We have now briefly traced the history of the idea of neutrality as it has been deployed in English-language liberal political philosophy over the past three decades.

As noted in the introduction, the idea did not meet with acceptance across the board. For some critics, such as the so-called communitarians, who rose to prominence in the 1980s, it was evidence of the misguided nature of liberalism,46 for others it stymied liberal goals.47 This thesis will not devote serious discussion to the communitarians, but will focus instead on the latter attack: criticism, in other words, from those writers who claim to share the commitment, the implications of which the prominent neutralists discussed in the thesis take themselves to be working out in their writings, to the paramount value of freedom and equality for all individuals within society, but who take the neutrality principle to be a misguided attempt at expressing these values


Before we get on to looking at what anti-neutralist liberals have to say, however, we need to fix, to the extent that this is possible, what exactly it is that neutralists are advocating, and why they do so. And here the picture gets complicated. I therefore devote this chapter to surveying a number of important formulations of the principle of state neutrality which have been offered by its advocates. An assessment of the perfectionist attack will have to wait until chapter three, after I have looked, in chapter two, in greater depth at the connections between formulating the principle of state neutrality and defending it.

As we have seen in the introduction, the neutrality principle is generally taken to mean that principle of political morality which requires the state, in a pluralist society, to maintain a position of neutrality towards those large-scale moral, philosophical, or religious frameworks typically referred to in the literature as conceptions of the good or conceptions of the good life, which claim the adherence of citizens. And the expression “pluralist society” is generally used to mean a society in which there is not widespread agreement on any such conception of the good, in contrast to the homogeneity which, we are told, characterised pre-modern western cultures and still, frequently, is said to characterise contemporary non-western cultures.

With respect to the conceptions of the good, between which the state is required to be neutral, it is important to note that they are nowhere taken to include all the moral values held by citizens. This is because the term “the good” has, as used by contemporary political theorists, acquired a quasi-technical sense, and in this context is standardly contrasted with “the right.” Roughly speaking, the good is understood as that which is worth pursuing, whereas the right is that which one is obliged to do.


So, for example, one might hold that pleasing God, experiencing pleasure, or expressing oneself artistically, is worth devoting one’s time and energy (or even one’s life) to, in which case these beliefs or attitudes would be part of one’s conception of the good. They could even be, as may be likely in the case of the first example, the entirety of one’s conception of the good. One might, however, if one is a good liberal, recognise that promoting that which is good or valuable is not the only claim on one’s moral attention. One might think, also, that one’s pursuit of the good ought to be limited or constrained by the obligation one has to take others into account. This might be because one recognises that others do not always value what one values oneself, and that it would therefore be unfair, or unreasonable, to expect them to sacrifice their own pursuits in favour of one’s own (should it come to that). Or one might hold that unlimited pursuit of what one held to be valuable would create an undesirable level of conflict with one’s neighbours.

And one might, of course, have quite different reasons for holding back from all out pursuit of any particular good. But the important point to recognise is that there is an aspect of morality which involves the acceptance of such limits. This aspect twentieth-century English-language moral philosophers have called the right, and its relation to, and apparent or alleged independence from, the good as a source of value has been the topic of much debate.

So, when liberal neutralists such as Rawls and Dworkin say that the state must be neutral between the various conceptions of the good life held by citizens, they do not mean that the state takes no position on how far citizens may go in promoting or pursuing the good, let alone that the state enforces no values whatsoever. On the contrary, the limits of the good are precisely what the neutral state enforces, and indeed it can be said to demonstrate its neutrality in doing so. Liberal neutralism can be understood as the view that the state should enforce the right, while standing aloof from the conflicts about the good.


This formulation is very general, however, and therefore admits of a number of ambiguities. And here is where the trouble starts, for the effect of these ambiguities is, as we shall see in the remainder of this chapter, that the principle has been interpreted in a number of ways, ways which may well conflict with one another.

My initial purpose in what follows is simply to offer a rough characterisation of the different interpretations of the principle of state neutrality. Questions about the plausibility of the principle will be left to chapter two, where I discuss what grounds there might be for adhering to any interpretation of the neutrality principle in greater detail, and where we will also see in greater detail how different grounds for doing so affect the stance one takes on the issues I discuss here in chapter one.

The first ambiguity I wish to discuss concerns what it means to say that the state must remain neutral between rival conceptions of the good. Thereafter I will look at two questions regarding the range of application of the principle, after which I will turn to what I take to be the crucial issue facing interpreters of the principle of state neutrality: the controversy regarding what it means to say that the state may not favour any conception of the good.

2.1 Rival conceptions of the good 


Neutralists raise no controversy (amongst each other, that is) if they claim that the state must be neutral between rival conceptions of the good. But we might wonder under what circumstances conceptions of the good might be said to be rivals. We might, in particular, wonder whether conceptions of the good which are largely unchallenged can be accurately said to have rivals. Obviously any conception of the good whatsoever could be challenged, but a good many of them are in fact not challenged, or not seriously challenged, at least within particular societies. The question then arises as to whether a state which intentionally and/or successfully promotes goods on whose value there is an overwhelming consensus does in fact take sides between rival conceptions.

Typical of one side of the argument is the view of Larmore, who holds that the ideal of political neutrality ‘demands only that so long as some view about the good life remains disputed, no decision of the state can be justified on the basis of its supposed intrinsic superiority or inferiority.’48 In fact contemporary constitutional democracies, the vast majority of which pay lip service to neutrality in some sense, frequently act on the basis of (relatively) uncontroversial conceptions of the good. This is apparent, for example, in the subsidisation of museums and galleries, or the special place monogamous marriage has in law; here the state clearly proceeds on the basis that the cultivated or the monogamous, life is particularly valuable.49

Against Larmore we have writers like Rawls and Dworkin, who express serious misgivings about the possibility of state endorsement of any conception of the good. Rawls remarks, for example, in A theory of justice that ‘the principles of justice do not permit subsidising universities and institutes, or opera and the theatre, on the grounds that these institutions are intrinsically valuable,’50 (although his position appears to soften in his later writings).51 Dworkin’s position is similar to that of the Rawls of A theory of justice, although he devotes considerably more effort to discovering allegedly neutral reasons as to why the state should indeed subsidise various forms of (uncontroversially valued) high culture.52


Ultimately, as we shall see in chapter two, this dispute has its roots in the reasons philosophers bring for advocating state neutrality in the first place. Here it suffices to say that if one is moved to support state neutrality solely out of, say, a concern for political stability, then it is indeed unclear why one would hold that the state should refrain from backing monogamous marriage. In a culture where marriage of this kind is taken almost unanimously to be the most valuable form of life for purposes such as the raising of children and the nurturing of serious sexual relationships, and where these purposes are almost unanimously regarded as worthy goals, state support for monogamous marriage will in no way threaten social stability – in fact it will in all likelihood contribute towards it.

If, on the other hand, one’s primary reasons for advocating neutrality lie in the fact that one particularly values ethical or social diversity, one might think that the society could do with a little more variety in its sexual or familial arrangements than that engendered by the predominance of monogamous marriages. One might also advocate state neutrality on the basis of a commitment to personal autonomy, in which case one might argue that in a society where monogamous marriage is legally enshrined the choices citizens make for or against it do not amount to genuinely autonomous choices.53 And so one might wish for the state to remain aloof from the conflict (if conflict it is) between monogamy, bigamy, and promiscuity, or that between heterosexuality and the various alternatives to it, leaving citizens to have maximum opportunity to arrive at their own conclusions about the respective values of these sexual strategies.

Whatever one’s position on how the neutral state should respond to consensus on the good, it cannot be defended without reference to the arguments for the neutrality principle. Because of this, we must postpone serious discussion of these alternatives until the second chapter, where I will examine the connections between arguments for neutrality and the various formulations of the neutrality principle in depth. For now I continue to survey these different formulations, turning next to the first of two questions concerning the range of the neutrality principle – questions concerning, in other words, which state actions, institutions, policies or processes must be neutral, and which need not be.

2.2 The range of the neutrality principle: comprehensive or narrow


The first way in which advocates of the principle of state neutrality come to different interpretations of the principle arose out of an ambiguity in the term “rival”. The second, as we shall see, arises out of the fact that the word “state” can be understood in different ways.

One might use the term “state” in discussions of the neutrality principle to apply strictly or primarily to the constitutional essentials of a society – those ‘fundamental principles that specify the general structure of government and the political process: the powers of the legislature, executive and the judiciary; the scope of majority rule’ and those ‘equal basic rights and liberties of citizenship that legislative majorities are to respect: such as the right to vote and to participate in politics, liberty of conscience, freedom of thought and of association, as well as the protections of the rule of law,’ to use Rawls’s formulation.54

And one might, in accordance with the use of the term “state” in this narrow sense, take the principle of state neutrality to apply only to such essentials, as do, for example, Rawls and Brian Barry,55 understanding the essential principles and rights to function like the rules of a game within which individuals and interest groups may legitimately attempt to promote their conceptions of the good. This view of the range of the principle I will refer to as the narrow neutrality principle.56


Principles and rights such as these do not, of course, exhaust the business of legislation. Much governmental activity involves making laws in contexts and with respect to matters less fundamental than those listed above, and which specify in a more fine-grained manner the rights and duties of citizens and their organisations. To many, if not most, the term “state” is just as applicable to legislation other than the constitutional essentials, and so, therefore, is the principle of state neutrality. This view we will call the comprehensive neutrality principle,57 and it rules out the promotion of the good in any governmental context.

It goes (almost) without saying that the reason why state neutrality of any range is required is that the state has immense influence over the lives of citizens: whatever reason one has for thinking a non-neutral state to be an evil assumes in the first place that the state has significant influence.

Furthermore, given that the state affects our lives primarily by means of legislation, and that constitutional provisions determine the nature and limits of other laws rather than vice versa, it is in the provisions of the constitution, rather than in the provisions of less fundamental legislation, that the state exercises its greatest influence. This is why the parties to the debate on this aspect of the range of the principle of state neutrality are divided in the way they are; this is why, in other words, there is no “the constitution can be partisan, but wider legislation must be neutral” party. The great influence of the constitutional essentials means that no one who was discriminated against by the constitution on the basis of their adherence to a particular conception of the good is likely to be satisfied by the reassurance that that all other legislation was neutral (were such a dispensation to be possible, which is doubtful), whereas one might certainly draw some comfort from a neutral constitution in cases where one’s conception of the good life was disadvantaged by the policies of the government of the day.


Rawls expresses this view when he writes that his aim, in Political liberalism,

is to consider first the strongest case where the political questions concern the most fundamental matters. If we should not honour the limits of public reason here, it would seem we need not honour them anywhere. Should they hold here, we can then proceed to other cases.58


We might sum this up by saying that if anything ought to be neutral, it should be the constitution.

Richard Arneson has suggested that ‘something approaching a consensus has formed around…[the view that state neutrality]… applies not to each and every policy the state pursues, but only to constitutional essentials and basic justice, or the principles that regulate the basic structure of society’59; around the principle of narrow neutrality, in other words.

As examples of important consensus-makers of this kind he cites Rawls, who writes that


 [o]ur exercise of political power is fully proper only when it is exercised in accordance with a constitution the essentials of which all citizens as free and equal may reasonably be expected to endorse in the light of principles and ideals acceptable to their common human reason60 

and Barry, who writes that ‘nobody is to be allowed to assert the superiority of his own conception of the good over those of other people as a reason for building into the framework for social co-operation special advantages for it’61 and that ‘at the point where basic principles and rules are being drawn up, no conception of the good should be given a privileged position.’62


Accepting the narrow principle would mean accepting that just as a neutral constitution does not forbid citizens from promoting the good in their individual capacities, it might also, under specified circumstances, allow some functionaries of the state to promote the good. This it might do, for example, by allowing local institutions to promote certain goods, while insisting that national institutions refrain from doing likewise. It should then be possible to interpret a constitution which conforms to the neutrality principle to allow, say, local educational institutions to decide which excellences or virtues their curricula will attempt to inculcate in schoolchildren, local governments to fund galleries and museums and sporting events of a particular nature, and so on. In all such examples, however, the important point to remember is that peripheral legislation, as we might call it, would be to some extent analogous to private action, in that it would be a sphere in which the good may be pursued within the limits set by neutral principles of right, enshrined in the constitution.

One might, of course, ask why the domain of political morality should be split into two like this. Is it not simpler, and more in accordance with liberal tradition, merely to distinguish between private morality and political, or public, morality? And if we do this, and if we think that the principle of state neutrality applies to political morality, then surely it applies to the whole of political morality?

Put very roughly, the problem seems to be that comprehensive neutrality seems too much, even for neutralists. (Obviously it is also too much for perfectionists, but they can see this as a reason to call the appropriateness of neutrality at any level into doubt.) Working out why writers like Rawls and Barry take the position they do involves a certain amount of speculation, as, as can be seen in the passages quoted above, both appear to favour the narrow principle out of caution; both passages express the suggestion that at least neutrality in the constitutional sphere can be defended, as if defending comprehensive neutrality would take them into treacherous territory, but should not in principle be ruled out. This way of proceeding cannot be straightforwardly taken as an argument for narrow, as opposed to comprehensive liberalism, but appears rather to be a strategy for defending neutrality per se.


Barry does offer hints as to why he adopts the cautious attitude when he writes that ‘public policy [not constitutional matters] will in many matters reflect some conception of the good,’ and further, to illustrate his point, that

 [i]t would be absurd to suggest that there is some way of determining a [public school] curriculum that is neutral between all conceptions of the good, and it is significant that those who support the idea of legislative (as against constitutional) neutrality have never attempted to lay out a neutral curriculum.63


This passage seems best interpreted to mean that it is impractical to demand that all essential state activity,64 at all levels, conform to the neutrality principle. Hence the cautious defence of neutrality; if its ambitions are scaled down, restricting themselves to the basic essentials, the accusation that neutrality of any kind is unachievable is less likely to stick. The practicability of a constitution which does not promote any particular conception of the good is at least prima facie plausible, regardless of the plausibility of an entire legal apparatus which is neutral in this way.

Is this a strategy that neutralists ought to follow? As we saw in the previous section,65 answering one question about the requirements of the neutrality principle may require already having answers to other questions about the meaning and justification of the principle. In the case under discussion, we need to know whether to interpret state neutrality to mean neutrality of effect or neutrality of justification; whether, in other words, a neutral state is one which ensures that nothing it does leaves anyone any worse off as a result of their adherence to any particular conception of the good, or whether the neutral state is one which does not to appeal to any particular conception of the good in justifying its policies, whatever the effects of these policies might be.66

Adopting the neutrality of effect interpretation does indeed render implausible the idea that the entire corpus of legislation, as opposed to the constitutional essentials, could be neutral. This is because ensuring that no one’s fate is adversely affected as a result of the effect any law or policy has on the adherents of any particular conception of the good would surely paralyse the state. Hence anyone who thinks that neutrality must be understood as neutrality of effect must prefer the narrow neutrality principle (if they are willing to stick with the principle of state neutrality at all), as requiring that nothing in the constitution leaves anyone worse off on account of their adherence to a particular conception of the good looks somewhat more achievable. (Whether one ought to adopt the neutrality of effect interpretation is another matter, of course; one I discuss in the section below entitled “Favouring a conception of the good”).


On the other hand, if the neutrality principle is understood as requiring simply that the state not appeal to any conception of the good in making law and policy, then comprehensive neutrality looks more plausible. Richard Arneson points out,67 in response to Barry’s claim that any education must appeal to the value of certain activities or experiences in comparison to others, that it is surely possible to justify educational goals in neutral terms. As Arneson puts it, ‘we can appeal to the idea that it is fair that every person have…opportunity to attain some reasonable threshold level of literacy and mathematical competence, and run public schools on this basis.’68 Furthermore, if one endorses Larmore’s position that state neutrality is not violated by legislation which promotes uncontroversial conceptions of the good, then it is even simpler to imagine how a policy of comprehensive state neutrality might successfully be applied to, for example, public school curricula.69

In the absence of compelling examples of the impossibility of justifying legislation neutrally,70 narrow neutralists face the accusation that there is something arbitrary about failing to extend the requirement of neutrality beyond the constitutional essentials. If there is a good case for state neutrality, and if it is just as feasible for legislation to be neutral as it is for the constitutional essentials, what grounds could there be for permitting state perfectionism on the legal periphery? This is not a question to which Rawls and Barry give further attention – indeed, one might expect Rawls’s response to be simply that peripheral legislation is not the focus of his work.

But perhaps there are such grounds – grounds for regarding peripheral legislation as sufficiently unimportant for state neutrality such that the state may legitimately promote the good by means of it. Obviously, no neutralist case for the narrow principle can succeed without first establishing that there is a case for any version of the principle of state neutrality. What the neutralist then has to do is explain why the reasons for adopting the neutrality principle in the first place – be they the importance of personal autonomy, the importance of stability, the impossibility of knowledge of the good, or any of a number of others – demonstrate the necessity of a neutral constitution or neutral principles of basic justice with that much more urgency than they do the necessity of neutral legislation.


Whether this is possible is a matter I will discuss in chapter two, when I come to discuss the cogency of the various arguments for neutrality in general. For the moment it suffices to note that neutralists divide into comprehensive and narrow neutralists, and that deciding which of the two strategies is the more promising depends, as with so much in this debate, on what reasons one has for adhering to the neutrality principle in the first place. In what follows I continue to survey the ambiguities in the formulation of the principle, turning next to discuss whether the neutralist state must refrain only from using coercion to promote the good.

2.3 The range of the neutrality principle: only coercive promotion of the good?

Another way in which the range of the principle of state neutrality is open to different interpretations concerns the issue of coercion; in particular, whether the principle requires the state to refrain from acting on the basis of a particular conception of the good only in cases where doing so involves coercing citizens, or whether neutrality is required even when the policy in question involves no coercive measures.

It is hardly surprising that anti-perfectionism and a particular concern with coercion are linked in the work of many writers; the state is distinguished from other actors in liberal71 political theory by its monopoly72 on the legitimate use of force,73 and there is a certain symmetry in holding that the state’s being distinguished in this way entails it being distinguished by the reasons for which it may act (or the reasons for which it may not act, in this case) as well. One might put it this way: for many anti-perfectionists, the state’s acquiring the right to coerce means it must lose the right to promote the good.


But if it is concern with the state’s right to coerce which is the primary impulse in ruling out perfectionist legislation, such neutralists must work out what to say in cases where the state appears to exercise its power without coercion.

Clearly, any state frequently resorts to coercion so as to ensure compliance with the law; any cases where it uses force or threatens to use force to compel citizens either to do or to refrain from doing something count as coercion. And neutralists are agreed that the state may not use such methods either to promote the good or to forbid the bad.74

But are all exercises of state power exercises in coercion? Citizens who accept the legitimacy of a law can certainly not be said to have been coerced into compliance with it. And although they may be aware that violations will result in punishment, in those cases where the legitimacy of the legislation is wholeheartedly accepted fear of punishment is not among their reasons for compliance.75 Furthermore, if we accept Raz’s stipulation that, for a threat to be a coercive one, it must be a serious threat, citizens who comply with a law because they wish to avoid the minor inconveniences that certain punishments would bring upon them – small fines, say – can also not be said to have been coerced into compliance by the state.76 Thus the promulgation of legislation which is complied with because citizens accept its legitimacy, or because they fear punishment which is not sufficiently serious for the threat of it to count as a case of coercion, cannot be described as an example of the state exercising its power coercively over those citizens.


We can conclude from this that the case for the legitimacy of non-coercive legislation of this kind – legislation that is non-coercive because it is accepted as legitimate by citizens – is the same as the case for the compatibility of the promotion of uncontroversial conceptions of the good by means of the law with the principle of state neutrality, as discussed in the section above entitled “Rival conceptions of the good”. For legislation which appeals to non-controversial conceptions of the good would not coerce citizens into compliance, regardless of what punishments were threatened for violations. And so those neutralists, such as Larmore, who are attracted to the view that state neutrality requires merely that the state refrain from promoting controversial conceptions of the good have reason to accept that non-coercive promotion of the good does not violate the principle of state neutrality.

The categories of non-coercive perfectionist legislation and non-controversial perfectionist legislation clearly overlap. But might there also be a category of controversial non-coercive perfectionist legislation? And, if so, what ought neutralists to say about it?

Not all exercises of state power are obviously coercive. A state may exercise its power through the medium of offers as well as threats. It may reward certain kinds of behaviour so as to promote the good. And it can, of course, reward those who pursue a controversial conception of the good, so as to promote this good.


One way of doing this would be to offer tax breaks to those who engage in or support favoured activities. Another would be simply to subsidise these activities. The state may also accord privileged legal status to forms of life which are considered valuable, as most contemporary constitutional democracies do when they agree to enforce certain kinds of contracts – monogamous marriage contracts, for example – and refuse to enforce others – agreements to sell body parts, and the like – and in so doing, aim at encouraging citizens that might not otherwise have done so to commit themselves to these putatively valuable forms of life and/or avoid those which are not valuable. Furthermore, the state might also use advertising or other public relations techniques so as to encourage valuable forms of life or discourage worthless ones. Such policies do not appear to coerce citizens into pursuing the good, and they certainly do not forbid citizens from pursuing the worthless, and so, it would appear, there is indeed a category of controversial non-coercive perfectionist legislation.

Waldron has suggested, however, that even if these allegedly non-coercive methods of promoting the good do not literally force one to get married or to take up painting, the distinction between coercive state action and non-coercive state action is illusory. As he puts it, ‘[the state‘s] supremacy ultimately springs from its command of considerable means of violence.’77

If Waldron is right, then any difference of opinion on what the principle of state neutrality requires of the state in cases where its power is exercised non-coercively is of little interest. But must we believe that methods such as subsidising, promoting, or according privileged legal status to, valuable forms of life are, despite their innocuous appearances, in fact examples of the coercive use of state power?


Waldron’s argument boils down to this: all of these state activities depend on the state’s collection of taxes. Given that payment of taxes is compulsory, and that failure to do so is punishable, frequently severely, those who do not wish to subsidise the allegedly valuable options for the benefit of which the taxes are collected are plainly coerced into supporting them. To the extent that they are controversial, then, they are also coercive, just as with more obvious forms of coercion.

Even in those cases where all the state does is establish special status for certain forms of life, as it does when it recognises monogamous marriage, Waldron argues, the laws establishing this status are what he calls ‘fragments of a legal system which is itself coercive,’ for the reason that, although they do not compel citizens to restrict themselves to any particular type of sexual relationship, let alone compel them to marry, they do affect greatly matters such as who has a right to what property, manner of financial support, and tax benefits, and other such issues, and these rights are upheld by coercive sanction,78 and in this way do indeed rely, ultimately, on the threat of coercion so as to promote the good of monogamous marriage.

A good deal of Waldron’s argument turns on the idea that what is actually problematic about coercive perfectionism is that it undermines the autonomy of citizens; for this reason even if his claim that all apparently non-coercive methods of promoting the good are ultimately coercive can be rebutted, he may still argue that their manipulative nature is sufficient for them to undermine citizens’ autonomy, and therefore for the principle of state neutrality to rule them out.


Furthermore, Raz, a leading defender of state support for valuable forms of life, but who, despite identifying himself as a perfectionist, nevertheless expresses adherence to Mill’s harm principle,79 concedes that, to the extent that the state relies on compulsory taxation to raise the funds necessary for subsidising valuable forms of life or according them privileged legal status, such strategies for promoting the good are coercive.80 Raz, like Waldron, thinks that the important issue is not whether such strategies are coercive or not, but rather whether they damage autonomy, and the key difference between his position and that of Waldron is that Raz thinks that such subsidies can in fact promote autonomy.

This is a matter to which I will shortly return, but for the moment I wish to remain focused on the question of whether all exercises of state power are necessarily coercive, given the guns lurking behind those subsidies for artists. This is because there are grounds for thinking they aren’t, whatever Raz concedes (and Waldron asserts).

Even if we concede that compulsory taxation does coerce those who do not share the conceptions of the good tax money is used to support, we do not have to accept that all uses of the state’s fiscal powers are coercive. This is because the state may promote the good by the use of tax breaks, as well as by collecting voluntary tax payments.


In cases of the first kind the state might, for example, make donations of certain kinds tax deductible and others not. Companies may then be faced with incentives to donate large sums of money to, say, artistic foundations, and in this way would be encouraged themselves to subsidise valuable forms of life. In cases of the second kind, the state might, as it in fact does in Germany, collect voluntary tax payments (effectively donations) on behalf of a small number of established religious bodies, presumably on the grounds that the practices these bodies promote constitute valuable forms of life.

Perhaps a case similar to that made by Waldron for the view that state support for monogamous marriage is ultimately coercive could be made against examples of the first kind; perhaps, in other words, one might establish that laws permitting such tax breaks can, ultimately, also be seen as fragments of a coercive legal system. One might do this by pointing to the web of legal relations in which such tax breaks are entwined, some of which would, ultimately, take the form of coercive prohibitions on certain uses of property etc. But it is quite implausible to classify the collection of voluntary church taxes as coercive in this way, at least if the performance of these collection duties is the only way the state privileges these particular religious bodies. So we do have at least one, and possibly two, examples of non-coercive perfectionism.

The question, of course, is what this conclusion implies for the debate between those neutralists who accept the legitimacy of non-coercive perfectionism and those who do not. One, obvious, implication, is that there is indeed a difference between the two parties; there is indeed something to fight about. But one may be forgiven for thinking that, unless a plausible rebuttal for Waldron’s argument that even subsidies and privileged legal status for valuable ways of life constitute coercion can be found, there is not a great deal to fight about. We’re left with the difference between a state that collects voluntary taxes on the behalf of valuable institutions and one which doesn’t. We may still ask, of course, which of these two views which serve as the basis for the fundamental political principles these two states adhere to squares best with the principle of state neutrality.


This depends, as with the other ambiguities in the neutrality principle discussed in this chapter, on what one’s grounds for adopting the neutrality principle in the first place are. And this is where we return to the question of autonomy, given its prominence as a basis for commitment to neutrality.81

Clearly, taking autonomy seriously means objecting to coercion. As Raz puts it, ‘[t]he contribution of autonomy to a person’s life explains why coercion is the evil it is.’82 But would non-coercive promotion of the good be any more acceptable?

Some writers, such as Waldron, suggest that any promotion of the good shows disrespect for citizens. He argues that by increasing the costs attached to pursuing the worthless, or decreasing the costs attached to pursuing the valuable (or both), the state prevents citizens from making decisions about the direction their lives ought to take on the merits of the options before them. Making decisions on the merits is a necessary condition of rationality, and rationality is, on most readings, a necessary condition of autonomy. So when the state promotes the good, argues Waldron, it makes the decision on behalf of citizens, or, to put it more moderately, it goes some way toward making the decision for them. In doing so, it fails to respect their autonomy.83


We might, as Waldron does, characterise such policies as manipulative, and, as such, just as damaging to autonomy as coercion. If, further, one adheres to the neutrality principle on the basis of the importance of autonomy, one will obviously incline towards the view that neutrality rules out any promotion of the good, as opposed to merely the coercive ways of doing so.

Other writers, however, including Raz84 and Sher,85 have called into question the view that subsidisation, or the according of privileged legal status to valuable forms of life, or even what Sher calls ‘the manipulation of the nonrational determinants of preference’86 are necessarily opposed to respecting citizens’ autonomy. Assessing the plausibility of the arguments of Raz and Sher would, however, take us into territory reserved for chapter two, namely, the question whether the principle of state neutrality ought to be adhered to at all.

For the moment it suffices to say that accepting that manipulation and/or coercion do not necessarily disregard the autonomy of the one manipulated or coerced is unlikely to leave one in the neutralist camp – at least not on the basis that neutrality is necessary for autonomy. If, on the other hand, one thinks that the principle of state neutrality can be derived from a commitment to personal autonomy, one is probably going to accept that non-coercive – manipulative, in other words – promotion of the good is enough of a threat for it to be proscribed.


Of course a commitment to autonomy is not the only reason why one might adhere to the neutrality principle. One might, for example, be a moral sceptic, and argue that seeing as there can be no reasons in favour of any particular conception of the good, the state is obliged to be neutral.

Although very few philosophers have taken this line of argument seriously,87 it still has sufficient life outside of philosophical circles. So we may note here that, assuming it is possible to take a consistent position of this kind, moral scepticism would not incline one to think that non-coercive perfectionism was any more justifiable than coercive perfectionism (assuming it was the only basis of one’s commitment to the neutrality principle). Opposition to the one would in all likelihood mean opposition to the other.

One might, somewhat more plausibly, think that state neutrality follows for what Sher calls “prophylactic” reasons;88 reasons such as the fear(s) that, in general, perfectionist governments run a higher risk of oppressing citizens, bringing about civil strife, or making citizens’ lives worse than neutral governments do.


Assuming for the moment that some such arguments succeed, we can say that where such neutralists would fall on the issue of whether the principle of state neutrality permits non-coercive promotion of the good depends on the role they believe non-coercive perfectionist laws play in bringing about the unfortunate consequences perfectionism in general is said to give rise to. Some such neutralists might agree that the principle of state neutrality permits non-coercive perfectionism and others might not, but whichever way they go depends on (quasi-)empirical claims about the effects of perfectionism.89

So, for example, one who thinks that perfectionism heightens the danger of tyranny, or the danger of those in power acquiring excessive influence or wealth, would need to assess whether a government which permitted only non-coercive promotion of the good would run a greater risk of fostering these evils than a neutral government. If the answer is yes, and fear of oppression is indeed the sole (or primary) impulse behind this commitment to the principle of state neutrality, then clearly all forms of perfectionist legislation must be proscribed. And so it goes, on similar lines, for arguments for neutrality from fear of instability or from the fear that perfectionist laws make the lives of citizens worse than neutral laws would do. We will look at the plausibility of such arguments in detail in chapter two, and this will enable us better to decide whether non-coercive promotion of the good really is compatible with the principle of state neutrality or not.

2.4  Favouring a conception of the good 

I wish now, however, to turn to what is perhaps the most important ambiguity in the formulation of the principle of state neutrality: what it means to say that the state favours a particular conception of the good.


Let us assume that the political arena can be understood as a conflict-ridden realm where the conceptions of the good endorsed by individuals battle for influence. Neutralists argue, unlike perfectionists, that the state must, in the face of these conflicts, ensure that no conception of the good gets any special favours, just as an unbiased sports referee favours no team.

Referees, of course, are frequently accused of bias. And states regularly face the same accusation. So what exactly is it that they are said to have done, or failed to do, when their neutrality is questioned?

2.4.1 Neutrality as neutrality of aim 

First, it is obvious that a referee who aims to help one side is not neutral. We might draw from this rough starting point the provisional conclusion that neutrality is a quality of the aims of those individuals or bodies entrusted with the kind of adjudicatory role referees and states are entrusted with (at least in liberal theory). We might conclude that to be neutral – in other words, to avoid favouring any side – is to act without any intention of helping or hindering either side.


In sport, however, even games officiated by the most scrupulously neutral referees usually end up with a winning side and a losing side. Likewise, certain practices, ideals, or forms of life, may “win” or “lose” under the auspices of a state which purports to be neutral, and where this neutrality is taken to mean that it aims neither to hinder nor to help any of them. The influence of, for example, particular religions or art forms may wax and wane. If neutrality is a matter of aim, we are not entitled to conclude from these “victories” and “defeats” that the state lacks neutrality. For just as the neutral referee is not obliged to ensure that every game ends in a draw, neither is the state obliged to ensure that all conceptions of the good fare equally well.

What the neutral referee is obliged to do – and this is what his neutrality, on the neutrality of aim view, consists in – is, by dint of the unbiased application of the laws of the game, ensure that only the deserving side wins, not the team that best secures for itself unjust advantages, as it might by cheating in various ways (which include receiving favours from the referee). Likewise, the neutral state is obliged to ensure, by the impartial application of the law, that only conceptions of the good which are entitled to do so90 gain influence, and not those ideals which are spread, say, by violence, or indeed by soliciting help from the state to facilitate their success.

Neutrality, in its core sense, as Waldron reminds us,91 implies keeping out of a quarrel between two (or more) other parties. But this is not, strictly speaking, what either a neutral referee or a neutral state does, given that they do intervene. This means that there must be something about the rules neutral adjudicators apply which allows us to say that neutrality consists in their impartial application, despite the fact that they have differential effects on different forms of life (or on different sporting virtues and vices). This feature is elusive, but it is frequently identified as the fact that the rules enforce the right (as opposed to targeting any particular conception of the good life), or, sometimes, that their aims are independent of any such partisan intentions.


The latter is Nozick’s position, which he outlines in a passage where he ridicules the suggestion that a state cannot be neutral if it prohibits rape, given that such a prohibition disproportionately penalises men.92 He concedes that such a prohibition does penalise men in this way, but argues that it cannot be construed as a non-neutral piece of legislation because targeting men (or even frustrating that class of men who are attracted by the prospect of raping) is not, in any known case, the aim of prohibiting rape. In saying that there are independent reasons for prohibiting rape he means there are (good) reasons for doing so which have nothing to do with the desire to privilege any particular conception of the good.

The sporting analogy with the prohibition of rape might run as follows: Forbidding punching on the sports field – assuming the sport in question is not boxing – has nothing to do with the desire to penalise whichever team happens to have superior pugilistic skills. There are good, “independent,” reasons for forbidding such actions, and so the referee who punishes the violent player who throws a punch is not thereby demonstrating his bias against violent players, or anyone else.

A similar argument is made by defenders of the neutrality of aim conception with respect to cases which do not involve prohibitions, but where policies with neutral aims lead to the decline of a particular form of life. Will Kymlicka presents the example of a state which promotes the English language at the expense of others for the reason that, in the state in question, English is the most common language, and that communication is therefore improved by everybody having a command of the language.93 Other languages might well begin to die out under such circumstances, and this may be experienced, at least by the last generation that speaks the language, as a cultural loss, but, on the neutrality of aim view, as long as the legislation does not aim at damaging the other languages, it cannot be construed as a violation of state neutrality.


Similarly, one might expect that cultures whose survival requires a certain amount of pressure on their members to conform – “closed” cultures, we might call them – are likely to suffer a mounting loss of membership and influence in societies where standard democratic freedoms are enforced. Members can no longer be pressurised to remain within the community, and, as a result, community practices fall into decline. This can be seen in the history of various religious and linguistic groups which have not made the transition to modernity with success. Here the sporting analogy would run as follows: A neutral application of the laws of most football codes will likely result in victory for the young, the strong, the fast, and the intelligent. This does not mean that the referee is biased against the old, the weak, the slow, and the stupid.

Now whereas no one (or almost no one) would be sorry to see the practice of rape decline, we might regret the demise of closed cultures under a neutral dispensation, a little as we might be sorry to see the humiliation of an ageing sports star by a new generation of players. As Rawls puts, ‘We may indeed lament the limited space, as it were, of social worlds, and of ours in particular; and we may regret some of the inevitable effects of our culture and social structure.’94 But we cannot, if we understand state neutrality as neutrality of aim, regard these differential effects as a sufficient condition for the attribution of bias to the state, and we cannot, therefore, require that the state intervene to restore the status quo ante (at least not on the grounds that neutrality requires this) any more than we think that a referee demonstrates impartiality by bending the rules to help the old star hold his own against the younger.

But is neutrality of aim how we should understand what it means to favour a particular conception of the good? On the one hand, any idea of neutrality which requires the state to be neutral between those who believe that raping others is part of the good life and those (everyone, rapists included, one would think) who hold that being raped makes their lives worse is clearly absurd; it is unclear what the attractions of neutrality are if they include a moral vision which refuses to discriminate between any ways of life. On the other hand, one might also think, as a speaker of a non-English language in the state of Kymlicka’s example, that any alleged neutrality in a state which takes steps which it knows will in all likelihood lead to the extinction of one’s culture is something of a sham.95 One might wonder, under these circumstances, why the intentions or aims of state actors are more relevant to the state’s neutrality than the fact that certain sectors of the population whose way of life is not illegitimate in any way are disadvantaged in ways fully foreseeable at the time of legislation by the laws these state actors make.


That one is inclined to wonder this kind of thing does not automatically mean that one is justified in wondering it, of course. But there certainly is reason to be dissatisfied with using the neutrality of the intentions or aims of state actors as the measure of the neutrality of a state. This is because of the mysterious nature of such intentions. Ascertaining the real intentions of individuals is difficult enough; they are frequently confused or unknown, and some would even suggest that they are in principle unknowable, although this seems exaggerated. But when we move to the intentions of institutions or corporate bodies, the problem multiplies beyond comprehension. It is genuinely unclear that we can talk, with any degree of sense, of collective intentions.

Furthermore, even if motivations were reasonably discernable, they are hardly ever pure. The English-speaking lawmaker may genuinely aim to improve communication throughout the land, while simultaneously wishing to damage the prestige of the languages he does not speak. How do we judge whether the legislation he helps, on the basis of these motivations, to promulgate is neutral – especially if he acts in concert with other lawmakers who are motivated in yet other ways (assuming one can ascertain what moves them all in the first place) – if the purity of his intentions is our yardstick?

2.4.2 Neutrality as neutrality of justification 

As a result of these difficulties contemporary neutralists generally do not formulate the principle of state neutrality such that state neutrality means neutrality of aim on the part of state actors. Some, such as Robert Goodin and Andrew Reeve, argue that, as a result of these difficulties, neutrality must be defined in terms of results, as they put it.96 We come to look at this option in the section entitled “Neutrality as neutrality of effect” below. But for the moment let us ask whether neutrality as neutrality of aim can be reformulated so as to shed the problem of the unknowability of the motivations of individuals and institutions, while retaining its intuitive appeal.


The most popular formulation of the principle of state neutrality, with respect to how the concept of favouring a conception of the good is to be construed, is what we will call the neutrality of justification interpretation. Larmore, for example, writes that

 [P]olitical neutrality consists in a constraint on what factors can be invoked to justify a political decision. Such a decision can count as neutral only if it can be justified without appealing to the presumed intrinsic superiority of any particular conception of the good life. So long as a government conforms its decisions to this constraint, therefore, it will be acting neutrally. There is no independently describable condition of society to be called “neutral” that the ideal of political neutrality requires a government to promote or maintain.97


Similar formulations can be found in the writings of Ackerman,98 Rawls,99 Sher100, and Waldron.101 The great advantage in replacing the aims of the state with the justifications put forward by a state for its actions is that these justifications are more readily discerned – they are in the public realm, so to speak – than its aims, if it can even coherently be said to have any.

The neutrality of justification interpretation understands neutrality to be a rule about what kinds of reasons the state may legitimately offer – over and above the normal requirements of reasonable plausibility – in justifying the legislation or policy it puts forward. This rule specifies that if a piece of legislation or policy cannot be justified except by appeal to its role in promoting a (controversial) conception of the good life, it is non-neutral, and therefore illegitimate, regardless of whatever other merits it might have.

So, to return to Kymlicka’s example, if the state were unable to justify its promotion of the English language at the expense of other languages on the basis that this would improve communication (or on some other neutral basis) then such legislation would not be neutral. In other words, if it turned out that promotion of English would not improve communication, or that the improvement of communication was not necessary, or did not outweigh reasons for promoting other languages equally, then the proposed legislation would be in trouble; an alternative neutral justification would have to be found. If this were not possible – if it were only possible to justify the legislation by appeal to the alleged superiority of the English language102 – then it could not be promulgated by a neutral state.


But, just as is the case with the neutrality of aim interpretation, the mere fact that the fortunes of forms of life, languages, cultures, religions and so on are differentially affected by neutrally justified legislation does not in itself call the neutrality of the state into question. Questions of justice, or questions of equality, may arise, but not questions of neutrality.

The appeal of neutrality as neutrality of justification is, on the one hand, that it accords with the core sense of neutrality as keeping out of a conflict by presenting us with the image of the state as referee, impartially adjudicating the rules of the contest between conflicting visions of the good life, letting the chips fall where they may, while, on the other hand, avoiding the obvious difficulties of neutrality of aim. It is unsurprising that the majority of writers on the topic adhere to it.

But, as we noted in introducing the question of the formulation of the principle of state neutrality, this consensus is not universal. Here the sporting metaphor is useful again, this time in that the point where it breaks down indicates why one might be unhappy with neutrality of justification as the correct interpretation of the principle of state neutrality. Sport, as a practice, unashamedly sets out to privilege certain (sporting) virtues, and the rules of most sports are set up precisely to allow these virtues to emerge triumphant over their corresponding vices. There is a point to the various sports, in other words. So, while a neutral referee favours no team, his interventions discriminate intentionally against certain qualities of teams – their age, their weakness, their slowness, their stupidity – for the reason that the laws of the game discriminate against these qualities. In fact one of the features of his neutrality is precisely his refusal to compensate for defects losing sides display. Pity is not neutrality!


But this is surely not how (most) liberals want to understand the laws of society. Sport is a circumscribed arena, and, largely, not a matter of life and death.103 In general, liberals – especially neutralist liberals – do not understand society as a forum for the triumph of virtue over vice. It is the fact that the individuals within a society do not agree on what the most important virtues and vices are that leads many liberals to advocate state neutrality in the first place. In this case a neutral referee looks more like one who adjudicates between different visions of virtue and vice, not one who impartially applies rules which are designed to allow the virtues to blossom unimpeded by the distractions of vice.104 So whereas the team that loses on the sports field does not, merely in virtue of the fact that it has lost, have a complaint against the neutrality of the referee,105 it may be that those whose conception of the good “loses” – whether we mean by this simply that it loses ground within society or, alternatively, that it is proscribed outright – do have such a complaint.

Raz has indicated precisely why such a complaint might be justified; because, he argues,106 the neutrality of justification interpretation excuses a nation which sells arms to one party during a military conflict and not the other, provided it (genuinely, one assumes) publicly announces profit to be the justification for the sales – profit, presumably, being a neutral, or independent, as Nozick would have it,107 reason for action. It is hard to see why the party to which arms are not sold ought to regard the nation which sells them as neutral – interdicting such shipments would seem perfectly justified – and yet it is also quite unclear why the profit motive should be understood, in general, as a partial or biased reason for action. The point is that it is biased in this case. And it is biased in this case, presumably, because of the effect it has on the conflict between the two warring parties. So a state that was genuinely neutral between all visions of the good might have to ensure after all that, say, languages other than English prosper, whatever the virtues of English as an agent of communication, and perhaps even ensure a broader “neutral” balance of power between various conceptions of the good life within society.

2.4.3 Neutrality as neutrality of effect 

These considerations suggest an alternative to the neutrality of justification view; the so-called neutrality of effect interpretation. It can be formulated in a number of ways. Wall and George Klosko put it as follows: the state should not do anything that has the effect – whether intended or not – of promoting any particular conception of the good or of providing greater assistance to those who pursue it.108 Raz presents a variant formulation which he claims is a paraphrase of the view taken by Rawls in A theory of justice, and which holds that if a state is to be neutral, it must treat ensuring for all persons an equal ability to pursue in their lives and promote in their societies any ideal of the good of their choosing as a goal which is lexically prior to any other.109


How would state neutrality of this kind work? The first point to note is that Raz’s formulation suggests that Larmore errs in commenting that there is no independently describable condition of society to be called neutral.110 There is; it is the state in which every citizen has an equal ability to pursue in their lives and promote in their societies any ideal of the good of their choosing. This is not a constraint on state decision-making, but an ideal state of affairs which the state is encouraged to bring about. Wall and Klosko’s formulation, on the other hand, does read like a constraint on decision-making, albeit of a different kind to that of the principle of state neutrality understood as neutrality of justification. Unlike Raz’s formulation, Wall and Klosko’s does not hold up any particular balance of power amongst conceptions of the good life as ideal, but merely forbids the state from altering the current distribution. The state may not do anything which tips the balance in any direction, but, apparently, it has no obligations to prevent the balance from tipping of its own accord, so to speak.

One might read Wall and Klosko’s version as an additional constraint on state action; additional to the constraint set out in the neutrality of justification formulation. On this reading, then, the state must not only refrain from justifying laws by appeal to (controversial) conceptions of the good, it must refrain from enacting neutrally justified laws which have the effect of promoting particular (controversial) conceptions of the good. (One might also imagine a version which allowed the promotion of the good under certain circumstances, provided that those disadvantaged were compensated.)

Alternatively, one might also read Wall and Klosko’s version as a separate constraint from the neutrality of justification formulation; it might rule out only laws which promote the good, whether intentionally or not, and have nothing to say on the matter of justification. Whether this is an interesting distinction or not depends on whether one thinks there could be laws which are justified on the basis of an appeal to a (controversial) conception of the good life, but which do not have any effects on the balance of power between conceptions of the good life. Although this sounds unlikely, one might wonder whether state support for churches in Europe fall into this category.


Raz’s formulation, however, appears to require the state to intervene where naturally occurring changes in the balance of power happen – perhaps even justified changes, such as those that are analogous to the way in which a superior sports team takes control on the field.111 As if Wall and Klosko’s formulation were not demanding enough! A state which is committed to ‘ensuring for all persons an equal ability to pursue in their lives and promote in their societies any ideal of the good of their choosing’ needs to pay close attention to the social mechanisms whereby citizens are prevented from pursuing and promoting ideals of the good, as well as those mechanisms whereby those choices are made in the first place.

Cultural, artistic, and religious choices – indeed choices of any kind – are heavily influenced by the extent to which they are in tune with the choices of others within a society. Choosing to pursue a minority religion is not, under any circumstances, as easy a choice as pursuing a majority one, even if it is not an impossible choice. (It may, of course, even be a choice that brings special rewards). One might be forgiven for thinking that many of these mechanisms are beyond the state’s control, and this is indeed the point of view of a number of commentators.112 Arguing that an ideal is unachievable is not necessary a damning indictment of that ideal – if that were so we could reject all theories of justice without bothering to consider their details – although the degree of difficulty involved in achieving it is a relevant consideration, as an ideal to which no approximation could be achieved is indeed one which ought to be rejected.113

Perhaps it is possible for a state to achieve the kind of neutrality envisaged in Raz’s formulation by maintaining the required balance in limited social areas, and it is quite probable that Wall and Klosko’s less demanding formulation could be followed, given a sufficiently restricted field of play. But the more worrying consideration with regard to both these formulations is the question why neutrality of this kind ought to be an ideal at all.


The idea that no one should find implementing their freely chosen conception of the good life more difficult than anyone else finds implementing theirs is clearly related to the ideals of equality and justice, but it is not in itself either an ideal of equality or an ideal of justice, and its relation to the core sense of neutrality is also unclear. It is not simply an articulation of the ideals of equality or justice for the reason that citizens might freely choose conceptions of the good which are themselves inegalitarian or committed to injustice; as do the rapist and the Nazi, to name just two examples. And clearly neither equality nor justice requires that rapists and Nazis find implementing their ideals as easy as anybody else does (although, as Raz suggests, rapists and Nazis might nevertheless, on the neutrality of effect interpretation, be owed compensation114).

Furthermore, requiring the state to intervene in people’s lives in the millions of ways necessary for maintaining equality between the countless (and overlapping) conceptions of the good life citizens adhere to presupposes a puzzling commitment to the equal value of, or the right to survive of, all these conceptions. It seems clear why a citizen might require the state to recognise his or her right to equal treatment or right to life; it is less clear why a conception of the good must be accorded the same (or similar) recognition. And, as we have also noted, it isn’t clear why such a commitment on the part of the state, even assuming it could be justified, should be designated neutrality.

It is for reasons such as these that Rawls distances himself from the ideal of neutrality altogether in Political liberalism, where he writes that ‘the term neutrality is unfortunate; some of its connotations are highly misleading, others suggest altogether impracticable principles.’115 He prefers to work with the notion of permissible conceptions of the good, by which he means those conceptions of the good which can survive in a dispensation in which the state makes no law on the basis of the alleged superiority of any one of them. This position, of course, reimports the (perhaps less worrying) difficulties associated with the neutrality of justification interpretation, but given that Rawls does not purport, in this passage, to be offering a characterisation of neutrality, the ways in which the neutrality of justification interpretation collides with the common understanding of the term are of less concern.


This discussion illustrates the difficulties, noted by Raz and Waldron, which arise when philosophers try to establish the “real” meaning of a term widely used in political (or any other form of) discourse. Raz, in particular, is eager to stress that the philosophical task is to articulate and defend a particular version of the term,116 and this task has been taken on here to the extent that some of the difficulties involved in the various positions described here have been illustrated. But our principal aims at this point are merely to come to an understanding, first, of the options available to the defender of the principle of state neutrality, and, second, of which arguments for neutrality are likely to underpin which formulations of the neutrality principle (regardless of whether or not these arguments are completely sound).

With this in mind, we can note that those who are drawn to the principle of state neutrality primarily on the basis that moral diversity, or individuality, is a good, as Mill is frequently interpreted as having been, are likely to hold that the state must take care that its actions help to bring about the right outcome; namely, some favoured distribution of conceptions of the good life. This need not, of course, be a balance of power in which all conceptions of the good are equally influential, but it nevertheless implies a concern with the effects of legislation, rather than the justifications thereof.

On the other hand, holding that the problem with perfectionism is that it violates equality by subjecting citizens to constraints they could not accept without abandoning their sense of their equal worth, as Dworkin does, leaves one open to formulate the neutrality principle either as neutrality of justification or neutrality of effect; for the claims that citizens’ sense of their equal worth is violated by their having to obey laws which cannot be justified without appeal to a conception of the good they don’t share and that citizens’ sense of their equal worth is violated by their having to contend with (neutrally justified) laws which discriminate against them on the basis of their commitments to particular conceptions of the good are both plausible. In fact Dworkin appears at different times to favour different formulations of the principle of state neutrality; in his discussion of conservatism in A matter of principle he appears to articulate a version of the neutrality of effect interpretation,117 while appearing to rely on the neutrality of justification interpretation during his discussion of the drawbacks of socialist economic decision-making in the same book.118


Dworkin has also been read as rejecting perfectionism on the basis of a commitment to personal autonomy,119 and one might, on this (widespread) basis, understand neutrality to mean neutrality of justification. One might think this because, for example, one thinks that citizens’ choices about how to live their lives – choices which are properly theirs to make – are being made for them by a state which legislates on the basis of a particular conception of the good (whether or not citizens are actually in agreement with that conception) and that this presumption on the part of the state threatens citizens’ autonomy. But a commitment to the value of personal autonomy is equally compatible with the neutrality of effect interpretation; one might think that the decision about, for example, which language to use in the state which promotes English is a decision that citizens are entitled to make for themselves, without the state nudging them in any particular direction, whatever its reasons for doing so.

The position of those who reject perfectionism for what Sher dubs “prophylactic” reasons120 – roughly speaking, for reasons such as the view that state neutrality insures against the state becoming oppressive, the view that perfectionist legislation increases the likelihood of social instability, or the view that the state is especially unlikely to have, or to be able to apply, knowledge of the good – with respect to how to construe the term “favour” could depend on how seriously they view the dangers they perceive state neutrality to ward off. If they view these dangers as extremely pressing, the neutrality of effect interpretation, being stricter, will appeal, assuming these neutralists believe it is practicable enough to serve as an ideal.

Clearly, if one believes that state neutrality is necessary to prevent the state from, say, administering laws cruelly or arbitrarily, using legislation to enrich members of government, or other such oppressive behaviour, one is likely to believe that requiring the state to ensure that every citizen has an equal chance of pursuing or promoting any conception of the good they choose is a stronger antidote to the evils of oppression (or instability or error) than merely requiring the state to refrain from legislating if the proposed law cannot be neutrally justified. A state which is obliged to be neutral in the neutrality of effect sense of the term, especially if this requirement is formulated the way Raz does it, really does have its hands tied.


Of course neutralists might also have reasons for leaving the state freer than this, despite their concerns about oppression and the like, and they might perceive the neutrality of justification formulation as providing a better balance between the constraints deemed necessary to prevent the state from becoming oppressive (or unstable or error-prone) and the room to manoeuvre which a state which is neutral in the neutrality of justification sense has (in comparison to the state which formulates the neutrality requirement it must adhere to in neutrality of effect terms, not, obviously, in comparison to the perfectionist state), and which they take as necessary for achieving other goals.

In summary, we should understand the question of what the term “favour” in the phrase “favouring a conception of the good” means to admit of two possible answers – the neutrality of justification formulation, which ought to be understood as a more plausible version of the neutrality of aim formulation, and the neutrality of effect formulation. As we have seen, both plausible formulations are not unassailable; the weaknesses of neutrality as neutrality of justification concern its apparent indifference to genuinely discriminatory (and even obviously non-neutral) behaviour, while the weaknesses of neutrality as neutrality of effect concern its impracticability and the fact that it is unclear why conceptions of the good life deserve the kind of respect or support which this formulation appears to give them. Obviously, though, these weaknesses cannot be dismissed on the basis of the brief commentary I have offered in this chapter.


2.5 Conclusion


In this chapter I have surveyed four different axes – the question of the meaning of the term “rival” in the phrase “rival conceptions of the good,” the question whether the neutrality principle applies to all legislation or merely to constitutional essentials and matters of basic justice, the question whether the neutrality principle applies to all ways of promoting the good or merely to coercive methods, and also to the question of the meaning of the term “favour” in the phrase “favouring a conception of the good” – along which different interpretations of the principle of state neutrality may be found. Making the optimistic assumption that there are only two possible ways of interpreting each of the four already leaves us with sixteen varieties of the principle of state neutrality. And I hardly need add that more may be found, given a sufficiently diligent search.121 But while it is important, in proceeding to the heart of the matter (and of the dissertation), namely, the arguments against the proposition that the state may not promote the good, to have a rough understanding of the cases for and against the most important ways in which the principle of state neutrality has been presented, the strengths and weaknesses of these cases, as I have noted in a number of places above, are closely tied to the strengths and weakness of the various arguments for adhering to any version of the neutrality principle. We therefore need to see chapter one as a necessary springboard for approaching the arguments for the principle of state neutrality investigated in chapter two, and not one whose subject matter is now left behind. I aim, in what follows, to make the connections between the formulation and the defence of neutrality clear.


Fußnoten und Endnoten

46  The most prominent of whom were Alasdair MacIntyre, Michael Sandel, Charles Taylor and Michael Walzer.

47  Raz’s The morality of freedom (Oxford: Clarendon Press, 1986) contains the most obvious case of such an argument, but the suggestion that state neutrality is a bogus ideal can also be found in the writings of the other liberal perfectionists discussed in this thesis.

48  Charles Larmore, Patterns of moral complexity (Cambridge: Cambridge University Press, 1987) p. 47.

49  Of course it might be possible to defend the legal enshrinement of monogamous marriage, and possibly even the subsidisation of museums and galleries, purely on the basis of right, although those who do so may be suspected of casuistry. My guess is that the enshrinement of monogamous marriage is likely to become increasingly threatened in contemporary Western democracies as its basis in a conception of the good which is no longer quite as widespread as it once was becomes clear, and right-based justifications of it become ever less plausible.

50  John Rawls, A theory of justice (Revised edition) (Oxford: Oxford University Press, 1999), p. 291.

51  In a rather obscure passage on page 215 of Political liberalism  (New York: Columbia University Press, 1993), he remarks that ‘it is usually desirable to settle political questions by invoking the values of public reason. Yet this may not always be so.’ Rawls uses the term “public reason” in his later writings to describe a form of justification which, amongst other features, appeals to no particular conception of the good. This passage ought therefore to be understood as suggesting that state policy with regard to constitutional essentials and matters of basic justice ought to be justified neutrally – in other words without reference to any particular conception of the good – after which it is also suggested that this stricture does not always hold. The circumstances under which it would not hold remain unclear..

52  See in particular the essay “Can a liberal state support art?” in A matter of principle  (Cambridge, Massachusetts: Harvard University Press, 1985), pp. 221-33.

53  See Waldron’s remarks to this effect in his “Autonomy and perfectionism in Raz’s Morality of freedomSouthern California Law Review 62 (1989), pp. 1098-1152 at pp. 1151-2.

54  John Rawls, Political liberalism (New York: Columbia University Press, 1993), p. 227. Obviously, one might formulate constitutional essentials and matters of basic justice differently. I present Rawls’s formulation here so as to give the reader a general sense of what characterises the distinction between such essentials and broader legislative functions.

55  Brian Barry, Justice as impartiality (Oxford: Clarendon Press, 1995), p. 161.

56  See Steven Wall and George Klosko’s “Introduction” to the volume they edit entitled Perfectionism and neutrality (Lanham: Rowman and Littlefield, 2003), p. 6.

57  See Steven Wall and George Klosko’s “Introduction” to the volume they edit entitled Perfectionism and neutrality (Lanham: Rowman and Littlefield, 2003), p. 6.

58  Political liberalism  (New York: Columbia University Press, 1993), p. 215.

59  Richard Arneson, “Liberal neutrality on the good: An autopsy,” in Steven Wall and George Klosko (eds) Perfectionism and neutrality (Lanham: Rowman and Littlefield, 2003), pp. 191-218 at p. 206.

60  John Rawls, Political liberalism (New York: Columbia University Press, 1993), p. 137. (The so-called principle of liberal legitimacy, which we encounter again in chapter three.)

61  Brian Barry, Justice as impartiality (Oxford: Clarendon Press, 1995), p. 160.

62  Brian Barry, Justice as impartiality (Oxford: Clarendon Press, 1995), p. 160.

63  Brian Barry, Justice as impartiality (Oxford: Clarendon Press, 1995), p. 161.

64  This argument assumes, of course, that the provision of a public school curriculum is not a function that the state can simply drop.

65  The section on the question whether the state should remain neutral between all conceptions of the good, or merely between disputed conceptions of the good.

66  I discuss this matter in greater depth in the section below entitled “Favouring a conception of the good”.

67  See his “Liberal neutrality on the good: An autopsy,” in Steven Wall and George Klosko (eds) Perfectionism and neutrality (Lanham: Rowman and Littlefield, 2003), pp. 191-218, at p. 210. 

68  See his “Liberal neutrality on the good: An autopsy,” in Steven Wall and George Klosko (eds) Perfectionism and neutrality (Lanham: Rowman and Littlefield, 2003), pp. 191-218, at p. 210. 

69  Some, such as Russell Keat in his Cultural goods and the limits of the market, have argued that the inculcation of certain cultural values (such as, for example, the idea that a life with art is superior to a life without it) promote autonomy in individuals. If one takes the – admittedly controversial – view that autonomy is a neutral value, Keat’s argument can also establish that a neutral public school curriculum is possible.

70  I do not address the matter here of the desirability of peripheral as well as constitutional legislation being neutrally justified, merely whether it would be possible. We will see in chapters three and four that liberal perfectionists do not consider such neutrality desirable; Raz in fact thinks that it damages personal autonomy. Furthermore, I come, in chapters two and three, to discuss the possibility that avowedly anti-perfectionist political moralities do in fact rely on (controversial) conceptions of the good. However the subject of these later discussions is the difficulty the possibility that these anti-perfectionist moralities in fact rely on unacknowledged conceptions of the good raises for any form of the principle of state neutrality, narrow or comprehensive. Given that the current matter under discussion is simply whether it is necessary or wise for neutralists merely to defend a narrow neutrality principle, as opposed to a comprehensive one, I do not raise deeper theoretical difficulties for neutralism in general at this point.

71  And any version of democratic theory.

72  Near monopoly would be more accurate, given that most democratic theorists, liberals included, recognise the right of citizens to defend themselves by means of force under some circumstances.

73  Raz is a notable exception to this view, and Waldron has suggested that Raz’s view that the state is distinguished rather by the generality of its claim to authority is linked to his view that taxation, for example, is not necessarily a coercive exercise of state power. See Jeremy Waldron,, “Autonomy and perfectionism in Raz’s Morality of freedom,” Southern California Law Review 62 (1989), pp. 1098-1152 at pp. 1139-40.

74  Some perfectionists, Raz included, hold a position quite close to this as well.

75  Here I ignore the possibility – far fetched, it seems to me – that all instances of apparently wholehearted commitment to the law are cases of self-deception, where fear is in fact a genuine motivating factor, albeit one which the actor cannot or will not acknowledge. This is not to deny that there are such cases.

76  See Raz’s comment at page 149 of The morality of freedom  (Oxford: Clarendon Press, 1986), that, in order for a case to be an example of coercion, avoiding the (credibly) threatened consequence of non-compliance must be, for the coercee, ‘a reason of great weight for not doing’ whatever it is that is being proscribed by the coercer.

77  Jeremy Waldron, “Autonomy and perfectionism in Raz’s Morality of freedom,” Southern California Law Review 62 (1989), pp. 1098-1152 at p. 1139.

78  Jeremy Waldron, “Autonomy and perfectionism in Raz’s Morality of freedom,” Southern California Law Review 62 (1989), pp. 1098-1152 at pp. 1150-2.

79  John Stuart Mill, on page 13 of On liberty (Indianapolis: Bobbs-Merrill, 1956), says, in an oft-quoted passage, that the only purpose for which power can be rightfully exercised over any member of a civilised community, against his will, is to prevent harm to others. His own good, either physical or moral, is not a sufficient warrant. He cannot rightfully be compelled to do or forbear because it will be better for him to do so, because it will make him happier, because, in the opinions of others, to do so would be wise or even right.

80  The morality of freedom find exact quote.

81  I discuss reasons for adopting the principle of state neutrality systematically and in detail in chapter two.

82  Joseph Raz, The morality of freedom (Oxford: Clarendon Press, 1986), p. 377.

83  As we noted above, Waldron believes in addition that such policies are coercive. But it is important to note that his argument for their damaging effect on autonomy applies equally whether such policies are in fact coercive or not.

84  See the section entitled “Autonomy and the harm principle” in Joseph Raz, The morality of freedom (Oxford: Clarendon Press, 1986) at pp. 412-9.

85  George Sher, Beyond neutrality: Perfectionism and politics (Cambridge: Cambridge University Press, 1997) at pp. 64-5.

86  George Sher, Beyond neutrality: Perfectionism and politics (Cambridge: Cambridge University Press, 1997) p. 36.

87  Bruce Ackerman, for example, discusses the argument for state neutrality from moral scepticism with a certain degree of enthusiasm on page 369 of his Social justice in the liberal state (New Haven: Yale University Press, 1980), but, given his coyness about the provenance of his neutralist convictions, it is difficult to know whether he genuinely endorses this argument.

88  See George Sher, Beyond neutrality: Perfectionism and politics (Cambridge: Cambridge University Press, 1997), pp. 106-39.

89  Quasi-empirical because predictions.

90  The question whether desert is involved where some conceptions of the good life prosper under a neutral dispensation and others do not is a difficult one, and falls beyond the scope of this dissertation, hence my use of the more neutral term “entitled to”, which encompasses luck as well as desert. Rawls denies that this is the case – see his Political liberalism (New York: Columbia University Press, 1993), pp. 195-200 – whereas Haksar, a perfectionist, writes as if this is indeed an assumption of Rawls’s position: see his remarks on what he calls the “choice criterion of value” in his Equality, liberty, and perfectionism (Oxford: Oxford University Press, 1979), pp. 206-25.

91  Jeremy Waldron “Legislation and moral neutrality,” in Robert Goodin and Andrew Reeve (eds) Liberal Neutrality (London: Routledge, 1989), pp. 61-83 at pp. 63-4.

92  Robert Nozick, Anarchy, state, and utopia (New York: Basic Books, 1974), pp. 272-3.

93  Will Kymlicka, Contemporary political philosophy (2ed) (New York: Oxford University Press, 2002), p. 344.

94  Political liberalism  (New York: Columbia University Press, 1993), p. 197.

95  One might also, of course, think that such a state is neutral, but that neutrality is less important than the survival of one’s culture.

96  Robert Goodin and Andrew Reeve, ‘Do neutral institutions add up to a neutral state?’ in Robert Goodin and Andrew Reeve, (eds) Liberal neutrality (London: Routledge, 1989), pp. 193-210 at p. 202.

97  Charles Larmore, Patterns of moral complexity (Cambridge: Cambridge University Press, 1987), p. 44.

98  Bruce Ackerman, Social justice in the liberal state (New Haven: Yale University Press, 1980), pp. 8-10.

99  John Rawls, “The priority of the right and ideas of the good,” Philosophy and Public Affairs 17, 4 (1988) pp. 251-76 at p. 262.

100  George Sher, Beyond neutrality: Perfectionism and politics (Cambridge: Cambridge University Press, 1997), p. 4.

101  Jeremy Waldron, “Legislation and moral neutrality” in Robert Goodin and Andrew Reeve (eds) Liberal neutrality (London: Routledge, 1989), pp. 61-83 at pp. 66-8.

102  Provided superiority means more than “better able to facilitate communication in this particular country”.

103  There are contrary voices, of course.

104  Ironically, the neutrality of justification interpretation could provide perfectionists, or even communitarians, with a theory of state impartiality (if not state neutrality): An impartial state, on this view, would be one which, without respect for persons, applied those laws which promote the favoured conception of the good (just as an impartial referee applies on the sports field, without particular respect for either team, the laws which are designed to promote those qualities which we value in sport).

105  It does not have a complaint against the neutrality of the rules either, despite them being discriminatory, in that the team’s participation in the contest assumes its willingness to be judged according to these standards.

106  See his The morality of freedom, (Oxford: Clarendon Press, 1986) p. 116, where he discusses Nozick’s version of anti-perfectionism.

107  Robert Nozick, Anarchy, state, and utopia (New York: Basic Books, 1974) pp. 272-3.

108  Steven Wall and George Klosko ‘Introduction” to Steven Wall and George Klosko (eds) P erfectionism and neutrality (Lanham: Rowman and Littlefield, 2003), p. 8. There is no indication in the text that Wall and Klosko actually endorse this formula as the correct interpretation of the principle of state neutrality.

109  Joseph Raz, The morality of freedom (Oxford: Clarendon Press, 1986), pp. 114-5. I leave open the question whether this is indeed Rawls’s position in A theory of justice. It is certainly not the position he takes in Political liberalism.

110  Charles Larmore, Patterns of moral complexity (Cambridge: Cambridge University Press, 1987), p. 44.

111  Assuming, of course, that the analogy can genuinely hold; that is, assuming that some conceptions of the good life can genuinely deserve to be successful in comparison to others. Some standard would have to be acknowledged for this to hold; even a standard as thin as “would be successful in competition with other conceptions of the good life in conditions under which none are privileged by the state”.

112  See, for example, Jeremy Waldron, who remarks on page 68 of his article ‘Legislation and moral neutrality” in the collection Liberal neutrality, edited by Robert Goodin and Andrew Reeve (London: Routledge, 1989), that what he calls neutrality of consequences interpretation ‘is a very difficult requirement to live up to, because it is so hard to predict what the effect of a law is going to be on lifestyles and mores.’ Similarly, Steven Wall and George Klosko suggest on page 8 of their introduction to the volume they edit entitled Perfectionism and neutrality (Lanham: Rowman and Littlefield, 2003), that ‘[i]t is impossible for a state to ensure that each person subject to its authority has an equal chance to pursue and realise his or her conception of the good.’

113  One of the reasons why impossible ideals ought to be rejected is because they can only be implemented partially, or selectively, that is in ways that are likely to violate the rule of law. One might put it more strongly and say that the more removed from the realm of possibility the ideal is, the more likely its attempted implementation will amount to injustice.

114  At pages 115-6 of The morality of freedom (Oxford: Clarendon Press, 1986).

115  John Rawls, Political liberalism (New York: Columbia University Press, 1993), p. 191.

116  See his remarks on pages 14 and 15 of The morality of freedom (Oxford: Clarendon Press, 1986).

117  See his A matter of principle (Cambridge, Massachusetts: Harvard University Press, 1985), p. 202.

118  Ronald Dworkin, A matter of principle (Cambridge, Massachusetts: Harvard University Press, 1985), p. 195. I owe this point to Jeremy Waldron’s discussion in his ‘Legislation and moral neutrality’ in Robert Goodin and Andrew Reeve (eds), Liberal Neutrality (London: Routledge, 1989), pp. 61-83 at page 68.

119  This is how Sher reads at least one of his arguments, for example. See his Beyond neutrality (Cambridge: Cambridge University Press, 1997), pp. 100-4.

120  See chapter 5 of Beyond neutrality  (Cambridge: Cambridge University Press, 1997).

121  Gerald Gaus provides a disturbingly long list of the diverse interpretations of neutrality. On page 138 of the volume Perfectionism and neutrality, edited by Steven Wall and George Klosko (Lanham: Rowman and Littlefield, 2003), he writes, in an essay entitled ‘Liberal neutrality: A compelling and radical principle,’ the following: Neutrality understood as a constraint on the sorts of reasons that may be advanced to justify state action is regularly distinguished from “consequential neutrality” – that the effects of state policy must somehow be neutral. Yet interpretations of neutrality are far more diverse than most analyses recognise. Neutrality is sometimes understood as a doctrine about the intent or aim of legislation or legislators, the proper functions of the state, the prohibition of the state “taking a stand” on some issues, the prohibition of the state enforcing moral character, or the requirement that the state take a stance of impartiality. Alternatively, neutrality can be understood as a requirement of a theory of justice rather than state action. There are also differences about whether neutral states (or theories of justice or legislators) are supposed to be neutral between conceptions of the good, particular sets of ends, comprehensive doctrines and conceptions of the good, particular or substantive conceptions of the good, ways of life, final ends, or controversial conceptions of the good. And it is unclear whether every principle of neutrality is inherently one of liberal neutrality, or whether liberal neutrality is a specific sort of neutral principle.

© 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: