Friday, 10 May 2013

How to justify a ban on the burqa (or anything else)

Bans on wearing the burqa and other face-covering religious garb (such as are under consideration or recently passed in several European countries) fall under a class of restrictions by government on the free choice of individuals over private matters. They thus have the appearance of being illiberal, of disrespecting people's natural rights to manage their own affairs in general, and to follow their own plan of life in particular. In fact, it is possible to justify such a ban in liberal terms. But not just any kind of ban will do.

Political debate about the burqa in the west is dominated by an unfortunate bigotry, a species of moral foolishness antithetical to liberalism. I have heard and read serious arguments for banning the burqa because it causes vitamin D deficiency (lack of sunshine), because people will try to rob banks dressed in burqas, because this is alien to our face-to-face culture, and so on. Such arguments are, respectively, trivial, stupid, and xenophobic (if not racist). 

Yet it seems to me that there are in fact plausible liberal arguments for banning the burqa (and various other things, such as addictive drugs), which focus on the harms that the burqa may do to the personal autonomy of particularly vulnerable women and girls. This line of reasoning is of course inconsistent with the libertarian interpretation of liberalism, in terms of the sanctity of individual liberty from the depredations of government restrictions, unless there is evidence that you are using your liberty to harm others in some significant way. The libertarian thus takes Mill's  'harm principle' seriously, if rather narrowly,

The only purpose for which power can be rightfully exercised over any member of a civilized community, against his will, is to prevent harm to others. (Mill, On Liberty)

The narrowness of the libertarian view has two dimensions. First, libertarians presume that people are already and always rational, and thus that their desires and opinions reflect their autonomous will. Yet while this is a good 'respectful' normative stance to take to other people in general, it is false in its particulars. We all know that everyone is less than fully rational at least some of the time (eg), and that some people are dysfunctionally irrational much of the time (such as the severely mentally ill, or drug addicts). What someone sincerely believes and feels that they want may not be consistent with what they would conclude if they had had the opportunity and capacity to consider their desires in the light of their other more important interests and values (i.e. what philosophers call practical reason). 

Second, libertarians presume that the only source of oppression that needs to be defended against is the government (vertical power). Yet we all know that the 'horizontal' subjugation of individuals by non-state actors, such as religious communities and families, can also be a significant source of oppression in people's lives. This is particularly significant because of the vulnerability of the development of personal autonomy. While the capability to reason autonomously about one's values and plan of life is essential to the idea of the human person in a teleological sense, this does not mean that it is 'natural' in a biological sense. Autonomy is something that develops from childhood on (hence the need for liberal parenting and education) and that may be constrained or deformed (for example, by mental illness or certain forms of parenting and education).

Like Mill himself, I therefore take individual autonomy rather than mere liberty as the foundational moral concern of liberalism. At the practical level, this means that liberals may question whether people's choices are necessarily free simply because the government isn't coercing them. Liberalism as a political doctrine utterly rejects the use of the apparatus of the state to subjugate citizens (the particular concern of libertarians). But it also requires protecting people from other forms of subjugation, such as unchosen entanglement in oppressive social relationships. In such cases well-designed government interference can be autonomy enhancing.

Turning back to the case of the burqa, feminists have argued that wearing it makes women socially dysfunctional and undermines their rational autonomy, and that it is even designed to do so. Women who wear the burqa have difficulty functioning adequately in society, since it imposes severe constraints on employment and general social interactions, and even things like eating in public. They are artificially separated - segregated - from other people (even other burqa wearers), and this seems to be exactly how the garment is supposed to work. 

Quite aside from making life generally more difficult, that separation reduces women's capability to benefit from the free interaction in a free society which, as Mill noted, is important for exercising and developing personal moral autonomy and a conception of the good life that is genuinely one's own. In the Western context, losing this opportunity for public interaction seems particularly significant because the very kind of communities and families that would require 'their' women to wear such clothing are likely to be exactly the kind of cloistered environments in which opportunities for free thought about values and personal exploration of different ways of living are especially (and deliberately) constrained.

In addition, it can be argued that the burqa is not only the outcome of misogynist thinking but appears deliberately designed to impose that perspective on its wearers. The justification for wearing the burqa is the characterisation of women as sexual objects by nature whose status in public life is of non-persons. Wearing a burqa from an early age reinforces and perpetuates that false thesis in the minds of those wearing it, and also, conveniently, in the minds of those who make them wear itThe wearing of the burqa itself thus plays a performative function in helping to make this illiberal ideology into reality, and it is arguably intended to do exactly this. 

Thus, the liberal case for a ban is that the burqa seems liable to systematically subvert women's rational autonomy - their freedom to think for themselves - in ways that go far beyond the practical restrictions it imposes as a particularly uncomfortable item of clothing.  

Now many readers will immediately object that these claims are empirically false. The burqa doesn't necessarily have these effects and it is perfectly possible for women to choose to wear it out of a free  and fully autonomous choice. Of course you're quite right. Nevertheless, it is also true that these problems do exist at least sometimes, and are ethically significant (at least to liberals). Welcome to public policy!

Fortunately, public policy analysis does not stop at identifying the harms of some practice and demanding that it be stopped, even if that is how newspapers often present it. Done properly it is concerned with adjudicating the costs and benefits of public intervention in liberal terms. Do the benefits of a ban, in preventing the harms that the burqa may do to some women, outweigh the harms that the ban might itself do to the general freedom (and freedom of religion) of others? 

There is nothing new about this requirement, which follows from the bluntness of law as a policy instrumentA law should be consistently applied - that is, treat all the cases that come before it in the same way - and this means that it is a limited instrument for dealing with issues that fall on a spectrum. A law is designed with a particular ideal target in mind, a specific sort of case considered of particular importance. If the law is well thought out, written and enforced then it should be able to adequately address that ideal target case. Yet it may well be a bad fit for many other cases that fall under its jurisdiction and must be treated in the same way. 

In considering whether to pass or revise a law one should therefore consider not only how well it deals with the ideal target case, but also the degree of 'collateral damage' it may itself introduce by mistreating other cases that fall under its jurisdiction. Sensible drug laws, for example, require making the right trade-off between preventing or minimising the harms that drugs can do to the functioning and rational autonomy of some people, and the harms that such restrictions or punishments impose on those who are quite competent to use them recreationally. Likewise for age-of-consent laws; anti-gambling laws; and so on. In many cases such scrutiny will reveal that the collateral damage is so great that, whatever its good intentions, this law or policy makes things worse rather than better (as for example with the US 'war on drugs'). 

The question of whether liberals can ban burqas therefore requires both an ethical evaluation and an empirical (social scientific) one. The first - ethical evaluation - concerns how we weight the harms burqas may impose on vulnerable women, and especially girls completely subject to their family's wishes, against the benefits autonomous women may gain from being allowed to wear them. How, for example, should we value an outcome where one hundred women are liberated from the burqa while one thousand women are prevented from following what they see as their freely affirmed religious duty? 

From the liberal perspective I think liberation should be weighted much higher, since being prevented from developing moral autonomy in the first place seems an orders of magnitude greater 'crime against the person' than relatively minor restrictions on how people may exercise their moral autonomy in public. Thus it seems to me that liberals have principled grounds for at least considering a ban, even if it means government interference in at least some women's exercise of individual autonomy about their freely chosen plan of life. Nevertheless, the sacrifice of the liberty of the many in order to protect the autonomy of the few is not something to be done lightly. It is a last resort that marks the limiting case of the ethical analysis.

The empirical analysis concerns the factual and logistical side of the case for public policy. Having decided what matters, we still have to perform a reality check to decide whether a law is appropriate and how it should be designed. Just how significant and prevalent are the ethical harms, or benefits, of the burqa? What are the mechanisms by which they are brought about? How could a ban be formulated to disrupt the harmful 'subjugation' effects of the burqa on vulnerable women with the least interference on the liberties of others? (For example, only for under-18s?) What positive social policies might helpfully complement a ban (or even be better by themselves than having a ban)? How might a ban itself impact the mechanisms under consideration - e.g. might dogmatic families and communities simply ban their women from leaving the house at all? 

The liberal justification for a ban on the burqa must demonstrate three things with this empirical analysis. That it would protect a non-trivial number of vulnerable women and girls from a significant harm to their autonomy; that the 'collateral damage' to voluntary burqa wearers is much less than the benefits to vulnerable women; and that only a ban can achieve this. But if these rather demanding conditions are met, then not only do we have good liberal grounds for a ban on the burqa, we would seem to have a duty to impose such a ban.


  1. Very interesting take on this problem. Would you think that the best solution would be a ban based on age?
    I guess that a secondary effect of a full ban would be that many women would not be able to leave their homes - I am thinking about very conservative and religious families - removing them even further from any interaction with society at large.

    1. A ban on under-18s would certainly seem easier to justify than a general ban, on the basis of the same evidence of harm to women's autonomy.

      This is because under-18s are often considered to be still developing their autonomy - to have values, interests and desires, but not the ability to judge them properly. On the one hand this means that one can justify paternalist interventions to protect/support their development without the evidential burden of demonstrating their lack of autonomy. On the other hand it means that their liberty is not as valuable as that of more mature individuals, since the exercise of 'freedom of religion' can't mean as much to people who are not yet mature enough to affirm their religion for themselves. Therefore the 'collateral cost' of a ban is low.

      This is the logic behind age-related bans on 'adult' activities from sex to alcohol consumption.

    2. I guess that has been said about many other groups including about women in the past. Still under 18s are in a tough situation where religion is concerned. They are completely dependent on their family for sustenance and many times their only option is to leave the home. Due to their age they are not entitled to the same levels of social acceptance thus have difficulty in finding work that allows them to survive. This is a good basis for some kind of action on society’s part.

      I understand that such ban would curtail the rights of those youngsters who freely chose to wear the burka. It is a matter of determining if any of these groups is overwhelmingly small and act accordingly.

      Maybe, on the other hand, this is a vanishingly small problem in the UK. I wonder if there are any numbers on this matter?

  2. The burka is evidence of primitive thinking, comparable to lily-feet in old China and lips stretched by wooden disks in parts of the old Congo. and even today, genital disfigurement of various kinds. Much of it is religious, primitive in itself. Moreover, the burka is a kind of sexual, in-your-face obscenity, a self-condemnation of a particular culture. Regardless of all that, I think the wearers and their masters should be let alone to signify those absurd cultural values. Actions speak louder than words.

    1. As a liberal I don't really believe in cultures' rights, and certainly not their right to inflict atrocities on 'their' own people. I start from and end with the moral significance of individuals.

  3. If the burka is truly an atrocity or if it encourages atrocities I stand with you. It is abominable in enlightened eyes, even appalling, but the victims of that persuasion do not seem horrified or express revulsion. I don't think the person so imprisoned has lost her moral significance but has had it sadly increased.

  4. Let me give you the feminist line since Mill. If this device is intended to create a state of subjugation and acquiescence to that state, then it is an attack on individual freedom and, indeed, individuality itself.

    Evidence of acquiescence to injustice cannot be taken as evidence that there is no injustice. Not unless one thinks that justice is only concerned with how happy people feel with their lives (cf Richard Layard), rather than with the relationships between people.

  5. You are a better arguer than I, so I suppose your argument is best, too, and I salute you for it.

  6. think you have the best argument you talk about respect for individualism which refers to respect for somebody else, in how you think, act and have their own beliefs

  7. I doubt libertarians oppose restraint of children's conduct only if their conduct would harm others. I think they'd also restrain children on the ground of harm to their own future prospects. That can be interpreted as harm to others, but only indirectly and then any conduct producing any kind of harm can be found to entail harm to others.

    On a ban of burqas, a virtually identical case is made for banning miniskirts and sexually related clothing on young girls who don't know the genderal meanings but wear it because they like being liked. Females wear minis or burqas because that is what men of their respective cultures overwhelmingly reward. The underlying determination is partly one of locating who has the bulk of the power (men of the respective culture) and how to get some of it (satisfy the men including sexually, reproductively, culturally, and religiously). What makes the two different is that one represents "us" and the other "the other", which is invertible except that we prefer our culture and we don't trust the judgment of the people who prefer the other culture, which suddenly makes the one ban something to discuss and the other subject to being called Stalinist. (I have a critique that the cultures are not equally good in secular terms, but that's a separate argument.) In the U.S., the clothing ban to be efficacious has to be limited to girls, because we assume adults can wield autonomy to societal benefit and so the laws on sex, gambling, and drug self-administration by adults are loosening up over recent decades. A ban generally doesn't work unless it is against a relatively small-minority view, because the argument for it is that parenting applying the clothing to be banned is abusive or neglectful, and that has been found to be nearly hopeless to apply except against low-income or oddball families with less legal or political power. A theologically-involved ban in the U.S. would be legally unstable unless and until it fails. History in North America tells of forced integration of native Indian children (as to clothing, language, family structure, law, religion, etc.) and the splits that caused: some assimilated but some did not and some wound up with children stolen as late as within our own lifetimes, with adults drunken, and with many living in poverty until casinos opened that legally could open only on Indian reservations; and we need to add the rapes by White male neighbors entering reservations secure that no one will convict them because jurisdictional limits are firm. We seem to have a history of underweighting collateral effects of similar kinds of bans when the bans apply only to people of "the other".


    1. You point out the significance of double standards in the real world where foolish or nasty policies are supported by social prejudices about class, race, or 'foreignness' etc.

      My post outlines specific ethical and empirical criteria that provide an objective test of policies that claim to restrict liberties in the name of protecting autonomy. That can help prevent a society's double standards from dominating political policymaking.

  8. how are we supposed to define moral autonomy? Why do you set the limit at under 18 years old? Who are you to decide when someone is acting autonomously or not? Wouldn't education of liberalism and self autonomy by more helpful than an outright ban?
    Besides, the banning of the burqa never seem to be for the reasons you put, they're just political manoeuvres at the expense of some minority, as you state in your second paragraph. Without stating that these bans don't meet your requirements I feel like your article is legitimising these essentially political actions
    Also can you apply the same approach to beauty pageants, where young girls are subjected to wearing make-up, dancing in scantily clad clothing by their mothers? shouldn't we ban discuss a ban on that too then, because these young girls may not have the moral autonomy to decide to partake in it?

    1. These are not well formed questions and do not engage with the post.

      Beauty pageants may be deplorable, but unless they inflict loss of liberty they hardly seem a relevant case.