“State neutrality is neither possible nor desirable.” Discuss.
The value of ‘state neutrality’ is frequently invoked in liberal political discourse, and an influential and sophisticated defence of it is provided by Rawls in Political Liberalism. Yet neutrality is in many ways a problematic notion, and it is often unclear both what it involves and why we should want it. The aim of this article, therefore, is to try to clarify these issues. First, I shall explore a few ways in which neutrality can be understood, so as to highlight what is particular to Rawls’s account, and anticipate some of the ways in which it might be criticised. I shall then consider Rawls’s position itself, asking: (1) what is his understanding of political neutrality? And (2) why does he think of it as a necessary component of liberalism? Although the Rawlsian position can be attacked in various ways, I shall focus on criticisms from a specifically feminist perspective, as I believe these cut to the heart of what is problematic about his brand of liberalism. Ultimately, I shall argue that neutrality is an incoherent ideal, its apparent attractiveness being parasitic upon the genuine values of equality and autonomy.
Firstly, we must ask what is meant by ‘state neutrality’. To be neutral, in general, means to not take sides, to not favour or assist one thing over another. One question that immediately arises from this is: what is the state supposed to be neutral between? The Rawlsian answer is that the state must be neutral between conceptions of the good – that is, comprehensive moral, philosophical and religious values and ideals – but it is important to note that this is not the only possible way of understanding neutrality. Nozick, for example, talks about neutrality between citizens, and perhaps we can also meaningfully speak of neutrality between groups, which might be defined by characteristics other than their conception of the good. It is not immediately obvious that neutrality between conceptions of the good entails, or is even necessarily consistent with, neutrality between persons or between social groups. Indeed, it should emerge in the course of this article that there is significant tension between the ‘conceptions of the good’ version of neutrality and the ‘social groups’ version, since the former conception disproportionately harms women as a group.
Assuming for now a Rawlsian line on the question of what the state is to be neutral between, a second potentially contentious issue concerns how we are to understand the demands of neutrality itself. Drawing partly on Raz’s discussion of this point, we can distinguish three ways that the idea of neutrality might be spun. First, neutrality can be thought of as essentially a matter of excluding as illegitimate appeals to certain kinds of reasons as grounds for political decisions. This is the understanding favoured by Rawls, whose position depends upon the idea that comprehensive – as opposed to political – principles cannot serve to justify state decisions (at least not on matters of fundamental justice). The distinction between comprehensive and political principles is a key component in Rawls’s thought, and is essentially a question of scope and subject matter. Political principles apply solely to the ‘basic structure of society’ – the political constitution and the principal institutions governing economic and social arrangements – whereas ‘conceptions of what is of value in human life, and ideals of personal character, as well as ideals of friendship and of familial and associational relationships’ all fall under the category of comprehensive principles. Later, I shall subject the political-comprehensive distinction to more scrutiny, but for now it should serve to give a sense of what Rawls is getting at.
The other two senses of neutrality are concerned not so much with the justification but with the consequences of government action, although each weighs the consequences in rather different ways. The first of these – what Raz calls the ‘principle of narrow neutrality’ – rules out any political action which makes a difference to the likelihood that a person will endorse one conception of the good rather than another, or to her chances of realizing her conception of the good, unless other actions are undertaken to counter-balance the effects of that action. This view makes the doctrine of state neutrality functionally equivalent to a strong anti-interventionism, since it places upon governments a duty to leave people’s conceptions of the good and chances of realizing them as unchanged as possible. The second ‘consequentialist’ understanding of neutrality – what Raz calls the ‘principle of comprehensive neutrality’ – asserts that governments have an overriding duty to ensure for all citizens an equal ability to pursue in their lives and promote in their societies any ideal of the good of their choosing. This principle does not necessarily imply non-intervention, since its concern with equality (at least in certain respects) between citizens may mandate government action, but it does share with the narrow principle and the Rawlsian principle a refusal to allow governments to act upon judgements as to the validity of conceptions of the good.
All three principles, therefore, are anti-perfectionist in that they deny that it is a legitimate aim of government to promote any comprehensive or non-political ideal above any other. These kinds of principles are often advanced as attempts to capture the core sense of the liberal ethos, and the political neutrality they describe is seen as prima facie desirable. Raz argues that this intuition is rooted in the confused notion that to act neutrally is necessarily to act fairly. This assumption, he thinks, can be easily dispatched by means of Montefiore’s example of the father who refuses to intervene in his children’s dispute, knowing that the older, stronger child will therefore win. It is plausible to say that the father is acting neutrally, but not that he is acting fairly. Therefore, neutral political concern cannot be justified by the inherent appeal of neutrality as such.
The advocate of political neutrality could respond, however, that this example is misleading, for in fact the father may only be acting neutrally according to the principle of narrow neutrality. Rawlsian political neutrality, they would argue, has a far more complex relation to the anti-interventionism that the narrow principle espouses, and Rawls can offer a correspondingly more sophisticated defence of neutrality. Therefore, it is to the Rawlsian position that I shall now turn. Rawls does not endorse the consequentialist understandings of neutrality in the first instance, but he does hold that commitment to certain features of them will inevitably result from the initial commitment to neutrality embodied by the veil of ignorance in the original position. This should become clear later.
First, though, it is necessary to lay out the justification for his position. According to Rawls, the state should deliberately refrain from acting on any comprehensive considerations and seek only to provide a neutral framework within which people can make their own individual choices. What motivates Rawls’s concern that his account of justice be political rather than comprehensive is his adherence to a social contract theory of political authority. If citizens are to be said to (hypothetically) consent to state power, the state must govern in accordance with a constitution that ‘all citizens can reasonably be expected to endorse in the light of their common human reason’. In other words, the conception of justice on which the government operates must be publicly justifiable – justifiable on the basis of reasons that all can endorse – and the object of an ‘overlapping consensus’. This is said to rule out comprehensive considerations because ‘the fact of reasonable pluralism’ means that conflicting conceptions of the good are an inevitable outcome of the exercise of human reason against the backdrop of liberal democratic institutions. We cannot expect all rational citizens to agree on a particular conception of the good, and therefore putative justifications which presuppose the validity of some comprehensive doctrine or other will not belong to the realm of ‘public reason’. So, on the Rawlsian model, the public justifiability requirement, and hence the demand for state neutrality between conceptions of the good, are seen to stem directly from a liberal understanding of the nature of politics and the proper relation between the individual and the state. As Rawls would say, the need for state neutrality between conceptions of the good follows from the view of society as a ‘fair system of cooperation between free and equal citizens’.
That it does so, however, depends on several other assumptions, which I shall now try to bring out. First, there is the distinction between the political and the non-political. The distinction is taken to apply to two types of things: (1) areas of life, or ‘institutions’ in the broad sense, and (2) conceptions of the good, including conceptions of justice. Clearly, there is an intimate connection between the two, in that those conceptions of justice and the good are taken to be non-political when they apply or refer to areas of life which are taken to be non-political. What is being invoked here is a dichotomy between the public political culture which persons participate in qua citizens, and the ‘background culture’ of civil society which persons participate in qua private individuals. The essence of Rawls’s political liberalism is the thought that justice as fairness (or any other political conception of justice) should not be applied to areas of life which fall into the category of ‘background culture’, and that principles which do apply to these areas should not be permitted in public political discourse. His argument for state neutrality can therefore be seen to depend upon a dichotomy between the political and the private, and the assumption that the two can be kept separate in considerations of justice.
A major threat to these assumptions is to be found in feminist discussions of the family. ‘Ideals of familial relationships’, it will be remembered, are paradigmatically non-political, and although Rawls suggests in A Theory of Justice that the monogamous family might count as a ‘major social institution’, the implications of this insight are not consistently pursued in Political Liberalism. Many feminists have argued, however, that the gendered family – i.e. the family embodying certain traditional gender roles – ‘constitute[s] the pivot of a societal system of gender that renders women vulnerable to dependency, exploitation, and abuse’ (Okin). For example, the gendered division of labour, which gives women primary responsibility for house-work and child-care, has far-reaching implications, raising practical and psychological barriers against women in all other spheres of life – including, it must be emphasised, the sphere of public political discourse.
In Justice, Gender and the Family, Okin argues that the expectation that a woman will fulfil the role of homemaker – even if she also works full time – makes her career aspirations secondary, her working hours (paid and unpaid) frequently longer, and her status and power (both inside and outside the family) lower, compared with her partner. In the workplace, the expectation that employees have someone at home taking care of domestic duties results in an ethos which favours men for access to more prestigious and better-paid jobs, with female workers concentrated in lower paid sectors which provide the more flexible working hours that women with domestic responsibilities require. Opportunities for promotion in these sectors are limited, and disadvantage at work makes a focus on home life more appealing to women. Frequently, this results in economic dependency on a male partner, which reduces women’s power to challenge the gendered division of labour. Furthermore, anticipation of their gendered role makes women less likely than men to invest in their career even before getting married or having children, and therefore makes the later side-lining of their career still more likely.
This last point ties in with a second aspect of the family’s impact on the political sphere: its role in social construction. Domestic life is where most of our early socialization takes place, and if the family is marked by gendered divisions of labour and power, this cannot but affect the expectations, attitudes, and socialized behaviour of both women and men. The family cannot be cordoned off from political life if, as Okin and others have argued, participation in the public sphere is affected – indeed, has its terms dictated – by people’s experiences as gendered individuals. Take the case of the judiciary. Here we have a classic Rawlsian ‘public institution’, dominated by men whose perspectives and conceptions of justice have to large extent been formed by their experiences within the family of the subordination of one sex to another. Their ‘public’ experiences, too, are inevitably affected by the consequences of the gendered family: for example, the token representation of women in many professions, the behaviour of other men and women shaped by the gendered family, and the gendered nature of dominant cultural representations. Here the connections between personal relations in the macro- and the micro-institutions of society are all too plain; sex bias has been shown to affect judicial attitudes towards women as defendants, victims and lawyers.
This seriously undermines Rawls’s claim that ‘non-political’ associations need not be governed by considerations of justice because ‘the prospective or continuing members are already guaranteed the status of free and equal citizens’. If injustice within the family has such far-reaching consequences as I have implied, then the family can be perceived as a separate sphere only insofar as equality between the sexes reigns within it. And the family is not the only ‘non-political’ institution with the power to affect the freedom and equality of citizens. For example, religious groups or cultural communities may serve a role analogous to the family in social construction, and Catharine Mackinnon has argued that gendered sexual norms and pornography impact upon women’s ability to speak freely in public life. It should also be emphasised that gender is not the only hierarchy whose ‘private’ enforcement affects ‘public’ participation, although it is a particularly pertinent one. Unger makes the general point: ‘men [or, we might add, women] whose everyday experience is one of submission or of predominance cannot be expected to think and to act as equals when they pass upon the affairs of the group’.
Leading on from this, we can say that, since genuine participation in public political discourse requires justice in spheres such as the family, and since such justice is not currently forthcoming, the result is that certain groups are systematically excluded from the public sphere, or granted entry on very different terms. On the basis of this, we can question whether conceptions of justice arrived at in the public sphere (thus constituted) may legitimately be called ‘neutral’. In Justice and the Politics of Difference, Young persuasively criticises what she calls the ‘ideal of impartiality’, which is encapsulated in the idea of a sphere of public reason which citizens engage in free from the baggage of comprehensive ideals and perspectives. In reality, she says, this ideal is impossible and merely serves to hypostatize the point of view of privileged groups into a supposedly universal position, functioning ideologically to make non-dominant perspectives appear illegitimate, biased and situated. She points out that flesh-and-blood decision-makers do not come to embody transcendental reason when they step into their decision-making role. Their ‘comprehensive’ desires and commitments do cease to exist and motivate just because they have been excluded from the definition of public reason; rather, they ‘lurk as inarticulate shadows’, belying public reason’s claim to neutrality.
An obvious response, however, would be to say that all we have demonstrated is the importance of making sure that comprehensive considerations are genuinely removed from political debate. This does not challenge the idea that they should be, and therefore does not pose a serious threat to Rawlsian neutrality. In other words, the suggestion is that we should simply try harder to achieve neutrality, rather than rejecting it as an ideal. Such a response underestimates the extent to which the dominant perspective can pass itself off as impartial, and for evidence of this we need only look at Rawls’s theory itself. I have already argued that the division between public and private is not gender neutral; the perception of a sharp dichotomy depends on viewing society from a traditional male perspective. The idea that justice as fairness should only apply to the ‘basic structure’ of society (excluding the family), and that certain reasons – those which concern ‘ideals of familial relationships’ – should be excluded from public discourse, is therefore based upon ‘an ideological division that lies about women’s shared experience and mystifies the unity among the spheres of women’s violation’ (MacKinnon). Yet these assumptions succeed in appearing ‘neutral’ because of their ubiquity within the male-dominated liberal tradition.
There is another, related, assumption which is similarly pervasive in liberal thought, and which further undermines Rawls’s claim to a neutral perspective. This is the assumption that state power, as exercised through the ‘basic institutions’, is the main thing likely to prevent citizens from being free and equal. Rawls does, of course, recognise that economic inequalities not deliberately created by the state can work to undermine citizens’ freedom and equality, which is why he argues that his egalitarian principles should apply to major economic institutions. When we turn to matters considered ‘private’, however, there is a barely-veiled assumption that government non-intervention is the measure of freedom. This Hobbesian view that freedom is the silence of the law is to be found, for example, in the idea that freedom of speech essentially means restricting government restrictions on speech, or that freedom in the domestic sphere is basically a matter of protecting the family from government interference.
MacKinnon argues that this is once again a gender-biased understanding of liberal values, for non-intervention can only respect freedom and equality if that freedom and equality already exist. As she puts it, ‘freedom from public intervention coexists uneasily with any right that requires social preconditions to be meaningfully delivered. If inequality is socially pervasive and enforced, equality will require intervention, not abdication, to be meaningful.’ To be fair, Rawls is by no means a through-going non-interventionist or an advocate of the negative state. He does, after all, emphasise that ‘the principles of justice must give priority to those basic freedoms and opportunities in background institutions of civil society that enable us to become free and equal citizens in the first place.’ However, since he does not take this to entail the relevance of the family to public political discourse and indeed to justice in general, his understanding remains ‘male-biased’ in that it ignores women’s experiences and serves to maintain the gender-hierarchical status quo.
By effectively ignoring the existence of pervasive power structures other than the state, this face of the public/private distinction creates the appearance of an important dichotomy between state action and state inaction, which in reality serves the interests of dominant groups. For the implication is that, while state action must be justified by public reason, state inaction does not carry the same burden of proof. This asymmetry can be observed in Rawls’s statement that: ‘In order to respect one another’s freedom and equality as reasonable and rational, citizens of constitutional democracies should not use the coercive power of the state against their fellow citizens except in ways that those subject to that coercion might reasonably be expected to endorse.’ (my italics) It is a key component of Rawls’s political liberalism that many non-governmental organisations are exempt from this restriction on the exercise of coercive power.
Bringing the argument together, we can say that Rawls’s doctrine of state neutrality is premised upon various assumptions relating to the distinction between the political and the non-political which are not themselves neutral. Once the issue of gender is brought into focus, Rawlsian neutrality is in fact shown to have much in common with the unjust father in Montefiore’s example, whose ‘neutral’ stance works to reinforce existing distributions of power. ‘Non-political’ spheres influence political ones, rendering political freedom and equality merely formal, and political discourse one-sided. Although Rawls rejects – as indeed he must – the ‘narrow principle of neutrality’ according to which non-intervention and neutrality are virtually synonymous, he nevertheless adopts elements of this view when he turns to the ‘private’ sphere of family, sexuality and religion. Having rejected the ‘narrow principle’, though, it becomes difficult for him to maintain the view that a state which perpetuates masculine domination is neutral in any meaningful sense. That apparent neutrality of Rawls’s position, I have argued, is a function of the social dominance of the traditional male perspective, which pervades society, busily framing the terms of the debate, and defining social meanings.
My claim is not particularly that genuine neutrality would involve a revisionist attitude to non-state institutions, for that would stretch our intuitive notion of neutrality a bit too far. Rather, I would suggest that Rawls’s failure to present a cogent account of neutrality is symptomatic of the inadequacy of that concept to deal with the complex power relations found in every area of life. Rawls states that his doctrine of neutrality follows from his commitment to free and equal citizens, but feminist critiques of the public/private distinction show that he assumes too quick and simple a connection between neutrality and personal autonomy, and between neutrality and equality. Clearly, neutrality has some conceptual links to both autonomy and equality, but since neutrality itself has been shown to be neither possible nor desirable, it is to these other values that we must turn.
