Against hands-on neutrality

In recent years, several theorists have defended a form of neutrality that seeks to equalise the benefits that state policies bestow upon citizens’ conceptions of the good life. For example, when state policies confer special benefits upon a conception that revolves around a particular culture, religion or type of sports, other cultures, religions or types of sports might be due compensation. This article argues that this kind of neutrality – which I refer to as ‘hands-on neutrality’ – cannot be vindicated, whether it takes a resourcist form or welfarist form. After explaining why states cannot simply avoid bestowing unequal advantages upon citizens’ conceptions of the good life, I identify several versions of hands-on neutrality and show that all of them fail to respect the moral equality of citizens. (In so doing, however, I do find that a different form of neutrality is appropriately applied under a more limited set of conditions, which is lottery-based.) I conclude by showing that in cases where theories of hands-on neutrality have intuitive force, more plausible rival principles can account for our intuitions.


Introduction
One important way in which societies can be diverse is that their citizens have different conceptions of the good life. As commonly understood, such conceptions consist of more or less consistent sets of beliefs about how one should live one's life, which might include religious, cultural and moral commitments but also preferences for specific forms of art, sports and intellectual activities.
At least since the 1980s, some liberals have argued that states ought to be neutral towards certain kinds of conceptions of the good (for early defences, see Larmore, 1987;Rawls, 1993). 'Neutral' has been understood to mean here that the justifications for state policies are ones that the adherents of different conceptions can accept after a reasonable amount of careful reflection (cf. Vallier, 2018). Exactly what this involves -or should involve -is a divisive issue. While proponents of this view concur that neutral justifications are not due towards those who oppose liberal democracy such as Neo-Nazis and Jihadists (Balint, 2015: 499), there are many points of contention among them, the most important one perhaps being whether members of the justificatory constituency ought to be able to accept state policies on the basis of shared reasons (for defences of this view, see, for example, Lister, 2013;Quong, 2010) or on the basis of their idiosyncratic but still intelligible reasons (for defences of this view, see, for example, Billingham, 2016;Gaus, 2010;Vallier, 2014). These disagreements notwithstanding, virtually all defenders of justificatory neutrality consider certain kinds of beliefs to be impermissible grounds for state (in)action; one might think of the beliefs that atheism is correct, that opera is intrinsically valuable, and that that a life of drinking beer and watching football is inferior to one devoted to athletic excellence. Even if these beliefs are true, the fact that not all adherents of suitably qualified conceptions of the good can recognise their truth upon reflection means that they should not inform policymaking. Or so the argument goes.
Whereas justificatory neutrality continues to be the most widely advocated form of state neutrality to date, a novel form of state neutrality has been defended in recent years by authors such as , Réaume (2017), Schutter (2017) and Pierik and Van der Burg (2014), which is the focal point of this article. Contrary to neutrality of justification, which we have just seen is concerned with the grounds on which state policies are justified, this kind of neutrality -which I, with Balint (2015: 498) refer to as 'hands-on neutrality' -is concerned with how the benefits that state policies bestow on citizens' conceptions of the good are distributed, which may take the form of any resource that makes it easier for a person to live in accordance with her conception of the good (for example, rights, subsidies, tax exemptions, material goods). Because of the presumed benefits of being able to live in accordance with one's conception of the good, which may be valued on different grounds including the fact that it promotes people's autonomy (see, for example, Patten, 2014: 127) and often their well-being and selfesteem, proponents of hands-on neutrality maintain that, to show equal respect to citizens, states should in many cases equalise such benefits among citizens' conceptions of the good.
To be sure, no proponent of this type of neutrality believes that the conceptions of the good of, for example, Neo-Nazis or Jihadists ought to be compensated when state policies confer greater-than-equal benefits upon other citizens' conceptions of the good. There is always a background assumption that hands-on neutrality applies only -or should apply only -to conceptions that are compatible with liberal democracy. Furthermore, some advocates of hands-on neutrality also require that the relevant conceptions have goals and commitments that are sufficiently valuable (I say more about such perfectionist requirements in the fifth section). For ease of reference, let us call conceptions of the good that are understood to fall within the purview of hands-on neutrality 'suitably qualified conceptions'.
Before distinguishing two different types of hands-on neutrality, it is worth considering some cases where principles of hands-on neutrality are commonly thought to apply by their proponents. Doing is important not just to get a sense of the practical implications of such principles. Since many defenders of hands-on neutrality endorse this type of neutrality based on the (in their view) implausible practical implications of rejecting it rather than on the basis of a more abstract set of philosophical principles from which principles of hands-on neutrality are then deduced, 1 such cases also play an important justificatory role. To start with, one might think of a case where a local government decides to build a skateboard park to promote a healthy lifestyle. According to Patten (2014: 122), the government in this example might need to provide, for example, swimmers and squash players with comparable facilities to avoid favouring the conceptions of the good of skaters over those of individuals who are into different types of sports. Another case can be found in societies where state organisations (for example, parliaments, courts, the police) use a single language to communicate. Insofar as the language that is being used belongs to one specific cultural group in society, many proponents of hands-on neutrality would maintain that the members of (some) other cultural groups ought to be compensated for this even when they are fluent in the administrative language (as, for example, virtually all Frisian speakers are in the Netherlands and virtually all Welsh speakers are in the UK). To see why, it should be noted that having one's language used by state organisations is one important way of ensuring that it remains widely spoken, which in turn helps the wider culture of which it is part to survive and flourish. A third case where many proponents of hands-on neutrality would maintain that this type of neutrality has force is when a state in a predominantly Christian society provides prisoners with access to a chaplain. Under these conditions, for example, Pierik and Van der Burg (2014: 501) have argued that non-Christian prisoners ought to be offered spiritual support that is tailored to their religious views as well, which might require that the relevant state hires Imams, Rabbis and so on.
As I have already alluded to, there are two different kinds of equality that a state committed to hands-on neutrality might seek to realise and, correspondingly, two general versions of hands-on neutrality: 1. Resourcist version: When state policies confer goods (for example, rights, subsidies, tax exemptions, material goods) on some suitably qualified conceptions that are not conferred on other suitably qualified conceptions or to a lesser degree only, then the state in question has strong pro tanto reason 2 for extending per capita equally valuable goods to the latter conceptions. 2. Welfarist version: When state policies confer goods on some suitably qualified conceptions that are not conferred on other suitably qualified conceptions or to a lesser degree only, then the state in question has strong pro tanto reason for conferring as many goods to the adherents of the latter conceptions as is necessary to allow them to benefit to the same degree as the adherents of the former conceptions (as determined by how much the relevant goods help them to act upon their conceptions of the good).
The most influential -and to the best of my knowledge only -expression of the resourcist version can be found in the work of . According to Patten's theory, which he calls 'neutrality of treatment', states that confer special goods on some citizens' conceptions of the good have strong pro tanto reason 'relative to an appropriate baseline' to extend 'equivalent form of assistance' to other suitably qualified conceptions (Patten, 2014: 112). The 'appropriate baseline' here is one where basic liberties are protected and where there is fair distribution of socio-economic goods (p.114). This means that it is not a departure of neutrality of treatment when citizens with antiliberal or anti-democratic conceptions of the good find it more difficult to act upon their conceptions under liberal-democratic laws and institutions than other citizens, given that the former's conceptions oppose basic liberties (p.109). Neither is it a departure from neutrality of treatment when redistributive policies that are necessary for securing socioeconomic justice make it harder for citizens with expensive tastes (for example, those with preferences for rare wines, first class rights, etc.) to act upon their conceptions of the good than those without expensive tastes (p.184).
Two further requirements must be satisfied in order for citizens' conceptions to fall within the scope of Patten's neutrality of treatment. First, the relevant conceptions must not be 'worthless' in the sense that the pursuit and attainment of their goals renders people worse off (p. 109). (While Patten does not give examples of conceptions that meet this criterion, we do not have to speculate what these might consists of as my later criticisms of theories of hands-on neutrality do not hinge on the precise scope of these theories.) Second, the relevant conceptions must have stakes within the same thematic domain as the conceptions that receive special benefits under a given state policy. Although Patten does not mention this last requirement explicitly, it is presupposed in the examples he gives of when neutrality of treatment is realised. To return to the example of the skateboard park, he notes that when a local government decides to construct such a park, it should do its best to offer proportional compensation to residents who are into different sports by providing 'a range of different kinds of comparable facilities', such as 'skating rinks, swimming pools, and squash courts' (p.122). Likewise, he notes that when a state accommodates one culture by providing public services in its language, it has strong pro tanto reason for providing public services in the languages of other cultures, just as a state that designates the holidays of one religious group as public holidays has strong pro tanto reason for giving members of minority of religions days off on their respective holidays (p.161).
What renders Patten's account of hands-on neutrality a recourcist one is that the 'equivalent assistance' that states have strong pro tanto reasons to bestow on citizens' suitably qualified conceptions of the good has the same per capita value as the goods that their policies impose on other conceptions of the good (pp.161-162). 3 This allows for the possibility that citizens with different conceptions will benefit from per capita equal support to unequal degrees. Such differences might arise, for instance, when the number or spread of individuals with the same conceptions of the good is such that some but not other groups will be able to enjoy economies of scale or afford certain lumpy goods.
Authors such as Pierik and Van der Burg (2014), Réaume (2017) and Schutter (2017), in contrast, have defended welfarist versions of hands-on neutrality. Characteristic of these accounts is that they are sensitive to how advantageous a given unit of resources is to people with different conceptions of the good, as measured by the extent to which receiving said unit would help them to live in accordance with their respective conception of the good. For example, Réaume (2017: 69-70) argues that when granting per capita equal resources to different types of sports allows softball players but not cricket players to recruit enough players to play a game, this might impose a duty on a local government to compensate the cricket fans by sponsoring a 'Cricket Appreciation Day' picnic or by hosting a 'match between out of town teams to attract participants by demonstrating the beauty and athleticism of the game'. In a similar vein, she (2017: 69), Schutter (2017: 77-78) and Pierik and Van der Burg (2014: 501) suggest that, relative to the number of domestic speakers, the languages of minority cultures might be due to greater public support than the languages of the majority cultures when giving these groups the same per capita resources does not allow the former's members to enjoy the same services or goods within their respective language as the latter's members. For example, receiving per capita equal resources might be insufficient for a minority to establish a school in its own language (schools being an example of a lumpy good) or to publish books in its own language at moderate cost when the fact that they have relative few speakers means that they do not benefit from economies of scale.
My aim in this article is to show that hands-on neutrality should be rejected, whether it takes a resourcist or welfarist form. As I argue, there never appears to be sufficient reason for states to compensate citizens for unequal distributions of policy-imposed benefits insofar as public money is being used to provide such compensation (as proponents of hands-on neutrality seem to presuppose). 4 After explaining why states cannot simply avoid bestowing unequal advantages upon citizens' conceptions of the good, including those that are suitably qualified (see second section), I consider several versions of hands-on neutrality and show that all of them fail to respect the moral equality of citizens (see the third section to seventh section). In so doing, however, I do find that a different form of neutrality is appropriately applied under a more limited set of conditions, which is lottery-based (see the seventh section). I conclude by showing that in cases where theories of hands-on neutrality have intuitive force, more plausible rival principles can account for our intuitions, including within the abovementioned sports case, governmental language case and chaplain case (see the eighth section).

Avoiding unequal distributions of policy-imposed benefits
As was mentioned, before turning to the problems with theories of hands-on neutrality, there is a prior question that needs to be addressed. Insofar as neutral treatment of citizens' suitably qualified conceptions of the good is important, then why can states not simply refrain from bestowing special advantages on any conception? Given the costs of planning and implementing schemes that compensate citizens' conceptions for unequal distributions of policy-imposed benefits, such hands-off neutrality may seem preferable.
The first thing to note here is that there might indeed be cases where it is both possible and desirable for states to avoid unequal distributions of policy-imposed benefits (for an argument to this effect, see Balint, 2017). For example, most secularists would argue that states with an established religion ought to cut their ties with this religion rather than try to extend (comparable) benefits to other religions by adopting a system of multifaith establishment. What is pertinent for our purposes is that even if hands-off neutrality can be vindicated in some contexts, there are many others in which it cannot. In societies with high levels of linguistic diversity, for instance, budgetary constraints may force states to offer public services in a limited number of languages only. Assuming that to allow for effective communication, the languages chosen for this purpose will need to be ones that a large proportion of the population speaks, such states will end up bestowing unequal benefits upon citizens' cultural conceptions of the good (Pierik and Van der Burg, 2014: 503;cf. Kymlicka, 1995: 115). Likewise, a local government that seeks to discourage youngsters from joining gangs by expanding the opportunities for them to do sports will ordinarily lack the resources to accommodate all sports that are practiced within its jurisdiction. Depending on which sports are supported, the conceptions of the good of some individuals will receive greater benefits than those of others.
But that is not all. Even when it is possible to repeal or to refrain from implementing policies that benefit citizens' conceptions of the good unequally without undermining legitimate public objectives such as the efficient delivery of public services or the reduction of crime rates among youngsters, authors such as Patten have suggestedplausibly -that when this means that no one's suitably qualified conception of the good will receive any additional benefits, there will often be decisive reasons for states to implement policies that benefit the conceptions of some citizens more than those of others (Patten, 2014: 170). For example, even when states could pick national days of rest that suit no particular religion within society (suppose arguendo that joint days of rest are necessary to reap the economic benefits of people taking time off work simultaneously), there seem to be good grounds for choosing the preferred days of some religious citizens to make it easier for them to worship. To accept this, we do not need to accept that choosing the universally inconvenient days of rest is Pareto-inferior. (Whether this is the case depends on whether one thinks that those who are denied special benefits are worse off simply because they are not treated equally.) All that it requires is that we accept that, in such cases, the reasons for accommodating some people's preferences override any reasons for giving them equal treatment.

The moral equality objection
If I am right that it is often impossible or simply undesirable for states to repeal, or refrain from implementing, policies that bestow unequal advantages on citizens' suitably qualified conceptions of the good, then it becomes important to know whether hands-on neutrality can be morally vindicated. The remaining sections will argue that this is not the case.
To show the problem with this kind of neutrality, remember that all versions of it are premised on the notion that when states policies bestow special benefits on some suitably qualified conceptions of the good, then the states in question have pro tanto reason for compensating other suitably qualified conceptions. Regarding the conceptions of the good that are due compensation, Martin (2017) has pointed out recently that proponents of hands-on neutrality make an implicit assumption, namely that the relevant conceptions must have stakes within the same thematic domain as the conceptions that receive the special benefits. Unless such domain restrictions are presupposed, it would be difficult to understand why Patten maintains that the construction of a public skateboard park gives local authorities reasons to fund other types of sports facilities as opposed to, say, places of worship or concert halls (see above). Similarly, it would be difficult to fathom why authors such as Réaume, Schutter and Pierik and Van der Burg maintain that states have reasons for compensating linguistic minorities rather than other groups for the use of the majority language by state agencies.
Before looking at versions of hands-on neutrality that lack these restrictions, I want to consider the problems faced by those that have them. The main problem plaguing these theories is that they do not respect the moral equality of those who care little, if at all, about the goods or activities that comprise the relevant domains (cf. Martin, 2017: 150). Consider those who dislike sports or who are simply indifferent to it. In societies where a local government constructs public football fields, hands-on neutrality might require these individuals to compensate those who prefer playing basketball or tennis over football by helping to fund basketball facilities and tennis courts through taxation. Or consider atheists who dislike religion in all its forms, and who, suppose further, do not care about which day of the week is designated as the national day of rest (for example, they do have a preference for Sunday on cultural grounds). In a traditionally Christian society where employers receive subsidies for giving Muslims time off for worshipping on Friday afternoons, these individuals would be forced to help fund such support.
The reason why people's moral equality is subverted in these cases is not, to be sure, that forcing individuals to pay for policies that bestow greater benefits on others' conceptions of the good -including on conceptions whose goals and commitments fall under domains in which they have no personal interests, such as that of sports or religion -is never justified. To illustrate, suppose that the construction of football fields is both an effective and an efficient way of tackling obesity within a particular neighbourhood. In this case, it would seem legitimate to force those who dislike sports to help fund such facilities through their taxes, assuming that governments have a moral duty to help create and maintain conditions under which those living within their jurisdiction can live healthy lives, one that derives from a more general duty to create and maintain a just and safe society.
Rather, the reason why the moral equality of the sports-haters and atheists is subverted in the abovementioned cases is that they are forced to contribute to the construction of sports facilities and the provision Friday-worship subsidies as compensation for the construction of football fields and the designation of Sunday as the national day of rest, respectively. While these forced contributions help to offset unequal distributions of policy-imposed benefits within specific domains (that is, that of sports and religion), they do so by creating another type of inequality that seems least as problematic, namely one between those with significant intra-domain interests and those without such interests, such as the sport-haters and atheists. Call these inequalities 'extra-domain inequalities'.
To see the problems with extra-domain inequalities, notice that if the sports-haters and atheists had not been forced to pay such compensation, they might have been able to spend this money on their own conceptions of the good. For example, any ornithologists among them might have been able to spend it on a new set of binoculars while any motorbikers among them may have had the option to spend it on a biking trip across the country. Given that these individuals have equally legitimate interests in living selfdirected lives as those they are required to compensate, forcing them to pay compensation in such cases seems to flout their moral equality.
Indeed, even when the money spent on mitigating intra-domain inequalities would have been taken from the sports-haters and atheists anyways, perhaps as part of fixed amounts of taxes that they and others ought to pay, the fact that this money could have been spent on goods that are central to the realisation and maintenance of a just and secure society -for example, environmental protection, road safety, a flourishing economy -suggests that their moral equality is still not being respected. The reason is not simply that these counterfactual expenditures might have personally benefitted them or those nearest and dearest to them, though this will often be true. Even when no such (direct) benefits would have been obtained, the fact that people can be reasonably expected to contribute to goods that help to achieve and maintain justice and security means that being forced to do so does not undermine their moral equality. For example, as helping fellow citizens to live minimally healthy lives is widely -and I assume here correctly -seen as a requirement of justice, we do not ordinarily think that people's moral equality is violated when a small part of their taxes is being spent on the development of medicines for those with rare genetic diseases even if some tax-payers neither personally benefit from this (at least not in any straightforward manner) nor have friends or relatives who do so. 5 The obvious response here for proponents of hands-on neutrality would be to say that compensating citizens for state-created intra-domain inequalities is a requirement of justice as well. However, as we have just seen that offering such compensation creates another type of inequality that seems at least as objectionable, namely extra-domain inequalities, it is far from clear that it is. After all, whatever justice requires, it is unlikely that it requires states to redistribute resources when the redistribution is no moral improvement upon the status quo ante.

Feelings of disrespect
The challenge faced by Patten and other proponents of hands-on neutrality, then, is to explain why intra-domain inequalities would be more problematic than extra-domain inequalities and, indeed, so much more problematic that creating extra-domain inequalities becomes justified to mitigate them, as when the abovementioned sports-haters are forced to compensate the basketball fans and the atheists are forced to compensate employees who are observing Muslims. Although Patten does not address this issue, he does provide various examples of state-created intra-domain inequalities 6 and notes that such inequalities are often 'consequential for the respect that individuals feel that they are receiving from others' (Patten, 2014: 169). To the extent that this is so, it might be argued that insofar as the creation of extra-domain inequalities does not give rise to feelings of disrespect or simply less often and/or to weaker feelings, for states to create such inequalities is justified to mitigate the intra-domain inequalities that their policies have produced.
One problem with this argument is that it is unclear whether state-created extradomain inequalities are less widely experienced as disrespectful than state-created intra-domain inequalities and/or as less disrespectful than the latter. To return to the Sunday rest example, it seems that many atheists would feel at least as strongly disrespected if their tax money were used to help Muslim workers take time off for worshipping as Muslim workers might feel as a result of being denied such financial support within a society where Sunday is the national day of rest. Not only does the provision of such worshipping subsidies (indirectly) force atheists to accommodate beliefs that they might consider to be hopelessly irrational, if not morally abhorrent, they may well feel that being required to fund these beliefs subordinates their interests to those of religious people. To see why, it should be noted that, had there been no believers within their society or simply no believers from different faiths, they would not have been required to make these sacrifices under theories of hands-on neutrality.
But that is not the only complication for the current argument. Even if state-created intra-domain inequalities are (generally) more consequential to whether people feel respected or not, it does not follow that it must therefore be justified to force those without significant intra-domain interests to address these inequalities. To see this, it should be noted that there are various cases where people wrongly perceive themselves to be disrespected. For example, some members of the American National Rifle Association feel disrespected by the implementation of stricter gun laws; some homophobes by the introduction or continued existence of same-sex marriage; and some creationists by legislation that requires evolutionary theory to be taught in schools. In these and other cases, the fact that people feel denigrated by state policies does not appear to provide reasons, let alone decisive reasons, for using public funds to compensate them. What this means for us is that even if state-created intra-domain inequalities tend to cause wider and/or deeper feelings of disrespect than state-created extra-domain inequalities -which I have questioned within the previous paragraph -this does not show that creating the latter inequalities must be justified to mitigate the former inequalities. Defenders of hands-on neutrality first need to establish that those who feel disrespected by statecreated intra-domain inequalities have good reasons for feeling this way. As I will argue in the sixth and seventh sections, there are good reasons for doubting whether this can be done.

Expanding the scope of beneficiaries
One way to respond to the moral equality objection that thematic domain restrictions of theories of hands-on neutrality raise is to abandon these restrictions. Doing so would mean that when state policies confer special benefits on some citizens' conceptions of the good, all the suitably qualified conceptions of citizens that do not receive comparable benefits will need to be compensated as opposed to only those with stakes within the same thematic domain.
As Martin (2017: 158) has pointed out, the problem with versions of hands-on neutrality that drop thematic domain restrictions is that they are bound to be prohibitively expensive. Let me illustrate this with an example. Consider Patten's claim that when a local government decides to build a skating park to promote the values of 'public health and community', it should compensate residents who are into different kinds of sports by providing a range of different facilities as well (Patten, 2014: 122). According to some estimates, an 8000-square foot skateboard park costs approximately US$360,000 and serves approximately 25,000 residents (Whitley, 2018). Now suppose that within a jurisdiction of this size, 60 per cent of the population does sports at least once a week, as is the national average in countries such as the Netherlands (Nederlandse Omroep Stichting, 2016). Suppose further that of the 15,000 residents who do sports regularly, 1000 individuals are active skateboarders (which seems a generous estimate given that skateboarding is practiced predominantly by young individuals and given that other sports, such as football and jogging, tend to be more popular in most contemporary societies). Based on these figures, we can calculate that if the government of this jurisdiction constructed the 8000-square foot park, it would bestow resources worth US$360 on each individual skateboarder. Insofar as the support given to other types of sports ought to be prorated as Patten's account of hand-on neutrality requires, this would mean that the government in question will need to spend US$5,040,000 (US$360 Â 14,000) on compensation.
If this is not prohibitively expensive already for a local government with a population of 25,000, it will almost certainly be when public support is extended to residents with suitably qualified conceptions of the good in which sports does not feature prominently, such as those devoted to the production of art or religious worshipping. Even when a significant number of residents within its jurisdiction have conceptions of the good that are intolerable and that are therefore ineligible for compensation, say a 1000 (imagine that there is a large community of Neo-Nazis), the relevant government would still have to spend another US$3,240,000 (US$360 Â 9000). In fact, not only would it have to make such payments for the construction of the skateboard park; it would have to do so for all policies that bestow unequal benefits on residents' suitably qualified conceptions of the good. 7 I have focused thus far on a recourcist version of hands-on neutrality. It is worth noting, however, that welfarist versions do not fare much better in terms of their feasibility. Whereas some individuals who are due compensation on welfarist accounts might have cheaper tastes than those who derive greater-than-equal benefit from state policies (for example, to promote the conceptions of the good of football players as much as the conceptions of the good of skateboarders are promoted when the skateboard park is constructed, less per capita resources will need to be spent as football fields are comparatively cheap), others will have more expensive ones (for example, to promote the conceptions of the good of ice-hockey players as much as those of the relevant skateboarders, more than per capita equal resources will need to be spent as ice rinks are comparatively expensive). Indeed, the fact that there are usually some citizens with very expensive tastes (for example, ones for drinking expensive wines, traveling the globe) suggests that a welfarist approach will frequently be more expensive and therefore more difficult to realise.
Hands-on neutrality does become more feasible once we add a perfectionist requirement to it. Such a requirement narrows its scope by restricting the conceptions of the good that are eligible for compensation for unequal distributions of policy-conferred benefits to those that make substantial contributions to people's flourishing. This would rule out, for instance, that sports such as lucha libre, roller derby and mud-wrestling would need to be compensated for the construction of the aforementioned skateboard park insofar as it is true (as is commonly assumed) that these activities have little value.
I think that even perfectionist versions of hands-neutrality are bound to be prohibitively expensive. To see why, it should be noted that almost all (perfectionist) philosophers believe that people can live valuable lives in a wide range of ways. 8 On most axiological accounts, for instance, lives devoted to intellectual inquiry, the production of art, athletic excellence, meaningful relationships, volunteering and the enjoyment of nature are all likely to be valuable without there being any obvious reasons for thinking that some are more valuable than others, whether this is because they are broadly equally valuable or because the values that they instantiate are incommensurable if not incomparable. But if this is correct, then it seems that a large proportion of citizens within contemporary societies have conceptions of the good with valuable goals and commitments, which renders it doubtful whether perfectionist versions of hands-on neutrality restrict the scope of conceptions that are eligible for compensation enough to make this type of neutrality achievable.

Partially realising hands-on neutrality
In response to this objection, proponents of hands-on neutrality might point out that just because an ideal cannot be achieved fully, it does not follow that states should not seek to realise it partially (cf. Balint, 2015: 498). One way in which they might seek to do this is by compensating only a subset of suitably qualified conceptions for unequal distributions of policy-imposed benefits.
To show the problems with this approach, consider the following analogy: Bicycle: In order to stay reasonably mobile, a family with under-aged triplets is forced to buy each family member a bicycle as it can no longer afford its car. Whilst children A and B dislike cycling, it so happens that the purchasing of the bicycles accommodates a longstanding wish of child C: To have her own bicycle. Since the parents are proponents of hands-on neutrality, they want to compensate children A and B for this by buying each of them their preferred gift. Due to the family's precarious financial situation, however, the parents are able to accommodate only one child, for which they have to lower the children's pocket money. The parents decide to introduce this cut and they end up buying B a pair of roller-skates. (Suppose arguendo that the price of the roller-skates equals that of a bike and that the roller skates makes an equally large contribution to B's conception of the good as the bike does to C's, so that B's has been neutrally treated relative to C according to both recourist and welfarist standards.) Some might doubt whether the parents' decision to buy the roller skates is really neutrality-promoting. For although child B is fully compensated for the benefits bestowed on child C, the result of this is that the inequality between child A and children B & C widens. Even if hands-on neutrality is satisfied to a greater degree, however, the parents' decision remains unjustifiable towards A. To see this, it should be noted that not only has the inequality between her and B & C increased, A now has less money to spend on her own hobbies. In the absence of reasons for thinking that these costs are fairly imposed on A (alongside any costs to her self-respect and psychological well-being that she might incur as a result of this inequality), we should conclude that A's interests have not been respected.
To be sure, we can imagine scenarios where it is not the case that A is left with less money to spend on her conception of the good than she would have had otherwise. Suppose that rather than lowering the children's pocket money, the parents took the money for B's roller skates from the family holiday budget. Here it is not just A who is forced to make sacrifices for B's gift but the entire family. Does this make a moral difference? The answer, I suspect, is negative; since there is no good reason for denying a gift to A rather than to B, the parents' actions seem to remain unjustifiable towards A.
I believe that we can extrapolate from this case. Insofar as states can choose between fully compensating a subset of citizens with suitably qualified conceptions for unequal distributions of policy-imposed benefits and not giving compensation to anyone, the latter option appears to be fairer given that it does not arbitrarily prioritise the interest of some citizens over the interests of others. This holds true regardless of whether those who with non-compensated suitably qualified conceptions would have kept the money that they are forced to spend on compensation (as in the pocket-money scenario) or whether this money would have been spent on some public or general good (as in the holiday-budget scenario). To return once again to the skateboard park case, what this means is that when a local government cannot compensate all (citizen) residents with suitably qualified conceptions of the good for the benefits bestowed on skateboarders when a skateboard park is constructed, it should refrain from compensating anyone rather than compensate only football fans and ornithologists, say. (Which, it should be noted, does not rule out that the government might have sufficient reason to build a football field and/or to subsidise a bird-watching tower when this serves legitimate public objectives; the point is simply that the construction of these alternative facilities would not be justified on compensatory grounds.) If the foregoing is correct, then insofar as partially realising hands-on neutrality is to be morally acceptable, it will involve giving less than equal compensation -as defined by either resourcist or welfarist standards -to all citizens whose suitably qualified conceptions do not benefit from a given state policy or simply not to the same degree as others. To go back to the bicycle analogy, this would be like giving a book voucher to children A and B to compensate them for the benefits that purchasing the bicycles has conferred on C's conception of the good (assuming that the monetary value of these vouchers is lower than that of a bike and that, as seems likely, the extent to which these vouchers help A and B pursue their conceptions of the good is smaller than the extent to which receiving the bike helps C pursue hers). To be more precise still, it would be like giving each of the two children a lollipop or something else that is very cheap, as compensating all suitably qualified conceptions within society for unequal distributions of policy-imposed benefits is likely to be so expensive for governments that only very modest forms of compensation will be affordable.

Objections against partially realising hands-on neutrality
So far, we have seen that the domain-specific restrictions of existing accounts of hands-on neutrality raise an equality-objection that can be avoided only by compensating all citizens with suitably qualified conceptions of the good for unequal distributions of policy-imposed benefits, not just those with stakes within the same domain as the conceptions on which the relevant states bestow special benefits. Building on Martin (2017), we have also seen that when the scope of beneficiaries is expanded thus, principles of hands-on neutrality cannot be realised fully because of the prohibitive costs that this would impose. As I went on to suggest, however, proponents of hands-on neutrality might reply that hands-on neutrality can and should still be realised partially. After considering different ways in which this might be done, I found that the fairest way is to provide all citizens with suitably qualified conceptions of the good with less than per capita equal compensation rather than giving some of these citizens full (that is, per capita equal) compensation and others nothing.
My next aim is to argue that states should refrain from partially realising hands-on neutrality in this way as well. To vindicate this claim, let us start by considering cases where granting even small amounts of compensation for unequal distributions of policyimposed benefits looks objectionable: The single-best way to improve the air quality within a given region is to close a local airport and plant a forest on its former location. The single-best way to help prevent children from drowning within a country is to make swimming classes mandatory. The single-best way to help feed the hungry within a particular neighbourhood is to subsidise a Sikh soup kitchen or Langar.
While these policies bring about goods (for example, clean air, safety, social welfare provisions) that might directly or indirectly benefit all members of society, they bring additional benefits to individuals with certain conceptions of the good. These include nature-enthusiasts (by creating new forests), ardent water-polo players (by increasing the pool of potential players) and Sikhs (by helping them to live in accordance with their religious conceptions of the good). Nonetheless, it seems odd, if not perverse that people with different conceptions of the good might be owed compensation for this. Indeed, this is true even if some of these policies render it more difficult for them to pursue their conceptions of the good; for example, they might cause the Nature-Enthusiasts Society to gain members at the expense of the Plane-Spotters Society; water-polo clubs to gain members at the expense of football clubs; and the Sikh community to attract converts from other religions. To say that these groups are owed compensation suggests that their members are treated unfairly by the abovementioned policies. Yet given that these policies help to realise pertinent public objectives (one that are central to the realisation of a just and secure society), namely providing clean air, protecting children from drowning and feeding the hungry, and use the single-best -that is, the most efficient, morally acceptable -means to do so, it is difficult to see how this can be.
In response, proponents of hands-on neutrality might object that I am attacking a straw man. They might point out that authors such as Patten consider state policies to be problematically non-neutral only when they impose unequal benefits on suitably qualified conceptions of the good relative to an appropriate baseline, which, on Patten's account, consists of a state of affairs where basic liberties are secured, socio-economic justice is realised and, more generally, where there is framework for 'fair opportunity for self-determination' (Patten, 2014: 114-118). What this means is that even when policies that are necessary for realising this baseline are 'especially accommodating of a particular conception of the good, no equivalent, offsetting policies are needed to re-establish neutrality ' (p. 118). However, if hands-on neutrality does not apply to policies that are necessary for what might be called the 'realisation of justice and security' for short, then it appears that no compensation is due for the three policies mentioned above. After all, by providing clean air, protecting children from drowning and feeding the hungry, respectively, each of these policies promote goals that are central to the realisation of a just and secure society.
While the scope restrictions under consideration render theories of hands-on neutrality more morally palatable (as well as more feasible), they raise doubts as to whether many of the examples given by Patten and others of when this type of neutrality applies do not similarly fall outside its scope. Is the construction of sports facilities not also necessary for realising a basic goal of justice, namely the promotion of physical health? And how can the state deliver public services effectively if it does not use languages that large segments of the population speak? Upon reflection, it looks like many, if not most, of the things that states do -for example, ensuring access to education and healthcare, stimulating the economy, fighting crime -contribute in essential ways to the creation and maintenance of a just and secure society. But if this is so, and if citizens cannot be owed compensation for the unequal benefits that such policies bestow upon their conceptions of the good, then theories of hands-on neutrality have a much smaller scope than many of their proponents acknowledge.
All this does not necessarily mean that they are invalid, of course. There remains the possibility that hands-on neutrality is appropriately pursued under the following conditions: i. When citizens' suitably qualified conceptions of the good benefit to unequal degrees from uniquely justified policies that pursue goals other than justice and security. ii. When citizens' suitably qualified conceptions benefit to unequal degrees from unjustified policies. iii. When citizens' suitably qualified conceptions benefit to unequal degrees from policies that are all justified but not uniquely justified, that is, that are interchangeable.
Let us examine these conditions in the order stated.

Policies that do not promote justice and security
Regarding (i), some might argue that justified policies that pursue goals other than justice and security are oxymoronic. On this view, states should stick promoting justice and security and otherwise leave people free to use their resources as they see fit. Whether this is true or not -clearly, a lot will depend on how these concepts are interpreted -let us assume for the sake of argument that such policies are not incoherent. Might states have duties to compensate citizens for the unequal effects of said policies? Since policies that fall under (i) are uniquely justified, meaning that there are no other policies that could pursue their goals equally well (as determined by their likelihood of success, efficiency, and any positive and negative side-effects that they might have), it is difficult to see how this can be. To focus attention, suppose that the singlebest way for a state to help citizens with lowbrow hobbies adopt more valuable pastimes is by regularly showing art-house movies free of charge within popular blockbuster cinemas. (Alternatively, one might imagine that promoting art-house movies is a necessary part of any package of measures with the greatest probability of promoting high culture among these individuals.) While promoting art-house movies -along other forms of high culture -might be a requirement of justice on some perfectionist theories, namely those predicated on the assumption that it is morally incumbent on states to create conditions under which citizens can live flourishing lives, see, for example, Raz (1988), many theories would regard this as something that is not necessary for the creation of a just society.
The question here, then, is whether the relevant state might have a duty to compensate citizens who attach little if any importance to these kinds of movies, or perhaps simply those who value other types of high culture insofar as compensating those with lowbrow pastimes defeats the state's goal of incentivising people to live more exacted lives (for example, the state might have a duty to compensate ballet-lovers by subsidising ballet plays, but not lucha libre fans by subsidising this type of wrestling because of its low value). As I already alluded to, I believe that the answer is negative. To say that compensation might be due suggests that there is unfairness in promoting arts-house movies over, inter alia, ballet. But given that -as I have stipulated -subsidising ballet would be a less effective or efficient way of promoting high culture among those with lowbrow hobbies, whether because these individuals already know to find their way to the blockbuster cinemas in which these movies are shown in this case, and/or because ballet is simply less interesting to them, it is unclear how this can be. Another way of making this point is to say that since ballet would have been publicly assisted had it had been more effective or efficient in realising this policy goal, there is no improper favouritism involved in the state's promotion of art-house movies and therefore no moral case for compensation.

Unjustified policies
Of course, states do not always implement policies that are justified. This raises the question: Is hands-on neutrality a cogent normative principle in cases where unjustifiable policies bestow unequal benefits on citizens' suitably qualified conceptions (ii)? I believe not given that the appropriate way to respond to such policies is to revoke and possibly replace them rather than compensate people for their unequal effects. To illustrate, suppose that promoting abstinence among youngsters is an ineffective or simply morally impermissible means for achieving a legitimate policy objective: Reducing teenage pregnancy rates. In this case, the proper government response would be to spend its budget for the pursuit of this goal on alternative ways of reducing teenage pregnancies (for example, on subsidising contraceptives and spreading knowledge about their use) rather than on compensating groups of citizens for the special advantages that the state's propagation of abstinence confers on those who oppose premarital sex (usually devout believers). Likewise, insofar as organising an annual football tournament imposes excessive costs on a population due to the stringent security measures that need to be taken, the proper response would be to end the tradition rather than to compensate citizens who are into different types of sports or past-times more generally.
This is not to deny that when groups have suffered disproportional burdens under an unjustified state policy, they might be owed reparations. For example, France's attempts to eradicate local languages during the 19th century as part of its nation-building efforts 9 might impose a duty on the current French state to remedy this evil by supporting the formerly oppressed languages, at least insofar as the relevant historical injustices have not (yet) been superseded. 10 What is important for our purposes is that this does not vindicate hands-on neutrality, as the reason why public assistance would be due here is that those local languages were wrongly suppressed, not that the actions of the French state bestowed unequal benefits on different linguistic groups as such. To illustrate, it should be noted that if the latter were true, then there would similarly be reason for the French state to compensate citizens whose mother tongue is Chinese or Japanese for the use of French as the language of government and administration. But this seems highly implausible.
What about cases where a state has a long-term goal to revoke public support that is unfairly granted to a specific group, say the majority culture or religion, but where it currently abstains from this because doing so is likely to result in violent protests or other highly detrimental outcomes? Under such non-ideal circumstances, there are good grounds for thinking that leaving the unfair privileges in place to avoid greater evils might be justifiable, if not morally required, and that this might actually justify, if not morally demand, that cultural and religious minorities be compensated for this. As in the previous case, however, such compensation would be due to rectify an injustice (in this case an ongoing one), not to remedy a policy-imposed inequality as such. Given as much, such cases do not vindicate hands-on neutrality either.

Interchangeable policies
This brings us to the final set of cases, namely those where a state can choose between interchangeable policies that are all justifiable -whatever this might involve exactlyand that favour different suitably qualified conceptions of the good (iii). By 'interchangeable', I mean that the relevant policies have an equal probability of realising their goals and use means that that are equally efficient and morally defensible, taking into account any positive and negative side effects that they might have. Consider some examples: Art museum vs opera house: A cash-strapped state seeks to stimulate the tourist industry in order to gain more revenue. The best way of doing this is to either build a new art museum, which would accommodate art fans, or to build a new opera house, which would accommodate opera fans. (Suppose that building both venues is impossible due to budgetary constraints whilst constructing smaller, less grandiose versions of each is not a viable alternative either as this would attract fewer tourists.) Sabbath vs Sunday: A country seeks to hold a legally binding referendum on leaving the European Union. For reasons of economic efficiency, the referendum can be held only on a Saturday or Sunday and not on both days. Whilst the expected public benefits of these options are equal, the choice for Saturday would accommodate observing Christians whereas the choice for Sunday would accommodate observing Jews.
I assume in these scenarios that for the relevant states to pursue neither alternative would be unacceptable as it would mean that their goals of promoting the tourist industry and of letting the demos decide about a major political issue will go unfulfilled. At the same time, the fact that the policies in question are interchangeable as far as their public benefits are concerned means that, by definition, there are no public or impartial reasons for choosing the preferred policy of one group of citizens over the preferred policy of the other group. Under these conditions, a more limited version of hands-on neutrality might seem appropriate. This version requires the relevant states to simply choose one policy and then compensate citizens whose suitably qualified conceptions of the good would have benefitted from the implementation of the other, interchangeable policy. For example, if the first state chooses to build the art museum, then it must (partially or wholly) compensate the opera fans for this choice, whereas if second state chooses to have the election on a Saturday, then it must (partially or wholly) compensate the Jewish population for this decision.
The main problem with this version of hands-on neutrality, I believe, is that it does not respect the moral equality of those who are indifferent as to which of the interchangeable policies are implemented. One might think of people with lowbrow hobbies who do not care about whether the art museum or opera house is built and of atheists and Muslims who do not care about whether the vote takes place on a Saturday or Sunday. Why should they be forced to compensate those whose preferred policy is not implemented? Assuming that they have equally legitimate interests in living self-directed lives, doing so seems to unfairly subordinate their interests to those of other citizens. (Which does not mean, of course, that forcing them to help fund the implementation of one of the disjuncts in each case is impermissible; while it is true that these contributions would help to bestow special benefits on other citizens' conceptions of the good, the justification for this is that implementing one and only one of the policies in question serves legitimate public objectives rather than that it bestows such special benefits, which means that people's moral equality is being respected.) What if only those whose preferred policy is implemented are required to compensate those whose suitably qualified conceptions of the good would have benefitted from the implementation of an interchange policy? In the cases at hand, this would mean that insofar as the art museum is constructed and the referendum held on a Saturday, only artlovers and Christians would be required to compensate the opera-fans and Jewish community, respectively.
Even when these groups can be accurately identified, this proposal remains unappealing. While people might care deeply about art and about their religion, there are other things in life that may matter more them, such as spending time with friends and family. This is apposite, for it means that when the art-lovers and Christians are faced with a choice between, on the one hand, compensating the opera-lovers and Jewish community for the implementation of their preferred interchangeable policies and, on the other, not having their preferred policies implemented but avoiding such compensatory payments (which would allow them to spend the money that is saved on visiting their family and friends or on any other goal that they may have), some might choose the latter. To the extent that this is the case, for states to force these individuals to pay compensation for the construction of the art museum and for holding the referendum on a Saturday, respectively, would fail to respect their autonomy.
It might be replied that while these costs to people's autonomy are undesirable, favouring the interests of some citizens by implementing their preferred interchangeable policy without compensating citizens whose preferred interchangeable policies are not implemented is undesirable as well. Indeed, some may argue that the inequality created by this constitutes a greater evil and infer from this that, in the absence of fairer alternatives, states should therefore pursue the limited version of hands-on neutrality under consideration.
Without trying to assess this comparative claim here, I think that there is a fairer alternative, one that realises a different type of neutrality. This type of neutrality, which has not been discussed within the literature on state neutrality to date, does not require states to equalise the benefits that their policies confer on citizens' suitably qualified conceptions of the good in either a resourcist or welfarist way. Rather, it requires that when different suitably qualified conceptions stand to benefit from different interchangeable policies, states give their adherents a per capita equal chance of having their preferred interchangeable policy selected within a lottery. This means, for instance, that when there are three million art lovers and one million opera lovers in the above example, the former should be given a 75 per cent chance of winning and the latter a 25 per cent chance. (To determine how large each group is, states might use demographic surveys or proxies such as visitor numbers of existing museums and opera houses; alternatively, they might use self-identification, for example by asking people to register online their preferences for either the art museum or opera house. While each of these methods has its own advantages and disadvantages, an assessment of these is beyond this article's remit.) How does this approach improve upon the alternatives just mentioned? Unlike theories of hands-on neutrality, it is more respectful of people's autonomy. First, it does not require citizens without stakes in the choice between the interchangeable policies to compensate citizens who do have such stakes. Second, it does not require those who are willing to forego the implementation of their preferred policy to avoid having to compensate other citizens to pay such compensation, which I have suggested would undermine their autonomy as well. At the same time, however, the lottery-based approach avoids the opposite evil of playing favourites among citizens by simply implementing one's group preferred interchangeable policy. Even when the choice for a particular interchangeable policy is based on the fact that it accommodates the preferences of the most citizens as opposed to the fact that it serves the policymakers' personal preferences, the use of a lottery seems fairer as it gives each citizen with stakes in the state's choice between interchangeable policies a chance of having their preferred policy implemented.
Yet, some might ask, is there not something unfair about the fact that the question of whose preferred interchangeable policy gets implemented is decided by sheer fortune? My answer is 'perhaps', but this still looks like the fairest way to proceed. As we have seen, whichever group of citizens is forced to compensate others for unequal distributions of policy-conferred benefits under theories of hands-on neutrality, there are always some citizens whose interested are being flouted.
Another objection does not oppose the use of a lottery as such, but rather the fact that under the proposed lottery, there will frequently be individuals who are more likely to have their preferred policy implemented under the proposed lottery because of differences in the number of citizens who favour specific interchangeable policies. In the above example, for instance, an art-lover is three times more likely to win the lottery than an opera-fan. My response to this is that these differences do not look objectionable once we consider that when the preferences for particular interchangeable policies are unevenly distributed, giving each policy an equal chance of being drawn would mean that the interests of some citizens are given no weight at all. On the contrary, treating people fairly appears to require in such cases that the lotteries be weighted.
Still another objection concedes that the use of weighted lotteries is appropriate when there remains a significant minority of citizens whose suitably qualified conception of the good would benefit from the implementation one of the interchangeable policies. This is arguably the case in the example I gave earlier, where among those with stakes in the choice between the opera house and art museum, 25 per cent are opera fans and 75 per cent are art fans. But suppose now that only 1 per cent of these individuals are opera fans and 99 per cent art fans. Some critics might say that, in such cases, the size of the discrepancy no longer renders it morally impermissible -and possibly morally required -for the state to adopt the policy that the overwhelming majority prefers even though from a public perspective, the policies are interchangeable (see my comments earlier in this subsection). On this view, simply giving the 99 per cent group a 99 per cent chance of winning the lottery is not enough as there would remain a small chance that its members will lose out, which would mean that the 1 per cent's preferred interchangeable policy gets implemented.
I think that the use of weighted lotteries is fairer in such cases as well. The reason is that, unlike the majoritarian approach just outlined, it gives moral weight to the interests of each citizen with stakes in the choice between the interchangeable policies by granting him or her a chance of winning the lottery while, at the same time, remaining responsive to the existing numerical inequalities by assigning the policy that is preferred by a greater share of citizens a greater chance of winning. However, even if I am wrong about thisand our critics might insist that when people are so heavily outnumbered, they cannot reasonably complain about the majority's preferred policy being implemented without prior lottery -using weighted lotteries would still be appropriate in various cases where the discrepancies are smaller. (Think again of ones where the distribution is 25/75, as well as of ones where the percentages are even closer to being evenly split.) Finally, for those who find the use of (weighted or unweighted) lotteries for choosing among interchangeable policies unorthodox and perhaps even outright odd, it is worth mentioning that lotteries are already used as an impartial way of allocating a variety of scarce resource, including university places, medical resources and monetary prices. Given as much, there is not any obvious reason for thinking that they could not become socially accepted as tiebreakers for choices among interchangeable policies.

Concluding remarks
If my criticisms in this article are cogent, then the prospects for hands-on neutrality look bleak. 11 Regardless of the specific version that is accepted, we have seen that there are always some citizens whose equal status this type of neutrality fails to respect. But if this is correct, then there remains an important question: Why are so many political and legal philosophers drawn to theories of hands-on neutrality?
Although I can only speculate here, my sense is that the cases that are commonly invoked to support such theories are misleading, including the ones mentioned in the introduction. To start with, consider the policy of some predominantly Christian states to provide prisoners with access to chaplains. This case is adduced by Pierik and Van der Burg (2014: 501) as one where principles of hands-on neutrality apply. Yet while it is certainly unfair for a state to provide Christian prisoners with spiritual support that is tailored to their religious beliefs but not prisoners with different religious backgrounds by hiring, for example, Imams and Rabbis, we do not need to buy into theories of handson neutrality to arrive at this conclusion. Since most contemporary societies are religiously diverse, including their prison populations, it seems that insofar as a religiously diverse state has the aim of providing spiritual support to prisoners, such support can only be effectively provided if it does not simply hire chaplains, but also Imams, Rabbis and so on. In so doing, benefits are bestowed on different religious conceptions of the good. Yet rather compensating the adherents of non-Christian conceptions for the unequal benefits that are conferred when only chaplains are hired, this policy is more plausibly described as one where religion-specific accommodations are made to realise a given policy objective, namely that of providing spiritual support to prisoners. That is, in extending similar services to other religious groups, states are not trying to off-set the non-neutral effects of a given policy, but rather implementing a policy that requires multiple religious groups (which include both Christian and non-Christian ones) to be accommodated to be successful.
Similar conclusions apply to Patten's claim that when a local government decides to build a skateboard park to promote the values of 'public health and community', it should compensate residents who value different kinds of sports -for example, football, baseball -by providing other types of sporting facilities as well (Patten, 2014: 122). Whereas I suspect that many would agree that a range of different sporting facilities ought to be provided here, this too does not show that hands-on neutrality is a sound principle. The reason is that merely constructing a skateboard park is unlikely to promote the values of public health and community as effectively or efficiently as providing a wider range of facilities. But if this is correct, then the problem here lies in how these values are being pursued rather than in the unequal benefits that constructing only a skateboard park confers. That is to say, while it is plausible to think that skateboarders are not the only group whose favourite type of sports ought to be accommodated, the reason is not that unequal distributions of policy-conferred benefits among citizens' suitably qualified conceptions of the good are always (pro tanto) problematic as proponents of hands-on neutrality maintain. It is rather that the local government has failed to implement a more effective or efficient policy that would have accommodated a wider range of sports.
As was mentioned, a final case that is often invoked to support theories of hands-on neutrality involves state organisations (for example, parliaments, courts, the police) using the language of one particular group of citizens to communicate. Many normative theorists have argued that, within linguistically diverse societies, there will frequently be some cultural minorities with their own language who ought to be compensated for the fact that their language is not used for political and administrative purposes, or simply not as extensively. 12 In the Netherlands, for instance, the Frisians -an ethnic minority in the north of the country -might be seen as having moral claims to compensation for the dominance of the Dutch language within state organisations. Again, however, it looks like more plausible rival principles can account for such moral claim-rights insofar as they exist. To illustrate, it should be noted that few, if any, proponents of hands-on neutrality are willing to accept that the Dutch state has a moral duty to compensate citizens who are native Japanese or Chinese speakers for the dominance of the Dutch language by supporting their respective languages. What this suggests is that any legitimate Frisian claims to public support for their language, as well as any similar claims that historical linguistic minorities in other countries might have such as the Welsh in the UK and Catalans in Spain, are not based on the fact that the suitably qualified conceptions of the good of many of their members benefit unequally from the way state organisations operate. They rather have different grounds, such as the fact that they are entitled to forms of nation-building or self-government for which the preservation of the local language culture is essential and/or the fact they have suffered historical injustice at the hands of the central government (think of the cultural and linguistic oppression that the Catalans have suffered).
While it is easy to appreciate why people might be intuitively drawn to theories of hands-on neutrality, then the various problems encountered by such theories along with the explanatory power of rival principles suggest that these intuitions should not be trusted.