3 DEFENDING THE PRINCIPLE OF STATE NEUTRALITY

↓52

When we turn to the question of why one might believe that the state ought to be neutral between comprehensive conceptions of the good life, we run into an apparent paradox. As noted in chapter one, state neutrality has come to seem necessary to many as a result of the increasingly fragmented nature of liberal democratic societies. Looked at historically, this necessity appears as an extension of the principle of religious toleration which was hammered out in early modern Europe.122 Looked at systematically, the necessity of neutrality appears to follow from a concern for the stability of divided societies, as well as from widely held commitments to values such as individual autonomy.

↓53

But whatever drives some contemporary thinkers towards the principle of state neutrality, it is of course not the case that all modern citizens wish for a neutral state. Many, if not all, take the conceptions of the good to which they adhere to be superior to their alternatives,123 and, unsurprisingly, these conceptions of the good are often thought to have political implications.

This is where the neutrality principle might appear paradoxical. Because if a neutral state appears to rely on one of these conceptions in defending its neutrality, it may be suspected, at least by those whose conceptions do not serve as the basis for legislation, of precisely the kind of partisanship it is intended to circumvent. In such cases n eutralism 124  might be thought of as a fighting creed, just as Catholicism, Islam, or various less clearly articulated visions of the good life which citizens hold are perceived – as a player; but in this case one who also claims the right to referee.

This appearance of paradox arises from a feature of democratic political theory which is less characteristic of theorising in other contexts: if one is at all committed to government by consent, as liberal neutralists, in virtue of being liberals, are, one wishes one’s political theorising to be accept ed by the general public. The mere existence of dissenters raises questions about the validity of one’s position. While it is no doubt the case that metaphysicians would be grateful for widespread acceptance of their views, they do not expect this, nor is their endeavour called into question by their tendency to address themselves primarily to specialists. The fact that not everybody accepts, say, the Kantian conception of the person, can be explained, without any appearance of paradox, by arguing that non-Kantians provide mistaken answers to metaphysical questions.

↓54

Error theories do not work quite so simply for democratic political theorists. If I claim to hold a position which is neutral, the fact that my position is not accepted by a sector of the public calls my neutrality into question. This, therefore, is the problem any advocate of the principle of state neutrality must deal with before going on to suggest why, whatever anyone currently thinks, there are good reasons for accepting the principle.

Concern about the appearance of partisanship has weighed sufficiently heavily on a number of philosophers for them to attempt to defend the neutrality principle in ways which at least do not presuppose any controversial comprehensive conception of the good life. We will see, in this chapter, what the prospects for such a justification are.

In the first part of the chapter I examine the logic of the attempts of a number of neutralists to formulate a defence of the principle of state neutrality which is itself neutral – in other words a defence which will take seriously the alleged need to achieve public consensus on matters of political theory, as opposed merely to consensus on the content of legislation. In doing so I will ask whether the “ecumenical” approach125 to the project of finding such a neutral justification for the principle of state neutrality, most obvious in the work of Ackerman, or the “deductive” approach126 to the task, as demonstrated in Larmore’s book Patterns of moral complexity as well as the writings of Rawls (who makes free use of both methods), holds out the most promise, and what the prospects for finding a neutral justification – a justification, in other words, that takes no position on the truth or falsehood of any comprehensive conception of the good life adhered to by citizens – of any kind for the neutrality principle are. I conclude, first, that consistency does not require that the neutrality principle be defended neutrally, and second, that the principle cannot in any case be so defended.

↓55

In the second part of the chapter I look further at what I call partisan defences of the neutrality principle; defences which express indifference to the question whether they are compatible with all (or a significant number of) the comprehensive conceptions of the good life currently adhered to within liberal democratic societies. These defences include all arguments which set out to offer convincing reasons for the principle of state neutrality regardless of whether their premises are actually held by a large proportion of citizens or not. In doing this I examine a number of common arguments for the neutrality principle, and conclude that none of them are conclusive.

3.1 A neutral justification for the principle of state neutrality?

We start, then, by asking what the logic of the demand for a neutral justification of the principle of state neutrality is. I will, to avoid unnecessary repetition, refer to this demand from here on as the principle of philosophical neutrality.

I understand the principle of philosophical neutrality to be the requirement that the principle of state neutrality be defended by arguments which take no position on the truth or falsity of any comprehensive conception of the good life. Its appeal arises from the apparent inconsistency of arguing for neutrality between conceptions of the good on a basis which assumes the truth (or at least the superiority) of precisely such a conception (or conceptions).

↓56

Whether or not this inconsistency is genuine (I will argue that it is not), its mere appearance leaves the principle of state neutrality vulnerable in contexts where obtaining consent for, or attracting commitment to, the principle is crucial, given that arguments for neutrality must avoid the appearance of partisanship in such contexts. Ackerman, Larmore, and Rawls all recognise, in different ways (which I will outline), this as a problem.

This discomfort is sometimes expressed in moral terms, sometimes in pragmatic terms. Rawls, for example, suggests both that the liberal requirement of public justifiability would be violated by a conception of justice which was not defended in terms that could be accepted by all citizens, and expresses the concern that defending his vision of justice – after this referred to as “justice as fairness” – in, say, Kantian terms, as he himself does in A theory of justice, runs the risk of alienating reasonable non-Kantians, and thereby threatening the stability of a well-ordered society of justice as fairness. Larmore calls the defence of the principle of state neutrality by appeals to ideals of the person that are themselves controversial ‘one of the damaging paradoxes of…liberal theory,’ and goes on to say that ‘[b]ecause liberalism is fundamentally a response to the variety and controversiality of ideals of the good life, it needs a justification of political neutrality that is itself appropriately neutral.’127

One might make the point against philosophical neutralism however, that, given that total moral neutrality would mean not standing for any norms whatsoever (assuming this were possible), a case must, and can, be made for seeing principles operating at different levels of abstraction, with neutrality in the face of disagreement appropriate at some levels (typically less abstract levels, such as the making of laws) and not at others (typically more abstract levels, such as defending jurisprudential theories or theories of political morality). We do not, after all, continue demanding ever more abstract levels of neutrality; there is no writer I know of who insists that the defence of the principle of philosophical neutrality must be neutral!

↓57

However, accepting that there is no hope of achieving total moral neutrality is compatible with arguing that it ought to be pursued at a higher level of abstraction than that of lawmaking: political theorising, for example, hence the principle of philosophical neutrality. Defenders of the principle must establish that there is as good a reason (or reasons) for responding to disagreement about which values it is appropriate to advance by means of the philosophy of law with the principle of philosophical neutrality as there is (or are) for responding to disagreement about which values it is appropriate to advance by means of the law with the principle of state neutrality. But first of all we will examine prospects for satisfying the principle of philosophical neutrality. If we find that this cannot be done, there is no need to raise this issue.

3.1.1 Ackerman’s ecumenical strategy

Ackerman, in his Social justice in the liberal state,128 offers one of the earliest attempts to articulate and defend the neutrality principle in a neutral manner. In doing so he makes a surprising move: he claims that he does not need to defend or reject any particular argument for the principle. This strategy I term, following Wall and Klosko, an “ecumenical” strategy.129

He can make this move because, he argues, there is a wide range of valid “argumentative paths,” any of which can lead one to conclude that the state must not favour particular comprehensive conceptions of the good life. Amongst the starting points of such paths he lists ‘realism about the corrosiveness of power; recognition of doubt as a necessary step to moral knowledge; respect for the autonomy of persons; and scepticism concerning the reality of transcendent meaning.’130

↓58

The reason Ackerman gives for wishing to remain neutral between these different arguments is the desire to avoid any claim of privileged access to the ultimate truth, a claim which, as he sees it, would result in a partisan defence of neutrality. Such a defence runs the risk of ‘[making] liberalism a hostage of a particular metaphysical system,’131 thereby contradicting ‘the essence of liberalism,’ which is ‘to deny people the right to declare that their particular metaphysic and epistemology contains the truth, the whole truth, and nothing but the truth.’132

He elaborates his strategy by noting that his

↓59

book aims, in principle, to be intelligible to all potential citizens of a liberal state. Given the very different people in this audience, however, any particular argument can be expected to persuade some only at the cost of turning others off. Thus, if I simply tried to tell you why I personally am a committed liberal, this would predictably divert some of you from arguments that you would find more compelling. Given this dilemma, it seems best to dispense with personal declarations of faith so as to better give you a sense of the different paths that can lead a thoughtful person to Neutrality.133

The appearance of partisanship is to be avoided by ensuring that no attempt is made to persuade anyone to accept the principle of state neutrality on the basis of premises which they do not share.

↓60

This is not to say that Ackerman says nothing in favour of the neutrality principle. He presents it to the reader as capturing the essence of the liberal vision of political authority, which, as he sees it, takes any ‘power structure [to be] illegitimate if it can be justified only through a conversation in which some person…must assert that he is…the privileged moral authority.’134 Liberal authority is justified rather by the free conviction of its necessity, produced through a dialogue of equals, on the part of its putative subjects. But this is not to argue for the principle of state neutrality; it is merely to assert its centrality to the liberal tradition, which, it is assumed (one expects), is the tradition of those addressed.

What exactly does this destination, to which all roads lead, look like? As with Rawls (and indeed with Kant135), it can be modelled by means of a thought-experiment, which explains how the exercise of power by one citizen (or a group of citizens) over another (or others) can be (or can fail to be) justified. But whereas for Rawls legitimate law (at least as pertains to the basic structure of society) must pass through an elaborate series of hoops involving the conditions of choice in the original position as well as three further stages, Ackerman’s demands are rather simpler. If a law136 cannot be justified without appeal to the unchallengeable moral insight (which we might call the clergyman’s fallacy, in deference to the historical origins of liberalism), or the inherently superior status of the one who exercises power in its name (what we might call the nobleman’s fallacy), then it is illegitimate. Only those laws which are justifiable without appeal to any comprehensive conceptions of the good life – neutral laws, if we understand neutrality to consist in neutrality of justification – will pass the test.

We can also detect an echo of the social contract tradition in Ackerman’s approach, despite his explicit rejection of it.137 It lies in the importance the notions of equality and consent have in both Ackerman’s minimal “moral proof procedure”138 and in the standard contractarian scenarios, including the set of conditions articulated by Rawls’s original position and the stages which follow it. Both Ackerman and (amongst others) Rawls, think that legitimate principles can only issue from a procedure which situates its participants initially as equals. Ackerman’s conversationalists are barred from appealing to inherent superiority of any kind in attempting to defend the particular powers they exercise, and Rawlsian parties in the original position are not distinguished by relative power or influence (or indeed in any other way).

↓61

Furthermore, both Ackerman’s rational conversation procedure and Rawls’s original position model the notion of consent, central to liberal political theory. If, as a result of a conversation conducted between equals, I am convinced by the reasons you offer for your particular powers, as opposed to being intimidated by your status or your threats, I can be said to have consented to them. Likewise, Rawls’s original position constitutes a hypothetical agreement between the parties behind the veil of ignorance.

What are we to make of Ackerman’s refusal to take sides on the question of which the best arguments for neutrality are?

The obvious suspicion is that, as soon as an ecumenical neutralist such as Ackerman encounters a sufficiently stubborn perfectionist – one who denies that there are any “argumentative paths” which establish the principle of state neutrality – he is going to have to dirty his hands and show why there is at least one. Ackerman might respond to this challenge, however, with the concession that it may prove necessary to follow a particular line of argument, but that doing so need not involve any commitment to the premises held by the perfectionist. It merely involves showing them how they have not yet noticed that their premises, whatever they are, commit them to the neutrality principle. The ecumenical neutralist might, for example, be able to show that if you think that virtue cannot be coerced into people, or that the autonomous life is the good life, then you will, if you follow the argument sufficiently rigorously, conclude that the state may not favour any particular comprehensive conception of the good life. Showing this, however, does not commit the ecumenical neutralist to the claim that virtue cannot be coerced into people or that the autonomous life is the good life. It commits them to logical consistency, but this can hardly be regarded as damagingly partisan.

↓62

We might ask, however, whether this confidence in the prospects of taking any given set of premises held by citizens, and finding an argumentative path from there to state neutrality, is warranted.139 Perhaps there are many widely-held premises of political morality which do not entail the neutrality principle. In saying this I am not merely raising the (doubtful) possibility that modern constitutional democracies conceal (large numbers of) people whose fundamental political commitments are utterly hostile to state neutrality. It is quite likely, rather, that many citizens of democracies believe that the state ought to be neutral with respect to most controversial questions of the good. This may not preclude, however, the possibility that many find themselves endorsing legislation in defence of the good on occasions when they think that values of great importance are at stake, that the defence of these values requires the exercise of state power, and that this defence is being impeded by those badly lacking in moral insight. Granting those lacking in moral insight a veto over legislation, especially in cases where they form a minority, might be thought to be unwarranted, possibly even disastrous. Hence, some citizens might not implausibly conclude that, on those occasions where, say, the state clearly does possess superior moral insight, and where the stakes are sufficiently high, it must exercise the power it has on behalf of the good.

What is the basis of the ecumenicist’s confidence that citizens whose commitment to the principle of state neutrality is limited in the way described, or perhaps lacking altogether, thereby reveal a failure to appreciate the implications of their own moral commitments? It would seem to be a conviction that the premises of political morality commonly held by citizens of contemporary liberal democracies all do, when correctly understood, rule out absolutely any state action on behalf of the good.140

But it is doubtful that this is so. I demonstrate in the second half of this chapter that arguments which indisputably rule out perfectionist legislation are very difficult to find, even amongst major neutralist writers. Clearly, if I turn out to be right about this, no amount of pointing to the argumentative paths will help Ackerman – at least not if he’s concerned with convincing his readers. But, for the moment, we merely need to ask how he might respond to this possibility. We need merely to ask, in other words, what he would have to do to rebut the claim that the principle of state neutrality cannot be vindicated.

↓63

Ackerman’s strategy must be to ask which premises of political morality the perfectionist accepts, and then attempt to show how state neutrality follows. The perfectionist may respond in two ways. She may say that, while she accepts premises a, b, and c, no unequivocal case for the neutrality principle can be made on the basis of these premises. Or she will accept that a case for the neutrality principle can be made on the basis of premises d, e, and f, but reject these premises. Ackerman’s  ecumenical strategy permits him to deal straightforwardly with her first response; he must redouble his efforts to persuade her that a case for the neutrality principle can indeed be made on the basis of premises a, b, and c. But it is imperative, if he wishes to remain ecumenical, that he succeed in these efforts. For if he fails he is forced, if he does not want to give up on the enterprise of making a case for the neutrality principle altogether, to attempt to persuade the perfectionist that she ought to accept one of premises d, e, and f.141

And arguing for a particular premise – arguing, for example, that the autonomous life is the good life, or some other such premise which might serve as the basis of an argument for the principle of state neutrality – is precisely what the ecumenical neutralist wishes to avoid. Ackerman is confident that he will not be cornered into doing this because he assumes widespread public acceptance of premises from which valid arguments for the neutrality principle can indeed be constructed. In this assumption he is, as I will show, optimistic.142

In summary, Ackerman’s confidence that he can remain neutral on the question of which arguments best establish the principle of state neutrality rests on the hope that sufficient widely accepted argumentative paths lead to that principle anyway. But the fact143 that many widely held views on political morality, views which are taken by many philosophers, including Ackerman, to be capable of serving as premises in valid argumentative paths to the principle of state neutrality, do not obviously establish that principle is a problem for him, because it obliges him either to criticise some of these argumentative paths or to defend other, valid, argumentative paths (or both) if he wishes to establish neutrality – at least if he wishes it to be established by argument. And as soon as he does this he is no longer neutral on the question of which particular argument best establishes the principle of state neutrality.

↓64

Waldron has adduced further reasons for rejecting ecumenical approaches of the kind Ackerman articulates.144 He thinks that we should reject the view that the neutrality principle should be given a neutral justification because he thinks that, strictly speaking, there is no such thing as the neutrality principle.

As he sees it “neutrality” is too heterogeneous a concept to serve as the focus of any one principle, and the importance of neutrality as such is in any case not what underlies commitment to the principles which are identified in the literature as variants of the principle of state neutrality. Neutrality’s attraction lies in its apparent capacity to assuage certain more fundamental concerns: the value of personal autonomy, the importance of experimentation for discovering which lifestyles are in fact the most valuable, the absence of an incontrovertible case for any particular conception of the good, and so on. It is not widely thought to be attractive per se, regardless of what the case for it might be.

Waldron’s initial premise is that one cannot articulate a principle without justifying it, or, to put it more colloquially, his attack on Ackerman may be paraphrased as “if you’re going to tell us what it is you’re defending, you’re going to have to tell us what your case for it is.” One does not need to accept the truth of this premise in all circumstances to see the force of Waldron’s point with regard to the neutrality principle. This is because, as we saw in chapter one, the principle can be interpreted in a great variety of ways, and because, furthermore, the variant formulations are, at least in some cases, closely tied to particular arguments for the principle. We saw, to mention merely one of the axes along which variations of the principle of state neutrality are possible, that one might understand the neutrality principle as requiring neutrality of aim, effect, or justification, and it is fair to say that which of these one takes up as a legislative body will have a significant impact on what laws are passed (the difference between neutrality of effect and the other two being the greatest). And yet it is also clear that not all of these formulations of the neutrality principle are equally served by the common arguments for the principle

↓65

Believing, for example, that ethical pluralism is important is likely to incline one to advocate neutrality as neutrality of effect (if it inclines one towards neutrality at all), as requiring legislators to formulate laws only on the basis of neutral aims is unlikely to guarantee any particular (allegedly) appropriate spread of ethical options in a society. Similarly, as Waldron points out, scepticism concerning the reality of transcendent meaning (to use Ackerman’s phrase) is likely to incline one to advocate the neutrality of aim or neutrality of justification interpretation. This is because the objection to perfectionism on the part of the moral sceptic is presumably that it is irrational to prefer one set of moral reasons over another as a basis for legislation.

And so because the arguments for state neutrality which appear in the literature are very varied indeed, as are the formulations of the neutrality principle, and because in many cases these principles have quite different policy implications, Waldron is able to draw the conclusion that there are particular neutrality principles, some of which are closely related to each other and others which are not, and all of which are justified by particular arguments.

One might summarise his position as follows: If one remains agnostic on which the best arguments for neutrality are, one, in effect, remains agnostic on what neutrality is. The neutralist must do some arguing: either she picks which version of the neutrality principle she finds the most plausible, and then defends it against arguments for the others, or she assesses the strengths of the various arguments and then defends the version of the neutrality principle which is implied by the best of the arguments.  Simply surveying the field, as Ackerman does, and suggesting that one does not need to defend any particular one of the arguments, leaves it unclear what principle is actually being recommended.

3.1.2 Rawls’s ecumenical strategy

↓66

Rawls, in his later work, is moved by considerations interestingly similar to those motivating Ackerman, while nevertheless raising the hope that he can avoid the pitfalls which mean we must reject Ackerman’s strategy. This is apparent in Political liberalism,145 where Rawls attempts to show that presenting justice as fairness as what he terms a political conception of justice can render it acceptable, under the right conditions, to a wide range of citizens who may differ greatly in terms of their other, non-political, commitments.

Unlike Ackerman, however, Rawls does not see himself primarily as addressing potential converts to the principle of state neutrality, who might be provoked to reject the principle if it is presented to them as the conclusion of an argument from premises which they do not accept. Rawls’s concern, rather, is with the possibility of an overlapping consensus on the acceptability of justice as fairness, which we can take to entail state neutrality, amongst citizens who adhere to diverse comprehensive conceptions of the good life. Rawls asks, in other words, whether a future, or hypothetical, “Rawlsian” state can defend its refusal to act on the basis of any particular conception of the good in a manner that is neutral between precisely these conceptions. And he asks this because he fears, for similar reasons to those of Ackerman, that a state which violates the principle of philosophical neutrality in its defence of the principle of state neutrality runs the risk of losing the loyalty of citizens who might otherwise have endorsed the latter principle. His hope is that a politically liberal state can rely on enough citizens to converge on a consensus endorsing neutrality on the basis of their own conceptions of the good – the so-called overlapping consensus.

The suggestion that citizens might not be able to converge on justice as fairness as the conception of justice which is to regulate the basic institutions of their society threatens the heart of Rawls’s case for it. This is because he had argued, in A theory of justice, that the superiority of justice as fairness over utilitarian and perfectionist conceptions rests, partially but crucially, on the fact that it offers superior stability, by virtue of the fact that it alone of the options discussed146 could function as the conception for a well-ordered society,147 and would for this reason be favoured by the parties behind the veil of ignorance.

↓67

Utilitarian and perfectionist conceptions of justice, if publicised as the conceptions which regulate the basic institutions of a society (as must happen if the society is to be well-ordered), would, argued Rawls, in all likelihood, fail to attract the support of all citizens, given that both utilitarian and perfectionist conceptions might mean lifelong disadvantages for some citizens for the sake of improving the average level of utility or perfection. But parties in the original position must, says Rawls, take into account the strains of commitment any conception is likely to impose on them once they enter a society; the parties dare not, in other words, choose a conception they could not live with if things turned out badly for them.148

Of course if the parties in the original position were permitted to pick a conception on the understanding that it need not be publicly acknowledged as the real measure of the institutions of the society, they might be inclined to choose utilitarian or perfectionist conceptions, given that they would know that citizens might be able to accept their unlucky lot in life, should it come to that, if the fact that their lot was a function of the real conception of justice which regulated the institutions of the society were concealed from them.149 The publicity condition on choice behind the veil of ignorance is of course designed to offset this possibility.150 It therefore counts against utilitarian and perfectionist conceptions of justice, for the parties in the original position, that their acceptance as conceptions which are to regulate the basic institutions of a society might, given the way we know humans to be, turn out to render those societies unstable. The serious disadvantages certain citizens might have to suffer under a utilitarian or perfectionist dispensation would be likely to turn them against such dispensations, damaging the stability of the society.

Justice as fairness, according to Rawls, does not suffer from this defect. The case Rawls makes for the superior stability of justice as fairness is a complex one, and I do no more than summarise it here.151 Roughly speaking, the two principles of justice as fairness are said by Rawls to give greater support to citizens’ sense of self-respect than do utilitarian and perfectionist views, in that the first principle acknowledges their equality in the matter of a number of important freedoms, and in that the second principle rules out material inequalities which do not benefit the worst-off. As a result, argues Rawls, it will be clear to the parties in the original position that, regardless of how things turn out for them in society, they will not be expected permanently to sacrifice their well-being or interests for the good of others. We can conclude from this, he says, that a well-ordered society of justice as fairness is more likely to win the loyalty of its citizens than a well-ordered utilitarian or perfectionist society, as citizens under justice as fairness have much less reason (if any) to think that they can improve their condition by establishing a new conception of justice as the principle which regulates the basic institutions of their society. A well-ordered society of justice as fairness is, in this sense, stable, and will, for this reason, recommend itself to the parties in the original position, whereas a utilitarian or perfectionist society which starts out well-ordered is unlikely so to remain.

↓68

In Part III of A theory of justice, Rawls elaborates on the theme of stability, explaining not only that citizens of a well-ordered society of justice as fairness come to acquire a sense of justice, that is, a disposition to act for the sake of justice, but that the exercise of this sense is compatible with, and can promote and realise the good of these citizens. In making this argument Rawls takes the good for each of us to be what it is rational for us to want, assuming that we have full information and have reflected critically on our ends, made them consistent, and decided on effective means for realising them. What concerns him here is the possibility that it is not rational in a well-ordered society to exercise and develop the sense of justice, as defined by justice as fairness, and to incorporate this virtue into one’s conception of the good; in other words that, in pursuing their conceptions of the good, citizens of a well-ordered society do not have reason to acknowledge the claims of the two principles of justice. Clearly, if this is so, then justice as fairness is utopian, and cannot provide the stability which allegedly renders it superior to utilitarian or perfectionist conceptions of justice.152

Rawls’s response, in A theory of justice, to this worry is the so-called congruence argument. Here he offers what he terms the Kantian interpretation of justice as fairness, in which he claims, first, that we are capable of autonomy by virtue of ‘our nature as free and equal rational beings,’153 and, second, that the original position can be construed as “modelling” this conception of the human person, from which we can conclude that the principles chosen by the parties in the original position can be interpreted as principles that we give to ourselves out of our nature. If this is so, then to act for the sake of the two principles of justice is to act autonomously in the Kantian sense. Rawls then argues that, given that it is rational for citizens of a well-ordered society of justice as fairness to realise their natures as free and equal rational beings, it is also part of the good of each citizen. The human good and the two principles of justice as fairness are therefore congruent, and it is therefore indeed rational in a well-ordered society of justice as fairness to exercise and develop one’s sense of justice and incorporate it into one’s conception of the good. Rawls concludes, as a result, that we need have no concerns regarding the stability of a well-ordered society of justice as fairness.

We are to imagine, then, a society in which everyone accepts the two principles of justice as fairness as those principles which are to regulate the basic institutions, in which it is public knowledge that the two principles are to serve this purpose, in whose institutions justice as fairness is consistently realised, and in which citizens are generally inclined to do what justice as fairness requires of them; a well-ordered society of justice as fairness, in other words.

↓69

In such a society, as in any free society, citizens may pursue various conceptions of the good life.154 In fact, because of what Rawls refers to as the burdens of judgement,155 there is good reason to expect a great variety of such conceptions. The free workings of the human intellect in a society which does not enforce any particular orthodoxy on the question of the good life will inevitably result in people reaching widely different conclusions about how to live their lives, and given that the case for any comprehensive conception of the good will never be completely clear-cut, even when all citizens are reasonable, the use of judgement cannot be avoided.

In A theory of justice, Rawls expected, as we have seen, that citizens of a well-ordered society of justice as fairness could, despite their adhering to different comprehensive conceptions of the good life, nevertheless converge on a commitment to the two principles of justice. By the time he came to write Political liberalism, however, he had come to think that this convergence might be threatened by the way in which the two principles of justice as fairness would be publicly justified.

Rawls’s worry here is that a public Kantian justification of justice as fairness would affect the stability of a well-ordered society of justice as fairness. Given the diversity of reasonable comprehensive conceptions in the society, the consensus on the principles of justice as fairness would not extend to the public justification offered for the principles and, as a result, commitment to the principles themselves might be weakened in the (presumably many) non-Kantians in that society. If this commitment, and with it the advantages in terms of stability which justice as fairness is said to offer, is to be maintained, a public justification which is not dependent on any one of the controversial comprehensive conceptions of the good life likely to be adhered to in a well-ordered society of justice as fairness needs to be found. And this is why justice as fairness needs to be justified neutrally – it needs to be offered, much as Ackerman does, as the conclusion of any number of arguments.

↓70

It is in response to this imperative that Rawls wishes to demonstrate that justice as fairness can be presented as a political conception of justice. A political conception of justice does not aim to encompass all the contexts in which we might speak of justice or injustice, let alone the good life. It is not offered, in other words, as a basis for social co-operation on the grounds that it is derived from any comprehensive (and putatively true) moral, philosophical, or religious vision of the good life despite its – hopefully – being compatible with many such visions. And it is this particular feature of a political conception – the compatibility of its justification, as well as its content, with a variety of comprehensive conceptions of the good life – which, Rawls argues, makes an overlapping consensus156 between adherents of various comprehensive doctrines, and the social stability consequent on this consensus, possible. We need, therefore, to examine the extent to which the political presentation of justice as fairness does render it such that a broad range of citizens will give it their assent.

What reasons do we have for thinking that presenting justice as fairness as a political conception would enable the Rawlsian state to take an ecumenical approach to justifying the principle of state neutrality? Do we have grounds for thinking, in other words, that such a state could argue for justice as fairness while remaining agnostic on the question of the truth or falsity of the conceptions of the good, or argumentative starting points, adhered to by citizens?

Ackerman expected, as we saw,157 that any premise158 would prove to be capable of serving as the starting point for a valid argument to the neutrality principle (given sufficient rationality on the part of those challenged to work out the implications of their starting points). In a similar manner Rawls expects that (at least) significant numbers of citizens will not need to be challenged to give up the comprehensive doctrines by which they live in order to submit to the demands of political liberalism. This is because, as Rawls envisages it, many citizens are likely to be reasonable, in the particular way in which he uses the term. This involves, firstly, their acknowledgement of the burdens of judgement. They do not, in other words, attribute deviation from their favoured conception of the good solely to the stupidity or malice of those with whom they disagree on these matters, and they conclude, consequently, that they cannot expect the state to legislate on the basis of, or enforce, their (or any other) doctrine of the good.

↓71

Assessing this expectation on the part of Rawls is not an easy matter, given that his position on the possibility of an overlapping consensus has prediction-like and argument-like aspects. Furthermore, the prediction-like aspect is rendered obscure by the fact that we should not read Rawls as suggesting that an overlapping consensus on justice as fairness will actually be achieved at some point in the future. But we can at least say about the argument-like aspect of the expectation that we do not have grounds for thinking that justice as fairness, even when presented as a political conception, is compatible with any and every comprehensive doctrine which is an option in contemporary liberal democracies.

This would not surprise Rawls, of course.159 But this does raise questions about whether his defence of political liberalism can be regarded as an example of an ecumenical strategy. The Rawlsian state (the state which enforces political liberalism, in other words) will, presumably, have to offer public arguments for justice as fairness (or some other conception of justice which is compatible with political liberalism) which are aimed not only at citizens who are reasonable, in the Rawlsian sense, but also at citizens who have not concluded that the conceptions of the good which guide their actions in the normal course of events are compatible with a political conception of justice.

Now obviously a Rawlsian state can remain ecumenical when justifying political liberalism to those who are part of the overlapping consensus. It can say to them, much in the way that Ackerman does, that it is has no interest in why they come to the conclusion that political values trump non-political values in the political sphere, or, to put the same point another way, how they get from their starting points to the conclusion that the state may not enforce any comprehensive conception of the good. As Rawls puts it in A theory of justice, ‘[j]ustice as fairness...[does not] try to evaluate the relative merits of different conceptions of the good...There is no necessity to compare the worth of the conceptions…once it is supposed they are compatible with the principles of justice.’160 And it may indeed be the case that the majority of citizens adhere to such conceptions, freeing the state from the burden of arguing for the liberal dispensation. But the breadth of Rawls’s ecumenicism is tested in its encounter with those outside the consensus.

↓72

If we judge by Lecture 4 of Political liberalism (the lecture on the idea of an overlapping consensus),161 the state may present two principal neutral arguments for political liberalism. The first simply involves the claim that what Rawls calls fair social co-operation on a footing of mutual respect is of great value, and the additional point that this form of co-operation would be threatened by legislation which could only be justified on the basis of a comprehensive conception of the good life. The way he puts it is to say that

the virtues of political co-operation that make a constitutional regime possible are…very great virtues. I mean, for example, the virtues of tolerance and being ready to meet others halfway, and the virtue of reasonableness and the sense of fairness. When these virtues are widespread in society and sustain its political conception of justice, they constitute a very great good, part of society’s political capital. Thus the values that conflict with the political conception of justice and its sustaining virtues may be normally outweighed because they come into conflict with the very conditions that make fair social co-operation possible on a footing of mutual respect.162

↓73

The ‘values that conflict with the political conception’ are “non-political” values – values that derive from the comprehensive conceptions of the good citizens live by, and they may conflict with the political conception if they encourage their adherents to press for a political order which expresses them, despite their less than universal acceptance. And when Rawls writes that these values ‘come into conflict with the very conditions that make fair social co-operation possible on a footing of mutual respect’ he takes it that mutually respectful co-operation between the moral, philosophical, and religious factions which are inevitable in a free society breaks down if the political order expresses views which are properly to be regarded as private, or non-political. Put more crudely, social peace (albeit of a strictly liberal kind) will be endangered by partisan laws, and this is why a politically liberal dispensation is to be preferred.

In making this point, Rawls challenges critics of political liberalism to ask themselves whether they really want to give up the great good that is a society in which people co-operate fairly with each other on a footing of mutual respect. Add to this challenge the second argument from Lecture 4: what we might call the appeal to reasonableness, and we have Rawls’s attempt at an ecumenical strategy laid out. He appeals to reasonableness as follows:

↓74

Since many doctrines are seen to be reasonable, those who insist, when fundamental questions are at stake on what they take as true but others do not, seem to others simply to insist on their own beliefs when they have the political power to do so. Of course, those who do insist on their beliefs also insist that their beliefs alone are true: they impose their beliefs because, they say, their beliefs are true and not because they are their beliefs. But this is a claim that all could equally make; it is also a claim that cannot be made good by anyone to citizens generally. So, when we make such claims others, who are themselves reasonable, must count us unreasonable. And indeed we are, as we want to use state power, the collective power of equal citizens, to prevent the rest from affirming their not unreasonable views.163

So the second half of the non-partisan case for justice as fairness consists in Rawls, or the Rawlsian state, challenging critics of political liberalism with the accusation that it is unreasonable to press for perfectionist legislation, given the inevitability of dissent from the values which are to serve as the basis of such legislation.

This is a more substantial strategy than that of Ackerman, in that Rawlsian ecumenicism does argue for justice as fairness, and thereby for state neutrality. Furthermore, it seems that in presenting justice as fairness as a political conception of justice which guarantees the great good of (a liberal) social peace, and in suggesting to those who lie outside the overlapping consensus that they are thereby unreasonable, the Rawlsian defends neutrality without having to dirty his hands and take sides on questions of the truth or falsity of the comprehensive conceptions of the good life citizens adhere to. The appeal to the importance of the political virtues does not obviously depend on the truth of any comprehensive conception of the good life. Nor does it appear to deny the truth of any comprehensive doctrine. And in pointing out that it is unreasonable to insist that one’s conception of the good be enforced by law, Rawls appeals to the uncontroversial fact that not all people of adequate intellectual standing and good faith have come to the same conclusions about how life ought to be lived. Pointing this out to someone is not to imply that their view is false. No stand on this question need be taken, it would seem.

↓75

But, as I have hinted at above, matters are not this simple. Why this is so becomes clear when we imagine in more detail, as we did in the case of Ackerman, the encounter between the neutralist and the perfectionist, as we might refer to those outside the overlapping consensus.

In discussing Ackerman, we noted that he might face two kinds of difficulties in dealing with stubborn perfectionists. The first possibility was that the perfectionist might indeed be committed to the premises which Ackerman takes to be potential starting points for an “argumentative path” to the principle of state neutrality, but nevertheless deny that neutrality is the logical end point of the path which begins with the premises they assert. The second possibility was that the perfectionist might reject the premises which Ackerman appeals to altogether, forcing him to argue for the truth of at least one, and thereby calling the ecumenical nature of his strategy into question. We will see how Rawls must deal with similar interlocutors.

With regard to the argument from the great value of fair social co-operation on a footing of mutual respect, the Rawlsian state may encounter perfectionists who do in fact value co-operation of this kind highly, but nevertheless think that, occasionally, this good stands in the way of the achievement of greater goods. Such a perfectionist will be mostly reasonable, in the Rawlsian sense, but will think that on some crucial issues the state must enforce the good over the reasonable objections of dissenting citizens. And we will see that encounters with moderate perfectionists such as these raise the question whether political liberalism, and therefore state neutrality, can be defended in an ecumenical manner.

↓76

Such perfectionists accept the facts of reasonable pluralism and the burdens of judgement: there is no need for the Rawlsian to get them to accept premises they don’t already accept. But what they don’t accept is that the facts of reasonable pluralism and the burdens of judgement imply that political values must always override non-political values. They do not, in other words, think that state neutrality is the logical response to the diversity of reasonable comprehensive doctrines. Clearly, the state must, in such cases, try to show such perfectionists why the principle of state neutrality does follow from the burdens of judgement. And in doing so it violates Ackerman’s standards of ecumenicity.

But the Rawlsian does have another arrow in his quiver: he will argue that Ackerman’s stringent degree of ecumenicity is neither achievable nor desirable, but that a lesser degree of ecumenicity is both. This lesser degree amounts to the political nature of the argument the state puts forward for its neutrality: it may have to take issue with the argumentative path followed by the moderate perfectionist who does not reach state neutrality from the burdens of judgement, but it does so without drawing on any comprehensive conception of the good life. And in doing this, the state accuses the perfectionist of being unreasonable, but not of adhering to a false doctrine.

Can the Rawlsian state clear this lower bar? Answering this question requires taking a closer look at what Rawls understands by the term “reasonable”. And when we do, it becomes clear that reasonableness, in the Rawlsian sense, is a little more partisan than we might have hoped. He says in Political liberalism that ‘[r]easonable persons…desire for its own sake a social world in which they, as free and equal, can cooperate with others on terms all can accept.’164 This sounds a good deal as if Rawls takes being reasonable to mean understanding society as a fair system of co-operation between free and equal citizens: precisely the view of society that political liberalism expresses. A little later on he goes on to say that

↓77

being reasonable is not an epistemological idea (though it has epistemological elements). Rather, it is part of a political ideal of democratic citizenship that includes the idea of public reason. The content of this ideal includes what free and equal citizens can require of each other with respect to their reasonable comprehensive views.165

Another way of putting this would be to say that being reasonable does not merely consist in having certain mental capacities or having access to certain knowledge: being unreasonable is not simply a failure of mental competence. The perfectionist may recognise the facts of reasonable pluralism and the burdens of judgement, but, for Rawls, this is insufficient. It is the perfectionist’s response to the facts that is inadequate, and, furthermore, it is inadequate for reasons that have nothing to do with the uncontroversial facts themselves. Being unreasonable, for Rawls, means failing to live up to a larger political ideal which includes, naturally, the ideal of reasonableness, but also, as he indicates above, a number of other notions from the lexicon of political liberalism, notions which, if they are accepted at all by perfectionists, are certainly not accorded the same weight they are in the vision of political liberalism.

↓78

The upshot of taking the reasonable to be part of a political ideal of this kind is that it allows Rawls to define the perfectionist, moderate or otherwise, as unreasonable, for it is precisely this understanding of the “political ideal of democratic citizenship” at the heart of political liberalism that the perfectionist calls into question. The Rawlsian notion of the reasonable is not, therefore, neutral ground to which the politically liberal state may appeal in attempting to persuade perfectionists to refrain from demanding that their favoured comprehensive doctrines be expressed in legislation. It is rather a particular kind of response to the burdens of judgement, one which expresses the values of political liberalism.

But having established that the appeal to reasonableness is not an appeal to ground which is neutral between the political liberal and the perfectionist does not establish, of course, that the notion of reasonableness is not a political notion, in the Rawlsian sense. It may well be that Rawls’s conception of the reasonable is one which, as part of the doctrine of political liberalism, applies only in the political sphere, and is elaborated in terms drawn entirely from ideas in public political culture of contemporary liberal democracies,

That this is not the case becomes clear when we look at how the Rawlsian state must confront non-moderate perfectionists – perfectionists whose premises must be challenged, as opposed to merely the conclusions they draw from premises they share with political liberals. In discussing this kind of encounter, Rawls asks us to

↓79

imagine rationalist believers who contend that these beliefs are open to and can be fully established by reason…In this case the believers simply deny what we have called “the fact of reasonable pluralism”. So we say of the rationalist believers that they are mistaken in denying that fact; but we need not say that their religious beliefs are not true, since to deny that religious beliefs can be publicly and fully established by reason is not to say that they are not true. Of course, we do not believe the doctrine believers here assert, and this is shown in what we do. Even if we do not, say, hold some form of the doctrine of free religious faith that supports equal liberty of conscience, our actions nevertheless imply that we believe the concern for salvation does not require anything incompatible with that liberty. Still, we do not put forward more of our comprehensive view than we think needed or useful for the political aim of consensus.166

In cases such as the “rationalist believer” the Rawlsian has even less purchase in appealing to the burdens of judgement – if the reasonableness of those who disagree with her position is denied, then, obviously, the next move to the reasonableness of political liberalism as a response to the facts of reasonable pluralism and the burdens of judgement cannot be made.

↓80

Now what is striking here is the admission that at least part of “our” comprehensive view has to be put forward in talking to the stubborn perfectionist. While the politically liberal state does not directly say that the comprehensive conception of the good life adhered to by the believer is false, this is implied by its actions, says Rawls. This is an extraordinary concession for Rawls to make, as it amounts to a recognition that political liberalism cannot be defended in a purely political manner against perfectionists who do not acknowledge the burdens of judgement. And this means that, at least with regard to non-moderate perfectionists, that the Rawlsian state does not justify state neutrality in a manner ecumenical enough to satisfy the relaxed requirements Rawls himself sets out, let alone Ackerman’s more stringent standards.

And when we consider why the Rawlsian state cannot do this, we see that it is because the way in which one responds to the burdens of judgement is in fact dependent on one’s comprehensive conception of the good life, and, further, that even perfectionists who do recognise the fact of reasonable pluralism must be confronted by the Rawlsian state in a non-political, and hence non-ecumenical, manner. This becomes clear when Rawls’s discusses the case of abortion. On this topic he says the following:

↓81

Suppose…that we consider the question in terms of three important political values: the due respect for human life, the ordered reproduction of political society over time, including the family in some form, and finally the equality of women as equal citizens…Now I believe any reasonable balance of these three values will give a woman a duly qualified right to decide whether or not to end her pregnancy during the first trimester…Any comprehensive doctrine that leads to a balance of political values excluding that duly qualified right in the first trimester is to that extent unreasonable…167

This passage should be understood as emphasising that the view that the state ought to treat abortion at any stage of pregnancy as murder is an unreasonable one, on the grounds that believing abortion to be the equivalent of murder is a non-political belief which might reasonably be disputed.

Clearly that belief is a view that may be reasonably disputed. And it is also clear that enforcing this view – making abortion illegal in most or all cases – is likely to hinder social co-operation on a footing of mutual respect. But one might accept both of these points and nevertheless think that abortion ought to be illegal in most or all cases because one thinks also that being reasonable and maintaining the social peace, however important, do not trump the need to prevent murder. This is in no way an incoherent position: it is a response to the burdens of judgement which is rooted in an at least partially non-liberal conception of the good. And the Rawlsian state’s response, whether it be the insistence on the premise that there is a fact of reasonable pluralism which must be acknowledged, as it would be in the case of the “rationalist believer” or non-moderate perfectionist, or the insistence that the burdens of judgement imply the principle of state neutrality, as it would be in the case of the moderate perfectionist, implies the falsity of the conception of the good to which the rationalist believer or the (mostly) liberal opponent of abortion rights is committed, because no one could adhere to that perfectionist conception of the good and take the attitude to the burdens of judgement which political liberalism requires.

↓82

In summary, then, Rawls’s appeal to reasonableness is not an ecumenical gesture. This is because the Rawlsian conception of reasonableness is a partisan one – it is in fact a moral conception, and one, furthermore, which is part of an anti-perfectionist vision of the citizen’s place in society – and not an uncontroversial starting point on which political liberals and their opponents can agree. The appeal to reasonableness involves denying the premises of any conception of the good which disputes the fact of reasonable pluralism and rejecting the argumentative path taken by any conception of the good which acknowledges the fact of reasonable pluralism, but accords the good of fair social co-operation on a footing of mutual respect less weight in relation to other goods than the Rawlsian liberal accords it. How one responds to the fact of reasonable pluralism turns out to be intimately connected to which comprehensive conception of the good life one adheres to.

Consequently, the apparently ecumenical defence of state neutrality which Rawls’s proposes in Political liberalism turns out in fact to have partisan elements. The politically liberal state cannot defend its refusal to legislate on the basis of any particular conception of the good without arguing for at least the view that the good of fair social co-operation on a footing of mutual respect ought to take precedence over any goods which might come into conflict with it or the view that reasonableness consists in responding to the burdens of judgement by rejecting perfectionism. The fact that one’s take on these two issues is a function of the conception of the good life one is committed to means that citizens cannot simply be left to work their way to the principle of state neutrality using argumentative paths they are already committed to. Their comprehensive conceptions of the good life, premises, or argumentative paths will, in many cases, have to be challenged by the Rawlsian state.

3.2 Non-neutral justifications of the principle of state neutrality

If it is neither possible nor desirable to justify the principle of state neutrality neutrally, what about frankly partisan justifications? Is there, in other words, a convincing case to be made for the principle of state neutrality on a basis which is not shared by all citizens of contemporary liberal democracies? Many such strategies have been suggested – Ackerman writes, for example, of ‘four…main highways to the liberal state: realism about the corrosiveness of power; recognition of doubt as a necessary step to moral knowledge; respect for the autonomy of persons; and scepticism concerning the reality of transcendent meaning.’168 He goes on to say that ‘[d]oubtless there are other paths as well.’169 I do not intend, in what follows, to adopt Ackerman’s schema in its entirety – the second highway seems to me to collapse into the first and the fourth seems to me wholly unconvincing – but will concentrate rather on the first and third highways, given the greater likelihood of their indeed leading us to the principle of state neutrality.170

3.2.1 The first highway: Realism about the corrosiveness of power

↓83

I deal first with a number of popular arguments for state neutrality which are premised on what Ackerman terms “realism about the corrosiveness of power” and conclude that they do not, in the end, establish sufficiently that the principle of state neutrality is the necessary response to this problem.

Roughly speaking, this realism might be paraphrased as the view that perfectionist governments cannot be trusted to promote the good. When one says that one fears that a perfectionist state cannot be trusted to promote the good, what exactly is it that such a state is thought to be in danger of doing? What evils, in other words, is state neutrality thought to be the best method of thwarting?

The first kind of misbehaviour occurs when the state misidentifies what the good in fact is, in which case it ends up promoting that which is worthless and (at least possibly) damaging to citizens’ lives. This concern is to be distinguished from moral scepticism – it is not the claim that there is no good which anybody, state or otherwise, might identify. In fact it is a kind of concern only available to moral realists, but it may be motivated by the belief that governments are particularly prone to judge moral matters incorrectly, given the multitude of temptations and interests they face. Raz alludes to this concern when he notes, as motivations for anti-perfectionism,

↓84

the dangers inherent in the concentration of power in few hands, the dangers of corruption, of bureaucratic distortions and insensitivities, of fallibility of judgement, and uncertainty of purpose, and the insufficiency and the distortion of information reaching the central organs of government.171

What are we to make of this concern? The first point to make is that there is little reason to believe that the state’s refraining from promoting its (possibly erroneous) vision of the good would guarantee that citizens would be free of any influences which promote or discourage the good. There are many ways that one’s life could go wrong without the state’s interference, and one might worry just as much that the state’s abstention from promoting the good would leave citizens vulnerable to other, possibly more malign, forces. So those who argue for neutralism on the basis that the state may misjudge the good need to show that the state is especially prone to this kind of misjudgement.

↓85

Why might one think that the state is especially prone to promoting misguided conceptions of the good? Mill suggests three reasons.172 The first one is that the state cannot know the needs and tastes of each individual sufficient for it to be any kind of authority on what his or her good is. Secondly, the standards for judging the successful promotion of the good on the part of the state are, or would be, so vague as to leave policymakers with little incentive to consider legislation carefully. And, thirdly, a state which successfully promotes even a valid vision of the good amongst the citizenry runs the risk of leaving them, at some future time, unable to discern the falsehoods within that vision, or falsehoods which come to replace that vision.

One might, for any one of these three reasons, think that the state best remain neutral, and that pursuit of the good is best left in the hands of individual citizens themselves, given their greater knowledge of their own predilections, the much higher stakes they have in their leading valuable lives in comparison to the state, and the likelihood that they will develop the kind of dependence on the state’s vision of the good that will leave them vulnerable, at a later date, to false values.

Sher has argued173 that Mill’s concerns are not misguided, and indeed make a convincing case for caution with regard to perfectionist legislation. But we need not conclude from this that the state must be bound by a rigorous principle of state neutrality. Holding instead that the state ought to be bound by a principle – which Sher dubs principle M – forbidding it from promoting any conception of the good unless that conception has been found to satisfy our usual standards of justification is just as likely, if not more so, to thwart the dangers of the state which misjudges the good.

↓86

Is there any reason to suppose that a state which observed Sher’s principle M would be more likely to disregard its lack of knowledge of the needs and tastes of each individual, and thereby to hinder citizens in their efforts to lead valuable lives, than a state which understood itself to be bound by the principle of state neutrality?174 Or to suppose that a state’s being bound by the principle of state neutrality is necessary to diminish the danger of false ideals being promoted by officials with little at stake? Or that a commitment to state neutrality would reduce the risk of future moral misjudgements on the part of the state more than a commitment to assessing conceptions of the good by means of our usual standards of justification before promoting them?

One misleading line of argument for neutrality needs to be dealt with first, and that is the claim that a principle such as M could not solve the problem of possible misjudgements on the part of the state, since it is the state itself which must apply the principle M. Given that it is the state’s vulnerability to error that raises the concern in the first place, so the argument goes, any response which suggests, in essence, that the state police its own errors (or, to put it more kindly, take its own fallibility into account), must be to that extent fallacious.

This argument shows too much. It shows too much because the claim that the state cannot be trusted to live up to the principles it is ostensibly committed to leaves it quite unclear as to why the principle of state neutrality should be any less vulnerable to such a move. If states which are supposedly committed to a careful weighing of reasons for and against before promoting a particular conception of the good cannot be trusted actually to act consistent with this principle, why should we assume that states which are committed to the principle of state neutrality can be so trusted? What is there about the principle of state neutrality which makes it less likely to be abused?

↓87

There may have been reason for optimism in regard to neutrality’s properties as a fail-safe principle in Mill’s day, but there is surely little reason now, given that we have all seen how easy it has been for states to commit every kind of abuse in the name of principles of right (such as justice) in the twentieth-century. But perhaps one does not need to deploy as sweeping an argument as the suggestion that states which promote the good will be tempted in every way to promote what is not in fact good, and that these temptations are faced to much lesser degree by governments bound to the principle of state neutrality. Perhaps we might rather find specific arguments for the special vulnerability of perfectionist governments to the failure to understand the intricate needs of citizens, or showing how perfectionist governments will be especially uncommitted to working out the consequences of their policies for the goods they are supposedly promoting, or showing how future generations are rendered especially rigid or complacent by perfectionist governments.

It is hard to see where such arguments are going to be found. There is no reason to think that, as Raz puts it, ‘one is more likely to be wrong about the character of the good life than about the sort of moral considerations which all agree should influence political action such as the right to life, to free expression, or free religious worship,’175 in which case we might say that a neutralist government is just as likely to leave its citizens vulnerable to false ideals than a government committed to Sher’s principle M.

This issue reveals the fundamental problem with postulating a neutrality principle as a way of preventing the state from erring. While there certainly is cause to worry that the state may err for the reasons Mill mentions, and any reputable state would do well to take its own fallibility in this regard into account in making law, there are many conceivable ways of guarding against the possibility that a state may promote false ideals of which the principle of state neutrality is only one. An independent argument for the principle of state neutrality is therefore necessary.

↓88

The second kind of pragmatic reason for forbidding the state from promoting the good is motivated by the fear of instability. Larmore’s “modus vivendi” liberalism, and the later work of Rawls can be read as following from the view that avoiding a society in which factions urgently desire control of the state so as to avoid having unwelcome conceptions of the good imposed on them requires excluding perfectionist considerations from lawmaking. Not surprisingly, many have feared that such a society would suffer from a dangerous degree of political instability.176 

The obvious difficulty any such argument faces is the fact that contemporary liberal democracies are neither strictly neutralist nor unstable. Substantive goods are openly supported by the states of North America and the European Union, to mention two primary loci of liberal democracy. Many European states, for example, have established churches, and all of them, as well as Canada and the United States of America, support the arts and monogamous marriage by means of taxation. And these are only the more obvious examples – a great deal of the fine web of law which governs people’s everyday lives is premised on the preferability of certain lifestyles over others.

The stability of contemporary liberal democracies is even more obvious than their lack of neutrality. Western Europe has seen remarkably little political turmoil since the end of World War II, and the United States and Canada have enjoyed an even longer periods of tranquillity. Furthermore, it is quite unclear that the causes of those periods of serious unrest in the recent history of the West can be traced to perfectionist legislation. It is true, of course, that the Nazis were perfectionists, but it was the odious nature of the substantive values they were committed to that pushed the world into war in the late 1930s and not the mere fact that they were committed to substantive values. And resistance to the Nazis can be explained by a host of factors other than objections to perfectionist government.

↓89

We must conclude, then, that there is little reason to think that the dangers which state neutrality is allegedly required to ward off, to the extent that they are genuine, are always better dealt with by a state which is bound only to legislate in a manner neutral between any conceptions of the good than they are by a state that is free to promote the good, provided it is also bound by the need to recognise various rights and procedures that form a standard part of the legal vocabulary of contemporary liberal democracies. It may be that, on occasion, state neutrality is an appropriate response to concerns of these kinds, but this fact is insufficient to establish a principle of state neutrality.

3.2.2 The third highway (1): Promoting autonomy as a ground for the principle of state neutrality 

I deal in this section with a number of influential arguments for state neutrality which rely on the importance of personal autonomy and also conclude that they do not, in the end, establish sufficiently that the principle of state neutrality is the necessary response to this problem.

The importance of personal autonomy is a frequently appealed to – possibly the most frequently appealed to – reason for the requirement that the state be neutral. The basic idea here is that everyone has a fundamental interest in leading an autonomous life, and that perfectionist legislation damages the ability of citizens to do so. It will be my contention that no such general principle applies.

↓90

It will be my contention, in other words, that although particular perfectionist laws may indeed damage citizens’ ability to lead autonomous lives, such laws may just as easily, if not more so, enable citizens to lead autonomous lives, and therefore the argument from the importance of autonomy to a principle of state neutrality fails.

Ackerman’s way of putting this is to say that one can reason one’s way to the principle of state neutrality by adopting ‘a conception of the good that gives a central place to autonomous deliberation and den[ies] that it is possible to force a person to be good.’177 One form that this argument takes moves from the claim that autonomous lives are of great value to the principle of state neutrality via the intermediate premise that more good (in the form of more autonomous living, in other words) results from the state’s refusal to promote any particular conception of the good than would from any alternative. Another way of putting this would be to say that autonomous lives (or autonomous choices) are good in the way that, say, beauty, or excellence, are, and therefore that, because the government must promote the good, the government must promote as much autonomy as it can, and this it does best by remaining neutral between conceptions of the good when legislating: the government’s promoting the most good requires it refraining from the attempt to do so.

The first troubling aspect of this argument, however, is that, while the claim that autonomy is valuable, and even the claim that autonomy is very valuable, is (rather obviously) plausible, the claim that only autonomy is valuable is (rather obviously) implausible. And this gives rise to two difficult questions for those who wish to argue from the value of autonomy to the principle of state neutrality.

↓91

The first question is whether autonomy is always reduced by state promotion of other goods. And if the answer to this question is positive, we might still ask why the state must always take the side of autonomy in any conflict between it and other goods. And even if autonomy is the most valuable good, we might nevertheless ask further why the state’s promoting a select set of goods other than autonomy could never end up promoting more good overall.

As long as the possibility of the state trading autonomy off against other goods exists, the argument from the value of autonomy to the principle of state neutrality will not work. But there may be ways of ruling out such trade-offs, and one tempting way of doing so would be to argue that goods other than autonomy depend for their value on being autonomously chosen. This position need not be interpreted as the dubious claim that the mere fact of being chosen autonomously confers value; it could be the claim that valuable goods acquire their value through an act of choice which recognises their potential independent value.

If one holds to this claim, which is not implausible,178 then one might equally plausibly argue that any state promotion of the good which overrides citizens’ autonomy cannot in fact succeed in promoting the good, as it removes the conditions under which genuine goods can arise. It is not obvious, however, that activities only become valuable in virtue of having been autonomously chosen. And the argument from the value of autonomy to the principle of state neutrality assumes that when a citizen takes up a form of life, or a potentially valuable activity, as a result of the state’s having promoted it in some way, this choice on the part of a citizen is necessarily heteronomous.

↓92

But this is dubious. Let us look at a number of ways in which the state may promote the good in order to establish whether these methods necessarily detract from the autonomy of those who choose the good as a result of these efforts on the part of the state.

Could those who take up a potentially valuable form of life (or reject a worthless form of life) as a result of, say, a state advertising campaign, be said have chosen autonomously? One might think that they cannot, given that, first, autonomous choice is made on the basis of sound reasons, and, second, advertising campaigns (generally) do not aim to persuade by means of argument. One might plausibly conclude that opting to pursue a valuable form of life is an autonomous choice only if the value of that form of life is the reason for it being chosen; doing so as a result of a persuasive advertising campaign is not usually understood as a response to good reasons.179

It is misleading, however, to suggest that if one’s initial choice to pursue a form of life is not made on the basis of sound reasons, then one’s subsequent pursuit of that form of life must be contaminated by this origin. This is because it is very frequently the case that valuable forms of life, pursued by citizens in what appear to be indisputably autonomous ways, were initially chosen on a less than rational basis. One might even suspect that all valuable practices, such as various arts, crafts, sports, or even religions, are initially taken up by their practitioners on the basis of, for example, admiration for prominent figures, the desire to impress their peers, an unmastered thirst for power or status, or any number of other forms of non-rational encouragement on the part of peers or authority figures. And yet it is false that such beginnings preclude an autonomous commitment to these practices in later life. It may be that non-rational choices are the necessary hooks which make it at all possible for adults to participate, whether autonomously or not, in such practices in later life. If this is the case, then a state policy which seeks to maintain an adequate range of valuable options for citizens to pursue, as Raz believes it should,180 must not be construed as threatening to autonomy. On the contrary, a state which fails to do so might be needlessly depriving citizens of many worthwhile avenues of endeavour.

↓93

The obvious rebuttal would be for the neutralist to concede that it may well happen that people respond autonomously to valuable options that they were initially introduced to by non-rational means, while arguing that these happy commitments are outnumbered by cases in which the initial method of persuasion does indeed render future pursuit of the option in question heteronomous, and that this means that the state should avoid promoting the good so as to avoid these more numerous unhappy outcomes. Put more simply, the neutralist might make a consequentialist argument that more good (assuming that autonomy is a good) results from a policy of state neutrality than from a policy of perfectionism, given the relatively low likelihood of autonomous choices being produced from the kind of heteronomous beginnings state promotion of the good induces.

But this seems much too pessimistic. Arguing that the an outright ban on any promotion of the good is necessary in order to forestall the loss of autonomy that such promotion inevitably entails requires arguing that the state is largely incapable of distinguishing between those cases where “manipulative” promotion of the good – say, through advertising campaigns which promote the arts, discourage dangerous drug use, encourage attendance at museums, participation in sports and various kinds of community service, and so on – supports long-term autonomous commitments to these forms of life and those cases in which it does not. There seems little reason to believe that once any state embarks on the promotion of forms of life uncontroversially accepted as valuable, such as those mentioned above, it will inevitably be tempted to promote them, and perhaps other, more controversial, forms in ways that damage citizens autonomy. This is not idle speculation, as most western states do aim to promote the good in precisely this way, and I know of no evidence to suggest that this has brought about a decline in the amount of autonomy in these societies.

A further problem for the argument from the value of autonomy is that it is surely impossible for a government to avoid affecting citizens’ tastes or conceptions of the good. As we saw when discussing the neutrality of effect interpretation of the principle of state neutrality in chapter one, it is simply impossible for a state so to calibrate its policies that the balance of power between all forms of life or conceptions of the good within a society remains exactly as it would be absent state action. If this is so, a discussion of the state’s influence over citizens’ preferences must turn into a discussion of whether it may do so intentionally or not, not whether it can be prevented from doing so or not.

↓94

But a state which refuses intentionally to influence the preferences of citizens in the direction of what it deems to be the good will not thereby increase autonomy in the society, as its unintentional influences will remain untouched; rather, the number of citizens who lead lives that are good in terms accepted by the state will drop, an outcome that should please nobody other than those who hold the state to be promoting false ideals of the good. Raz makes a similar point – emphasising the way in which individuals cannot always create valuable options without state support – when he says that ‘anti-perfectionism in practice would lead not merely to a political stand-off from support for valuable conceptions of the good. It would undermine the chances of survival of many cherished aspects of our culture.’181 I come to discuss this point in greater detail when I outline Raz’s defence of perfectionism in chapter three.

Whatever we have said about the claim that the state can create more autonomy by refraining from using non-coercive, non-rational, methods than by using them does not, of course, necessarily apply to the same claim about coercive methods. This, after all, is the crux of the neutralist’s case: it is in the state’s use of coercion to promote the good or discourage the bad that the neutralist sees the greatest threat to citizens’ autonomy.

But is the argument from the value of autonomy against coercive perfectionism any stronger than the same argument against non-coercive perfectionism? What, after all, is the objection to coercion other than its preventing its victims from responding to the reasons that apply to them?182 And, if this is so, in what way does it differ from the argument that non-coercive perfectionism reduces the overall amount of autonomy? Is it not possible that initial coercion could produce citizens who respond autonomously to the potential good of genuinely valuable forms of life?

↓95

The obvious example would be laws against the use of narcotics. Some citizens may refrain from narcotics out of fear of punishment, and not because they recognise the dangers of narcotics. As such they cannot be said to be living (with regard to this choice, at least) autonomously. But it can hardly be denied that they may well lead lives a good deal more autonomous as a result of their being forced to refrain from narcotics; more autonomous, in all likelihood, than their peers who autonomously choose to devote themselves to regular drug consumption.

I conclude, then, that arguing from the value of autonomy to the principle of state neutrality will not establish that principle, as it is not the case that promotion of the good on the part of the state always decreases citizens’ autonomy. As I argued above, it is possible, under not uncommon circumstances, for perfectionism to increase citizens’ autonomy, especially if citizens’ autonomy is considered over the long run. Furthermore any consequentialist argument which aims to show that a neutral state increases the overall good in a society must account for the place of goods other than autonomy. And this cannot be done in such a way – at least not if the argument is a consequentialist one – as to show why autonomy should always take precedence over other goods.

3.2.3 The third highway (2): Respecting autonomy as a ground for the principle of state neutrality 

One need not, however, defend autonomy as a good to be maximised. It may be that getting from the importance of autonomy to the principle of state neutrality is best done by understanding respect for autonomy as a constraint which rules out non-neutral lawmaking.

↓96

This is a popular route for neutralists – so popular, in fact, that I cannot hope to examine all possible variations on it. As a result I look principally in what follows at the work of Rawls, who, as I have indicated in the introduction, is both unquestionably the most important contemporary liberal neutralist and a formidable opponent of consequentialist styles of political argument. I examine whether a convincing non-consequentialist case for the principle of state neutrality can be gleaned from his writings.

Rawls, as we have already seen, sets up a decision procedure known as the original position in his landmark work A theory of justice, the purpose of the original position being to model the principles of justice that free and equal citizens would agree on, given the right kind of circumstances. Naturally questions will arise as to why exactly Rawls sets up the original position the way he does, and these are precisely the questions we will need to deal with in working out why Rawls believes that free citizens, whom we can understand as autonomous citizens for our purposes, would choose principles of justice which are not to be based on any particular conception of the good. In other words working out why, for Rawls, respect for autonomy means state neutrality, means, at least with regard to A theory of justice, working out why the original position is set up the way it is. And this I do in what follows.

But first a small detour. The immediately puzzling aspect of any argument which rules out promotion of the good under certain circumstances is the fact that it requires one to ignore what would normally be considered good reasons for action; and in the case of arguments from respect for autonomy to the principle of state neutrality the state is required to ignore such reasons entirely.

↓97

How might such a requirement arise? Constraints which forbid us from promoting the good do make sense under some circumstances: it is uncontroversially illegitimate, for example, to extract organs from an unconsenting, living, person, no matter how much good would thereby be promoted. Such constraints, Sher points out, can be justified by the Rawlsian argument that ‘one person’s losses cannot be offset by the gains of others’183.

But, as he goes on to explain, torturing for the greater good is not on all fours with promoting the good in general, as the latter, in theory at least, is an effort to promote the good in the lives of all citizens, not in some (such as those whose suffering is preventing by the torture) at the expense of others (such as the one whose torture prevents the suffering of others). Expanding on this, we might say that state promotion of the good cannot be understood as the thwarting of citizens’ interests as torturing someone can be understood as thwarting his or her interests. This is because, as well as having an interest in leading autonomous lives, citizens also have an interest in leading good lives, in which case the state’s failure to promote the good might well be understood as thwarting citizens’ interests.

Let us turn now to Rawls. In what follows I will look at arguments presented in both A theory of justice and Political liberalism which are understood by Rawls to make the case (although seldom explicitly) for a neutral state, beginning, naturally enough, with those presented in A theory of justice.

↓98

What is Rawls’s case for the design of the original position which is simultaneously a case for the principle of state neutrality? To be precise, the question we have to ask is why Rawls denies the parties in the original position knowledge about the conceptions of the good, given that it is ignorance of this which leads them to assent to principles neutral between such conceptions. In A theory of justice, Rawls describes the veil of ignorance as ensuring ‘that no one is advantaged or disadvantaged in the choice of principles by the outcome of natural chance or the contingency of social circumstance.’184

Given that principles of justice which favoured certain kinds of personal qualities or talents would be precisely principles which advantaged or disadvantaged citizens on the basis of “natural chance” or “the contingency of social circumstance”, we might think that principles which favoured certain conceptions of the good are ruled out for the same reason. It might be that Rawls thinks of the conceptions of the good people adhere to as being unchosen in the way that talents are, and therefore undeserving of favour.

I would caution against interpreting Rawls in this way, however, as tempting though the analogy may be, he surely cannot think of conceptions of the good in the same way as he thinks of talents. And this is because of the significance he takes our ‘capacity to form, to revise, and rationally to pursue a conception of the good’ to have.185 It would be quite unclear why this capacity would be of any importance if it were simply the product of contingency: in fact it would be quite unclear how anybody could revise their conception of the good if that were the case.

↓99

Furthermore, it would also be unclear why any beliefs at all would be permitted to influence the choices of the parties in the original position, as Rawls gives us no reason to think (nor could he, one must assume) that beliefs about the good are products of contingency, whereas their other beliefs are not. And clearly, some beliefs are required behind the veil of ignorance, otherwise the parties could have no grounds whatsoever for choosing the principles they do. We can conclude therefore that Rawls’s reasons for excluding conceptions of the good from the knowledge available to parties in the original position lie elsewhere.

On page 560 of A theory of justice 186, Rawls says that

↓100

 [w]e should not attempt to give form to our life by first looking to the good independently defined. It is not our aims that primarily reveal our nature but rather the principles that we would acknowledge to govern the background conditions under which these aims are to be formed and the manner in which they are to be pursued. For the self is prior to the ends which are affirmed by it; even a dominant end must be chosen from among numerous possibilities. There is no way to get beyond deliberative rationality.187

This suggests that what is important about conceptions of the good – what, in other words, makes them unfit to influence the choices made behind the veil of ignorance – is not that they are the products of contingent factors beyond anyone’s control, but rather that they are external to the self.

But the problem with understanding Rawls to exclude conceptions of the good on this basis is that his claim that the nature of the self is revealed by the principles it would choose behind the veil of ignorance, and not by its ends is that we do not thereby have an argument for denying the parties in the original position knowledge of the conceptions of the good they will adhere to. Naturally, seeing the self in this way leads to Rawls’s designing the original position the way he does, and, unsurprisingly, this design produces principles of justice which forbid the state from acting on the basis of any particular conception of the good. But the progression from Rawls’s vision of the self as prior to its ends to the design of the original position to the principle of state neutrality does not satisfy us if we want to know why we ought to see the self in this way, and if we do not know why we ought to see it this way we will not have been shown why the state should respect the principle of state neutrality.  

↓101

In any case, in attempting to understand why the claim that the self’s being prior to all its ends implies that conceptions of the good cannot be taken into account when principles of justice are being established, we run up against the problem of why one might think that the contingency of one’s conceptions of the good should rule them out from consideration behind the veil of ignorance.188

The problem is that one’s conceptions of the good are not the only beliefs whose failure to “reveal our nature”, as Rawls understands it, is indicated by the fact that our selves remain what they are despite changes in our conceptions of the good. This is in fact true of any beliefs we might have, including those beliefs that Rawls would deem straightforwardly to be “knowledge”, and therefore knowable behind the veil of ignorance. But ruling out all these beliefs from the deliberations the parties in the original position take part in would deprive them of any basis for choosing principles of justice.

And so if Rawls were to argue from the contingency of conceptions of the good, or their changeability, to the principle of state neutrality, he would in both cases show more than he would want to: by ruling out conceptions of the good from the deliberations behind the veil of ignorance he would not only deprive the parties in the original position of what they need to establish principles of justice which are premised on the good; he would deprive them of what they need to establish any principles of justice at all.

↓102

But perhaps these are not in fact Rawls’s reasons. Writing about the arguments for neutrality that might be found in A theory of justice, Raz says that

 [t]o vindicate Rawls’s position one requires convincing reasons…for excluding moral and religious beliefs from the information available behind the veil of ignorance…A theory of justice contains hardly any explicit argument for the exclusion of moral and religious beliefs from the original position. Such argument as there is turns on the need to secure unanimity, the need to have, in the original position, one viewpoint which can be the “standpoint of one person selected at random” which excludes bargaining and guarantees unanimity.189

↓103

So perhaps we can get to the principle of state neutrality (via A theory of justice) this way: We must place knowledge of the good behind the veil of ignorance for the parties in the original position, because failing to do so would make it impossible for them to reach any agreement, with the result that ‘we would not be able to work out any definite theory of justice at all’190  But this surely cannot, in itself, be sufficient for us to accept a principle of state neutrality, as it is quite possible to imagine configurations of the original position other than the setup Rawls actually uses which provide us with an agreement, without denying the parties in the original position knowledge of their conceptions of the good.

They could be, for example, provided with knowledge of a minimal number of well-supported values, much in the way that they are provided with a certain amount of empirical knowledge. Or we might note, as Sher has, that ‘nothing [said by Rawls in A theory of justice] shows why we should not altogether dispense with the contractarian premises about the good life.’191 and that doing this would produce conclusions no less determinate than those reached by the Rawlsian parties.

Furthermore, Raz has argued that even being denied knowledge of one’s conception of the good would not suffice to establish a neutral constitution: as he puts it, the original position

↓104

may yield an agreement to establish a constitutional framework most likely to lead to the pursuit of well-founded ideals, given the information available at any given time. Ignorance of one’s particular moral beliefs will not exclude this possibility, since the parties in the original position know that they have moral ideals. They accept, in other words, “a natural duty” to pursue the best-founded moral ideal.192

Contrary to Rawls, therefore, a principle of state neutrality cannot be deduced from the choices parties in the original position make: they might well refrain from adopting a ‘particular perfectionist principle as a constituent of their doctrine of justice,’193 given their general knowledge of human fallibility, but there is no reason why they could not ‘accept a doctrine of justice including an agreed process for determining which perfectionist principle should be implemented in the state’194 – if we (and they) are to consider matters of the good as potential subject matter for rational deliberation, that is. And Rawls is of course eager to evade charges that he is a moral sceptic.195 We might just as well ask ourselves, when thinking about justice, what sort of process the parties in the original position would want to implement for deciding which conceptions of the good ought to serve as the basis of legislation. This conclusion is, of course, welcome to perfectionists.

↓105

Of course A theory of justice is not all there is to Rawls’s work. And it may be that, despite the difficulties of establishing where exactly the case for the principle of state neutrality is to be found in this first book, things become clearer once we turn to Political liberalism, where Rawls presents his theory of justice in a new manner, taking it to be the conclusion of a process, valid for contemporary liberal democracies alone, in which its principles are “worked up”, as he puts it, from widely-held assumptions about political morality which are, so to speak, part of the public culture of such democracies. It might be that this strategy makes clearer why a commitment to respect for the autonomy of individuals requires the state the refrain from making law on the basis of any particular conception of the good, as Rawls sees it.

In Political   liberalism Rawls emphasises, once more, his view that a conception of justice must be ‘as far as possible, independent of the opposing and conflicting philosophical and religious doctrines that citizens affirm’196 – neutral between conceptions of the good, in other words. And in Political liberalism he no longer argues for this neutrality on the basis of the original position. It is therefore worth asking why, according to the later Rawls, ought we to make these demands of a conception of justice? Why neutrality?

Raz’s answer is that Rawls advocates state neutrality because he takes it that ‘social unity and stability based on consensus – that is, achieved without excessive resort to force – are valuable goals of sufficient importance to make them and them alone the foundations of a theory of justice for our societies.’197 As Raz interprets Rawls, perfectionist societies must either be unstable, or they must coerce their citizens into stability.

↓106

We have already seen that the claim that unless the laws of contemporary liberal democracies are free of perfectionist legislation, instability will ensue, lacks foundation.198 But let us grant Rawls this premise for the moment, so as to reveal further problems which arise from the way in which Political liberalism sets up the case for state neutrality. If, as Rawls suggests, any conception of justice which is not derived from the shared values of the public political culture is illegitimate (leaving aside for the moment the question of what makes it illegitimate), we must ask why justice as fairness is derived from a controversial conception of the good; the supreme value of uncoerced stability, as Raz puts it.

This apparent dependence on the value of uncoerced stability upsets the later Rawls’s case for state neutrality in way similar to the way in which the argument for the neutrality principle from the value of autonomy is upset.199 If we are to derive the neutrality principle from the original position, the setup of which is itself premised on a certain vision of the autonomous and equal individual, but we are told that this vision of the person derives its validity from its place in the public culture of contemporary liberal democracies, then we need first of all to know why holding such a place in the culture confers validity on the vision of the autonomous person. The answer Rawls must give us, if Raz is correct, is that failing to base our vision of the person on the public political culture would mean basing it, instead, on a controversial conception of the good life. And this, so Raz’s interpretation goes, is ruled out as a basis for legislation on the grounds that it is a recipe for instability.

But here, once again, we have to ask whether the rejection of uncoerced stability is as pervasive a part of the public political culture of modern liberal democracies as all that. It is true that there is no general enthusiasm for instability, but, as Sher remarks, it is still unclear ‘why we should never regard some sacrifices in stability, or some amounts of coercion, as reasonable prices to pay for suitably large amounts of other goods.’200 And here we see that the trade-off problem, which we had hoped arguments from the respect for autonomy (as opposed to the value of autonomy) could sidestep, arises again. It may be that the principle of state neutrality can be derived, via the public political culture, from the value of uncoerced stability, but we are not told why uncoerced stability should take precedence over other, or combinations of other, values.

↓107

One of the important innovations of Rawls later writings201 is the so-called “principle of liberal legitimacy”. This principle, which states that

our exercise of political power is proper and hence justifiable only when it is exercised in accordance with a constitution the essentials of which all citizens may reasonably be expected to endorse in the light of principles and ideals acceptable to them as reasonable and rational202

↓108

is derived, first, from the fact that we do not choose our relationship to the politics of our society, but are simply born into it, and, second, from the fact of the state’s overwhelming power.203 Rawls takes this principle to require that, as we saw in chapter two, conceptions of the good be excluded as a basis for reasoning about constitutional essentials or basic questions of justice.204

So we know from the principle of liberal legitimacy, that no constitution which cannot be reasonable endorsed by all citizens can be legitimate. We do not as yet know, though, as Sher accurately points out,205 that this means that a legitimate constitution cannot be premised on any conception of the good. And this is because we do not yet know what exactly it is that citizens can or cannot reasonably endorse.

Rawls’s view is that citizens, or their comprehensive conceptions of the good life, are reasonable if they do not expect the state to coerce others into compliance with it; if they, in other words, accept the principle of state neutrality. But this is of course not an argument for the principle of state neutrality, for we first need an argument as to why Rawls’s particular vision of reasonableness ought to be accepted: we might, quite reasonably, one would think, regard a reasonable person, or conception of the good life, as one for which there is a good case, a view which does not have obvious neutralist implications. We still do not have a convincing reason for abandoning, in the political sphere, what is obvious in our personal lives; that we ought to promote the good.

↓109

***************


Fußnoten und Endnoten

122  A process described in the “Introduction” to Rawls’s Political liberalism (New York: Columbia University Press, 1993).

123  Failing to do so may leave one unable to justify one’s own adherence to these values to oneself.

124  Or liberalism, were we to see it, as many do, as inextricably linked to the principle of state neutrality.

125  I owe this term to Wall and Klosko. See Steven Wall and George Klosko (eds), Perfectionism and neutrality (Lanham: Rowman and Littlefield, 2003), p. 11.

126  I also owe this term to Wall and Klosko. See Steven Wall and George Klosko (eds), Perfectionism and neutrality (Lanham: Rowman and Littlefield, 2003), p. 11.

127  Charles Larmore Patterns of moral complexity , (Cambridge: Cambridge University Press, 1987), p. xiii.

128  New Haven: Yale University Press, 1980.

129  Steven Wall and George Klosko (eds), Perfectionism and n eutrality (Lanham: Rowman and Littlefield, 2003), p. 11.

130  Bruce Ackerman, Social justice in the liberal state (New Haven: Yale University Press, 1980), p. 369.

131  Bruce Ackerman, Social justice in the liberal state (New Haven: Yale University Press, 1980), pp. 356-7.

132  Bruce Ackerman, Social justice in the liberal state (New Haven: Yale University Press, 1980), p. 357. Liberalism does not in fact (or ought not to!) deny anyone the right to declare that their particular metaphysics is the ultimate truth, but rather the right to oblige others to act as if this were so.

133  Bruce Ackerman, Social justice in the liberal state (New Haven: Yale University Press, 1980), p. 360.

134  Bruce Ackerman, Social justice in the liberal state (New Haven: Yale University Press, 1980), p. 11.

135 

 In the essay ‚Über den Gemeinspruch: „Das mag in der Theorie richtig sein, taugt aber nicht für die Praxis,“, translated as ‘On the common saying: “This may be true in theory, but it doesn’t apply in practice,”’ by HB Nisbet in the collection edited by Hans Reiss entitled Kant’s political writings (Cambridge: Cambridge University Press, 1970), pp. 61-92, Kant remarks at page 79 that

[we] need by no means assume that this contract (contractus   originarius or pactum   sociale), based on a coalition of the wills of all private individuals in a nation to form a common, public will for the purposes of rightful legislation, actually exists as a fact, for it cannot possibly be so…It is in fact merely an idea of reason, which nonetheless has undoubted practical reality; for it can oblige every legislator to frame his laws in such a way that they could have been produced by the united will of a whole nation, and to regard each subject, in so far as he can claim citizenship, as if he had consented within the general will. This is the test of the rightfulness of every public law. For if the law is such that a whole people could not possibly agree to it (for example, if it stated that a certain class of subjects must be privileged as a hereditary ruling class) it is unjust.

136  Ackerman talks rather of “the exercise of power,” but I take law to be a subset – perhaps the most important subset – of those cases in which citizens exercise putatively legitimate power over one another.

137  Bruce Ackerman Social justice in the liberal state (New Haven: Yale University Press, 1980), p. 6.

138  Thomas Scanlon suggests this idea in his ‘Contractualism and utilitarianism’ in Amartya Sen and Bernard Williams (eds), Utilitarianism and beyond (Cambridge: Cambridge University Press, 1982), pp. 103-28.

139 

On page 12 of Social justice in the liberal state (New Haven: Yale University Press, 1980), Ackerman expresses this confidence:

Not that it is completely impossible to reason yourself to a rejection of Neutrality. Plato began systematic political philosophy with such a dream; mediaeval churchmen thought there were good reasons to confide ultimate secular authority to the pope. Only they recognised – as modern totalitarians do not – the depth of the reconceptualisation required before a breach of Neutrality can be given a coherent justification. It is not enough to reject one or another of the basic arguments that lead to a reasoned commitment to Neutrality; one must reject all of them. And to do this does not require a superficial change of political opinions but a transformation of one’s entire view of the world – both as to the nature of human values and the extent to which the powerful can be trusted to lead their brethren to the promised land.

140  It is also compatible with the view that it does not matter how citizens are brought to support the principle of state neutrality, but merely that they can be. Since this is an illiberal position, I will not accuse Ackerman and others of defending it.

141  Assuming he agrees that it serves as the basis of a valid argument for the neutrality principle, which in all likelihood he will, given his confidence that all roads lead to the neutrality principle.

142  See the section entitled “Non-neutral justifications of the principle of state neutrality” below.

143  Rendered plausible in section entitled “Non-neutral justification of the principle of state neutrality”
below.

144  In Robert Goodin and Andrew Reeve (eds), Liberal n eutrality (London: Routledge, 1989), pp. 61-83, at p. 69.

145  (New York: Columbia University Press, 1993).

146  The other conceptions of justice on the table, so to speak, in A theory of justice, include utilitarianism, perfectionism, and various mixed options.

147  Rawls takes, in Political liberalism, a well-ordered society to be a society united in its political conception of justice, and in which this political conception of justice is the focus of an overlapping consensus of reasonable comprehensive doctrines. This is in contrast to his view in A theory of justice, where he takes a well-ordered society to be a society united in its basic moral beliefs. An ideal such as this, he came to believe, is no longer tenable in contemporary liberal democracies, given their moral, philosophical, and religious diversity.

148  The “strains of commitment” are discussed in §29 of A theory of justice (Revised edition) (Oxford: Oxford University Press, 1999), pp. 153-60.

149  It is hard to see, however, why parties which lack a conception of the good would wish to do this.

150  See §29 of A theory of justice (Revised edition) (Oxford: Oxford University Press, 1999), pp. 153-60.

151  See §29 of A theory of justice (Revised edition) (Oxford: Oxford University Press, 1999), pp. 153-60 for Rawls’s full discussion

152  See, in particular, chapter 9 of A theory of justice (Revised edition) (Oxford: Oxford University Press, 1999), pp. 450-514, for Rawls’s full discussion of this issue.

153  John Rawls, A theory of justice (Revised edition) (Oxford: Oxford University Press, 1999), p. 222.

154  Provided that their doing so does not cause the principles of justice as fairness to be violated.

155 

 Rawls offers a list of the sources of reasonable disagreement, which he terms “the burdens of judgement”, on pages 56-7 of Political liberalism (New York: Columbia University Press, 1993), noting that items (a) to (d), as he puts it, apply mainly to the theoretical uses of our reason and (e) and (f) apply to the reasonable and the rational in their moral and practical use.

a. The evidence – empirical and scientific – bearing on the case is conflicting and complex, and thus hard to assess and evaluate.

b. Even where we agree fully about the kinds of considerations that are relevant, we may disagree about their weight, and so arrive at different judgements.

c. To some extent all our concepts, and not only moral and political concepts, are vague and subject to hard cases; and this indeterminacy means that we must rely on judgements and interpretation (and on judgements about interpretations) within some range (not sharply specifiable) where reasonable persons may differ.

d. To some extent (how great we cannot tell) the way we assess evidence and weigh moral and political values is shaped by our total experience, our whole course of life up to now; and our total experiences must always differ.

e. Often there are different kinds of normative considerations of different force on both sides of an issue and it is difficult to make an overall assessment.

f. Any system of social institutions is limited in the values it can admit so that some selection must be made from the full range of moral and political values that might be realised.

156  Rawls introduces the idea of an overlapping consensus in the “Introduction,” pp.xv -xxxii of Political liberalism (New York: Columbia University Press, 1993) and discusses it at length in “Lecture 4,” pp. 133-72 of the same book.

157  See the section above entitled “Ackerman’s ecumenical strategy”.

158 

To be fair, Ackerman does concede the remote possibility that there may be starting points out there which could not lead one to the principle of state neutrality. See, for example, the following remarks at page 12 of Social justice in the liberal state  (New Haven: Yale University Press, 1980).

Not that it is absolutely impossible to reason yourself to a rejection of Neutrality. Plato began systematic political philosophy with such a dream; mediaeval churchmen thought there were good reasons to confide ultimate secular authority to the pope. Only they recognised – as modern totalitarians do not – the depth of the reconceptualisation required before a breach of Neutrality can be given a coherent justification.

159  See his discussion entitled “Is justice as fairness fair to conceptions of the good?” at pp. 195-200 of Political liberalism (New York: Columbia University Press, 1993).

160  John Rawls, A theory of justice (Revised edition) (Oxford: Oxford University Press, 1999), p. 94.

161  John Rawls, Political liberalism (New York: Columbia University Press, 1993), p. 157.

162  John Rawls, Political liberalism (New York: Columbia University Press, 1993), p. 157.

163  John Rawls, Political liberalism (New York: Columbia University Press, 1993), p. 158.

164  John Rawls, Political liberalism (New York: Columbia University Press, 1993) pp. 49-50.

165  John Rawls, Political liberalism (New York: Columbia University Press, 1993), p. 62.

166  John Rawls, Political liberalism (New York: Columbia University Press, 1993), pp. 152-3.

167  John Rawls, Political liberalism (New York: Columbia University Press, 1993), fn. 32, pp. 243-4.

168  Bruce Ackerman, Social justice in the liberal state (New Haven: Yale University Press, 1980), p. 369

169  Bruce Ackerman, Social justice in the liberal state (New Haven: Yale University Press, 1980), p. 369

170  My discussion is in fact more indebted to the schema George Sher presents in Beyond neutrality than it is to that of Ackerman.

171  Joseph Raz, The morality of freedom (Oxford: Clarendon Press, 1986), p. 427.

172  John Stuart Mill, On liberty (Indianapolis: Bobbs-Merrill, 1956), p. 93.

173  George Sher, Beyond neutrality: Perfectionism and politics (Cambridge: Cambridge University Press, 1997), pp. 131-8.

174  George Sher, Beyond neutrality: Perfectionism and politics (Cambridge: Cambridge University Press, 1997), pp. 131-8.

175  Joseph Raz, The morality of freedom (Oxford: Clarendon Press, 1986), p. 160.

176  Of course the overwhelming power of one faction within such a society would stabilise it. But this is hardly the kind of solution liberals look to.

177  Bruce Ackerman, Social justice in the liberal state (New Haven: Yale University Press, 1980), p. 11. One need not, of course, endorse both clauses. One could believe that it was indeed possible to force people to be good, but that that was nevertheless forbidden.

178  It is, as Sher points out, analogous to Kant’s view that actions acquire independent moral value only if performed for the right reasons, namely recognition of their potential independent value. See George Sher, Beyond neutrality: Perfectionism and politics (Cambridge: Cambridge University Press, 1997), p. 59.

179  Of course an advertising campaign could merely present a sound argument to those who have not yet encountered it. I do not refer here to that kind of argument.

180  See Joseph Raz, The morality of freedom (Oxford: Clarendon Press, 1986), p. 372.

181  Joseph Raz, The morality of freedom (Oxford: Clarendon Press, 1986), p. 162. Here Raz is thinking in particular about practices such as opera, but his point in fact covers a wide variety of cases.

182  We will see in chapter three, when I discuss Raz’s view of coercion, that the way I have put it here is a simplification.

183  George Sher, Beyond neutrality: Perfectionism and politics (Cambridge: Cambridge University Press, 1997), p. 74.

184  John Rawls, A theory of justice (Revised edition) (Oxford: Oxford University Press 1999), p. 11.

185  John Rawls, “Kantian constructivism in moral theory,” Journal of Philosophy 77, 9 (1980), pp. 515-72 at p. 525.

186  (Revised edition) (Oxford: Oxford University Press, 1999).

187  John Rawls, A theory of justice (Revised edition) (Oxford: Oxford University Press, 1999), p. 560.

188  See George Sher, Beyond neutrality: Perfectionism and politics (Cambridge: Cambridge University Press, 1997), p. 82, for a similar argument.

189  Joseph Raz, The morality of freedom (Oxford: Clarendon Press, 1986), pp. 124-5.

190  John Rawls, A theory of justice (Revised edition) (Oxford: Oxford University Press, 1999), p. 140.

191  George Sher, Beyond neutrality: Perfectionism and politics (Cambridge: Cambridge University Press, 1997), p. 83,

192  Joseph Raz, The morality of freedom (Oxford: Clarendon Press, 1986), p. 125.

193  Joseph Raz, The morality of freedom (Oxford: Clarendon Press, 1986), p. 126.

194  Joseph Raz, The morality of freedom (Oxford: Clarendon Press, 1986), p. 126.

195  John Rawls, A theory of justice (Revised edition) (Oxford: Oxford University Press, 1999), p 328.

196  John Rawls, Political liberalism (New York: Columbia University Press, 1993), p. 175.

197  Joseph Raz, “Facing diversity: The case of epistemic abstinence,” Philosophy and Public Affairs 19, 1 (1990), p. 14.

198  We can safely conclude, as well (although I do not discuss this in any detail), that the reason contemporary liberal democracies are stable is not (at least not to any worrying extent) because of large amounts of coercion.

199  See pp. above.

200  George Sher, Beyond neutrality: Perfectionism and politics (Cambridge: Cambridge University Press, 1997), p. 85.

201  By this I mean those writings after and including his “Kantian constructivism in moral theory,” The Journal of Philosophy 77 (1980), pp. 515-72.

202  John Rawls, Political liberalism (New York: Columbia University Press, 1993), p. 217.

203  John Rawls, Political liberalism (New York: Columbia University Press, 1993), pp. 216, 135-6.

204  John Rawls, Political liberalism (New York: Columbia University Press, 1993), pp. 216, 135-6.

205  George Sher, Beyond neutrality: Perfectionism and politics (Cambridge: Cambridge University Press, 1997), p. 85.



© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme.
DiML DTD Version 4.0Zertifizierter Dokumentenserver
der Humboldt-Universität zu Berlin
HTML-Version erstellt am:
13.10.2011