in their weakness is only a matter of ‘hope’ (PL 65; c.f. CW 573). As I have shown elsewhere (see section on exclusion), the “unreasonable” are subject to violence rather than negotiation or inclusion, and therefore, relations between the “reasonable” as a solidified in-group and the “unreasonable” as a string of excluded groups can only take the form of a permanent power-struggle or a semi-concealed war. The euphoria (in Barthes’s sense) of insiders’ “civic friendship” is bought at the cost of closure against the “unreasonable”. (“Public rules” and the like would tend to strengthen the closure and intractability of the “normal” in-group and render them increasingly unresponsive to outsiders, endangering and undermining dialogic attempts to negotiate arrangements in cases such as psychological difference. In practice, entrenched groups of rule-obsessed people who bar themselves from such dialogue - for example, judges and bureaucrats - are a menace to the different).

The “reasonable” defines the borders of the logic of place which constructs the in-group Rawls desires. Justice is a system to protect the reasonable from the unreasonable. For instance, ‘[c]itizens’ claims to liberties, opportunities, and all-purpose means are made secure from the unreasonable demands of others’ (CW 371). They are not, of course, made free from the demands of others: in the passage in question, Rawls is defending one such demand, i.e. the idea of “responsibility for ends”. Rather, they are protected from a certain type of other, and this other is defined characterologically, through discourses of exclusion. Basically, the other does not accept the reasonableness of the in-group and shows flaws such as a lack of self-discipline (CW 370). In general (c.f. section on reciprocity), Rawls systematically favours one characterologically-defined group, namely the “normal” or petty-minded (possessive individualists?), and he constructs his categories in reference to this group (usually reified as an ideal of the person referring to this group’s supposed ego-ideal). Demands by others on this group are unreasonable, whereas demands by this group on others are reasonable (regardless of content). Thus, Rawls expects far too much from (one might say makes “unreasonable” demands on) anyone excluded from his model (e.g. the psychologically different): such people are expected to give up anything and everything at a moment’s notice should one’s needs clash with the system’s imperatives. On the other hand, he expects very little of those included: they are not to give up any privileges or any wealth (it is “unfair” that they “pay for” others’ difference, conceived as characterological inferiority), and are even entitled to a wholly exclusive social system (via “responsibility for ends”) and a right to excessive wealth (because of their demand for “incentives”). There seems to be for Rawls no irony in his use of the label “reasonable” for a system biased in this way, i.e. a system which confronts excluded individuals with an intensive territorialisation denying access to any space whatsoever. (The idea of “reasonable expectations” closes the system around its self-perpetuating self-valorisation).

The “reasonable” has a nodal significance because it regulates the field of legitimate force. As Rawls puts it, ‘when constitutional essentials are involved, political power, as the power of free and equal citizens, is to be exercised in ways that all citizens as reasonable and rational might endorse in the light of their common human reason’ (JAFAR 84). Also, ‘political power is only legitimate when it is exercised in accordance with a constitution (written or unwritten) the essentials of which all citizens, as reasonable and rational, can endorse in the light of their common human reason’ (JAFAR 41). Coercion is justified as long as it stands up to public scrutiny (CW 293), presumably via public reason which is defined in terms of its reasonableness. In other words, the “reasonable” comes closer than any other concept to filling the position of the master-signifier or nodal point in Rawls’s theory. It defines a shared essence of “everyone” which justifies power being used, i.e. which makes it “non-oppressive” be appealing to something “no-one” can object to - and by doing this, it also defines the “no-one” who objects. The “reasonable” is the concept which squares the circle, allowing coercive power to be seen as somehow reciprocal or equal, as the collective power of those who it coerces. It is the means to construct an exclusionary “us and them” and thereby to create the social unity of the “us” which Rawls terms “civic friendship”. (This is a situation where persons, or in international relations, peoples, feel affinity with each other and are prepared to enter into some kinds of relations involving sacrifice and trust. See e.g. LN 112-13, CW 579. It means reassuring others of the existence of a “we” - CW 594 - and therefore maps easily onto the Laclauian model of “us and them”. It has, however, a strange qualifier - see below on the individual deciding what is reasonable - since civic friendship is supposed to result from the fact that others act on the basis they think one will find reasonable, regardless of whether one finds it reasonable or not). This approach confirms the Laclauian model of the operation of mainstream politics (a model I agree with as a description of the mainstream, but disagree with as a naturalisation of this logic), but Rawls also seems to think that his particular master-signifier is somehow different from all the rest: it is not equivalent to, say, constructing an “us and them” on the basis of religion or of (say) Kantian liberalism. In my view, the qualitative difference Rawls assumes between these different kinds of domination is overstated. Rawls’s “political” conception has many of the characteristics of a partially comprehensive doctrine - indeed, it is nothing but one particular part of a particular grouping of comprehensive doctrines which can be used to establish limited translatability between them in relation to a particular sphere. One might even say it is a religion of sorts, i.e. a form of “statolatry” or worship of the state. As with any master-signifier, those who cannot speak within the space constructed by this fixed point are excluded and oppressed, and mutual comprehensibility between insiders and outsiders is damaged rather than aided. (One might add that religious states rarely impose a fully comprehensive doctrine; they typically only impose core elements of religious loyalty. In other words, they perform a gesture similar to Rawls’s but with a different content to the ideology specified as “reasonable”). The emphasis on non-existent formal differences between a political conception and a comprehensive doctrine is a cover for the way Rawls in fact privileges a particular content by putting it in an asymmetrical position vis-à-vis other belief-systems.

Any semantic unity which might appear at first sight to operate in Rawls’s concept of the reasonable due to his definitions or delineations of it turns out on closer inspection to be illusory. This is partly because there is a proneness to circularity in his arguments (e.g. the definitions themselves, or their adoption, depend on the meaning of the idea of the reasonable being external to them and already being known, yet they are used to define the reasonable). This is not to say that it lacks a unity - it certainly creates delineations which construct a boundary between an in-group and an out-group - but this unity does not seem to be on an officially-declared basis. Furthermore, the reasonable has many diverse meanings and characteristics in terms of its uses in Rawls’s own theory. I shall now examine some of these, before setting out my own view of its probable operational basis.


One way of interpreting the idea of the “reasonable” is to link it to social order. In other words, someone is “reasonable” if she or he acts in such a way as to make possible a cooperative society. The aim of political theory is to reconcile people to the social order; there are, however, in Rawls’s terms, ‘limits to reconciliation’. One can only be reconciled to social reality if one acknowledges the same general facts as Rawls. ‘To be reconciled to a social world, one must be able to see it as both reasonable and rational’, so if one is a fundamentalist who sees the present as a ‘nightmare’, one is beyond the possibility of reconciliation (LN 126-7). Thus, the unreasonable seems to be a self-constructing category, and the reasonable expresses the best possibility of cooperation. In Rawls’s own terms, a reasonable person is someone who is prepared to offer reciprocal terms to others (see below) and who wants to take part in a fair system of cooperation (in Rawls’s sense), and one of the reasons this may be emphasised is that it is necessary to produce social cooperation. He refers to a ‘political aim of consensus’, and he consciously avoids putting forward more of his own views than are necessary to achieve such a consensus (PL 153; c.f. below on overlapping consensus). Furthermore, the agreement is supposed to be on constitutional basics, and is not to be conditional on the economic and social effects of these basics (CW 481). The overall aim of ‘reasonable’ action is ‘reaching reasonable and workable political agreement’ for the good of ‘peace and concord’ (PL 63), and a doctrine is reasonable if it allows ‘a reasonable balance of political values’ (PL 243). If an overlapping consensus can be formed which supports a reasonable balance of political values, Rawls rhetorically asks how there can possibly be any objection to it (PL 244). This implies that those who are constructing the model care more about the former than the latter, suggesting that, to be “reasonable”, one must have a primary concern with social order.

In one passage “reasonable” is directly used to imply that it means views which forward the possibility of cooperation: Rawls hopes, he says, that a ‘constitutional democracy’ would ‘force groups to advocate more compromising and reasonable views if they are to be influential’ (PL 431). In another, he treats it as pragmatic: something is reasonable if it is useful for the purpose of constructing an agreement on political justice (e.g. CW 427). The reasonable is supposed to allow differences to be narrowed: even if ‘we’ cannot always reach agreement, ‘we should be able at least to narrow our differences’ and ‘come closer to agreement’ through political theory (PL 118). A political conception is not reasonable if the well-ordered society it would produce would be unable to generate a sufficient sense of justice (JAFAR 89). If a conception causes instability, it needs revising (PL 141). Further, ‘it is by the reasonable that we enter the public world of others’ (PL 114 - note that one enters this world, rather than constructing it). The respect citizens are supposed to have for “legitimate law” and “public reason” seems to derive from the role such respect has in constructing an otherwise impossible social unity. The reasonable offers a ‘benchmark’ for distributing benefits and burdens which is linked to the idea of society as a system of reciprocal cooperation (PL 300). Furthermore, the ordering function is the main concern in considering the components of the reasonable, which seems to suggest that Rawls thinks it is primary among the concerns of all reasonable people. For instance, public reason is assessed, not as a political morality, but as regards its suitability for a constitutional regime (PL 254), or in other words, its usefulness for a certain kind of state. Rawls also claims that any reasonable political conception places restrictions on ‘permissible’ comprehensive doctrines and operates in an exclusionary way (JAFAR 153). One could also add the occasions where Rawls rules out particular options (e.g. the fair value of civil liberties - see the section on inequality) on grounds of it causing instability. Rawls seems to declare anything to be “unreasonable” if it would lead to instability or if it would render impossible the kind of social order he supports. For instance, it is ‘unreasonable’ to believe in equality of outcomes.

The reason this cannot be the full meaning of the “reasonable” is multiple. Firstly, the view that a reasonable overlapping consensus is likely to produce a stable social order is largely speculative, as is the idea that this consensus is best formulated in a liberal way. The division between the reasonable and the unreasonable leads, as I have noted above, to a constant state of semi-concealed warfare between the two. It is by no means obvious that the reasonable can produce a more stable unity than some grouping of unreasonable groups, or a mixture of the two. Every master-signifier performs an ordering function; there is no specific reason to prioritise Rawls’s own master-signifier in this respect. Therefore, the reasonable defined simply as an ordering drive would be empty of content: there would be no way to derive a particular social order from it. (Someone who puts social order before all else could in principle agree to any social order). Some of Rawls’s general facts refer back to the idea of the reasonable anyway (e.g. the “fact of reasonable pluralism”), while others are contestable in terms of their empirical validity. It is only if one accepts a whole string of naturalisations that one can see Rawls’s theory in terms of reconciliation to inevitable realities. Secondly, no basis is given for the primacy of social order itself. Rawls clearly wishes to derive a monopoly right to use coercion from the idea of the “reasonable”, yet there is no inherent reason why order should be more desirable than, say, freedom or need-satisfaction. Regarding both of these criticisms, Rawls would probably defend his view by claiming that it is something that “all reasonable people” accept. In other words, the ordering discourse justifies itself on the basis of its being reasonable, and if it is itself assumed to found the reasonable, the argument would tend to become circular. A similar problem arises regarding the desirability of social cooperation (in the particular sense Rawls gives to this term). This, too, might be typified as something “reasonable” people want, and this is perhaps why Rawls values it. If so, one could not then define the “reasonable” by reference to a desire for such a social order: this would again involve circularity.

While its ordering function cannot for this reason operate as a semantic definition of the reasonable (at least not of the open, inclusive kind Rawls explicitly advocates), this function (minus ideological ideas such as cooperation and reciprocity) is clearly part of its actual operation, so should remain part of its definition on some level - not, I would suggest, as a logical component, but as part of the psychological drive behind the concept. Reasonable persons are partly defined by their valuation of social order above all else - their desire for ‘reasonable and workable political agreement’, ‘peace and concord’. (In my view, the systematising or ordering drive is primary in Rawls’s work, and he is prepared to invalidate or suppress anything which stands in its way. However, this drive needs to be supplemented by others to construct a positive content). While Rawls’s conception of the reasonable does not express the singular path to social peace, Rawls certainly has a tendency to label anything unreasonable if it undermines a particular type of social order which he supports. Stirner remarks that the state seeks to ‘direct… desire to it alone, and to content that desire with what it offers’, so as to ‘tame’ desire (The Ego and Its Own, trans. S. Byington, London: Rebel Press 1993, p. 312). One might, therefore, say that being “tame” to a particular kind of master, rather than “feral” or “wild”, is one of the characteristics of Rawlsian “reasonableness”. This taming function is also internal. For Rawls, people must be able to ‘subordinate their power to reasonable aims’, or else life is meaningless (LN 128), and the reasonable and the rational are a way of orienting people who are otherwise faced with a chaotic mess of possible ends (JAFAR 3). There is also an extent to which the “reasonable” is defined by submission to a particular social system, and hence by an ordering discourse of a slightly different kind. I shall also return to this issue later, as regards the issue of “legitimate law”.


One distinction between the “reasonable” and its Other is that the “reasonable” is expressed in the form of an overlapping consensus. This means that reasonable doctrines do not simply stand alone, but enter into a limited alliance with other doctrines to produce a shared public space around ideas all the doctrines can accept as reasonable. The ability to permit or even encourage the growth of a reasonable overlapping consensus is for Rawls one of the central criteria by which his theory should be assessed (JAFAR 12). According to Rawls, political consensus emerges as an overlapping consensus of reasonable doctrines, each of which endorses the consensus ‘from its own point of view’. So long as the resultant consensus on justice does not jeopardise ‘citizens’ essential interests’, such a consensus is ‘the most reasonable basis of social unity available to us’ (PL 134), and reasonableness offers ‘the deepest and most reasonable basis of social unity available to us in a modern democracy’ (PL 391). An overlapping consensus takes the form of the endorsement of a freestanding political conception by people who adhere to a variety of different comprehensive doctrines (e.g. Kantian liberalism, Millian liberalism, intuitionism, free-faith Christianity and a partially comprehensive conception could all endorse political liberalism). At one point, Rawls suggests that the development of an overlapping consensus relies on the lack of comprehensiveness of many people’s comprehensive doctrines, so an allegiance to the political conception can arise more-or-less directly (PL 168). (This means that he relies on the incoherence of “common sense” to ensure “reasonableness”). The characteristic of entering into an overlapping consensus is therefore among the features which seem to specify which doctrines are reasonable. The freestanding political conception is assumed to lead to a particular kind of social unity, such that one’s identity does not change if one converts from one comprehensive doctrine to another (PL 32). In Rawls’s view, the idea of an overlapping consensus is an important step towards ‘realism and stability’ (PL 38). He is also of the view that agreement by everyone on a comprehensive doctrine would render it ‘oppressive and stifling’, so that an overlapping consensus is also valuable for any particular doctrine (CW 425).

An overlapping consensus operates to turn a partial similarity into a political order. For Rawls, ‘a partial similarity of citizens’ conceptions of the good is sufficient for political and social justice’, because, by emphasising the similarities, one can gain agreement. ‘It is enough that citizens view themselves as moved by the two highest-order interests of moral personality… and that their particular conceptions of the good, however distinct their final ends and loyalties, require roughly the same primary goods’ (CW 361). An overlapping consensus is necessary because a democratic society always involves many different comprehensive doctrines as the inevitable result of the exercise of human reason in the context of free institutions (PL 4). An overlapping consensus is a historical phenomenon which Rawls accounts for by a historical narrative. According to Rawls, a liberal modus vivendi tends to shift people’s beliefs in the direction of liberal principles, evolving into an overlapping conception (e.g. PL 163). One should note that this narrative is entirely speculative. In particular, Rawls’s account misses the extent to which the emergence of a liberal overlapping consensus required, not a simple mutual reconciliation between doctrines, but a change in their actual content. In particular, the rise of liberalism was not simply a becoming-tolerant on the part of Catholicism, but depended on (or caused) the breakdown of the medieval variant of Catholicism to a sufficient extent that it could no longer itself operate as an overlapping consensus. (For example, the spread of world exploration and trade undermined the illusion, pervasive in the middle ages, that all human beings are born Christian and that non-Christian peoples are corrupted by devils).

Rawls outlines the characteristics of the social unity he seeks in a passage in Political Liberalism. This unity is, firstly, to be built around one or more liberal political conceptions. Secondly, it is to be endorsed by all reasonable comprehensive doctrines, at least in one of the versions it offers. In one passage, he specifies that the doctrines he seeks to incorporate in a consensus are reasonable doctrines (PL 144). Thirdly, it is to be based on a conception which renders public questions decidable. If all citizens see the conception as reasonable, political society is ‘stable for the right reasons’, and this is ‘usually the most we can expect’ (PL xlix-l). (There is, of course, a difficulty which I shall discuss later with the claim that an overlapping consensus is endorsed by all reasonable comprehensive doctrines; this claim might be tautological and true by definition). Rawls hopes, in fact, that people can agree to accept any of a number of political liberalisms. ‘Citizens will differ as to which of these conceptions they think the most reasonable, but they should be able to agree that all are reasonable, even if barely so’ (LN 14; c.f. CW 578). Although in some contexts Rawls specifies that the overlapping consensus must be liberal, in other cases he refers only to a need for an overlapping consensus (e.g. LN 16). Indeed, his arguments for a liberal overlapping consensus tend in fact to concentrate on the importance of having an overlapping consensus as opposed to a publicly-affirmed comprehensive doctrine. The role of a political doctrine is to prevent the imposition of a comprehensive doctrine (JAFAR 189). The declared aim of Rawls’s conception is inclusion. At one point, he says that fair terms of cooperation are those which mean one is prepared to cooperate with all members of society (PL 303).

The problems with viewing the idea of an overlapping consensus as a core characteristic of the reasonable are twofold. Firstly, the overlapping consensus is itself specified by reference to the reasonable, firstly because it is rendered as a legitimate overlapping consensus because it is endorsed by reasonable doctrines (e.g. PL 144), which therefore presumably are defined externally to it, and secondly because the preference for an overlapping consensus rather than an attempt to establish public unity around a comprehensive doctrine is itself justified on the basis of its being “reasonable”. At one point, Rawls affirms that, when he uses the term “overlapping consensus”, he means a consensus of reasonable doctrines (so presumably a consensus of unreasonable doctrines would not qualify). However, he also suggests such a consensus can be deduced in reference to the original position (LN 32).

Secondly, as with social order, so with overlapping consensus, the mere fact of a consensus does not specify its content. Rawls does not deal adequately with the fact that many different overlapping consensi may be possible, each including a different range of included conceptions and a different series of excluded ones; he simply assumes that an overlapping consensus would be liberal. (Rawls may claim he wants his consensus to include “everyone”, but he also affirms that “unreasonable”, i.e. excluded, doctrines and people will always exist). It is possible, for instance, to discuss medieval Catholicism or contemporary Iranian political Islam as varieties of overlapping consensus, to the extent that the publicly-affirmed doctrine is not exclusive of the existence of some differences between ways of life. In medieval Catholicism, public prohibition of heresy and apostasy did not prevent the existence of multiple forms of worship carried out by different monastic orders, and it did not prevent the incorporation of folk beliefs which seem superficially to contradict church doctrine. Normal practice was for the church to subsume heterodox beliefs as a tolerated substrate or as orders, unless the beliefs exceeded certain limits in their public discourse and thereby became “unreasonable” (disorderly in relation to the power of the church as an institution). In Iran, the operation of an Islamic state nevertheless permits the existence of controversies between different groups all of whom accept the primacy of Islam, such as divisions between “traditionalists” and “reformists”. One might also, rather more positively, speak of an anti-capitalist overlapping consensus making possible the organisation of demonstrations such as Seattle. Diverse groups with different ideologies and tactics are able to come together in a common space, with some degree of “civic friendship”, due to a sufficient space of similar (mainly negative) beliefs. One also finds the idea of “overlapping consensus” in Marxist strategic theory in the concept of the “united front” (e.g. Trotsky’s “united front against fascism”), and in some anarchist texts (e.g. the idea of “panarchy” in the pamphlet Consent or Coercion?). Estlund rightly notes that even a highly exclusionary sect such as the Branch Davidians could classify itself as the insular group of the “reasonable” (although here, there is only one comprehensive doctrine, and so not a proper overlapping consensus). One might, indeed, say that any system of “articulation”, in Laclau’s sense, is an “overlapping consensus” of sorts. Thus, there is no reason to assume that liberalism is somehow exclusive in its claim to be able to form an overlapping consensus, and therefore, the mere fact of enabling such a consensus is not a decisive case for liberalism. In a liberal overlapping consensus, the consensus is anyway incomplete: some doctrines fall outside the consensus, and so are apparently excluded from the “civic friendship” it establishes. (A liberal overlapping consensus also tends to prohibit any social link which leads to extensive solidarity with the “unreasonable”; this includes the discouragement of institutions liberals support, such as the family and charities, when these become too intense or altruistic).

It is not to be ruled out that a different overlapping consensus might repress values which Rawls sees as central, or might leave Rawls and his supporters among the ranks of the “unreasonable”. (This may or may not be true in some of the cases outlined above, in the sense that some of them are not exclusionary in the same way as Rawls’s, but it is also clear that Rawls could not exercise his reactive desire to construct an overarching regime of power in those cases where explicit exclusion does not operate. In other words, he would be prevented from imposing his views, so the oppressive part of his beliefs would be, in a sense, “excluded” or denied actualisation). However, this is not an argument against such conceptions on the grounds Rawls accepts as legitimate. That an individual who happens to be a Kantian liberal wishes to be part of an overlapping consensus which does not exclude Kantian liberals is understandable, but this is equally true of some groups Rawls excludes. To rig his theory so as to include his own preferences does not give it any claim to universality. If the overlapping consensus of medieval Catholicism seems intolerable to Rawls, this is partly because, within it, he would be among the excluded rather than the included. He could only use this as a legitimate basis for his opposition to it if he also opposed exclusion in general. Otherwise, he is simply replacing one system of domination with another. Rawls seems to avoid even considering the possibility that he could ever himself be in the position of being one of those who is excluded and “unreasonable”, and who therefore has no recourse to reasonableness, reciprocity or repressive “we”. (The anonymous author of Consent or Coercion? argues that panarchy would be a tolerant and inclusive system - one might say, a true overlapping consensus - whereas statists like Rawls only allow ‘a mish-mash of strictly enforced rules that come out of battles fought on the elite turf of the official political process’ - 19. Also, tolerance of others’ views, including tolerance so as to avoid “civil strife”, does not necessarily require an overlapping consensus).

One might even find complex cases where (for instance) there are a series of four groups, ABCD, and three possible overlapping consensi (XYZ), among whom consensus X is acceptable to A and B, consensus Y to B and C and consensus Z to C and D. (For instance, suppose that A is an anarchist, B a left-wing trade unionist, C a conservative liberal and D a religious fundamentalist. Consensus X would be a variety of syndicalism, Y a variety of political liberalism and Z a moderate religious state). There could also be a tripartite model where three groups (ABC) could agree to three different consensi (A and B to X, B and C to Y, A and C to Z). (An example would be that A are liberals, B Marxists and C conservatives; A and B could agree about some issues of individual rights and the ideal of equality, A and C about the form of the state, B and C about the need to oppose capitalist monoculture). If the reasonable is defined by reference to acceptance of an overlapping consensus, the exact content of the reasonable and the unreasonable would depend based on arbitrary variation in preferences for each particular consensus. One can no doubt also find complicated cases where someone finds one form of political liberalism acceptable, but not another form, or where someone finds a particular overlapping consensus acceptable only on condition that it is interpreted in a particular way. In these cases, the boundary between the reasonable and the unreasonable would be extremely fluid, depending on contingent variations: a particular decision on interpretation or between different political liberalisms would not simply alter the balance of forces between reasonable groups, but would alter the classification of groups as reasonable or unreasonable. In other words, it is not simply that “all citizens” decide something: the content of the category “all citizens” is altered depending on what they decide. (As I shall discuss later, Rawls tends to assume, via the idea of “legitimate law”, that the desire to submit to any and all possible versions of a liberal overlapping consensus must be absolutely primary for someone to be “reasonable”. This undermines his claims that a few shared beliefs could create an overlapping consensus of the kind he wants, because it is precisely not enough that people share the same conception of the person and desire the same goods; it is also necessary that they prioritise the desire for social order above all else, sufficiently to establish the “shared” views in a hierarchically superior position to the differing ones). One should keep in mind that Rawls is dealing with crucial questions which determine the distinction between legitimate use of violence and utterly illegitimate oppression and tyranny, as well as much else of great importance.

Thus, if Rawls is to give specific priority to a liberal overlapping consensus as opposed to any other type, this consensus must have a referent other than its simply being a consensus, and this presumably means that the idea that it is a consensus of “reasonable” doctrines operates in exteriority to the consensus itself, i.e. that the concept of the “reasonable” is not defined exclusively by endorsement of an overlapping consensus but also contains reference to the specific content of a particular consensus or of particular doctrines. On some occasions, he specifies that a good overlapping consensus is one which subsumes all doctrines likely to thrive. This is crucial to his definition of the role of political theory, since it determines the possibility of achieving “stability of the right kind” (CW 421). On another occasion, he defines a reasonable overlapping consensus in terms of its subsuming all reasonable comprehensive doctrines and its ability to construct stability by constructing an ‘enduring majority’ of such doctrines. A political conception is reasonable if it is endorsed or supported by reasonable comprehensive doctrines (PL 391). Indeed, to qualify as an overlapping consensus, a political conception must win the adherence of all reasonable doctrines which persist over time (PL 15). In another passage Rawls asserts a “fact” that democracy can only rest on majority support for reasonable conceptions (CW 425). In other words, an effective overlapping consensus must involve a numerical majority. However, as I shall argue below, Rawls resists the methodological implications of suggesting that one can first specify which doctrines are reasonable and then assess an overlapping consensus relative to them (whether the means of specifying is content-based or numerical). Thus, there does not seem to be a prior content of “reasonable comprehensive doctrines” definable independently of the overlapping consensus they form (either in terms of their being the most numerous or in terms of the specific characteristics of doctrines). This means that one cannot define or legitimate the consensus by reference to the fact that it is endorsed by the reasonable doctrines - the doctrines are reasonable at least partly because they endorse the consensus, and so the argument becomes circular.

However, the only other alternative is that the overlapping consensus is specified, not by its adherents, but by the specific beliefs it holds. In other words, its defining feature is that it embodies a certain series of particular beliefs which Rawls himself happens to think are fundamental. If Rawls were to adopt this position (which I suspect is where his logic leads), the boundaries between a political conception and a comprehensive doctrine become extremely vague.

If Rawls’s aim is to maximise the number of people and/or doctrines included in the overlapping consensus, he has provided no argument for preferring his own principles. One could note that the expansion of the range of potential imperatives to which one must submit to be “reasonable” contracts the range of doctrines/persons which/who could accept the consensus, and vice-versa (i.e. there is an inverse correlation between inclusion of doctrines and extent of state control). Rawls’s theory, via the idea of legitimate law and the defence of the work system (e.g. “responsibility for ends), in fact legitimates quite a lot of potential imperatives. It is therefore unlikely to maximise inclusion, or for that matter “stability for the right reasons”, as it is possible to conceive of more open systems which would include more (e.g. to extend provision of a basic income to cover surfers). Rawls is only able to maximise (or rather, guarantee) inclusion of a particular group (the petty-minded), and this cannot be considered an argument for preferring his principles independently of some other argument establishing the privilege of this group. The replacement of political appeals to comprehensive doctrines with a political doctrine and public reason is only effective in maximising inclusion if it produces a wider range of agreement. If, for instance, it proves harder to dull hostilities through an overarching appeal to civility and stability than Rawls assumes, his conception could end up being more exclusionary than some comprehensive doctrines (since it declares all outside its limits to be valueless).

In any case, Rawls admits that he cannot guarantee that political liberalism could produce an overlapping consensus. ‘Many doctrines are plainly incompatible with the values of democracy’, and people may decide that moral, philosophical or religious values outweigh political ones (JAFAR 37; c.f. PL 386). In other words, he cannot prove the reasonableness of his own conception, if by this he means its ability to construct an overlapping consensus. It is able to praise itself by such criteria only on the basis of circularity: political liberalism is the only kind of political conception able to produce an overlapping consensus capable of subsuming those comprehensive doctrines which are compatible with political liberalism.


One way in which Rawls specifies the “reasonable” is by specifying a list of beliefs which “reasonable” people are assumed to share. This set of beliefs is conceived as somehow different from the ordinary run of contentious views, so that one invalidates oneself by failing to endorse them. These beliefs can be specified in a short list, which varies slightly between different renditions (and which may also include unspecified additions, such as a belief in the social necessity of incentives, which do not arise explicitly in discussions of the “reasonable” but which are elsewhere used to differentiate between “reasonable” and “unreasonable conceptions). Particular facts can be specified in such a way that ‘it is unreasonable not to recognize’ them (JAFAR 197).

On one occasion, the beliefs which reasonable people hold are specified as an acceptance of the existence of pluralism, an acceptance of the existence of moderate scarcity and an acceptance of the possibility of gain from social cooperation (CW 445). On another, the list includes the fact of reasonable pluralism (not quite the same as the fact of pluralism), the burdens of judgement, the fact of moderate scarcity and the possibility of gains for all from fair social cooperation. In this version, one must accept each fact ‘together with the full consequences of this’, as specified by Rawls himself (JAFAR 196-7). The fact of reasonable pluralism is not a truth-claim, let alone a fact, yet Rawls treats it as if it is. For instance, he refers to it as a limit to the possible (LN 12). In another passage, being reasonable means ‘recognizing and being willing to bear the consequences of the burdens of judgment’. The consequences include not seeing one’s own doctrine as having any special claim on others, and thinking it unreasonable to suppress reasonable views (PL 58-60). In another passage, the two characteristics of the reasonable are willingness to propose and abide by fair terms of cooperation (a tautology, since the fairness of the terms is defined by their reasonableness) and acceptance of the burdens of judgement along with their consequences, notably the idea of public reason (PL 54; c.f. PL 375, 395). (I deal with public reason separately below). The burdens of judgement are taken to be a logical consequence of the facts of oppression and of reasonable pluralism (PL 54). In another passage, he says that reasonable comprehensive doctrines are defined by the criteria that they ‘recognize the burdens of reason’ (i.e. of judgement), ‘accept the fact of pluralism’ and ‘accept freedom of thought and liberty of conscience’ (CW 487; notice that the last of these is a conclusion, rather than a premise; see the section below on specific beliefs which define the reasonable). In yet another, he lists acceptance of the burdens of judgement, offering fair terms of cooperation and affirming reasonable comprehensive doctrines as the criteria for reasonable persons (LN 87). Elsewhere, he specifies the burdens of judgement and reasonable pluralism (JAFAR 191). The very indeterminacy of the definition - the oscillation between different criteria which change slightly on each rendition - suggests that such ideas are not actually the basis of the idea of the reasonable, but are attempts to rationalise retrospectively an alignment which has already been constructed elsewhere.

One harmful consequence of the idea of the “burdens of judgement” is that it removes any possibility of unconditionally opposing oppression, even when one is directly oppressed. The “burdens of judgement” are supposed to cast doubt on whether one is actually oppressed or not, or rather, in Rawls’s affect-blocked language, whether one’s “claims” are really “stronger” than others’. Anyone who posits an unconditional demand can be dismissed due to this principle, which equates unconditionality conveniently with intolerance.

The fact of reasonable pluralism is distinct from the fact of pluralism in that, whereas the latter simply asserts that there are many comprehensive doctrines, the former asserts that several of these doctrines are reasonable, and that there are always several different doctrines which are reasonable. Hence, ‘political liberalism takes for granted not simply pluralism but the fact of reasonable pluralism’ (PL xx), and pluralism is only accepted as an inevitable fact if it remains within liberal limits (PL 304). (Rawls’s “realism” is thus a “realism” with arbitrary limits established by his master-signifier). Although political disagreement is ‘accepted as a normal condition of human life’ (TJ 196), it is not clear how ordinary pluralism can derive a conclusion of reasonable pluralism on the level of fact alone. Indeed, on closer inspection it emerges that one is not supposed to simply endorse a fact in the usual, empiricist sense. Recognition of pluralism is supposed to be related ‘to the inevitable outcome of free human reason’, i.e. to be an analytical/derivative claim, not a claim about the ‘brute forces of the world’, i.e. an empirical claim (PL 37). It involves, not a statement of fact, but a value-position; in other words, it amounts to little more than a belief that the reasonable is plural. The word “fact” is misleading.

The so-called fact of reasonable pluralism is taken to be a central basis for the idea of the reasonable. ‘Once we accept the fact that reasonable pluralism is a permanent condition of culture under free institutions, the idea of the reasonable is more suitable as part of the basis of public justification for a constitutional regime than the idea of moral truth’ (PL 129). On the other hand, the treatment of “unreasonable” doctrines is determined, not by the fact of pluralism, but by the principles of justice (?LN 16). There appears to be a link between the idea of reasonable pluralism and a positive valuation of debate (agonism?). Hence, if society ‘has vitality and spirit’, debate between comprehensive doctrines should be ‘without end’ (PL 383). Most often, however, Rawls links reasonable pluralism to the “burdens of judgement”.

The “burdens of judgement” in particular tend to be a sceptical device (though a selective one): they are used to silence disputes about truth by a repressive reduction to an idea of the equal vulnerability of all to the burdens of judgement (PL 62). However, Rawls denies such scepticism. He does not think that the burdens require hesitancy or scepticism in one’s own beliefs, nor that they rule out the possibility of correct views (PL 62-3). This would seem to involve a position of subjective self-contradiction or internal striation/repression: one is to believe that one is correct, but one is nevertheless not to manifest this belief. “Truth” becomes a kind of repressed secret. There is no way that a person who is neither sceptical nor hesitant about a belief could nevertheless accept, for non-tactical reasons, that they should hold back from the pursuit of such a doctrine to its fullest. Clearly Rawls’s approach implies some degree of hesitancy and scepticism, at least to the extent that the views contained in the comprehensive doctrines are to be taken as less certain than those contained in the public political conception (which are taken to be implicitly immune to the burdens of judgement).

The “burdens of judgement” are a series of hazards involved in the process of correct and conscientious reasoning and judgement which lead to ‘grave difficulties’ in weighing claims and which therefore lead to differences in judgements (PL 55-6). (If Rawls took this further, it would seem to problematise the entire model of “weighing”, “correctness”, etc., which precisely assume a standpoint of objectivity). Rawls lists six sources of such hazards: complex and conflicting evidence; disagreements about the weighting of different considerations; ‘indeterminacy’ and vagueness of all concepts in some ‘hard cases’; influences of experience on judgement; different normative considerations; and limits to the range of political values any social system can realise (PL 56-7). Reasonable persons may differ within a range set by the indeterminacy of concepts (JAFAR 35). On another occasion, he adds that many differences result from ‘different standpoints’, stating that he wants to view ‘deep and unresolvable differences’ as ‘a permanent condition of human life’ (CW 329). Rawls wants everyone to ‘acknowledge that all are equally subject’ to the burdens of judgement (JAFAR 197).

The “burdens of judgement” include a variety of possible causes of differences in normative judgements, some of which are genuine reasons for hesitancy, others of which involve positing the limits of the existing social system as absolutes. The idea, for instance, that, for institutional reasons, ‘social space’ is necessarily limited so one must ‘restrict’ which ‘moral and political values’ can be realised (PL 57) is an example of an essentialisation of the existing striated form of social organisation. Another instance of such an essentialist conception is when Rawls states that ‘many hard decisions may seem to have no clear answer’, yet ‘some decision must be made’ - one is ‘forced to select’ (PL 57). To the extent that the possibility of mistakes and issues of contingency and standpoint relativity operate, they problematise any kind of top-down or striated political organisation, not only those which rely on comprehensive doctrines. The problem is that Rawls projects the authoritarian “need” for a master-signifier - the idea, for instance, that “a decision must be made” - into his understanding of contingency. (Who exactly “forces” one to “select”?)

Another problematic element in the “burdens of judgement” is Rawls’s endorsement of repressive tolerance (which Rawls terms the ‘democratic idea of toleration’). For Rawls, different ideas arise from ‘different perspectives’, and ‘many of our most important judgments are made under conditions where it is not to be expected that [reasonable persons]… will all arrive at the same conclusion’. He gives little basis for believing this, except that he asserts that it is ‘unrealistic’ and leads to ‘mutual suspicion and hostility’ (PL 58). Further, people are to ‘recognize that to insist on their own comprehensive view must be seen by the others as their simply insisting on their own beliefs’. Because ‘there is no shared public basis’ for assessing such beliefs, truth-claims come to be reduced in public to the status of mere opinions (PL 128, 247). Indeed, it is clear that he cannot resolve such conflicts, but simply projects an affect-block on top of them: people are not to care enough about such differences to cause disorder. This requires, however, the opposite of toleration: the core orientation to “order” is to be prior to any specific “opinion”. It is clear that in practice, it would lead to a state which tries to impose its own “balance” on conflicts, even when this seems unreasonable to the conflicting parties. In other words, reasonableness comes to mean submission to the state. (Repressive tolerance de-activates the content of political beliefs and is therefore a form of oppressive discourse: it is a way of ensuring that particular kinds of questions are not listened to, because they are displaced behind a barrier which renders them mere “opinions”).

The “burdens of judgement” are to be distinguished from illegitimate causes of disagreement, such as ‘bias, self- and group interest, blindness and willfulness’, which also play a role in actual cases of conflict (PL 58). However, he rules out by decree the possibility that all disagreements result from such origins as bias and irrationality (PL 54-5). There are thus two categories of disagreement: those due to the burdens of judgement and those due to heteronomous/pathological causes. Rawls does not provide any basis for determining whether a particular disagreement is due to the one or the other. This is crucial because the discursive status of the two within his theory is quite different: differences due to the burdens of reason are to be subsumed within the political conception, whereas those due to bias are to be invalidated as “unreasonable”. Also, since Rawls does not establish how one distinguishes between the two types of disagreement, he does not provide a basis for knowing that the former type exist at all. Therefore, belief in the “burdens of judgement” is an act of faith. It is as if Rawls wants agreement to be impossible, because otherwise, he lacks the premises for the rest of his theory. (In a different passage - PL lx - Rawls says there are three types of conflict: conflicts between comprehensive doctrines are mitigated but not solved by political justice; conflicts based on status issues such as gender and class can be mitigated or even solved; but conflicts based on the burdens of judgement are irreducible).

There is also a large danger that Rawls justifies biased, self-interested and bigoted actions by their possible misrepresentation as results of the burdens of judgement. One could argue, for instance, that the evidence of the innocence of Mumia Abu-Jamal and Leonard Peltier is so obvious that differences in judgement on this matter must reflect heteronomous concerns, namely racism, “telishment” and statist arrogance. (In other cases, the causes might be corruption, self-interestedness, stupidity/misinformation, etc.). However, how does one show this to be the case? It would be too easy for a complacent liberal to simply put the difference down to “burdens of judgement”, dismissing the resultant “miscarriage of justice” as an unfortunate necessity resulting from imperfect procedures, and thereby providing an excuse for state racism. This is a particular problem given that Rawls also wants reasonable persons to assume that others are honest (CW 478-9), i.e. to adopt a naïve position rather than a hermeneutics of suspicion. ‘First, the political discussion aims to reach reasonable agreement, and hence so far as possible it should be conducted to serve that aim. We should not readily accuse one another of self- or group-interest, prejudice or bias, and of such deeply entrenched errors as ideological blindness and delusion. Such accusations arouse resentment and hostility, and block the way to reasonable agreement. The disposition to make such accusations without compelling grounds is plainly unreasonable, and often a declaration of intellectual war’ (CW 478). The bias would resultantly be towards assuming that every disagreement, including those which result from prejudice and other heteronomous causes, are in fact results of the burdens of judgement. The de facto logic of the imperative to assume others to be honest seems to be: reasonable disagreement is possible, therefore all disagreement is reasonable. However, this is not Rawls’s strict position (it would rule out any concept of unreasonableness), and anyway does not follow. A hermeneutics of suspicion may be necessary to expose entrenched biases; Rawls’s tabooing of it clearly makes it easy for bias, delusion, etc. to disguise themselves under the formal veil of “public reason”. (A good example would be a situation in which white bigots think anti-crime crackdowns are sufficiently valuable to justify racist policing as a tolerable evil. They might then claim that the experiences of oppressed minorities have “less weight”. In such a context, this group could construct a racist system based on views which formally appeal to public reason at every stage). An accusation cannot be rendered invalid simply because it may arouse “resentment and hostility”: the genuinely biased are as likely to react in this way as the affronted. In other words, Rawls does not show that an assumption of sincerity is in any way justified; he simply anathematises a hermeneutics of suspicion because it is inconvenient for his goal of systematisation. (A related problem is how to tell “scientific” issues, which yield “general facts”, from issues which are subject to the burdens of judgement. To clarify, Rawls clearly sees the issue of incentives as somehow above the burdens of judgement, and is therefore prepared to dismiss egalitarianism as “unreasonable”).

Perhaps one should even go a step further, and suggest that the concern to be “reasonable” is a blocking discourse which produces an aversion to originality and critical thought (which could disrupt the “stable” consensus or lead in subversive directions). Rawls certainly has an aversion to those who are too “controversial” or take reflexivity too far for his liking.

One finds a similar passage in The Law of Peoples. Of the “fact of oppression”, Rawls says: ‘Some may find this fact hard to accept. That is because it is often thought that the task of philosophy is to uncover a form of argument that will always prove convincing against all other arguments. There is, however, no such argument. Peoples may often have final ends which require them to oppose one another without compromise…. Political liberalism begins with terms of the politically reasonable and builds up its case from there’ - it is therefore, says Rawls, the only way to achieve peace (LP 123). One should note here that Rawls directly embraces a sceptical position, and also that the reasonable is taken to be a founding dogma.

The “fact of oppression” (e.g. JAFAR 84) seems to be part of the conception of the reasonable, even though it does not arise on the lists. (One might suggest that it is a consequence of the fact of reasonable pluralism). This fact is outlined as follows. ‘A public and workable agreement on a single general and comprehensive conception could be maintained only by the oppressive use of state power’, and in a way that is unstable and undermines the ‘constitutional essentials’ of liberalism (CW 425; c.f. JAFAR 188). The oppressive use of state power which would be necessary to maintain the rule of any comprehensive doctrine would have very harmful effects. It would encourage ‘official crimes’, ‘inevitable brutality and cruelties’ and ‘the corruption of religion, philosophy and science’ (JAFAR 34). A reasonable person rejects the oppressive use of state power (i.e. its use against other reasonable persons/doctrines). ‘Reasonable persons think it unreasonable to use political power… to repress other doctrines that are reasonable yet different from their own’ (CW 612, LN 16). Reasonable persons do not wish to impose their own comprehensive doctrines on others (JAFAR 192). They therefore believe it is unreasonable to impose their own comprehensive view (LN 138). The “fact of oppression” acts as a label for a rigid binary: either one uses the state oppressively, or one looks for a political conception of justice (CW 434). (The phrasing of a term such as ‘the suppression of reasonable doctrines is oppressive’ is pseudo-descriptive; in fact, it seems to be what John Searle calls a declarative. The assertion that reasonable persons think something is supposed to make it the case. The use of declaratives, which depends on power-asymmetries, recurs in Rawls’s work, suggesting he assigns himself a “legislator” role). (As regards the idea of “imposing” one’s comprehensive view, see below on “legitimate law” and also the section on exclusion: for Rawls, one can “interfere” with others, i.e. impose a view, simply by not obeying reasonable principles, and there is no difference between imposing a view and resisting legitimate laws).

There are a number of problems with this version of the reasonable. Firstly, it implies, like the idea of a prior set of reasonable doctrines, that one can specify the reasonable prior to the overlapping consensus. In other words, the reasonableness of the consensus can be tested by specifying which people are reasonable, and then establishing empirically whether or not they endorse the consensus. (If the consensus can be deduced from the “shared” beliefs, this should produce the same result as such an investigation). However, as I shall show below, Rawls resists any possibility of such an empirical investigation. Secondly, why does the imposition of this particular set of beliefs have a different significance to the imposition of any other set? Why is it qualitatively different to demand that everyone accept “the fact of reasonable pluralism” and to demand that everyone accept (say) “the fact that Jesus rose from the dead” or “the fact that Mohammed is the prophet of Allah”? Thirdly, the construction of this special category of facts seems to contradict certain of the facts themselves. If the burdens of judgement apply to everyone at all times, then one must be as uncertain of one’s endorsement of an overlapping consensus as of one’s endorsement of a comprehensive doctrine. Since one’s belief in (say) the fact of reasonable pluralism is open to the possibility that one has made a mistake in calculations, it cannot be a valid basis for coercion. It is no more “ reasonable”, based on the burdens of judgement, to set up a coercive political “consensus” than to impose a comprehensive doctrine. Rawls might respond to the second and third problems by reference to the idea of public reason (see below), but as I shall show, this idea has its own difficulties and would not overcome the problem of the burdens of judgement.

Another problem is that certain of the “facts” are not truth-claims at all. For instance, the “fact of reasonable pluralism” already requires a belief in the concept of the reasonable, not simply an empirical claim. While an acceptance than many comprehensive doctrines exist is an empirical position, the claim that several of these doctrines qualify as reasonable is a value-judgement and not a truth-claim. In a sense, beliefs such as “reasonable pluralism” and the “fact of oppression” (which amounts to a value-position that state power is oppressive if used to impose a comprehensive doctrine but not if used to impose a political conception) involve a circular justification of the reasonable: the reasonable is defined by its acceptance of itself. There is a problem with declaring that, while the fact of reasonable pluralism requires one to accept and tolerate differences, the fact of pluralism as such does not. One can conveniently ignore or deny the full version of the fact, i.e. the version which is actually a truth-claim, yet one is compelled to accept and act on the ethically-qualified version, a version which is not actually a fact at all.

The same circularity applies to various characteristics related to “fair” terms of social cooperation (see below on character-traits), since these criteria already depend on the concept of fairness which is derived from them.

The “fact of oppression” or “oppressive use of state power” (the only instance of Rawls using the term in his work) relies on an undefined distinction between oppressive and non-oppressive uses, and this distinction is a matter of judgement, not fact. The imposition of an overlapping consensus in a context where “unreasonable” persons/doctrines exist depends on the coercive use of state power, by Rawls’s own admission. Presumably, therefore, the oppressive character of the imposition of a comprehensive doctrine using state power results from the fact that it does not only suppress “unreasonable” doctrines but also “reasonable” ones. The claim that the suppression of reasonable pluralism must rely on oppressive use of state power is therefore not a description but an ascription. If it is also taken as a defining feature of reasonableness, it is therefore circular. (The idea of reasonableness and the “fact of oppression” are also implicit in the distinction, implied in The Law of Peoples, between attempts by outlaw states to spread their culture and attempts by liberal and decent societies to make all other societies “well-ordered”). (Chantal Mouffe also criticises Rawls for his belief that a final order can be achieved in which non-oppressive coercion is possible and in which non-compliance can be taken as proof of irrationality or unreasonableness. For Mouffe, such exclusion is necessarily “political” and experienced by the excluded as violence. See ?RP 28-30).

Especially if one deducts from the list the claims which are quasi-circular (but perhaps even if one does not), the beliefs Rawls lists are insufficient to produce the conclusions he reaches. (Minus the quasi-circular claims, one is left mainly with the dogma that “cooperation”, i.e. reciprocity, is good, the burdens of judgement and belief in moderate scarcity). In other words, his operative conception of the reasonable cannot be derived logically from the minimal list of beliefs which “reasonable” persons may be assumed to have. To take an example, the version formulated on CW 487 (accepting pluralism, freedom of thought and conscience, and the burdens of judgement) does not in any way require a pro-state position. Therefore, Rawls’s concrete theory would exclude conceptions which are by his own definition reasonable, by insisting on subordination to the state.

In particular, there is no way in which the need for a political conception can be derived from the existence of pluralism and/or the burdens of judgement. One would also need both a basis for suspending belief in the burdens of judgement in the case of the political conception, and a reason for wanting the existence of an overarching state power. (The burdens of judgement also seem to be suspended in the cases of the borderline between the reasonable and the unreasonable and in the case of the so-called “general facts”). In particular, the burdens of judgement plus moderate scarcity (sufficient resources to satisfy basic needs but not all the presently operative desires) would not preclude a rhizomatic society without any kind of state organisation. It is clear from Rawls’s discussions, however, that he feels an irreducible need for a master-signifier - a moment of “decision” - and also that he wishes to include bigots and petty-minded people among the “reasonable” (whereas the status of altruists and prefigurative activists is less clear). Logically, this gesture is unreasonable if the burdens of judgement are accepted. Rawls insists that the burdens of judgement only authorise a limited space of variance ‘where reasonable persons may differ’ (PL 56). However, he does not specify this space, and the idea of the burdens of judgement would not seem to authorise it. Rather, if (for instance) people necessarily disagree about the “weight” of different claims, it is a priori impossible to construct a single just order which assigns claims a proper weight. If the burdens of judgement are accepted, their limits apply just as much to disputes between “reasonable” and “unreasonable” as they do within the world of sameness.

The specific beliefs which are conditions of being “reasonable” are generally quite contentious. The possibility of gains from social cooperation is asserted and is based on a counterfactual, so there is no good reason to accept it (especially since “cooperation” is a technical term and so could not directly include, for instance, benefits of the division of labour as instances of its benefits). It could, indeed, be taken as a dogmatic act of faith in liberalism: an assertion of its universality in spite of its exclusionism (i.e. that it leads to benefits for “all” even though it involves domination by some over others).

One could add as regards the burdens of judgement that Rawls takes his case too far. While he may be right to claim that everyone is vulnerable to these burdens, it is by no means clear that all are equally vulnerable, as Rawls claims (JAFAR 197). Such an equality in falsity is necessary for Rawls to derive the conclusions he seeks (e.g. regarding the silencing of truth and the endorsement of repressive tolerance). An example of a case in which Rawls is wrong is the following: X knows that Y’s position is wrong because Y is engaged in a blatant logical self-contradiction, or because X is aware of decisive evidence which falsifies Y’s position. X remains subject to the burdens of judgement, with the result that X’s positive position might also be false, and X is aware of this. However, X can be relatively certain that Y is wrong, and this makes X’s position much more likely to be right. (There is perhaps a marginal possibility that X’s judgement of Y’s falsity is also wrong, but this seems to be a peripheral possibility). In this circumstance, there is no good reason for X to accept the primacy of Y’s position in public discourse - for instance, to accept Y’s victory via majoritarian or juridical procedures. Also, availability/knowledge of evidence and precision/clarity of concepts can vary a lot between different discourses and different participants in a debate. (I have in mind cases where majority or elite/juridical opinion is grossly misguided due to the insidious operations of hegemonic apparatuses. For instance, in my view, people have every right to fight against the victimisation of asylum-seekers, and to do so in defiance of so-called “legitimate law”. I don’t see why this right should be affected by the pervasiveness of vicious racist beliefs, because it is easy to demonstrate that these beliefs are grossly false. To invoke the “burdens of judgement” in such a case is like invoking them in the case of Holocaust denial: it is an excuse for levelling all statements to the same low level, with the effect of aiding bias and “injustice”. While all are subject to the burdens, it is not the case that all are equally subject to them).

In fact, the special status of so-called political conceptions is reinforced in the case of the burdens of judgement, from which they seem to be excluded. The pluralism of comprehensive doctrines is taken to render the imposition of one such doctrine immoral and unreasonable, at least partly because of the coercion necessarily involved, yet the pluralism of “public reasons” or of attitudes to politics is still subject to the same imposition. Thus, Rawls replaces the phrase “no salvation outside the church” with the phrase “no salvation (or order) outside the liberal state”, replacing one dominant system with another. He recognises a field of differences only on condition that it be quilted by a master-signifier, and in the case of this signifier (the reasonable and the conception of the person), difference is suspended. (NB how this singularity is supposed to be necessary to make all the differences possible). He divides his beliefs up into those he thinks he can demand that everyone accept and those he thinks he cannot (in practice, between conditional and unconditional demands, sometimes separated by the trope of “balance”). Thus, the political (the state) plays the role in Rawls’s theory that religion plays in absolutism. If liberalism seems in practice less oppressive than absolutism, this is only because of contingent factors (such as the number of liberals in existence and the range of contingent ways of life which liberalism happens to permit) from which Rawls cannot strictly speaking draw any advantages.

CAN ONE JUDGE ONESELF TO BE REASONABLE? (Reasonableness as self-assessment exercise)

One major difficulty with the concept of the reasonable is that it is an externally-directed concept (i.e. it involves positions which take others into account and should be acceptable to others, or at least to others of a certain type), yet it is determined solely by internal criteria. At one point, Rawls says that something must ‘be seen to be reasonable by other citizens’ in order to be reasonable (PL 243). ‘Insofar as we are reasonable, we are ready to work out the framework for the public social world, a framework it is reasonable to expect everyone to endorse and act on, provided others can be relied on to do the same’. It sets terms which are to be publicly recognised as ‘grounding our social relations’, and therefore sets up a public world (PL 53-4). Citizens are supposed to take each other into account as having reasonable conceptions, albeit on the assumption that others endorse the political conception (PL 387), and to appeal to ‘public values that we might reasonably expect others to endorse’ (PL 253). ‘We try to appeal only to principles and values each citizen can endorse’ (JAFAR 41; c.f. JAFAR 92). A ‘successful’ theory must be acceptable to others as well as oneself (JAFAR 27), and one should use arguments ‘that others are able to accept’ (JAFAR 118). A reasonable person has a goal ‘to live politically with others’ based on views ‘all might reasonably be expected to endorse’ (PL 243). Rawls aims to construct ‘a mutually acceptable point of view’ which gives a ‘shared basis’ for judgements so ‘each cooperates… with the rest on terms all can endorse as just’. He therefore starts from premises everyone who fits his model of the person ‘may reasonably be expected to share and fully endorse’ (JAFAR 27).

However, Rawls’s theory is hopelessly non-dialogical. The “reasonable” seems to be deducible by an individual in reflective equilibrium, and this requires that it has no concrete reference to others. Even at its most extensive, the reasonable only means the individual observer assessing how affected others would react: ‘even though we think our arguments sincere and not self-serving when we present them, we must consider what it is reasonable to expect others to think who stand to lose should our reasoning prevail’ (JAFAR 116). Repeatedly and in different contexts (international as well as domestic), Rawls declares that something is reasonable if one believes (or ‘sincerely think[s[‘) that other people will (or can ‘reasonably be expected’ to) accept it (e.g. LN 57, PL 226, 253). A particular person will have an internal ‘criterion’ or ‘test’ which determines what reasonable others ‘can reasonably be expected to endorse’ (PL 226). Each person must have principles and guidelines she/he believes are ‘what others… might also reasonably be expected reasonably to endorse’ (CW 581). Coercion is proper as long as one sincerely believes that the reasons are sufficient and that others might reasonably accept them (CW 578). One has met one’s duty of civility (and, by implication, one is reasonable) as long as one thinks others could accept one’s premises and conclusions (CW 594), and one fulfils this duty if one explains one’s coercive acts by reference to what one regards as the most reasonable political conception (CW 576). ‘Citizens are reasonable when, viewing one another as free and equal in a system of social cooperation over generations, they are prepared to offer one another fair terms of cooperation according to what they consider the most reasonable conception of political justice; and when they agree to act on those terms, even at the cost of their own interests in particular situations, provided that other citizens also accept those terms. The criterion of reciprocity requires that when those terms are proposed as the most reasonable terms of fair cooperation, those proposing them must think it at least reasonable for others to accept them, as free and equal citizens, and not as dominated or manipulated, or under the pressure of an inferior political or social position’ (CW 578). Reciprocity as a criterion of reasonableness means one must think others will find a proposal reasonable in order to make it (LN 14).

In one passage, Rawls states that ‘[t]he overall criterion of the reasonable is general and wide reflective equilibrium’ by a citizen in civil society. Wide equilibrium arises if one has considered and ‘weighed the force’ of different conceptions, and therefore implies ‘wide-ranging reflection’. In other words, it remains radically individual. However, he assumes that it can nevertheless operate intersubjectively, because the singular point of view of the citizen in reflective equilibrium is recognised by everyone (PL 384-5). This idea rests upon an implicit assumption that people are identical, so that adopting the same imaginary image produces the same result. There is no reason whatsoever why one person’s wide reflective equilibrium will end up the same as another’s. (Rawls refuses to argue with others on their own terms, insisting instead on the possibility of reproducing the same singular argument in each case).

Thus, the other-regardedness seems to be reduced to sameness: others are conceived as psychological clones of the self, and actually-existing others are deemed “unreasonable” the moment they become incomprehensible from the position of sameness. The self assesses whether others “might reasonably accept” something - but if this assumption is incorrect, they can legitimately be coerced, on the basis of what the self decides they should accept. Furthermore, it is ‘to be expected’ that people will have different criteria as to what actually counts as reasonable. For Rawls, it is enough to have (the empty form of) the criterion of reasonableness, because this ‘alone imposes very considerable discipline on public discussion’. This discussion will then determine which conception is the most reasonable (PL 227). This is a variety of form-imposition, an insistence above all that the form of the master-signifier (the reasonable) be retained, whatever content is to fill it. This ensures that there is always a dominant group, whatever particular conception of sameness it uses to determine the limits of the “reasonable”. The net effect is that Rawls demands that people have a desire to act in ways which are reasonable to others, but that he also suggests that this need not include any desire or requirement to consult or negotiate with others: the self is to decide unilaterally what it is reasonable for others to accept. Indeed, he hints that he thinks the self’s discourse must use this form of expression to be legitimate. In this form of discourse, everything rides on what one can and cannot “reasonably” expect others to accept - a debate in which others may or may not actually participate, and from which they may even be excluded. (The reference to the conception of the person is indicative, because Rawls assumes this provides a specific content which “everyone can be expected to endorse”).

The trick is, of course, that Rawls allows people to coerce others into doing what they can “reasonably be expected to endorse” (PL 140, JAFAR 190). In other words, Rawls excuses domination, provided the person who exercises it believes she/he is acting on the basis of universality conceived in a particular way. ‘If we can make the case that there are adequate reasons for diverse reasonable people jointly to affirm justice as fairness as their working political conception, then the conditions for their legitimately exercising coercive political power over one another… are satisfied. The argument, if successful, would show how we can reasonably affirm and appeal to a political conception of justice as citizens’ shared basis of reasons, all the while supposing that others no less reasonable than we may also affirm and recognize the same basis’ (PL 390). (I should not have to add here that domination is, of course, not exercised by “reasonable” citizens over each other, but rather, is exercised by the “reasonable” in-group over excluded out-groups).

Rawls specifically denies using his theory in the way I here suggest. He suggests that a reasonable political conception must not be an account by the powerful of how they can convince themselves they are acting properly. It must, rather, address ‘each citizen’s reason’ so it ‘appeals… to citizens viewed as reasonable and rational’ (JAFAR 186). However, he does not explain how one avoids the situation where the powerful convince themselves they are acting properly by convincing themselves that their power is a product of “each citizen’s reason”. After all, it is still one’s own decision whether a particular objection is a result of a citizen’s “reason” or of something else, and the terms of address to this unitary “reason” are still determined internally. It is only the fantasmatic image of sameness which could convince one that one is indeed addressing others’ “reason” through Rawls’s procedure.

I have repeatedly suggested that in practice, the “reasonable” operates as a simple invalidatory discourse, but this is not Rawls’s intent in formulating it. He thinks that there must be reasons independent of any particular disagreement for labelling others as unreasonable, irrational or not conscientious, and these reasons ‘must be in principle recognizable by those who disagree with us’ (PL 121). It is not clear, however, how something can be “in principle recognizable”, or how one can know it is. One can tell in dialogue if someone does in fact recognise a reason, and one can tell from discourse analysis if someone uses a particular type or form of discourse, but both of these are concrete analyses and cannot be determined ‘in principle’. In other words, the self who is barred from simply using disagreement as a basis for anathematisation can nevertheless use disagreement alone as such a basis provided she/he also maintains that ‘in principle’ others would recognise the basis (i.e. provided she/he reformulates the position in universalist terms, in a “hegemonic” gesture in the Laclauian sense). In practice, the form of discourse Rawls encourages is a form which consists of claims to know a priori what “all reasonable people” believe. (There are also certain other ideas which are supposed to limit the extent to which one can use the idea of the reasonable to invalidate any view one happens to agree with: one has to consider how one would view one’s view from others’ standpoints, one has to believe it is reasonable in good faith, and so on. However, these criteria all seem to be interior and non-falsifiable: Y can neither know whether X has carried them out correctly, nor whether X has carried them out at all). (The only kind of manoeuvre which is substantively prohibited is the direct assertion of self-interest or comprehensive doctrine as common interest. For instance, a capitalist could not use the discourse of the “reasonable” in relation to a claim founded only on her/his desire to become rich. However, she/he could reformulate the claim in reference to the growth of the national economy and a need for capital accumulation, and it would then be impossible to show that the view is unreasonable. Of course, the capitalist might know very well that the poor would utterly reject the claim. Nevertheless, she/he could abstractly claim that she/he could “reasonably expect” the poor to accept the claim - even if in practice they prove to be “unreasonable”. Anyone can posit any claim as “reasonable” so long as she/he has access to sufficiently sophistic language strategies; the advantage would seem to be with those who are most skilled in the manipulative use of language. I could construct arguments that all my own positions reflect things that people “would” accept if only they were reasonable, just as Rawls can - the difference being that I seek a less manipulable basis for argument). The operation is something akin to Kant’s categorical imperative, i.e. one is to judge one’s position as if it were a Supreme Court judgement (PL 254). (What if one’s judgement is that the courts should be disbanded?)

It seems, therefore, that X can declare her/himself to be reasonable in the following way. If X thinks X’s belief is so obvious that Y “can be expected” to accept it, X can declare this view to be reasonable, and therefore also declare X’s imposition of it to be reasonable. There are limits to how far X can do this, since X must retain a commitment to an overlapping consensus (which would seem to rule out X declaring all her/his views to be definitive of the reasonable). However, it does not place any barriers on X’s making assumptions that all other people are like X, and therefore does not prevent X from exercises in repression of difference whenever X does not understand it. One could even find a case where X knows Y will reject demand A, yet X claims that Y would accept the demand if Y were reasonable; X takes into account what X can “reasonably expect” Y to accept, and takes no account of Y’s actual position because X believes Y to be unreasonable (in this case or in general). It is not clear how X’s position in this case differs significantly from the situation which would pertain were X to simply be unreasonable towards Y (i.e. if X had no interest in whether Y could reasonably be expected to accept A).

Žižek puts an objection to Rawls as follows. ‘When the trial by “veil of ignorance” tells me that, even if I were to occupy the lowest place in the community, I would still accept my ethical choice, I move within my own fantasy frame - what if the other reasons within the frame of an absolutely incompatible fantasy?’ (Enjoy your Symptom p. 109). The illusion here is, again, the illusion of sameness constructed by the misleading emotion of empathy. Rawls wrongly believes that it is possible to figure out what others will think of one’s reasoning (CW 443), a position which amounts to a denial of radical difference. The idea of the reasonable is supposed to guarantee some mutual comprehensibility: ‘we think the balance [resulting from any application of the reasonable] can be seen as at least not unreasonable in this sense: that those who oppose it can nevertheless understand how reasonable persons can affirm it. This preserves the ties of civic friendship’ (PL 253). In other words, the reasonable is a repressive mutual comprehension between the “normal” which prevents any of them denouncing the others as unreasonable.

Žižek’s criticism is to the point. Indeed, the insularity of the reasonable seems to be a characteristic derived at least in part from characterological superiority. Rawls explicitly links the legitimacy of the authority of judges to their personal sincerity. In other words, they embody some kind of decent system of law so long as they sincerely and reasonably believe that they are acting on an idea of justice and not merely exerting force. A judge is publicly defending ‘society’s injunctions’ provided only that she/he meets this subjective criterion when making decisions in court (LN 66-7). In other words, it is not systems of domination to which Rawls objects, but particular types of psychology. Thus, by having the correct orientation as specified by Rawls, one can self-certify as reasonable.

It is not clear why X should accept that Y’s view is reasonable simply because Y believes it to be a view that others can reasonably be expected to accept. Such acceptance (even reluctantly) by X could only occur either if the gesture of pseudo-empathy itself convinced X to be prepared to submit to Y, or if X valued the empty form of the proclamation of reasonableness above its content.

Part of the problem is that the reasonable operates with the term at both sides of the equation. The reasonable act reasonably ‘provided others do so’ (PL 54), and reasonableness is necessarily something which operates on both ends of a relation, only arising when both self and other are reasonable (PL xliv). Reasonableness is reciprocity, but only reciprocity with other reasonable persons (PL xlvi). X is reasonable if X acts reasonably towards Y, if and only if Y is also reasonable, and this means that X is in a sense able to decide the meaning of the term in each instance of application. Since the significance of any act in establishing X’s reasonableness is dependent on the status of others in relation to whom X is acting, the term becomes a referent for an exclusive in-group, rendering its individual-attributive grammar misleading. Indeed, I suspect that the idea of being reasonable to others which is part of the self-conception of the reasonable is fantasmatic, and possibly even mythical. There is supposed to be a commitment to improving life for everyone through cooperative effort (PL 54), but of course, the nature of cooperation is defined intensionally: it may, in fact, be nothing more than the core of Rawls’s own “fantasy frame”. (Indeed, I would question the significance of the idea of social cooperation as having positive benefits. I suspect that Rawls is pursuing security, not satisfaction, and that the idea of positive benefits of cooperation is a fantasmatic supplement to this).

A situation could arise as follows: X cannot see how Y and Z could view conception A as reasonable, given X’s knowledge of X’s own needs and circumstances (because conception A is on some level a radical denial of or threat to X). However, X can well understand how Y and Z could agree to conception A and establish mutual comprehension on its terms. In this sense, X can see how Y and Z could conceive A to be reasonable, yet X could not accept that A is reasonable or that Y and Z are acting reasonably. X may be put in this position even if X meets the minimum requirements of Rawls’s conception of the reasonable (i.e. X is prepared to cooperate with Y and Z in a fair system which is reasonable to X, as understood by X, and X also accepts the burdens of judgement within the range of the reasonable as understood by X). An example would be a form of inadvertent discrimination. For instance, a situation where X requires social openness as a prerequisite for social activity, but Y and Z both believe that the need for openness is outweighed by concerns about security, and so set up a highly closed and striated space which is effectively inaccessible or less-than-usually accessible to X. One could add the assumption that Y and Z also require that X work, and that this is made difficult or impossible by the way they arrange social space. This situation would seem patently unreasonable to X, even if X were entirely reasonable in the sense discussed in this section. In practice, Rawls’s approach would tend to lead to X being labelled as unreasonable, clearly contradicting the idea that the reasonable is any kind of internal attribute. (One could also add that there is no way to determine whether an individual is reasonable if the individual has never had an opportunity to enter into relations she/he might reasonably expect others to endorse and which she/he can also endorse; for instance, someone who is psychologically different and is confronted with an already-constructed normalist system, or a Black Panther activist responding to unreasonable racist treatment through militant direct action. The same can be said for cases where social circumstances render “reasonableness” impossible in other ways. In particular, the idea of moderate scarcity implies that there is a point at which urgent need renders people unable to be “reasonable”).

This is not a basis for a stable consensus, since all that is shared is the empty desire to be “reasonable” and/or to construct a “reasonable” social system. Its actual content may vary drastically between different individuals. (It may, however, be a basis for consensus if the desire for unity is assumed not only to be shared but also to be primary, i.e. if X will accept Y and Z’s construction of what X “can reasonably be expected to accept”, even if X thinks it is unreasonable; for instance, in the example above, provided that X is prepared to accept being excluded and put in an untenable position because X values X’s own social position less than X’s commitment to social order. This is not, of course, a position of reciprocity; it involves a form of self-sacrifice). What Rawls ignores is the possibility that people may each have, in good faith, conceptions of what others should reasonably accept which others find intolerable or unreasonable.

In social terms, the confusion involved in the idea of authorising oneself to determine what others “can reasonably be expected” to believe is a recipe for domination. People such as judges can claim in a substitutionist way to be the carriers of the “claims of others”, and can then invalidate any claim by others on the grounds that it is “unreasonable”. In practice this means that they use the term “unreasonable” to invalidate any defiance of their own will. From the imaginary godlike standpoint of the agent of perfect balance, they can then pick and choose which unconditionals are to “count”. Reasonableness plus procedure equals elite domination. (One case that comes to mind involves a judge who extends the drive for gleichschaltung so far as to attack a man who had not yet built a shed because he was ill. When the man accused him of breaching his human rights - a perfectly valid accusation - the judge accused him of being “obsessive” and suggested he seek counselling - and then threatened the man into apologising when this triggered a predictable outburst! The Kafkaesque absurdity here is that the judge’s own claims are as “obsessive” as those of any possible defendant: the judge was also acting on an unconditional drive, albeit a reactive one with no substantive basis). An example of Rawls supporting such a discourse is that the only limit he puts on decisions by the Supreme Court is that judges must believe in good faith that rational and reasonable citizens can reasonably be expected to endorse their judgements (PL 236). In other words, they are not externally limited at all, and those who are “taken into account” are passive and voiceless in the entire process.

In any case, the obligation to be “reasonable” to others seems to be limited to others one judges to be “reasonable” (i.e. who have the same capacity to be “reasonable”, i.e. a similar mode of thought), and Rawls remains slippery on how this is determined. Since one can, in effect, be unreasonable to people or doctrines which are unreasonable, without problematising one’s status as a reasonable person, one’s own definitions may easily simply reaffirm one’s own conception by invalidating opponents as “unreasonable”. (Rawls does not, for instance, seem to problematise the way his position on inequality might appear to the worst-off: they may well feel the idea of “incentives” to be a self-serving excuse for the activities of the rich). The problem is, again, that the reasonableness of the insular group as a whole cannot be established on the basis of standards which refer internally to this group (for instance, that X acts on a particular conception if and only if Y does).

The problem is that it is impossible for any individual to clearly intuit what might or might not seem reasonable to others. Others’ needs, desires and psychological structures may or may not be the same as one’s own. Indeed, it is ‘reasonable’ (in the sense of ‘realistic’) to assume that one will not have the ability to intuit others’ intuitions, and to assume that others will not necessarily conform to a particular psychological model. In this sense, the reasonable is unreasonable.


At times, Rawls resorts to the mythical construction of “reasonableness” as a character-trait. At one point, Rawls lists criteria for a ‘reasonable moral psychology’. These include the ability to acquire and act on principles of justice, a readiness to cooperate in just/fair institutions provided others do so, a preference for basic rights and liberties and the ability to trust others who act fairly (as well as a number of specific beliefs) (CW 445). In another passage, the ‘psychology of the reasonable’ is rendered identical to the idea of reciprocity: one is reasonable if one responds in kind. ‘[T]he reasonable generates itself and answers itself in kind’ (JAFAR 196). There are also other occasions where Rawls establishes an equivalence between reasonableness and reciprocity (CW 316, PL 302). Also, Rawls here assumes that people can be reasonable, so much so that he is prepared to use the assumption in historical explanations (JAFAR 196, 198). He supplies a list of specific characteristics of reasonable persons, including the two moral powers, cooperation in just institutions based on guarantees that others will also, preparedness to ‘do their part in just and fair institutions’, trusting others who cooperate, and recognising four general facts and their consequences (JAFAR 196-7). Reasonable persons relate to others in a particular way, assuming that others are honest and expecting disagreements and differences of opinion (CW 478-9). They are also fairminded and open to compromise (JAFAR 190). Political liberalism only needs to address ‘doctrines that reasonable citizens affirm’, as opposed to those which arise from ‘people’s understandable tendency to view the political world from a limited standpoint’ such as self- or class-interest (PL 37). In an early paper, Rawls identifies reasonableness (apparently used in a narrower sense than in his later work, and without a differentiation from the rational) as including elements such as a willingness to use inductive reasoning to find reasons for choices, open-mindedness, a willingness to reconsider beliefs, and some kind of attempt to compensate for one’s own bias and prejudice (CW ?12-13). Virtues such as ‘reasonableness’ and the ‘duty of civility’ are supposed to be necessary in order for reasoned discussion to take place (PL 224). In one passage, the virtue of reasonableness is counterposed to the idea that ‘to have character is to confront’ or that ‘[t]o be is to confront’ or that language is or should be ‘warfare’. Such assumptions are presumed to involve a failure to see the ‘great public good’ of the virtue of reasonableness (JAFAR 118).

I call this construction mythical because the actions which express the supposed trait are clearly situational and complexly mediated. It is not clear that one can refer to an outcome such as “a readiness to cooperate in just institutions” as a psychological as opposed to a situational variable. Other problems with this approach include its implicit privileging of the “reasonable” as a group: this group in effect engages in “special pleading”, since its domination is not in the interests of the “unreasonable”. In effect, the characterological rendition of the idea of reasonableness naturalises intersubjective relations by constructing them as the outgrowth of individual psychological traits, while simultaneously positing as unquestionable the superiority of a normalised in-group.

Furthermore, the operation of “reasonableness” is too situation-dependent for it to be a character-trait. If a reasonable person only acts reasonably when others also act reasonably, it is not akin to concepts such as “greed”, “laziness”, “generosity” and “extroversy” which might be said to designate character-traits. A reasonable person on Rawls’s model is not always reasonable; she/he is quite capable of being unreasonable towards unreasonable persons. Furthermore, she/he is tolerant of the “unreasonable” unconditionality of the state, which is not at all required to compromise with those outside its limits. Rather than being reasonable by virtue of being open to compromise, she/he becomes reasonable by belonging to an in-group or conforming to a particular social system if and when it exists. She/he is then open to compromise with those people who also conform. This is a question of political/ideational alignment, not character.

The ambiguities of “reasonableness” rub off on the other characterological concepts. For instance, as regards “fairmindedness”, this concept does not specify when one should accommodate to others and when one should (or need) not. Rawls only specifies that one shows the quality when making the decision (PL 217), without giving criteria as to how one knows whether the quality has been shown.

If the “reasonable” is to include anything which is not a reference to a “limited standpoint”, the vexed question of the boundary between the universal and the particular becomes crucial. Rawls’s theory does not seem to include all possible doctrines which could couch their claims in terms of the universal, so he would need some criterion for differentiating the two types which does not rely on the principles of justice. It is by no means obvious that Rawls’s conception of the person, to take one example, is an expression of a universal interest, or that rejection of this conception is “unreasonable” in the sense of being based on group- or self-interest.


Perhaps more to the point is to suggest that one has to conform to Rawls’s idealised model of the person to qualify as reasonable. In one passage, he defines ‘reasonable persons’ as persons who conform to his model of being free and equal, 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). (“Free” includes “responsible for ends”, autonomous in the Kantian sense, i.e. free from heteronomous desires, etc., while “equal” means, among other things, having a sense of self connected with the idea of being a citizen). Similarly, in another passage, Rawls claims that his views are reasonable ‘for those who conceive of their person as it is represented in the procedure of construction’ (CW 355). In other words, being reasonable is coterminous with accepting the whole of Rawls’s ethical theory.

This is often how the term seems to operate, yet it would rule out any appeal to the idea of the “reasonable” as a justification for this theory itself. It would tend to render Rawls’s process of reasoning tautological, because the fact that (for instance) his conception is able to construct an overlapping consensus of reasonable comprehensive doctrines would be true by definition.


One strange characteristic of Rawls’s concept of the reasonable (or of “reason”) is that it seems to have little to do with “reasons” in the plural. A view can be reasonable, even if one can give no reasons for it (CW 9). It is also a starting-point from which arguments proceed (LN 123), rather than a reasoned conclusion. Furthermore, since it generates itself and answers itself in kind (JAFAR 196), the reasonable is based on a closed and exclusionary system of sameness. Thus, “reason” is not only a mythical rendition of “reasons” - it tends to become their opposite and to transmute into doublespeak. Whereas “reason” was historically constructed in opposition to “dogmatism”, Rawls seems to fuse the two: a reasonable view can be, and in some senses must be, dogmatic. Another aspect of dogmatism arises because the reasonable is taken to be something which grounds social relations and which it is reasonable to expect everyone to endorse (PL 53-4). However, it is also taken to be a matter of fact that unreasonable persons and doctrines exist, so that, when one takes it to be reasonable for everyone to endorse it, one is engaged in a denial of what one knows to be true. Thus, the gesture of exclusion - of declaring a borderline between the reasonable and the unreasonable - is dogmatic, and involves a performative denial of the existence of certain kinds of difference. Furthermore, the reasonable does not ground social relations if these are also backed up by coercion against the unreasonable. The relations are, rather, grounded in the act of exclusion itself.

Another hint at dogmatism arises when Rawls says that reasonable comprehensive doctrines either have ‘a certain looseness’ or are ‘not… fully comprehensive’, so there is enough ‘leeway’ for them to ‘cohere loosely’ with a political conception. He does not care where this looseness comes from - even if it comes from, for instance, a desire to submit to whatever is normally done - as long as it exists (PL 159-61). In other words, he demands of comprehensive doctrines a submissiveness towards the political core. It is not that adherents of comprehensive doctrines make up their own minds to be reasonable; it is that they submit to the reasonable as a prior set of dogmas.

One could compare also the following passage. ‘the balance of reasons itself rests on judgment… The best worked out political conception cannot overcome these limits; nor are they defects, as they lie in the nature of our [sic] practical reason. In political philosophy, as elsewhere, we must rely on judgment as to what considerations are more and less significant, and when in practice to close the list of reasons. Even when judgment is unanimous we may not be able to articulate our reasons any further’. The final appeal is simply to ‘due reflection’ (JAFAR 134), with ‘due’ unspecified. In other words, he ultimately asserts a right to make judgements without reasons.

In one passage, Rawls formulates the reasonable in an explicitly dogmatic way. ‘Political liberalism offers no way of proving that this specification is itself reasonable. But none is needed. It is simply politically reasonable to offer fair terms of cooperation to other free and equal citizens, and it is simply politically unreasonable to refuse to do so’ (LN 88). Here one sees a definitional/assertive dogmatism (“no salvation outside political liberalism”, so to speak) which demonstrates a persistence of the “unreasonable” as the extimate and quasi-disavowed foundation of the “reasonable”. Treated in this way, Rawls’s dogma is not differentatable from those of the doctrines he labels “unreasonable”; it is assumed to be primary only because it affirms this particular dogma, to which Rawls happens to be committed.

There is another passage where Rawls effectively asserts ‘acceptability’ to be the final word in political argument (JAFAR 27-8). This idea assumes the existence of an in-group who determine what is or is not acceptable, i.e. who decide what they will and will not accept and who then project this via a declarative statement into a discursive universality. In this context, the others with whom one engages can be nothing but clones of the self. Further, the insular group is sometimes defined explicitly through its relation of otherness to an “irrational” other. As a reason for trusting the judgements of so-called competent judges (an early incarnation of the insular category of the “reasonable”), Rawls says that ‘one could hardly be expected to prefer judgments made under emotional duress, or in ignorance of the facts by unintelligent or mentally sick persons, and so on’ (CW 18). In other words, the in-group establishes its right to impose its view on the psychologically different, because it alone valorises its own judgement as superior to that of its other.

In practice, the “unreasonable” seems to become an irrational limit, undefined and open only to quasi-tautological specifications, to how far thought can challenge reality. One becomes “unreasonable” if one is “extremist” or nonconformist. One becomes “reasonable” if one dogmatically supports the status quo (or rather, a “barely just” status quo). A basic human reaction to an inhuman situation may find itself labelled as “unreasonable”, whereas a complex construction of a system of domination is “reasonable”. It becomes “unreasonable” to ask why one must accept systems of domination such as the statist bulldozer and the capitalist system of work and inequality - especially if this “why?” initiates a refusal to submit. Inclusion is dependent, either on holding a specific series of Rawlsian dogmas, or on being prepared to play along in a public “let’s pretend” game based on these dogmas (as in the Lacanian notion of fetishism).

Sometimes, Rawls seems to establish his conclusions first, and then accept arguments because they conveniently produce the conclusions he wants. For instance, he admits that claims about limits to the possible (e.g. the “fact” of reasonable pluralism) are conjectural, defending them nevertheless on the grounds that they nevertheless open up the possibility of the social forms he advocates (LN 12). This is hardly a good excuse, since they also close discursive space via the idea of the “unreasonable”.

Another occasion where insularity tends towards dogmatism is in the idea that the fact that democratic peoples esteem and value their democratic institutions proves that these institutions have value as a ‘great social good’ (e.g. JAFAR 201). It is as if such peoples are invulnerable to the burdens of judgement. It would also leave open the possibility that Rawls’s enemies, such as “outlaw states”, would also be shown to express a ‘great social good’ merely because they themselves value their own system. Estlund’s term “insularity” contains a presumably unintended reference to a very different model: Badiou’s idea of insularity as an identity as an ‘island of law and liberty’, as if somehow cut off from the world system (Ethics p. 33). It is in such discussions of “democratic peoples” that Rawls comes closest to establishing the identity of the two uses.

One should compare how the logic of the liberal state presents itself as “reason”, even though its basis is often irrationalist. (For instance, the reasoning of juries is concealed so as to maintain an illusion that court decisions are based on a rationally-reached certainty).


In some occasional remarks, the “reasonable” is taken to be a characteristic of an unspecified and quasi-metaphysical entity known as “reason”. For instance, the reasonable is unique (and therefore, presumably, primary) because of ‘its tie with reason’ (JAFAR 196). In another passage, reasonable persons are assumed to ‘share a common human reason’ (CW 476). Elsewhere, Rawls states that a conception must ‘win its support by addressing each citizen’s reason’, or else it ‘is not reasonable in the first place’ (PL 143). In another passage, rights and liberties are justified on the grounds of a right to apply reason (PL 417). Reason operates as a quasi-agentified characteristic. Rawls claims that ‘reason… is self-originating and self-authenticating’ (PL 100), suggesting that it may be a form of dogma. (It is also unclear how “reason” relates to and arises from desire).

This kind of comment displaces the problems of interpretation from “reasonable” (adjective) to “reason” (noun), but do not seem to affect it otherwise. In particular, the issue of distinguishing “reasonable” and “unreasonable” returns in the problem of distinguishing a “common human reason” from a common set of imposed dogmas. There have, after all, been many hegemonic ideologies in world history, any of which could in principle term itself the “common human reason” of a group of “reasonable” persons, whereas the term “reason”, in the way Rawls uses it, implies its singularity. In other words, the issues are displaced, not resolved.


One suspicion I have, which I hinted at above in the section on order, is that for Rawls, reasonable persons are specified by their submission to a particular kind of state. He is unwilling to declare this openly, because it would tend to locate his theory symmetrically in relation to those of absolutists, but it emerges in his establishment of links between reasonableness and the idea of “legitimate law” (or acceptance of a particular constitutional scheme). He admits that he is mainly prepared to use force against people who do not accept the constitution or laws (JAFAR 182).

The distinction between “legitimate law” and “oppressive use of state power” is a central dividing-line. For Rawls, laws supported merely by force are ‘grounds for rebellion and resistance’ (LN 88). However, the elements he adds to “mere” force are fairly insubstantial (e.g. the requirement that those who use force believe internally, in a non-verifiable way, that they are acting reasonably, and that they are prepared to frame excuses in terms of “public reason”). In practice, the division seems to run along the lines of whether the force is part of an archetypal liberal state or not. In theory, however, Rawls refers the distinction back to his idealistic framework: law is legitimate when it is ‘addressed to’ rational beings (CW 118), and constitutional democracy reflects what is known in ‘foreseeable conditions’ (PL xix). I suspect that this formula is ideological, in the sense that it covers the operative criterion with other criteria which are not operative. In other words, a liberal-democratic state is automatically assumed to be the outcome of a situation where people decide basic questions on the basis of public reason, are prepared to offer terms they reasonably expect reasonable others will endorse, and so on. Operationally, the satisfaction of such criteria can be read off directly from acceptance of the liberal state (c.f. the slippage between “liberty” and “system of liberties”). Such quasi-deliberative ideas therefore form a fantasy-frame providing a mythical legitimation for an actual power regime. (For instance, deciding between different truth-claims is supposed to be publicly impossible, and the ordering discourse - in Rawls’s rhetoric, the drive for “civility”, to avoid “strife”, for “workable arrangements”, etc. - is supposed to dull specific concerns sufficiently to enable generalised submission. However, the fantasmatic expression of this refers to ideas such as the “burdens of judgement” and “reasonable pluralism”). One can observe the fantasmatic operation of Rawls’s ideal categories in a passage where he specifies the reasonable as it operates through a string of increasingly concrete stages. In the original position, the most abstract, the reasonable permits a lot, and one can observe it permitting less and less as one draws closer and closer to actual people (PL 340). I also suspect that the comprehensive doctrines which enter a Rawlsian overlapping consensus do not do so as participants; they are subsumed/overcoded into an overarching structure. Their main role is to provide fantasmatic legitimations for the forced choice imposed by the liberal state. As Rawls puts it at one point, ‘the just draws the limit, the good shows the point’ (JAFAR 141).

Rawls claims that ‘a democracy necessarily requires that, as one equal citizen among others, each of us accept the obligations of legitimate law’ (CW 590). Any law supported by others on grounds of public reason is to be ‘binding on citizens by the majority principle’ and to be ‘reasonable and legitimate law’ (CW 606). Institutions ‘framed… to yield just legislation’ are to limit valid claims (JAFAR 150). Hence, ‘reasonable citizens understand [the idea of legitimacy] to apply to the general structure of political authority. They know that in political life unanimity can rarely if ever be expected on a basic question, and so a democratic constitution must involve procedures of majority or other plurality voting to reach decisions. It is unreasonable not to propose or accept such arrangements. Thus, citizens recognize the familiar [sic] distinction between accepting as (sufficiently) just and legitimate a constitution… and accepting as legitimate (even when not just) a particular statute or a decision in a particular matter of policy’. Crucially, this model of legitimacy requires acceptance of majority decisions even when these are unjust (for instance, one can oppose war yet must accept the state’s right to go to war). Demanding consistent justice becomes itself unreasonable. ‘It is unreasonable to expect in general that human statutes and laws should be strictly just by our lights’ (PL 393). On the other hand, it is ‘reasonable’ to expect people, ‘within certain limits’, to ‘abide by’ constitutional essentials (CW 121). He hints that somebody is reasonable if they support, or can be brought to support, constitutional essentials (JAFAR 189). He claims that it is necessary that in a just society, everyone ‘recognize… legitimate law’ and ‘accept the obligation not to violate it’ (CW 595). Indeed, a duty to obey unjust laws is taken by Rawls to be a general fact he has to account for. ‘The moral grounds of legal obligation may be brought out by considering what at first seem to be two anomalous facts: first, that sometimes we have an obligation to obey what we think, and think correctly, is an unjust law; and second, that we sometimes have an obligation to obey a law even in a situation where more good… would seem to result from not doing so’. Rawls calls such an obligation a ‘familiar situation’ (CW 119). It is, he suggests, alright to have a procedure which ‘decides whose opinion is to determine legislative policy’, even though this cannot render it true (CW 121). Rawls uses the term “fact” for such dogmas, which clearly reflect an asymmetrical and irrational privileging of the state over all others and the unconditionality of the state even in relation to its own legitimatory basis (another example of the way in which Rawls turns into natural facts in the case of the normal the kind of imperatives he invalidates as unreasonable in the case of the abnormal). (Rawls’s explanation for this fact is that it is justified because one is only compelled to obey, not to agree. ‘The vote does not result in a statement to be believed’. Were belief necessary for obedience, one could not give an undertaking to obey - CW 120-1. If it is true that statist imperatives are not statements to be believed, then the state is truly Kafkaesque; it has no rationality in it at all, and Rawls’s entire Kantian schema falls down. In practice, the imperative to obey is in fact an imperative to believe, but in a fetishist way: an internalised obligation to obey implies that one knows very well that a law is wrong, yet still one keeps acting in line with it, i.e. the Lacanian formula for fetishist belief. The distinction between obedience and belief is simply an affect-blocking discourse; it involves a disavowal of one’s own political agency. In practice, submission to majorities involves the “bad faith” of a doctrine which expresses itself in conditions of repressive tolerance but caves under pressure from others. It involves an unpreparedness to believe that one’s views are right, and a trusting of the state - or at least the rightness of the existence and function of the state - above oneself - or at least above the rightness of any other belief).

Reasonable persons are supposed to have a ‘sense’ (a reified term for a belief) that the ‘imperfection’ of laws and institutions is a necessity (PL 242). (To operate as a “sense”, it would have to be a part-unconscious libidinal investment). One could also link the idea of legitimate law to Rawls’s rather strange assumption that one will obey something unless one has a good reason not to (TJ 405). A primordial, pre-ethical obedience seems to be part of what it means to be “reasonable”.

Thus, “reasonable” persons seem to be those who put proceduralist conformism ahead of ethical concerns, provided that they are acting within a more-or-less liberal state. They are people who do not render the liberal state conditional on anything, and who do not even use “justice” as an exterior standard to hold the state to account. They are people who lack the dignity of holding their own beliefs; rather, they alienate their ethical judgements into exteriority in order to avoid conflict. (Again, Rawls fails to establish that people in the original position would agree to such draconian limits: clearly those who wish first of all to guarantee certain higher-order interests, and who demand, for instance, that their comprehensive doctrines be left intact could not accept a distinction between justice and legitimacy). Presumably, reasonable persons are therefore persons who value the mere existence of “legal” repression above any concerns about the content of this repression (perhaps with a few exceptions; see below). They may “see themselves” as “free and equal” and as “politically autonomous”, but if so, they are deluded. In fact, they alienate their ethical selfhood to the state. Rawls binds ethics a priori to the state, and insists that resistance normally stop short of taking on the state.

Majoritarianism only sneaks into Rawls’s account in the small print, so to speak, and its basis in his theory is unclear. In other words, why should reasonable citizens believe that a majority of reasonable citizens will reach more reasonable conclusions than a minority? (One should keep in mind that they must hold this view strongly enough to justify coercion on its basis). There is no logical way to derive the necessity of majority voting from the non-existence of unanimity, especially in the case of unconditional needs and desires. (The majority, when it coalesces as, say, a normalist or ethnocentric group, is very likely to pursue a total territorialisation of social and physical space which is grossly intolerant towards all others, but which would nevertheless be able to articulate its claims in terms of “public reason”. Even if the majority is “unreasonable”, it could fake “reasonableness” to secure dominance).

On one occasion he specifically derives belief in majoritarian procedures from an assumption that reasonable persons must support a specific social system. ‘It is reasonable to expect a variety of opinion in political judgments generally, and it is unreasonable to reject all majority voting rules. Otherwise liberal democracy becomes impossible’ (LN 87). The structure of this argument implies that it is unreasonable to accept any view which would have the outcome that ‘liberal democracy becomes impossible’. In other words, it implies that the “reasonable” as an exclusionary group are defined by their commitment to a particular regime of power. (The first half of the first sentence alludes to the “fact of reasonable pluralism”, but this “fact” is insufficient to make sense of the rest of the statement). In another passage, he says that he only concerns himself with political conceptions ‘reasonable for a constitutional democratic regime’ (CW 585). In other words, the existence of constitutional democracy is taken as a good, prior to any commitment to reasonableness, and it determines what is and is not to count as reasonable. On another occasion, he specifies those among whom he wishes to narrow disagreements (i.e., in his usual terminology, the reasonable) as ‘those who affirm the basic principles of democratic institutions’ (JAFAR 134). Elsewhere, he says that it is the consistency of doctrines with ‘a democratic polity’ which determines whether he is prepared to trespass on their space (CW 615). Further, reasonable comprehensive doctrines are defined by having a coherent and consistent ordering of values and recognising the essentials of a liberal democratic regime (LN 87); ‘all reasonable doctrines affirm such a society with its corresponding political institutions’ (CW 608). ‘The basic requirement is that a reasonable doctrine accepts a constitutional democratic regime and its companion idea of legitimate law’ (CW 574). ‘Political liberalism also supposes that a reasonable comprehensive doctrine does not reject the essentials of a democratic regime’, and that a doctrine which does not accept such essentials is irrational or mad and should be contained (PL xviii-xix). ‘On the other hand, comprehensive doctrines that cannot support such a democratic society are not reasonable’, because unless they support such a society, they cannot guarantee reciprocity or basic rights (CW 608-9). At one point, Rawls works as if he can derive a theory compatible with all reasonable doctrines by drawing on existing ideas supposedly fundamental in democratic culture (PL 25). Majority voting is ‘reasonably taken as fixed, as correctly settled once and for all’ (PL 151). On one occasion, acceptance of ‘legitimate law’, ‘political authority’ and ‘a reasonable democratic constitution’ form one of three defining features of reasonable persons, while acceptance of constitutional essentials is the one criterion for a comprehensive doctrine to be reasonable (CW 613). In some passages, the concept of decency functions as an equivalent to legitimacy. What Rawls terms imperfections are supposedly tolerable as long as they lead to a ‘decent political order’ (JAFAR 4). (The idea of “legitimacy” seems to be a restatement of the idea of the “barely just” present in the first edition of A Theory of Justice but removed from the revised edition). Liberal democracy is assumed a priori to be reasonable: ‘we start from the conviction that a constitutional democratic regime is reasonably just and workable, and worth defending’ (JAFAR 37, although Rawls also writes here of trying to make it appeal to ‘reasonable’ people, who must therefore exist independently of whether they support liberal democracy). He takes a similar position in IR, stating that a society of peoples is reasonably just if it follows a reasonably just Law of Peoples (LN 5; c.f. LN 29).

The reason Rawls endorses the idea of legitimacy is that he feels an irrational desire for social closure to be established through a decision. Hence, the role of legitimacy is ‘to authorize an appropriate procedure for making decisions when the conflicts and disagreements in political life make unanimity impossible or rarely to be expected’ (PL 428). In order for this to operate, one must have the condition that a decision ‘must be made’ (PL 428), as well as the burdens of judgement/fact of reasonable pluralism (whereas Rawls is only able to account for the latter).

One should also notice that Rawls’s discussions of the nature of violence are heavily overshadowed by his model of legitimate law. He takes any instance of forceful resistance to legitimate law to be the same as trying to impose one’s own conception by force (CW 606; c.f. PL 138), when clearly there is a massive gulf between the two (since the relationship between different agents is radically different). Although the case he cites (militant anti-abortionism) happens to involve people who probably are trying to impose elements of a comprehensive doctrine (Christian fundamentalists), the “precedent” set by Rawls’s position is dangerous, and mainly operates against the direct action movement (who are not trying to impose a conception in the sense Rawls suggests). To treat direct-action activists as if they were illiberal statists is a blatant fallacy. A similar view occurs in another passage, when Rawls suggests that the only alternative to state violence is private violence. ‘I stress here that the Law of Peoples does not question the legitimacy of government’s authority to enforce the rule of democratic law. The supposed alternative to the government’s so-called monopoly of power allows private violence for those with the will and the means to exercise it’ (LN 26). Notwithstanding that government violence generally does far more harm (in the form of massive-scale wars and genocides) than private violence (never mind whether Rawls would defend such violence; it is rendered possible by the institutions he legitimates), this ignores other options apart from state and private dominance. Indeed, it naturalises the existence of dominance in some form. (The absence of a state would not require laissez-faire private violence, since it would not preclude the existence of ethical restrictions, self-defence, solidarity action or non-state reciprocal arrangements. People without a state might also lack the means or the will to engage in private violence on a regular basis).

The situation regarding Rawls’s views is complicated by his insistence on discussing his model primarily in relation to intolerant others and his tendency to evade other issues of difference. There is certainly an implication that he wants “democratic” institutions obeyed unconditionally, and a preparedness to resist (universal) ‘democratic law’ for (particular) specific reasons is a major taboo (see CW 589).

One should not overstate the case, because there are also limits to the exercise of legitimate law, beyond which it becomes illegitimate. The binding to obey unjust laws ‘stops at certain limits’ (CW 180). ‘Laws cannot be too unjust if they are to be legitimate’ (PL 429). For instance, ‘no one is expected to put his or her religious or nonreligious doctrine in danger’ (CW 590), and courts can sometimes make ‘unreasonable decisions’ (JAFAR 147). However, the borderline is disturbingly unclear (at one point Rawls says he cannot discuss the degree of deviation allowed from justice by the idea of legitimacy - PL 390 - and at another point he refers to its ‘undetermined range’ - PL 428). It rests on a distinction between tolerable and intolerable quantities of injustice which Rawls is reluctant to define or specify. When given, the specifications are vague. For instance, in one passage Rawls argues that unjust laws should be tolerated to the degree necessary to make democratic politics possible, and no further (CW 181 - notice how Rawls is tolerant of the injustice of laws but intolerant of the injustice of resistance). The meaning of this standard in practice is unclear, because the acceptance or non-acceptance of a specific law is unlikely to involve the direct dissolution of democratic politics as such, and extrapolations regarding the generalisation of a category of acts are necessarily speculative. (It is also not clear whether it is “structurally” significant, readmitting an unconditionality external to the state, or whether it is simply a reassertion of the insistence that the social order must be roughly-speaking liberal to be legitimate. In other words, it may simply mean that Rawls is reaffirming liberalism against a more expansive majoritarianism which would also affirm the right of majorities to impose, for instance, a state religion. It is certainly clear that majorities do not have an unconditional right to impose anything and everything in Rawls’s theory, and that this could be conceived as a limit to “legitimacy”). Rawls would presumably oppose protesters’ right to enforce freedom of movement by dismantling police barricades and fighting back against the police, for instance in the Red Zone in Genoa. However, there is no way to establish whether such resistance would make liberal democracy impossible, or whether, on the contrary, it would strengthen the population against unaccountable organisations such as the G8. (Even if one could justify the state, there would be a strong case for keeping it in its place by constantly treading close to the line at which it would break down, so as to maximise freedom; there is no reason why a case for the existence of the state should also entail a case for its smooth or frictionless functioning). In another context, Rawls claims that law is legitimate and ‘politically (morally) binding’, and ‘to be accepted as such’, so long as lawmakers follow public reason (CW 578). In other words, obedience is conditional only on the formal correctness, by Rawlsian standards, of politicians’ rhetoric. Also, Rawls elsewhere claims that specific laws need not be justified by reference to public reason (JAFAR 91), further weakening this restriction. Even in the case of religious intolerance, Rawls does suggest that people who hold “unreasonable” doctrines, or whose doctrines arise in conjunctures which force them to act “unreasonably”, should be expected to give up their doctrines, since these would have “no weight” according to his schema. In general, ideas such as “legitimate law” seem to operate as affect-blocks. As an element in language, they massively strengthen the state by providing let-outs in relation to specific “injustices”.

This aspect of the “reasonable” shows most clearly the statist biases in Rawls’s theory. As long as the state meets a few minimal and unspecified criteria, it is taken to be “always right” in matters of force, even when it is wrong. (Opponents can therefore be right in substance and yet still be wrong when they act consistently on their beliefs). Rawls’s pervasive statism also problematises the special status he claims for his doctrine, because any kind of state can (and would) establish a dividing line for inclusion and exclusion based on submission to its own type of political regime. Rawls cannot, therefore, establish differentiations between himself and other statist doctrines on this basis. (Any benefits for individuals from this particular kind of state are secondary if the reasonable is defined in this way). In addition to eliminating his claims to a special status, this model also constructs Rawls’s defence of the liberal state in a circular way. It is by definition true that reasonable persons, defined as persons who accept a liberal constitutional order and “legitimate law”, embrace this kind of social order; it is a case neither for the social order nor for the choice of this particular group as one’s theoretical focus. Rawls has succeeded only in demonstrating that a liberal state is the most satisfactory political order possible among people who all accept the essentials of a liberal state.

My objections on the level of human dignity, unconditionality of people versus the state and unconditionality of active versus reactive desire are too complex to discuss here. I shall also not dwell on Rawls’s patent failure to demonstrate the benefits of legitimate law - complete with the resultant injustices and indignities - when seen from the standpoint of the original position. (In my view, to have basic dignity is, in a sense, to be a “stickler”; also, a “no sticklers” rule is intolerable because it strengthens the state to an enormous degree, making it hard for people ever to say “ya basta” no matter what happens). Suffice to say that there is a large difference between concern for others and preparedness to submit to an overarching framework of generalised control. It is worth noting also that Rawls tends to construct the reasonable so that, while its fantasmatic frame is constructed around ideal and ideational notions (preparedness to tolerate others, character-traits favouring fairness, belief in a string of general facts such as reasonable pluralism, etc.) which are difficult or impossible to operationalise and which are vaguely defined, the basis for determining whether individual X is reasonable or unreasonable tends to become behavioural. X’s reasonableness or unreasonableness can be determined by reference to X’s acceptance or defiance of a particular regime of power. The actions drive the conclusions and the conclusions drive the premises in terms of their reasonableness. ‘Whether this or some other understanding of public reason is acceptable can be decided only by examining the answers it leads to over a wide range of the more likely cases’ (PL 254). In other words, as long as the “conclusion” of people’s reasoning is the same as Rawls’s, they are “reasonable”, even if, like the quack mathematician who just happens to get the right answer, their basis for the conclusion is nonsense. (c.f. his explicit statement in LP, p. 68, that it is the outcomes of a belief-system which qualify for inclusion in the Society of Peoples, regardless of the origins of these outcomes). In this way, he fixes thought, asserting a denkverbot that it must not cross certain boundaries. In practice, therefore, the issue is not a binary between imposing a model of truth and accepting a plurality of reasonable conceptions. It is about the primacy of ideas or “behaviour”. Rawls in practice allows violence against those who are deviant on the level of social action, and his implicit attack on the idea of truth operates as a way of insulating this “public” violence from ethical and epistemological critique. Another effect of the de facto behavioural emphasis is that doctrines deemed reasonable tend to be colonised by a conformist urge, because the “costs” (libidinally as well as materially) of deviance are so great. In any case, the implication of this aspect of the reasonable is clear: Rawls’s commitment to the liberal state is prior to his category of the reasonable and his theory of justice. It is a commitment for which he does not account, and which in all probability is neither “rational” nor “reasonable” (in the sense of having other claims which support it, and in the sense of being open to dialogue or compromise). In other words, this is the one point on which Rawls is a “stickler”; it is the master-signifier of his theory. (It is “unreasonable” to posit active desires as unconditional, but not to posit the systematising drive as unconditional). Also, such aspects render the content of the overlapping consensus prior to the reasonableness of the doctrines which endorse it. It is therefore no surprise that it proves possible for Rawls to articulate these doctrines into such a consensus; if he claims this possibility to be evidence for either the doctrines or the consensus, he is engaged in circular argument. Rawls’s concept of reasonableness is also dependent on a rightist suspension of the ethical: while particular people and groups are barred from being “sticklers” (in Larmore’s vocabulary) and from positing unconditionals (which Rawls tends to wrongly identify exclusively with rightist concerns), this bar is dependent on the state itself being “unreasonable”, being a stickler and positing unconditionals. For instance, police are inherently “sticklers”, insisting on their own victory at all costs (“you’ve totally got to take people on - see Intro) and entering into “negotiations” (for instance, during sieges) on the assumption that others will alter their positions and police will not. Their stance cannot be assumed to be reasonable from the perspective of those police fight; their position is inherently non-reciprocal. However, it is presumably those who resist the police who Rawls would term “unreasonable”. (One could make a case that any master-signifier is inherently “unreasonable”, since it involves an attempt to construct a singular core discourse and impose it on everyone).

One could add a couple of extra points regarding related aspects of Rawls’s theory. Firstly, he suggests that reasonable persons accept a conception of objectivity as established by the point of view of ideal reasonable and rational agents, and therefore as never established by any particular agent’s view. No individual’s or group’s belief that something is just or reasonable makes it just or reasonable. ‘[I]t is definitive of reasonable agents’ that they recognise five criteria of objectivity, including this one and the resultant view that ‘the judgment of any agent… may be mistaken’ (PL 111-12). This seems to be a restatement of the idea of the burdens of judgement, except that, in this case, an ideational basis is given for the suspension of scepticism in the case of state institutions. An idealised position is taken to escape the otherwise total logic of fallibility, and this position is presumably identified with the state.

The state also sometimes finds itself hiding behind a special ethical sphere known as “political values”. Reasonableness is then specified as an acceptance that political values are not overridden in most circumstances, even by greater values (CW 609). A reasonable doctrine establishes a reasonable balance of political values; ‘not all reasonable balances are the same’, but a view is unreasonable if it ‘cannot support a reasonable balance of political values’ (though it may not necessarily be pervasively unreasonable) (PL 253). (This implies also that it is “unreasonable” not to accept the metaphor of balance). Rawls tries to avoid claiming that he renders political values primary over all else, even though this is logically necessary for the role he assigns them in his theory. However, he does state that the existence of the ‘special domain’ of the political and its outweighing of other concerns expresses the ‘very great values’ it carries. This is because he assigns them a status of founding social existence itself (PL 138-9), a clear confirmation of my claim that he treats the state as a master-signifier. (Societies pre-existed the state, so it is logically impossible to conceive of the state as having this kind of grounding role, except by ignoring inconvenient evidence).

Incidentally, if reasonable citizens necessarily support a coercive state, it would seem to follow that they either cannot foresee any circumstance in which the state could act against them (an illusion), or else they have a demeaned self-conception which allows them to conceive of circumstances in which they could legitimately be coerced (presumably involving a double-bind of some kind). In either case, this constitutes a breach of Rawls’s model of the person.

Rawls inverts the relationship between the state and the ideas which support it, claiming that political life in fact arises from a desire to live with others on ‘terms that all reasonable citizens can accept as fair’ (PL 98). This idealist misreading of the actuality of state practice is necessary for Rawls to perceive obedience to a liberal state as somehow embodying the “higher” ideals he espouses.


The grammar of Rawls’s discussions of the “reasonable” tends to construct it as a characteristic of persons or doctrines, even though this contradicts the way he usually specifies it (i.e. by reference to the liberal institutions or overlapping consensus which defines whether doctrines are liberal). For instance, he expresses a hope (not a tautological certainty) that justice as fairness might be endorsed by ‘all reasonable comprehensive doctrines’ (PL 375). He also wants to limit what a political conception can say, with reference to reasonable doctrines (a limit which is implicit in the distinction between comprehensive doctrines and political conceptions). In his “Reply to Habermas”, he states that a political conception should avoid ‘any idea… which not all reasonable doctrines could reasonably be expected to endorse’. One must, for instance, believe it does not violate others’ ‘deepest religious and philosophical commitments’. This argument is supposed to have a specific referent, since Rawls also asserts that ‘Habermas’s comprehensive doctrine violates this’ (PL 389-90). Hence, Rawls wants to use the concept of the reasonable so as to establish a basis for his own theory vis-à-vis others. He also wishes to use it to specify a role for his own theory, by which his success can be measured. That an overlapping consensus be endorsed by actual opposing views which are likely to thrive is necessary to produce stability of the right kind, one of Rawls’s main goals (CW 421). He says similar things elsewhere. ‘[T]hose holding a reasonable comprehensive doctrine must ask themselves on what political terms they are ready to live with other such doctrines in an ongoing free society’ (PL 392). A political conception is reasonable only if it is endorsed by reasonable comprehensive doctrines and therefore justified ‘at the tribunal of each person’s understanding’ (PL 391). That most reasonable persons accept a view renders it reasonable (LN 67; c.f. “Outline for a Decision Procedure…” on competent judges), and Rawls’s own theory is problematised if, and only if, it is rejected by reasonable comprehensive doctrines (PL 65-6). However, the problem is that the reasonable are not required to take account of the doctrines of the unreasonable (e.g. CW 614), so only the borderline between the reasonable and the unreasonable can give a positive content to the requirement that a doctrine not violate a commitment to rely only on views reasonable others could endorse.

If the concept of the reasonable is really to be taken as a characteristic of doctrines (or persons), as opposed to an inference about doctrines or persons projected onto them as a result of their status in relation to a particular theory, it should be possible to specify reasonable doctrines or persons independently of their support for a particular political conception. It should, further, be possible to construct a list of actual or possible reasonable doctrines against which the reasonableness of an overlapping consensus or a political conception can be assessed. However, this is an approach Rawls avoids at all costs. Setting up a political conception in order to include actual views would make it ‘political in the wrong way’, although it should avoid putting barriers in the way of those who could potentially accept it, and it should be framed so as to win support (JAFAR 37; c.f. JAFAR 188). The political doctrine must not be tailored to gain the allegiance of existing comprehensive doctrines (CW 491). Rawls also supposedly tries to avoid being ‘arbitrary and exclusive’ by defining reasonable doctrines in a ‘loose’ way (PL 59).

It is in his rejection of the method of examination in the light of actual or theorised comprehensive doctrines that Rawls shows that the category of the reasonable is derived from his political conception (i.e. a doctrine is reasonable if it reaches conclusions which accord with Rawls’s own on a range of issues specific to the political sphere: legitimate law, public reason, the value of an overlapping consensus, and so on). However, he continues to suggest that this is not the case, because he wishes to derive the validity/reasonableness of his political conception from its ability to gain acceptance from reasonable doctrines (requiring that he avoid the circularity involved in deriving its validity from its acceptance by those doctrines which happen to accept it). Rawls is here trapped in an inescapable contradiction, because the only way to avoid circularity would be to establish the kind of conceptual gap which would authorise exactly the kind of empirical assessment he repeatedly rejects. Unless Rawls is able to specify the reasonableness of persons and/or doctrines independently from their acceptance of his political conception, he cannot claim that this conception is reasonable or otherwise valid on grounds referring to the conceptions it can subsume or from which it can gain acceptance (e.g. its ability to avoid “the oppressive use of state power”, to construct an overlapping consensus of “reasonable” doctrines or to take account of “reasonable pluralism”). Indeed, independently from some other basis either for the validity of the political doctrine or the reasonableness of the persons/doctrines who/which accept it, such arguments become mere evidence of insularity, if even that. (Any doctrine could claim to avoid the “oppressive use of state power” in the sense of not using state power to suppress doctrines which agree with it, etc.). As I have already suggested, the tautological/circular operation of the concept of the reasonable does not preclude its having a specifiable referent, and I suspect Rawls is specifying the reasonableness of his conception on the basis of undeclared criteria of his own. However, he also seems to have standards which separate reasonable from unreasonable doctrines which usually operate in a connotative or mythical way. If, for instance, a theory should operate so that someone who has an affect-blocked character-structure, a strong superego, a nominal commitment to reciprocity and a tendency to naturalise “incentives” may nevertheless end up declared as “unreasonable”, this would probably make the view seem intuitively unacceptable to Rawls in reflective equilibrium.

Hence, Rawls is prepared to provide lists of particular conceptions he considers as reasonable, such as Kantian and Millian liberalisms and free faith religious doctrines (PL 169-70). He sometimes uses such specifications as the basis to claim that he must not adopt some positions as part of the political conception. For instance, the political doctrine must not affirm the idea that there is an independent order of moral values, as this would exclude rational intuitionism, which is a reasonable doctrine (PL 95). He implies that the reasonableness of doctrines can be seen empirically when he states that ‘the history of religion and philosophy shows that there are many reasonable ways in which the wider realm of values can be understood so as either to be congruent with, or supportive of, or else not in conflict with, the values of the special domain of the political as specified by a political conception of justice. History tells of a plurality of not unreasonable comprehensive doctrines’ (JAFAR 190). His theory also depends on the claim that he allows ‘ways of life fully worthy of human endeavor’ and ‘sufficient scope’ for such activity (CW 386). ‘[J]ust institutions and the political virtues expected of citizens would serve no purpose - would have no point - unless those institutions and virtues not only permitted but also sustained ways of life that citizens can affirm as worthy of their full allegiance’ (CW 449). (This does not, however, suggest that the worthwhile character of a doctrine renders it reasonable - see CW 463). He also states that his political conception ‘must be acceptable to a plurality of opposing comprehensive doctrines’, in order to be liberal and in order to fit the facts of the matter (CW 493). A reasonable overlapping consensus not only has to include all reasonable doctrines, but also has to put these in an ‘enduring majority’, guaranteeing stability (PL 391). In the preface to Justice as Fairness: A Restatement, he says that ‘we must keep track of different points of view’ in order to ‘have any chance of gaining the support of an overlapping consensus’ (JAFAR xviii). A valid political conception must be able to ‘gain the support of a diversity of reasonable comprehensive doctrines’ found in a well-ordered democracy, especially those with significant and enduring support (JAFAR 184). Another argument which suggests a responsiveness to actual doctrines is when he claims there are more doctrines in the world than in any one society (LN 19), a claim linked implicitly to his claim that a world government would necessarily be oppressive. (Fewer comprehensive doctrines could only lead to a less oppressive society if the reasonable is defined to include existing doctrines). All of these claims imply certain specifications about reasonable doctrines, and the claim about “history” suggests that the reasonableness of doctrines can be directly perceived or intuited. (It is worth noting that the context of this quote is an attempt to show the possibility of a liberal overlapping consensus, so the reasonable doctrines cannot logically involve a reference back to such a consensus). Another instance where Rawls avoids establishing circularity is in the case of his argument for a right to conscientiously refuse military service. This right seems to be the result of an attempt to include pacifists and Quakers within a liberal overlapping consensus which they could accept, but would be required to reject if this concession was not made. (This implies an opening onto a string of claims Rawls otherwise rejects: that comprehensive doctrines or persons have a right to some kinds of unconditional claims which must be taken into account by a tolerant doctrine. Once this string of claims - which might include a right of the hungry to steal and a right of protesters to resist police violence - is accepted, the state itself, in its intolerant unconditionality, tends to be rendered “unreasonable”. It is arbitrary to establish a right to conscientious refusal of military service but not also to insist on a right to conscientiously refuse jury conscription, compulsory occupation orders, compulsory education - so-called truancy - and other state-imposed “duties”). His position that metaphysics are acceptable within a political doctrine as long as they are uncontroversial and minimal (e.g. CW 429) also depends on a reference to the actual content of doctrines, i.e. whether doctrine disagree about such metaphysics as are used.

In one passage (in “The Domain of the Political and the Overlapping Consensus”), Rawls expresses both contradictory views simultaneously: firstly, the view that an overlapping consensus must be endorsed by ‘each of the main’ comprehensive doctrines likely to endure (not even limited to “reasonable” doctrines in this passage) and by a majority, and secondly, the view that principles of justice must not be amended to fit in with existing conceptions (CW 473-5).

He also sometimes specifies unreasonable doctrines, but his tendency is always to identify this genus with a particular species, namely right-wing authoritarianism. Hence, he lists religious fundamentalism, aristocracy, the divine right of kings and instances of autocracy and dictatorship as examples of unreasonable doctrines (CW 609). On another occasion, the list includes religious intolerance, slavery, voting limited on grounds of property, and male-only suffrage (CW 579). This clearly does not demonstrate the extent of his concept of the unreasonable.

In one passage, he suggests that reasonable comprehensive doctrines have certain formal features (involving philosophical and moral views, covering the whole of life and organising values), as well as specifying that they must include ‘recognized values’ and form an intelligible world-view (PL 59). This suggests that he is using criteria related to his own empathy with doctrines to specify them as reasonable (i.e. they are reasonable if Rawls finds them intelligible and if they include values he recognises).

One can also add in terms of Rawls’s method that a conception which defines the reasonable in a circular way could not be the result of the deliberations of persons in the original position who are not specified as reasonable in this sense. In other words, a conception of the reasonable which is defined by acceptance of a political conception could only be accepted in the original position if the parties are already assumed implicitly to hold the conception. This means that Rawls’s decision procedure either assumes this, or the conclusions do not follow from the premises. The flaws in the argument from the original position actually seem to fall into both of these categories. On the one hand, the highest-order interest seems to be an interest in the existence of a Rawlsian type of state, thereby already projecting the conclusions into the premises. On the other hand, there are displacements in terminology and argument which are also flawed. For instance, the parties’ concern for their basic liberties would only lead to Rawlsian conclusions if the parties primarily sought formal liberties, whereas Rawls assumes they seek actual liberties. Furthermore, the parties’ ignorance of which specific comprehensive doctrines are held by those they represent would not preclude introducing considerations of specificity and inclusion, as Rawls seems to assume it would. Rather, it would simply require that the parties not engage in special pleading on behalf of any one particular doctrine. One could examine the issue slightly differently by posing the question: would the parties in the original position be sufficiently committed to a “right”/privilege of employers, statists, etc. to impose uniforms that they would forfeit the liberty to hold the purple-jumper conception should this turn out to be the conception they/those they represent hold when the veil of ignorance is removed? A “no sticklers” rule could only be deduced from the original position if the parties are not motivated rationally but already have a primary irrational commitment to generalised subordination to a single centralised doctrine. Furthermore, a Rawlsian theory, including the first principle, could only be deduced if this primary commitment already involved a preference for a liberal state - in which case the argument from the original position is superfluous; all Rawls has proven is a tautology, i.e. that those who prefer liberal states prefer liberal states.

Why is Rawls so determined to avoid reference to specific conceptions prior to the political conception? His official excuse (that this would make his theory “political in the wrong way”, requiring a substantive process of bargaining between different groups) is only part of the story. Rawls’s theory is a theory of closure: it tends to give what Rawls terms “enabling rights” to the state and to induce pressures to conform, and its primary commitment is to systematisation, rules and procedures (c.f. section on exclusion). It is an overarching regulative conception and not a minimal conception which simply protects people from others’ intolerance. If the “reasonable” were specified as a characteristic of doctrines prior to the political conception, this would impose extensive duties of openness on the social system and the state, and would turn concerns about openness into external standards which could delegitimate a particular state. Furthermore, it could do so on the basis of reference to particularities which happen to be reasonable (including empirical cases as well as analytical ones). To take once more the “purple jumper” example (from the exclusion section), an approach which specifies reasonableness prior to the political conception it accepts would be likely to see such a conception of the good as entirely reasonable and to demand its inclusion as a prerequisite for the validity of a political conception (since the conception of the good involves only minimal demands on others and is compatible with the existence of a wide range of contradictory doctrines). It is only by constructing the political conception as a primary good, prior to and definitive of the reasonableness of conceptions, that one could render the conception unreasonable (under what Larmore terms the “no sticklers” rule). This prior conception is expressed in an image of an essence of the person, and it is only via this mythical manoeuvre that Rawls can avoid taking actual doctrines into account.

Rawls directly endorses a position that the “reasonable” is definable by reference to a prior group in his early essay “Outline of a Decision Procedure for Ethics”. Here, he defines reasonableness of a principle as residing in its being able to ‘explicate’ existing judgements of certain kinds, to win the allegiance of competent moral judges, to convince such judges in cases of moral perplexity and to ‘hold its own’ in the minds of such judges when it fails to accord with their existing views (CW 10-11). Although this assumes a prior group - those who are later termed the “reasonable” - it in some ways simply displaces the issue onto the concept of “moral competence”. If someone could be labelled as morally incompetent if they do not accord with particular principles, the argument would again become tautological.

In any case, Rawls only aims to be acceptable to reasonable comprehensive doctrines ‘so far as possible’ (PL 114), implying that even included doctrines have no overarching claims. It also seems to be enough that the political conception be ‘expressly designed’ to include reasonable comprehensive doctrines (JAFAR 185-6), whether it actually does so or not.


The “reasonable” is not an empty concept in Rawls’s theory; it has an active theoretical role. It is used to establish the validity of particular concepts which Rawls puts forward as part of justice as fairness. For instance, the criterion of primary goods is to be accepted because it ‘seems the most feasible way to establish a publicly recognized objective and common measure that reasonable persons can accept’ (TJ 81). On another occasion, he claims that a doctrine which refused a right to first-trimester abortions would be unreasonable, because ‘any reasonable balance’ would include such a right (although this does not necessarily mean that those who oppose this right are otherwise unreasonable) (PL 244). All reasonable doctrines agree that one cannot “deserve” native endowments (JAFAR 74). Also, if ‘we’ are reasonable, ‘we’ accept that social-positional interests are not a valid reason for accepting a principle of justice (JAFAR 18). ‘It would stand to reason’ in a decent society that something like the Nazi boycott of Jewish shops is ‘a blatant violation of freedom of religion’ and conscience (LN 22). It is also supposed to be unreasonable to constantly improve the wellbeing of the worst-off (JAFAR 64) or to demand that specific associations operate the difference principle in their internal affairs (PL 261). He also specifies that the idea that there is no salvation outside the church is unreasonable (JAFAR 183), and specifies a set of ‘reasonable’ schooling requirements, which are actually quite extensive (JAFAR 156-7 - basically, it is “reasonable” that one must learn to be cooperative and to be a good wage-slave). Meanwhile, the first principle of justice is declared to be reasonable because supported in reflective equilibrium (JAFAR 105-6). Such statements seem to imply that the “reasonable” is equivalent to Rawls’s personal views on particular subjects, although he tends to resist this conclusion. (This is not surprising, because it would remove the distinction between a political conception and a comprehensive doctrine). In the case of abortion, he later amends his position under pressure on this subject. His point, apparently, was simply that a doctrine ‘runs afoul of public reason’ if it is unable to present itself in terms of public political values (CW 605-6).

In one passage in “Justice as Fairness: Political Not Metaphysical”, Rawls defines what is ‘reasonable for us’ in terms of its fit with certain ‘provisional fixed points’ present in ‘our political culture’, such as religious tolerance and opposition to slavery (CW 393). In this case, the reference to an insular group is explicit. However, different “points” could be “fixed” or made nodal so as to construct a variety of different versions of “we”, and Rawls does not provide any basis for preferring the liberal “we” to any other type (c.f. the discussion below on public reason).

The only characteristics of persons he specifies as limits on political conceptions involve ensuring that these remain within a space of normality which permits the ordinary operation of capitalistic logics. Hence, rules are not to be too complex, are not to impose ‘excessive transaction costs’ and are not to ‘burden citizens with requirements of knowledge and foresight that they cannot normally meet’ (PL 268, although the existing law would seem to fall down on these criteria). This seems to be a formal requirement regarding the relationship between people and rules, i.e. it involves a model where capitalistic/serialised relations between people are retained and where the role of rules is as a reified separate space. In this way, ethical action is reduced to obedience, individuals do not have to concern themselves directly with ethics when making day-to-day decisions, and the rules operate as a fantasmatic supplement of capitalistic seriality (see section on “basic structure”).

He also gives examples of reasonable disagreements. For instance, he states that there is reasonable disagreement about the causes and effective responses to unemployment (CW 478). In this case, the assertion that disagreement is “reasonable” operates as a decision on inclusion: presumably, Rawls includes those who blame unemployment on idleness instead of economics, at the expense of potentially excluding the unemployed themselves. (It is not clear why such a characterological view is “reasonable” in Rawls’s terms). This is an instance where Rawls’s idea of reasonableness is a case for repressive tolerance. Another example of a supposedly reasonable disagreement is over the issue of nuclear weapons, an issue which ‘cannot be removed from politics’ because it involves contestation between reasonable people (CW 436). It is, again, a mere act of assertion to state that someone can endorse M.A.D. and still be reasonable. (This very position of “neutrality” in fact puts Rawls on the side of the pro-nuclear lobby, for two reasons: firstly, because the case against nuclear weapons depends on arguments about the irrationality of the other side which are incompatible with the position that they are “reasonable”, i.e. with repressive tolerance, and secondly, because such an assignation to the sphere of political disagreement renders the only effective resistance to nuclear weapons, i.e. direct action, a form of intolerable imposition of a comprehensive doctrine in Rawls’s rhetoric. In practice, this leaves the task of “balancing” up to “democratic” politicians and their friends in the military-industrial complex, with predictable results. It is no coincidence that the “necessary” injustices of “legitimate law” always tend to strengthen the state and its closest capitalist allies). One could also note that Rawls’s own version of “sensible” policies include benefit cuts, raising the retirement age and limiting ‘extensive terminal medical care’ (CW 581) - in other words, that the “reasonable” seems to involve a capitalist logic of penalising the worst-off. (c.f. his view that egalitarianism is unreasonable - e.g. CW 262).

Such instances seem, however, to operate in ways which are secondary in the sense of assuming the reasonable to already exist prior to them; they are not definitions of the reasonable. Nevertheless, they imply it has a structured content in spite of its often tautological formulation.


One of the features of the “reasonable” is that it involves appeals only to public reason in public. Public reason is a particular field of reasoning, roughly coterminous with “common sense”, (or ‘fixed points’ in ‘our [sic] considered values’ - PL 381) which is supposed to be shared by all reasonable citizens (as distinct from reasoning specific to comprehensive doctrines and “controversial” theories, which is not). ‘The most reasonable political conception for us is the one that best fits all our considered convictions on reflection and organizes them into a coherent view. At any given time, we cannot do better than that’ (JAFAR 31). Rawls assumes that ‘average reasonable citizens’ all share the same ‘commonsense knowledge’ (JAFAR 121). Thus, public reason seems to be a systematised version of common sense. However, this ideology on which we supposedly cannot improve has more than intellectual significance. According to Rawls, ‘the coercive power of government’ should only be invoked through public reason (CW 603). Hence, ‘the exercise of coercive political power… is to be justifiable to all in terms of their free public reason’ (JAFAR 141). Such a basis for coercion is necessary if people are to respect each other (JAFAR 91). To be reasonable is to recognise that coercion can only be justified by public reason and not by a comprehensive doctrine (PL 226). There is a close relationship between public reason and reasonableness. In one passage, Rawls typifies a willingness to stick to public reason as ‘reasonableness and fair-mindedness’ (JAFAR 91), and in another he uses these terms to refer to adherence to common sense methods and procedures (CW 484). In another, he states that individuals, citizens, associations and governments can be said to ‘act reasonably and responsibly’ if they have a ‘recognized way of reasoning about what is to be done’, with different kinds of reason suited to different spheres (JAFAR 92). One has the virtue of reasonableness if one uses public reason, provided one is also fairminded and open to compromise (JAFAR 190).

Public reason is to be a more-or-less closed sphere which judges itself to be reasonable ‘as judged by public reason alone’. It provides ‘scope within which each doctrine can accept, even if reluctantly, the conclusions of public reason’, tending to manipulate doctrines in such a way that they tend to become reasonable instead of unreasonable. Whether in fact it is acceptable to doctrines is an issue ‘that transcends public reason’ and with which Rawls refuses to engage (PL 246). Also, ‘the forms of public reason are always several’ (CW 583). What matters is that ‘everyone can recognize’ the ways of reasoning involved in public reason (CW 326). (If “everyone” means “all reasonable persons”, all the usual problems arise: what “everyone” believes turns out merely to be what the observer decides defines them as “everyone”, etc.).

On one occasion, Rawls insists that public reason must include a series of his favourite liberties and rights and an idea of reciprocity, even if not that of justice as fairness (CW 581-2). Hence, ‘the content of public reason is given by the principles and values of the family of liberal political conceptions’ (CW 584). Similarly, he elsewhere specifies that he constructs his conception by assuming ‘the reasonableness of the tradition of moral philosophy’. This is, of course, a questionable assumption to say the least, yet Rawls’s only excuse is to plead necessity: ‘where else can we start?’ (PL 259). (One could reply a great many things, from analysis of situations through to discourse analysis).

One major function of the idea of public reason is to foreclose the public space to a range of conceptions unable to subsume themselves in common sense. For instance, Rawls argues against socialisation of the means of production on the grounds that the arguments for it are not general enough; they can only be convincing to ‘us and a few likeminded others’. ‘Philosophical debate alone is most unlikely to convince either side’, so the issue must be set aside in order to construct a shared public culture. The two principles of justice are to act as a ‘common court of appeal’ which is supposed to resolve by default what rational argument cannot achieve (PL 338-9). Of course, given Rawls’s statism, the outcome must be either that means of production are “socialised” or that they are not; therefore, his own approach no more provides a basis for unity than if he supported one side or the other. Presumably, he hopes that the desire for unity at any cost is sufficiently strong to compel submission on the part of whichever side loses out at the ‘shared’ court of appeal.

Public reason is only supposed to be binding in relation to a very limited range of decisions, reflecting a rigid division between public and private (see e.g. PL 215). (Strictly speaking, one is only required to use public reason in relation to debates about constitutional basics, and not in all public affairs - see PL 235). It seems, therefore, to construct a split subject who must reach two sets of decisions in two different sets of terms, and who, furthermore, Rawls insists must do so in good faith, without being hypocritical or insincere (see e.g. PL 242, where Rawls insists sincerity in public reason is possible). (What, then, should one do if one privately reflects that something is unjust yet one cannot articulate this in the public discourse?). Rawls assumes that associations with particular doctrines in fact have a crucial role in making public justification possible (PL 389-90). He also allows people to introduce comprehensive doctrines into public debate provided this strengthens the ideal of public reason (PL 247-8).

I shall not go into the problems with such a reliance on common sense in any detail here, as this is a broader subject in Rawls’s work worthy of a fuller treatment. I shall only mention a few issues especially relevant to the issue of the “reasonable”. Firstly, “public reason” does not have any characteristics which clearly specify why it should be viewed as anything but a partially comprehensive doctrine which happens to be very widespread and to occur in a number of different articulations. Secondly, the content of public reason is dependent on the overlapping consensus, and would therefore vary in the same ways as this consensus. Indeed, any particular overlapping consensus would have its own variety of “public reason” consisting of the grouping of concerns or discourses which happen to unite its participants (religious dogma in the case of medieval Catholicism, opposition to capitalist oppression in the case of anti-capitalism, the absence of state power in the case of panarchy, etc.). One should, therefore, write not of “public reason” as a singularity, but of public reasons relating to different overlapping consensi. (Rawls half-recognises this, since he asserts a different kind of public reason in the international sphere - LN 68). In this context, the specification of the liberal state as the content of public reason is a formally unaccounted-for gesture which is in effect a matter of arbitrary privileging. Thirdly, common sense is not as inclusive as Rawls assumes. It only provides a basis for terms of social cooperation others can accept if it really is a universal and unchallenged ideology. However, it performs no such role, so its use is just as “unreasonable” as relying on any other specific philosophy or conception of the world. Those who do not agree with “common sense” beliefs are excluded just as much as if a comprehensive doctrine was used. Rawls assumes reasoning is limited to common sense - e.g. ‘the limits set’ on reasoning (JAFAR 92) which is already a part of it. In fact, common sense may be imposed on some people as an alien and oppressive form of reasoning. For instance, Rawls’s theory would tend to render a Barthesian semiotician, who refused to use “common sense” ways of arguing on principled (and maybe true) grounds would be “unreasonable”, even though in effect taking the same stance as a Christian who refused to accept a comprehensive secular doctrine or an atheist who refused to accept a state religion.

Fourthly, public reason is irrational, because it puts widespread beliefs beyond question. It tends to fix and reify existing beliefs in such a way as to render them invulnerable to criticism, and it puts their mere popularity ahead of concerns about their accuracy, consistency and inclusiveness. (The issue of consistency can be very important. For instance, it may well be “common sense” to condemn the September 11th attacks, but not to condemn similar actions by western military forces. In this case, a demand for ethical consistency undermines any possibility of a reliance on “common sense”, especially since “common sense” shows itself to be insidiously ethnocentric in this way. Prioritising commonality over consistency tends to favour bigots). It turns the way things appear to the normal/commonsensical (from the outside) into an almost Godlike standpoint. Finally, public reason may be, not simply short of truth, but actually false. For instance, it may be a widespread “public” view that “prison works”, even when widespread evidence shows that it does not. Since the evidence is “controversial”, it would be inadmissible in public, and those who know and believe it would not even be allowed to argue on it (at least regarding basic constitutional questions), let alone to insist that the unnecessary suffering caused by this failing system be stopped.

One could construct an even more extreme, yet hypothetical, world where the majority share some delusion - for instance, they do not accept the existence of cats. Suppose that, in this world, only X would admit that there is a cat. X can see, hear and feel the cat, and so is sure it exists. Nevertheless, “public reason” would oblige X to act as if it did not exist - for instance, to step on its tail. If X refused to do this, and even more so if X tried to prevent others from stepping on its tail, X would be “unreasonable”. It is not clear, however, that such a person would be “unreasonable” in other senses; for instance, X may well be prepared to offer others terms of cooperation X believes they ca reasonably be expected to accept, and so on. The problem here is a problem of truth, not reasonableness. In other words, Rawls’s theory becomes Kafkaesque the moment a patently untrue belief enters public reason. (Is it “reasonable” to expect others to act on an untrue belief?) (Rawls’s discussions of truth are unduly prejudiced by his emphasis on the particular case of claims to moral or religious truth. He ignores instances where truth and falsehood can be determined in some way, and where the resultant controversy of a true, or probably true, view is a result of prejudice and dogma). Other examples would include: a society with a Ptolemaic conception of the world; a society with a “common sense” religious conception of nature (e.g. an Evangelical image that God would not allow the environment to collapse) amidst ecological crisis; etc.

Indeed, Rawls admits that his reliance on common sense means that public reason will ‘contain at least some error’, excusing this on the grounds of the supposed greater good of the ‘practical’ task public reason performs (CW 328-9). However, he evades the depth of the problem by mainly seeing this ‘error’ in terms of an incapacity to appeal to the whole truth in politics. He does not seem to see a difference between seeing public reason as ‘but a part of the truth’ and relying ‘on our [sic] present commonly based and shared beliefs’ (CW 329), even though he has no basis for assuming the latter to be even a part of the truth. The arguments he gives for public reason are generally arguments against the position that politics should be derived from the whole truth (e.g. PL 216-19), and do not consider the possibility that public reason is actually false. (The misleading collocation “commonsense knowledge” is one of the ways Rawls conceals the possibility that common sense might involve false ideas). Rawls explicitly states this in one passage. ‘Since the beliefs we attribute to the parties [in the original position] are those of common sense… and of science when not controversial, there is a decent chance that most of those beliefs are accepted for good reasons’, i.e. are not false consciousness in the Marxian sense (JAFAR 121-2). This is a very dubious assumption to make, and one which certainly does not answer Marx, let alone Gramsci or Lukács.

Put simply, common sense is a philosophy like any other (though less coherent than most). If the “reasonable” are defined in reference to common sense, the result is that the reasonable are a doctrinally-constructed group; to refer to “the reasonable” is symmetrical with referring to “Christians” or “Kantians” or “Muslims” or “utilitarians” in terms of its ideational structural location. It is simply that one particular ideology/philosophy/worldview is given an ontologically privileged status. It is not a “political conception” distinct from “philosophies”, but one particular (partially) comprehensive doctrine.

(The analogy of exceptions to admissibility of facts in court is a bad one, since this involves protection against state abuses - for instance, the banning of evidence gained under duress, and of facts likely to prejudice juries - whereas public reason involves the legitimation of state power; also, the limits in public reason are far more sweeping. The “public” discourse would operate as a generalised block against “publicly” criticising the state or the status quo on a whole range of grounds deemed inadmissible. A closer analogy to admissibility in court would be a specific taboo on, say, racially-motivated voting).

The reason I do not take seriously the idea that the “reasonable” is constructed by public reason is that the construction of public reason itself involves a reference-back to the reasonable. In other words, the reasonable do not rely on public reason simply because they happen to believe in it. They rely on it because it provides the best possible focus for a public political conception. While it may well be true according to Rawls that all reasonable persons use public reason, this is because all reasonable persons are pursuing such an alleged consensus. Public reason may well be part of the operational classification of the reasonable, but it must be assumed to be an outcome rather than a premise of reasonableness. For instance, public reason shows reasonableness because it expresses an ideal of the citizen and a willingness to settle matters in a way others will find reasonable (JAFAR 92), and the role of public reason is to ensure that public inquiry is reasonable (among other things) (JAFAR 91). In other words, the reasonable must be defined prior to public reason, because it is used as a basis for adopting public reason.

I would also add that Rawls does not really answer the question he raises regarding what one should do if public reason cannot decide a question (PL 240-1). He simply insists that it is a great value regardless of whether it sometimes fails or not. There is also a crucial problem regarding Rawls’s assumption that the aim of justification to others requires reference to shared or common beliefs (PL 100-1, CW 394), a claim which is simply untrue (since one appeals to others based on what they believe or desire, not necessarily on the basis of what one believes or desires oneself). This assumption is of no small importance, since it carries the idea of a common essence or an already-existing unitary discourse. (This issue also links to Rawls’s concept of objectivity - see PL 116). It is not a belief Rawls holds consistently, since in “The Idea of Public Reason Revisited”, he advocates argument from ‘conjecture’ (i.e. from immanent critique and interpretation of views one does not share) (CW 591). The “need” for a “shared” conception has origins which are not fully articulated, and may involve the pursuit of a more substantive unity than Rawls admits to pursuing (i.e. he wants a “shared” master-signifier constructing an image of society as a molar totality, something which is not strictly necessary for goals such as resolution of resource-use conflicts).


I would summarise my analysis above as follows:

* The “reasonable” does not involve a substantively demonstrable distinction between different kinds of political doctrines. Terms such as “political conception”, “public reason” and “overlapping consensus” are an ideological expression of the fact that Rawls is actually imposing a single (partial) ideology in the same way as if he were imposing a “comprehensive doctrine” (especially a “partially comprehensive doctrine”, perhaps even a “religion” of statolatry). The beliefs he imposes are not comprehensive because they refer only to particular issues (as do, for instance, most religions), but they do not involve any qualitative characteristic which sets them aside from other kinds of belief-systems. The reasonable are an ideological/discursive in-group.

* The “reasonable” is not specified by substantively greater levels of tolerance or preparedness for dialogue than could arise in an “unreasonable” doctrine. “Reasonable” persons are not required to be other-regarding in any concrete sense, or to enter into dialogue, or to compromise with or even take into account actually-existing others. They can classify themselves as “reasonable” on the basis of thought-operations internal to themselves, and they can pursue the resultant beliefs more-or-less intransigently. (Effects of apparent tolerance arise from contingent tolerant features of the doctrine the “reasonable” happen to affirm, especially its use of repressive tolerance as a way to curb conflict. Also, the liberal consensus may also allow an effective modus vivendi which in practice tolerates groups which are strictly speaking “unreasonable”).

* The “reasonable” cannot be reduced to anything which is “generally” affirmed. Rather, it consists of a set of beliefs which are “generally” affirmed by an in-group which defines itself as the only relevant group in assessing whether the views are affirmed.

* The “reasonable” is not a characteristic of persons or doctrines, because it cannot be deduced prior to the beliefs they hold (i.e. one cannot say that a political conception has gained the support of reasonable comprehensive doctrines, as if these exist and can be specified independently of it). Rather, it is a label used declaratively to designate an insular group, and this group is constructed alongside or after the construction of the political conception it affirms. (It seems to have to include and exclude specific named groups, but this is because it involves a commitment to “common sense”, which would specify these groups as inside or outside liberalism/reasonableness)

Rather, the “reasonable” can be specified as having the following operative characteristics:

* A desire for “order” and systematisation

* Preparedness to submit (especially on a “behavioural” level) to liberal states, including their laws and constitutional procedures; and preparedness to accept the necessity of anything which is necessary for the existence and operation of liberal states and related social systems (i.e. capitalism or something similar)

* Belonging to an insular group specified by a dogmatic commitment to certain specific aspects of liberal ideology (specified as a “partial” group of liberal ideas, not the content of any particular liberal theory)

* Acceptance of aspects of “common sense” conducive to liberalism (so one can endorse and engage in “public reason”)

* A fantasmatic identity as reasonable in the sense of offering others terms they can be expected to endorse (and maybe also including other aspects of the Rawlsian ideal of the person), i.e. which identifies the reasonable in some sense with everyone

“Being reasonable” is therefore a falsely universalised way of saying “being a liberal”. Reasonableness involves a commitment to core aspects of liberal ideology, and in this sense, it is a label for an adherent of this ideology (similar to “Christian”, “Marxist” or “conservative”). Rawls cannot therefore demonstrate why one should be a liberal, and he assumes that liberals alone have a right to form states and use force to establish the primacy of a particular type of person and a particular style of argument. He can only conceal the issue of domination rhetorically. The “reasonable” is a hegemonic device in the Laclauian sense: it does not make a case for liberalism, instead simply placing it in the place of the universal. (It also contains a gesture of constructing a primary, disavowed other necessary for its existence, in the form of the intolerant right. This force is necessary to conceal the operation of liberalism as a limiting and dominatory system against active and radical projects). The decision to privilege this particular ideology/discourse over others - to declare a qualitative difference between “oppressive use of state power” and “legitimate law”, between imposing a “political conception” and imposing a “comprehensive doctrine”, between “reason” and “fanaticism”, between intolerance of pluralism beyond the reasonable and intolerance of pluralism within the reasonable, etc. - is a dogmatic and arbitrary gesture for which Rawls provides no argument. The “reasonable” is the signifier of a dominating in-group. Rawls’s argument ultimately comes down to the assertion that Rawls (or “political liberalism”, or “justice as fairness”) believes this and not that, and a hope that the reader will collude in this dogmatism, finding it subjectively appealing. The “rationality” of his arguments is in the last instance as much about show as substance.


The space equivalent to the reasonable in domestic politics is covered in I.R. by the concepts of the reasonable and the decent. As in domestic politics, these involve a consensus established as valid because of its acceptance by an exclusionary insular group which authorises itself to make universal claims. Indeed, Rawls’s advocacy of such a position is stronger in The Law of Peoples than anywhere else. This is relevant to his more general concept of the reasonable because there appears to be a relationship of equivalence (at least on a formal or structural level) between his derivation of domestic principles of justice and his derivation of the law of peoples. As a background concern, one should note that Rawls denies any value to individuals as such in constructing the Law of Peoples at all. Only “peoples” have claims at this level (which could lead to a block on solidarity with oppressed minorities). Thus, he draws a distinction between internal and external reasonableness. An internally unreasonable society may nevertheless be treated as reasonable in I.R. (LN 83-4). The fit between political liberalism and a reasonable Law of Peoples is supposed to confirm the reasonableness of political liberalism, even though the reasonableness of political liberalism seems to be a precondition for the derivation of the Law of Peoples and its reasonableness (LN 23), another instance of circularity.

The main difference between the two is an extension of the space of inclusion to include so-called “decent” societies, particularly “decent consultation hierarchies” (e.g. religious states with some internal pluralism, civil rights and representative procedures), which are not themselves liberal (and whose advocates would presumably be deemed “unreasonable” within a liberal society, potentially leading to contradictions: if Rawls’s “domestic” argument holds, decent non-liberal societies are guilty of the oppressive use of state power, and it hardly seems justified to form a stable overlapping consensus with those one terms oppressors). These societies are specified in a way which suggests that they are necessarily heavily statist, and in fact borderline liberal in all but a few areas. For instance, they include “rule of law” and ‘a system of law… such as to impose bona fide moral duties and obligations… on all persons within the people’s territory’, including procedural limits non action. Their society must be be a ‘scheme of cooperation’ in Rawls’s special sense (i.e. a logic of place), involve systematised notions of right and wrong and endorse a principle of reciprocity, at least in I.R. (Included “peoples” seem to have to be organised as, or rather dominated by, states). Elites are supposed to avoid being ‘corrupted’ by wealth and power, and must ‘reply to’ conscientious criticisms. Leaders may insist on a final say, but must reply to objections with reference to a theory of why the government can ‘impose duties and obligations’. (Notice that this reply is a form of repressive tolerance; it requires only formal acknowledgement, not any substantive response to grievances and certainly not any empowerment of ordinary people against leaders). They must also endorse free trade and have an idea of the common good, and see people as decent, rational and capable of moral learning. (In other words, they must have all the core structural principles of state-centred moral psychology Rawls constructs in his own theory - for instance, a slave morality, submissiveness towards “rules”, petty-mindedness in interpersonal relations and naturalisation of capitalism). Rawls specifically avoids asserting that such societies (which look very like liberal societies, but without liberal ideology, i.e. they are liberal in form but not content) actually exist, merely saying that, if they do, they should be included in the Society of Peoples. The argument for their inclusion mainly rests on a (falsely-exaggerated) binary held up to the reader’s intuition: either one includes such societies in a Law of Peoples or one declares permanent war on them. It also includes a claim that the inclusion of such societies would aid their transition to liberalism whereas exclusion would impede it (presumably in distinction from other types of society such as outlaw states, which presumably would not be similarly affected; also NB the contradiction with his earlier claim that the Law of Peoples does not exist so as to make all peoples liberal), and a quasi-circular justification of their inclusion (i.e. they should be included because they happen to support the same Law of Peoples as liberal societies, a law which is constructed precisely on the assumption that their support is to be encouraged). (LN 59-77). (One also finds the repetition of the theme of pluralism as a basis for extending inclusion - LN 10 - another instance of the recognition of pluralism only when it is convenient: the “fact” of pluralism also includes the existence of those societies Rawls does not include, so there is no good reason why those he includes are relevant and these others are not, especially since “decent” doctrines and their advocates are not covered by the “fact of pluralism” when it comes to their position within liberal societies). As in domestic politics, so in I.R., there is a gap between the fully reasonable and the fully unreasonable (LN 74).

Rawls does not specify the range of the concept of the “decent” exactly - he only discusses the case of a “decent consultation hierarchy”, and then only to show that a society can be decent without being liberal - but it seems likely that a people who have no state and who on the whole reject state forms, especially if they are also predominantly anti-capitalist (e.g. the populations of Bougainville and West Papua and the liberated areas of the Chiapas, should the oppressor governments be fully repulsed from these territories), are insufficiently close to the liberal model, even though their way of life is in large part harmless to others and they are not at all equivalent to a violently expansionist religious state. One sees, therefore, the repetition via the Law of Peoples of what Rawls accomplishes in the case of anarchism in domestic politics: by the construction of a misleading binary, a particular system of domination is constructed as absolute, and those who oppose it are labelled as intolerant and intolerable (in this case, presumably, as outlaw states). One could easily imagine a U.S. government using a Rawlsian rhetoric to justify an invasion of the Chiapas or of Bougainville, because these areas lack a “rule of law” and “free trade”, and must therefore be in the grip of tyrannical and dangerous despotisms.

Rawls explicitly states that, because (for instance) human rights are part of the law of peoples (defined by its being acceptable to liberal and decent peoples), they are ‘universal’ and have a moral force ‘binding on all peoples and societies including outlaw states’ (LN 79-80). Furthermore, these groups are given an asymmetrical right to appoint themselves as the judges of universality, simply because of characteristics specific to their particularity. Because human rights are shared by all liberal and decent societies, they can be demanded of all societies without exception (LN 111 - notice also that this is circular, since liberal and decent societies are partly defined by their endorsement of human rights). Because it is accepted by all liberal and decent peoples, the Law of Peoples should become the universal medium of international political debate (LN 85). ‘As we have worked out the Law of Peoples for liberal and decent peoples, these peoples simply do not tolerate outlaw states. The refusal to tolerate those states is a consequence of liberalism and decency’. If the arguments for political liberalism and the Law of Peoples are ‘sound, then liberal and decent peoples have the right, under the Law of Peoples, not to tolerate outlaw states’ (LN 81; he then adds arguments that they have this right because outlaw states are violent, affect the international system, and so on). One should notice in particular two crucial points. Firstly, the “contract” model of social relations constructs a residual position of rightslessness, which places the excluded in a moral vacuum. Secondly, this vacuum is used as a pretext for relations of domination (which, one should recall, are not instances of solidarity with the victims of, say, human rights abuses or wars of aggression, but are occasioned by the breaking of a system of rules). Such relations arise because a system set up by liberal and decent societies for liberal and decent societies and in consequence of their own liberalism and “decency” is taken as legitimate even when used against those by and for whom it was not created. (One should add, as with domestic examples, that the boundary of this consensus varies depending on exactly which states are included and excluded; in a different circumstance, states now termed “outlaw” could construct the same exclusionary insularity so as to declare liberal states to be “outlaw” and “aggressive” - as indeed they are, if they act as Rawls advocates).

Rawls assumes the universality of the Law of Peoples is such that outlaw states are not indifferent to it and can be won over to it, although the means of winning them over seem to amount to economic and military coercion (LN 93). Furthermore, all rights are dependent on inclusion (as is typical of Rawls’s theory). For instance, a society only has a right to self-defence if it has something worth defending. Whether they have such a “something” can apparently be decided externally (more specifically, as human rights plus a non-aggressive foreign policy) (LN 92). A liberal or decent people always has this right, even if it has dirty hands (LN 94). However, such societies also seem to be given the right to dominate others, on the grounds that these others have nothing worth defending (while, a cynic could add, they have plenty worth “attacking” - for example, oil supplies to grab, a cheap workforce to exploit, or historic treasures to loot and then smuggle abroad). In any case, Rawls thinks there is no basis for peace with expansionist societies, except as a modus vivendi (LN 123).

The gesture of establishing the asymmetrical primacy of liberal and decent “peoples” is explicitly ethnocentric. However, Rawls wants to deny such ethnocentrism, providing an extended rebuttal of the accusation (which, in great part, simply amounts to a claim that the particular views he advocates, such as “reciprocity”, are in fact universal). ‘[L]iberal societies ask how they are to conduct themselves toward other societies from the point of view of their own political conceptions. We must always start from where we now are, assuming that we have taken all reasonable precautions to review the grounds of our political conception and to guard against bias and error. To the objection that to proceed thus is ethnocentric or merely western, the reply is: no, not necessarily. Whether it is so turns on the content of the Law of Peoples that liberal societies embrace. The objectivity of that law surely depends not on its time, place, or culture of origin, but on whether it satisfies the criterion of reciprocity and belongs to the public reason of the Society of liberal and decent Peoples’. The Law of Peoples is universal because accepted by liberal plus decent societies, and because it offers all societies reasonable conditions and ‘a relation of fair equality’. ‘They [i.e. opponents] cannot argue that being in a relation of equality with other peoples is a western idea! In what other relation can a people and its regime reasonably expect to stand?’ (LN 121-2). (This may also be rooted in ideas that all humans share a common human condition understood by liberal theory. For instance, Rawls claims that the choice ‘of what to do with life is always there’ regardless of culture - TJ 362. In another passage, he refers to reasonable and unreasonable differences between people - LN 54-5 - suggesting that the reasonable is supposed to be a universal limit on difference). Rawls does not realise that there might be anything questionable in the assumption of the universality of ideas of reciprocity and public reason or of his particular versions of equality, fairness and reasonableness, or in the insularity of the liberal/decent coterie (which, anyway, is constructed by liberals themselves on terms of what they find “decent” in other societies, i.e. liberal societies seem first of all to construct a Law of Peoples and then to decide who else is “decent” enough to be admitted). It is, of course, the western observer who dictates what relations others can “reasonably” endorse, and whether their actual conduct has rendered them “unreasonable”. In addition, even supposing that all peoples sought something under the label “equality”, this does not mean that they would be seeking the relations specified by Rawls’s technical use of this term. One would have to take into account implicit ethnocentrism in the idea of “reciprocity” and the model of the person it implies, in the idea of “responsibility for ends”, in the construction of individuals as carriers of interests, in the image of the self as primarily a vulnerable being, in the idea of ethics as external law, and so on.

One should notice several further features of this argument. Firstly, Rawls’s depiction of the international system is formal and legalistic. In practice, this is misleading, and so-called “outlaw” states do not arise in a void. If Rawls’s presentation involves the assumption that a society’s political structure arises internally, and if this is a logical deduction from his concept of “peoples”, this concept can rightfully be termed a Barthesian myth. In practice, “outlaw states” arise either as a direct result of meddling by imperialist (often liberal) powers (e.g. US support for Saddam Hussein and for the Afghan mujahideen), or as a variety of resistance to oppressive imperial power operations (e.g. the various Stalinist regimes in countries peripheral to the global system). To view them outside this context, and place them in a moralising discourse focused on external structures, is highly misleading and mythical. Furthermore, it is dangerous, since it can strengthen the discursive position of precisely those core imperialist powers which are the source of many of the problems in the world system.

Secondly, Rawls does not in any way connect problems of state violence to the existence of states. His moralising binary between reasonable and unreasonable states allows him to dismiss the violent and oppressive practices of “outlaw” states as the product of an excluded other. This allows him to conceal the linear connections between the two types of state, via the structures and practices of state microphysics and infrapolitics. Rawls’s idealist misreading of state power leads to an endorsement of states which is to a great extent constructed through the operation of myths.


Post a Comment

<< Home