Hijab ban interview: Why a secular court should not tell a citizen what her religion is

On March 15, the Karnataka High Court upheld a hijab ban for educational institutions. The court argued that wearing the hijab was not an essential religious practice in Islam and therefore would not be protected by the Constitution.

The case has thrown into sharp relief how courts use the essential religious practices test. In effect, a secular court decides what is “essential” to a complainant’s faith and then, on that basis, does or does not award Constitutional protection to her beliefs.

Speaking to Scroll.in, Tarunabh Khaitan, professor of public law and legal theory at the Faculty of Law at Oxford University argues that a secular court should not attempt to objectively interpret faith. The judiciary would be much better off subjectively approaching religion, taking into account what adherents themselves believe in, subject to logical tests.

This would prevent the state from arbitrarily taking away religious freedoms, argues Khaitan. This approach would have seen the hijab ban struck down.

Edited excerpts below.

What problems do you find with the essential religious practices test as it stands right now given that it is criticised so much? Is there a different way to protect religious freedom which will also allow for a progressive interpretation of faith?
There are two dimensions to whether your freedom of religion has been breached. We first need to ask what is the scope of the right and whether the state’s act in question has infringed that scope, and then we can come to whether the infringement is justified because of public order, morality or health.

The broader you draw the scope of the right, more the number of acts of the state the right will catch. So the first issue is how broadly do you understand this scope.

With respect to religious freedom, there are two broad approaches to defining this scope: one is the subjective approach and the second is the objective approach.

Let us understand the subjective approach first. Because religion is so complicated, adherents tend to take a personalised approach to religion. And this has been well established by several social science studies: that religious adherence shows enormous diversity and pluralism, not only between religions but also within religions. In fact, my supervisee Radhika Agarwal is working on a thesis that claims that in multi-faith contexts like India, people not only pick and choose from within a particular religious context, but frequently pick beliefs and practices from across religions.

Now, if you are serious about protecting freedom of religion, then your understanding of what religion means will have to be subjective. This does not mean anything goes, there will still have to be some evidentiary benchmarks: we will still need to know that the claim is being made sincerely and that the adherent is making a plausible claim.

Your religious claim cannot be so outlandish that it would seem implausible to a reasonable person who is similarly situated as you are. For example, some people in the United States called themselves Pastafarians, worshipping the Flying Spaghetti Monster who apparently created the universe after getting drunk. Their purpose was to mock religion generally. They would fail the sincerity test, and probably the plausibility test as well.

The subjective approach does not allow the court to sit in judgment over what this individual claimant’s religion is. It demands that it should take their word for it, subject to the plausibility and sincerity tests.

But that subjective approach casts the scope of religious freedom extremely wide. Because of this exact worry, the Supreme Court has taken an objective approach. Instead of asking how the claimant adheres to her religion, the objective approach instead conducts two distinct inquiries: first, the theological inquiry asks what the official doctrine of the religion says (determined by courts using evidence of experts or by interpreting texts), second, the sociological inquiry looks into what most of the claimant’s co-religionists believe or practice.

So, if a particular practice or belief is either determined to be unsupported by theological doctrine or not widespread amongst the claimant’s co-religionists, the court will find that it does not fall within the scope of religious freedom, even though the claimant sincerely and plausibly thinks it does. Her subjective understanding of her own religion is irrelevant under the objective approach. The objective test significantly narrows the scope of the right to religious freedom.

But the court does not stop here. It narrows the right even further by asking whether this belief or practice is also essential to that religion. This is also determined objectively by reference to both the prevalence of the practice and the official doctrine. Thus the essential religious practices test casts the scope of the right to freedom of religion extraordinarily narrowly.

The essential religious practices test is also problematic because it is disrespectful to people whose rights are supposedly being defended here. The court takes upon itself to tell religious people that they are wrong in their interpretation of their own religion. This is simply not what secular courts can or should do.

It is much better for the court to adopt the subjective approach and accept a sincere and plausible religious claim at face value and shift the bulk of the inquiry to the justification stage.

An example should clarify: assume if somebody says “I do not want to pay taxes, because my religion says I should not pay taxes to the state, but only to God”. In that case, what the Supreme Court’s objective essential religious practices approach will either say that is just your subjective interpretation of religion, but objectively your religion does not forbid payment of tax. Or it might say that most of your co-religionists happily pay taxes so it cannot be against your religion. Or it will say, even if it is forbidden by your religion, this prohibition is not essential to your religion.

Basically, applying the objective approach, the claim will be dismissed at the scope stage, and the question of justification will not even arise as the court will not find any infringement with the right to begin with.

However, if the claim is a sincere and plausible one, what the court should do is accept that the claimant’s religious freedom is engaged simply because she sincerely and plausibly believes this. However, the court will tell the claimant that whether this will exempt you from taxes will depend on the state’s justification of the infringement of her religious freedom. As part of this justification inquiry, the court will weigh the pros and cons of the public interest in everybody paying their taxes and the public interest in defending the claimant’s religious freedom. It is highly likely that a wise court would conclude that the restriction on religious freedom imposed by taxes is a justified infringement on religious freedom.

So, in this case at least, the outcome of the case under the objective as well as the subjective approach is likely to be the same, but the latter is a lot more respectful of religious freedom. It is very different from the insulting claim that the objective approach requires: that you do not really know what your religion is, and we, as judges of a secular court, will tell you. That’s essentially what the court told the hijabi women in the Karnataka case.

It is not wise for secular courts to get into the business of interpreting anybody’s religion, majority or minority.

The state certainly does not know best when it comes to religion. This extreme kind of paternalism in matters of religion is infantilising. It is one of the key reasons why we have not completed a transition in our jurisprudence from thinking of people as citizens rather than subjects.

How does a subjective approach change the court’s interpretation of religious freedoms? Would it change the outcome of the hijab case?
Under the subjective approach, where you cast the net wide, most of the jurisprudential heavy lifting is done at the justification stage, rather than the scope stage. The scope is very permissive, it is a very low threshold, most sincere claimants should be able to jump over it. The real issue is whether the restriction on my religious freedom is justified. And for that, the court has to use what is called the proportionality test, which is: is the restriction proportionate? Here is what it means.

All of us occasionally face a dilemma when we have conflicting things that we want. Since we cannot have everything, we need to find the right balance. So, an example would be: say your family’s income is limited and you have to make a choice between sending your child to an excellent but expensive school but not being able to afford any health insurance for the family. Or sending your child to a cheap school but getting the best health coverage available. Or sending your child to a reasonably good school and getting a reasonable level of health insurance for the family.

Most of us would think that the third choice strikes the right balance. This is the reality of human existence: we cannot have everything we want or need. What we can do is optimise based on what’s possible.

Proportionality is a test to determine whether things that are in conflict are indeed things worth having, and if so, whether the state is striking the right balance between them. To explain, let me change my example a bit: Say the choice was between having a very expensive foreign holiday and what kind of school your child should go to. I think most people would think that even putting the holiday on the scales vis-a-vis education is illegitimate. So, unlike health, where balancing health against education seems reasonable, balancing what is a luxury with what is a necessity seems unreasonable.

So here is the first stage of the proportionality inquiry: as soon as you have satisfied the scope test, you are already assuming that there is something valuable, at least on one side of the equation. Because freedom of religion is important as a fundamental right, a person who is claiming her subjective right to freedom of religion is asking for something valuable and we want to be the kind of society where we respect people’s decisional autonomy in religious matters.

This is where we need to get into a deeper inquiry. What is the value or good on the other side? Are there also legitimate items that should be put on the balancing scale of proportionality against a valuable fundamental right? I will explore some possibilities, without doing an exhaustive summary of the hijab case.

A clearly illegitimate aim behind the hijab policy would be a prejudice-motivated goal of disadvantaging Muslims. Such a goal is so thoroughly illegitimate that it is even worse than the holiday analogy, which is illegitimate as a goal only in relation to a fundamental necessity. It is not wrong per se. But the goal of disadvantaging a community due to rank prejudice is illegitimate on its own. It has no redeeming feature.

Another conflicting goal is uniform policy. On the one hand, we have a fundamental right to religion. On the other hand, we have a uniform policy, and there may be some good reasons for schools to have one. In this case, the purpose of the uniform policy is inclusion, a legitimate goal. But is a uniform policy that is being used for an exclusionary purpose useful, if its consequence is some girls not coming to school at all? This part of the inquiry will be taken up in the subsequent steps of the proportionality test.

Gender equality comes up next. Now, gender equality is a very important good in our constitutional framework. So if there is a genuine conflict between religious freedom and gender equality, it would be a case like education versus health rather than education versus holiday.

In gender equality, the first step is clear. Religious freedom and gender equality are legitimate objectives, nobody can fight that. Then you ask: does the measure in question actually serve gender equality, the thing that you say you are achieving through this uniform policy?

Is the measure in question actually suitable to serve that policy?

First, let us go back to our education and health example. Suppose with the money that you saved from sending your child to a second-rate school, instead of buying quality health insurance, you used it to reconstruct your house to align with some sort of supernatural astrological principle which you believe keeps your family healthy.

Now, while both education and health are legitimate objectives, the mechanism you are using to take from education and enhance health is unsuitable. In the sense that, yes, you are taking away from education, but what you are giving to health is not actually going to health. That is unreasonable and unsuitable.

That is where the gender equality argument falls in our current example.

Gender equality cannot be achieved by coercing women to not get educated. You cannot make women equal by reducing their freedom. You can only make women equal by expanding their options and autonomy.

There was a very good case called Anuj Garg v Union of India where a law prohibited women from serving as bartenders. That law was challenged as breaching gender equality. The state tried to justify it saying we are actually protecting gender equality because drunk men sexually harass women in bars.

But the court said that is not a good enough justification because the mechanism is unsuitable to achieve gender equality. If you really want gender equality, you do not do that by restricting women’s choices. You achieve it by ensuring the safety of women.

Exactly the same argument will apply here. What the state is trying to do here in the name of equality is reducing the option of hijabi women rather than expanding them.

After the legitimacy and the suitability stage, the third is the necessity stage: do you actually need to infringe this right in order to achieve your purposes? Maybe your family has a stash of money sitting somewhere, which you can use to make sure that you will provide both a first-class education and a first-class health care to your family.

So is there another way of doing this, which in this case, will translate that is there a way of both ensuring that the religious freedom of the girls is not violated and the uniform policy is respected? Here it would obviously be to make some kind of reasonable adjustment. We can ask: are there exceptions being made to the uniform policy elsewhere? If there are, for example, turbans being allowed for Sikh men, then that already suggests that whatever the value of the uniform policy is, the value is not disserved by making some minor adjustments.

Thus, adjustments, like wearing the same colour hijab as the uniform etc., serve the purpose of inclusion better as students can also see both the sameness and the differences between them. In other words, refusing to make minor adjustments to the uniform policy is not necessary to achieve the relevant purpose of inclusion – to the contrary, a uniform policy that allows minor adjustments is more inclusive.

And the final stage in the proportionality test is balancing. Once you have the legitimacy, suitability and necessity question, you will then ask: is this too much interference for too little gain? Is the cost you are paying in denying these women quality education too high a cost for a very small gain in maintaining uniform policy, even if it is a legitimate policy.

Had the court carefully drawn to each of these elements, there was only one possible outcome: the exclusion of hijab-wearing students from educational institutions is clearly disproportionate.

How do you deal with situations like what happened in Karnataka, where students started wearing saffron scarfs, stoles and turbans in school saying that if hijab is allowed, their religious clothes should be as well?
At least on the basis of media reports, it seems to me that what some of these students were doing was weaponising the saffron scarfs, and that to my mind, would clearly fail the sincerity test. That context would also have raised grave doubts in the court’s mind about the legitimacy of the college’s claims as well, i.e. whether the weaponisation of the restriction on religious freedom is beginning to look a lot like the persecution of a minority. If the actions are motivated by prejudice, then we are in very easy territory, because then the state is clearly breaching the non-discrimination guarantee and the equality guarantee and is acting on prejudice and hostility.

Of course, that will turn on evidence the claimants can present. I think speeches given by ministers, members of the ruling party, the timing of this move close to elections – all of those things will be relevant in determining what is motivating them.

Note however that proving malice on the part of public authorities is one of the hardest things to do in constitutional law. But in this case, it is at least within the realm of plausibility that the colleges have not been acting on bona fides.

In this present case, how can you also make arguments based on expression, privacy, autonomy and non-discrimination based on Articles 19, 21, 15 and 14?

How do you understand these rights which you get as an individual along with rights, such as religious rights, which you get as parts of a group, but both seem to be working together? A common argument is that this kind of claim will open floodgates of litigations? How do you tackle that?
Let us start with the right to freedom of expression. Humans give meaning to a variety of activities, including actions, which are called speech acts, that convey meaning. It can include gestures, like the act of putting red vermillion within a ritualise setting on a woman’s forehead is a speech-act that signals the finalisation of a marriage.

We express ourselves through the food we eat, the dress we wear, the gods and goddesses we pray to or do not pray to, etc. So it is, without doubt, the case that wearing or not wearing the hijab is an expressive act.

Now, it may be that your interest in a particular expressive act is stronger or weaker. If it’s related to religion, you may have a very strong interest.

You can have a strong expressive interest even if it is individual to you. If you are really concerned about climate change and you want to wear a t-shirt to school saying do not burn coal, that may perhaps have a very strong emphasis, even though it does not have a group dimension.

But just as we talked about in the religious case, this bogeyman about floodgates should not arise: the wide scope of a right can still be contained at the justification stage.

In the first step, you may want to wear a particular t-shirt to school since it is a part of your expression of freedom. While that claim may be true, that is the beginning of the inquiry, that is not the end of it.

You then get to the justification stage. You ask: well if we do not permit this, what do we get? If we permit it, what do we lose? So, the justification analysis is a way of making sure that you do a proper cost-benefit analysis.

You justify a breach of rights if there is sufficient value to be gained. You cannot say what you like. You cannot shout fire in a crowded place. People may die.

This is what it all boils down to. Do you want to live in a country where the state needs to give reasons to take away your freedoms or live in a country where the state can take away your freedoms arbitrarily.

You have also written that an indirect discrimination claim is better suited here. How can an indirect discrimination claim be made in this case? And how likely was it to stand in a court of law?
Indirect discrimination was argued in the case, but the court dismisses it in two sentences. The court says there is no discrimination here, because the initial policy is the same as applied to everyone. But the entire point of indirect discrimination is you apply a facially neutral policy to everybody, but the burden of the policy falls differently on different groups of people.

The idea of indirect discrimination can be explained through the famous parable of the stork and the fox. The fox invites the stork for dinner and offers a meal on a shallow plate, which the stork is unable to eat given its long beak. And then the stork returns the favour by inviting the fox to dine from a tall pitcher with a very narrow mouth.

But, in the real world, not all storks can reply in kind, because there are power differentials. Here, foxes decide the rules and the storks often lose out.

A classic example is having stepped access to buildings, without any ramps: this is indirectly discriminatory against disabled people.

So in this case, a uniform policy that bans the wearing of anything on your head is indirectly discriminatory not just against Muslims but also against Sikhs. If the policy only bans the wearing of the hijab, then that’s not even indirect discrimination, but direct discrimination because it will only affect Muslims.

Direct discrimination will be based on the evidence that is available to establish an intention to discriminate. If the petitioners can show that there is singling out of hijab-wearing women, based on circumstantial evidence, they may succeed in establishing direct discrimination. But this can often be hard for claimants to prove in courts.

The beauty of indirect discrimination is that it has nothing to do with intention. All it requires is to show who is the policy affecting the most. If the group being disproportionately hurt by this policy is a protected group, then that is indirect discrimination irrespective of intentions.

Thus, indirect distribution is an easier case to make because it is very clear, whatever your intentions might have been, that the group that has been hurt here is hijabi Muslim women. And because religion is a ground for non-discrimination, they are a protected group.

Which claim do you think was the most likely to stand in court? Should the petitioners have relied on other claims than primarily relying on the essential religious practices?
I think the indirect discrimination claim should be the most likely to succeed because it is so blatantly obvious about what is going on, who has been affected.

But I think under current Indian jurisprudence claims under Articles 14 [equality], 15 [prohibition of discrimination], 19 [freedom of speech] and 21 [life and liberty] were all likely claims to stand in court because the proportionality analysis will fail in each of those cases.

But the religious freedom (Article 25) claim, under the current jurisprudence, could go either way. That is because the current jurisprudence is faulty and the court should revisit the essential practices test and adopt a more subjective approach.