Religious and related discrimination
The President of the Employment Appeal Tribunal gives us his insights into the issues that have arisen in relation to religious discrimination and human rights, and the impact that these might have on the development of the law under the Religion or Belief Regulations.
On this page:
Race discrimination and religion
The Religion or Belief Regulations
The scope of the Regulations
Meaning of religion or belief
When is discrimination direct and when indirect?
Accommodating differences
Conclusions.
This article is based on the second annual Harvey Memorial Lecture given by Mr Justice Elias.
A few years ago, I wrote an article with a colleague of mine, Jason Coppel, arising out of the dismissal of Glenn Hoddle as the England football manager. Some of you may remember Hoddle – a brilliant, inspirational and successful footballer; an uninspiring and unsuccessful manager.
He gave an interview with The Times talking about his strong religious beliefs. These included the belief that you reap in this life what you have sown in an earlier incarnation. Your karma works from another lifetime. So Hoddle said that he believed that people with disabilities were made that way because of past misdemeanours.
The political world was outraged. The sports minister, Tony Banks went on television to say that Hoddle’s views were “totally unacceptable. If his theory is correct, he is in for real problems in the next life. He is doomed to come back as Glen Hoddle.” The PM appeared on television on The Richard and Judy Show and said that it was “difficult” for him to remain in his job. Others leapt to demonstrate their righteous views. And so Hoddle was sacked.
I was shocked, but for a different reason. Did we really live in a society where someone could be dismissed for expressing in a temperate way his or her religious views? Did freedom of speech count for nothing? Must a figure in the public domain keep his religious views a secret lest his beliefs cause offence? Would anyone even suggest that Richard Dawkins should be sacked as an Oxford Professor for writing about religion in a way that causes very real and deep offence to many religious people?
It is not even as if Hoddle’s beliefs were out of the religious mainstream. Members of the Hindu religion, many Buddhists and those of other religions share this belief. Could the FA advertise the England manager’s job and say “Hindus and Buddhists need not apply?” If not, why not, if Hoddle’s dismissal was justified?
Yet that would surely be quite scandalous, infinitely more objectionable than Hoddle’s temperate expression of his views. Or is there some unspoken assumption that it is acceptable for foreigners to hold such views, but not white English Christians? One would hope that the Religion or Belief Regulations would make this unlawful. After all, it is not even as if he was being discriminated against for a manifestation of his beliefs. It was for the beliefs themselves.
When we wrote the article, the Regulations were not in force. The purpose was to explore the protection, if any, that might have been afforded to Hoddle under art. 9 of the European Convention on Human Rights. That draws a distinction between freedom of thought, conscience and religion, which is unqualified; and the freedom to manifest one’s religion, which is subject to the qualification that it may be subject to: “such limitations as are prescribed by law and are necessary in a democratic society in the interests of public safety, for the protection of public order, health or morals, or for the protection of the rights and freedoms of others.”
I start with a consideration of this provision because it will inevitably have a considerable influence on the construction of the Religion or Belief Regulations. Of course, in principle, art. 9 applies where there is an interference with the freedom of religion irrespective of whether there is also discrimination.
It is art. 14, read with art. 9, that protects against unlawful discrimination with respect to rights conferred by the Convention. But, in practice, in most cases an infringement of art. 9 does potentially involve discrimination. The complainant usually is being treated differently to those who hold different religious views.
Our conclusion was that the human rights courts have made a rather poor fist of protecting religious freedom. Three of the leading cases in the employment field are Ahmad v London Borough of Hackney and another [2001] EWCA Civ 812, Stedman v United Kingdom [1997] EHRLR 545, 23 EHRR CD 168 and Konttinen v Finland [1996] ECHR case no.249/49/94. These are three decisions of the European Commission on Human Rights in which employees were alleging that they were required to work hours which, in accordance with their religious beliefs, were incompatible with their religious duties (they were Muslim, Seventh-day Adventist and Christian respectively.)
In Ahmad and Konttinen they had originally signed up to these hours and were wishing to change them, but in Stedman, it was the employer who was insisting that she should change her hours. In each case, the Commission held that their claims were not arguable because their religious convictions were not even engaged; the cases never even got to the European Court of Human Rights (ECHR).
This Convention jurisprudence was considered by the Court of Appeal in Copsey v WWB Devon Clays Ltd [2005] IRLR 811. Mr Copsey was a practising Christian and objected on religious grounds to working on Sundays unless it was an absolute emergency. The employers were faced with an upturn in demand which required all employees to work a seven-day shift. Attempts were made to accommodate Mr Copsey’s interests but to no avail. Ultimately, he was dismissed and claimed unfair dismissal. He alleged that, in assessing this, the employment tribunal should have taken into account the rights conferred by art. 9. The Court of Appeal held that, given that the employers had acted reasonably and explored the alternatives to dismissal, the dismissal was fair and art. 9 would have made no difference even if applicable.
There was a difference of view as to whether the article was even engaged. Mummery LJ analysed the Convention jurisprudence and concluded that these three cases constituted a clear line of authority that the Convention was not even engaged in a case of this kind because the employee was always free to resign in order to manifest his religious beliefs. “If you don’t like the heat, get out of the kitchen,” was essentially the response of the ECHR jurisprudence. Lord Justice Mummery observed that it was hardly compatible with the fundamental character of art. 9 for it to be so restricted, but he felt constrained to follow the jurisprudence.
Lord Justice Rix was not quite so pessimistic about the effect of the Commission’s decisions. He felt that while they established that an employee could not allege an interference with his art. 9 rights if he has freely entered into the contract and accepted the restrictions, a different principle applied where, as in this case (and indeed in Stedman), it was the employer who was seeking to change the contractual terms.
The rationale for the distinction was that whether there is an interference with the right depends on the extent to which an individual can reasonably expect to be at liberty to manifest his or her religious beliefs. Someone who has freely accepted the limitations cannot expect the same freedom as someone who accepted the job on the understanding that it is compatible with their religious obligations. On the facts, however, the interference with the right was proportionate because the employers had investigated the possibility of accommodation and acted reasonably in requiring him to work.
Lord Justice Neuberger observed that the decisions of the Commission were “arguably surprising and the reasoning hard to follow”. On any view, the protection afforded by art. 9 is very narrow, and it seems likely that even on the analysis adopted by Rix LJ, the employer would not have been expected to vary the terms of employment so as to accommodate the employee in the first place.
In the later decision of R (on the application of Begum) v Denbigh High School [2006] UKHL 15 the House of Lords affirmed that the approach of the ECHR was indeed to conclude that there was no infringement of art. 9 at all where a person voluntarily puts himself or herself in a position where the religious practice or observance is not accommodated and there are other means of manifesting the practice or observance without undue hardship or inconvenience. In that case, the majority of the House of Lords (Lords Bingham, Hoffmann and Scott) held that a schoolgirl who refused to wear the school uniform and insisted on wearing the jilbab could not establish that her art. 9 rights had even been infringed.
They accepted that the wearing of the jilbab was a manifestation of her rights, but she knew when she went to the school what the rules were; and there were other schools she could have attended which would have permitted her to dress as she wished. The minority (Lord Nicholls and Baroness Hale) were inclined to the view that there was an infringement.
However, all their Lordships were sure that any interference was justified in the circumstances. It leaves open the question of whether there would have been an interference with the rights if the school had changed the dress code whilst she was a pupil. Stedman and Rix LJ in Copsey would suggest not, and that is surely the better view; the circumstances are then very different.
Race discrimination and religion
The absence of religious discrimination legislation in the UK did not leave all victims of religious discrimination bereft of remedies. In some cases the link between religion and ethnic origin is such that protection for the former is achieved under the guise of protection for the latter.
In the seminal case of Mandla v Dowell Lee [1983] IRLR 209 the House of Lords laid down the principles for determining whether a group can be described as being of a common ethnic origin or not. In that case it found that Sikhs were a separate ethnic origin and therefore discrimination against someone for wearing a turban (in that case a boy at school) could constitute indirect discrimination that would require justification. So discrimination against Sikhs as such will constitute direct race discrimination. Jews have also been held to be a distinct ethnic group so the same principle applies to them, but Rastafarians not.
It is also possible for other religions to bring themselves within the race legislation, but only indirectly. Neither Muslims nor Christians share a common ethnic origin and therefore discrimination against them will not be direct race discrimination. Exceptionally, however, it may be indirect. For example, discrimination against Muslims may constitute indirect discrimination against Asians or perhaps persons of Pakistani or Bangladesh origins, although this requires a finding of disparate impact. It will then of course be for the employer to justify the treatment.
This covert protection via the race legislation is hardly a satisfactory way of protecting against religious discrimination, quite apart from its limited scope. First, the discrimination which a Muslim feels for, say, being prevented from wearing the veil or being given time off to pray, is religious discrimination. Similarly, the turban is a religious symbol for most Sikhs and discrimination on that ground is more a religious than a racial slur. The law, therefore, may provide a remedy but it does so without engaging with the real complaint.
Second, the operation of indirect discrimination in these contexts can be wholly arbitrary. A Muslim from a predominantly Muslim country may establish that a rule adversely impinging on Muslims adversely affects persons of his or her national origin, whereas another Muslim, who is victimised in precisely the same way but comes from a country where Muslims are in a minority, will not. The employer will have to seek to justify the treatment in one case but not the other.
The Religion or Belief Regulations
The 2003 Regulations are intended to give effect to the European Directive establishing a general framework for equal treatment in employment and occupation. They must of course be construed consistently with that Directive. They follow the familiar pattern of discrimination law, with protection from direct discrimination, indirect discrimination, victimisation and harassment.
In R (on the application of Williamson and others) v Secretary of State for Education and Employment & Others [2005] UKHL 15 Lord Nicholls said: “Religious and other beliefs and convictions are part of the humanity of every individual. They are an integral part of his personality and individuality. In a civilised society individuals respect each other’s beliefs. This enables them to live in harmony. This is one of the hallmarks of a civilised society.”
(He went on to add, however, that much inhumanity has been perpetrated in the name of religion, although as Stalin, Hitler and Mao show, lack of religious beliefs is certainly no more a guarantee of enlightened humane attitudes.)
Before considering some of the difficulties posed by the Regulations, I wish to make some preliminary observations about them. There are three features of the Religion or Belief Regulations to which I wish to draw attention and which, to a greater or lesser extent, distinguish religion from discrimination on other grounds.
The first is that a person’s religion (or lack of it) is often not apparent; unlike sex or race, it will generally not be immediately identifiable. Sometimes it is, of course; it will be reasonably obvious that certain persons are, for example, Sikhs or Jews because of the headgear they wear, although even then it may be their racial or ethnic groups rather than their religious convictions that will be apparent. But often an employer will have no idea of someone’s religion or belief.
The kind of direct religious discrimination that is perhaps more likely to arise in those circumstances is from an employer who firmly holds a particular religious belief and seeks to recruit exclusively from those who share those religious views. This will amount to discrimination against all those of a different or no religious persuasion and, as I will indicate in a moment, such discrimination falls within the purview of the Regulations.
The second feature is that frequently, from the employee’s point of view, it is not only the belief but also the manifestation of that belief, in rituals and practices, which demands consideration. In all sorts of ways there are religious obligations that may conflict with the requirements of the job. The examples I have given about the need for time off for worship are obvious illustrations. Others include wearing religious jewellery or symbols.
An important question is how far the employer is obliged to respect the conscience of an employee who feels that his religious duties override his duties to his employer. How far must he positively accommodate the needs of the employee?
The third feature is that in this area, more than in any other area of discrimination, there is frequently a clash of conflicting rights. This is inevitable once beliefs are protected. There is no limit to the potential range of beliefs, rational or irrational. A minority may have beliefs that significantly depart from the views of the majority.
So the law has to regulate the priority to be given to these conflicting rights. The classic example in the context of religion is the place afforded – or more accurately not afforded – to women and gay men in certain religions. There is no avoiding the need in a pluralist and multicultural society to make difficult choices that will, in particular contexts, give priority to one claim over another. Fashioning the ground rules for deciding how such conflicts should be resolved is highly contentious, not least because there is no accepted priority of rights. The Regulations do provide, within very narrow limits, for religious concerns to trump other human rights. So, for example, it is lawful for various religions to refuse to accept women priests, or those who are gay.
The scope of the Regulations
The Regulations as originally drafted have been amended by the Equality Act 2006. Some of the changes are of real significance. The most important is the change to the grounds of discrimination being rendered unlawful.
First, the original Regulations defined religion or belief as being: “any religion, religious belief or similar philosophical belief.”
That has now been amended to make it plain that the reference to religion or belief includes a reference to lack of religion or belief. So any uncertainties arising under the original formulation about the protection afforded to atheists, agnostics or members of secular or humanist societies are laid to rest. Moreover, it is not just the strong anti-religion lobby – the Richard Dawkins’ of this world – who are protected. The indifferent or uninterested equally cannot be the subject of discrimination.
The importance of this is that, save where a specific exception applies, an employer who seeks to recruit only those who hold a particular religious belief will thereby be discriminating against all who do not hold that belief, whether they are adherents to some other religion or to none. The law is therefore brought into line with art. 9 of the European Convention which, according to the European Court of Human Rights in Kokkinakis v Greece [1993] ECHR 20 is: “a precious asset for atheists, sceptics, agnostics and the unconcerned”.
A further change, which perhaps was intended to reinforce this point, is that instead of discrimination applying simply “on grounds of religion or belief”, the amended section 3(1)(a) provides that: “A person ‘A’ discriminates against another person ‘B’ if on the grounds of the religion or belief of ‘B’ or of any other person except ‘A’ (whether or not it is also ‘A’’s religion or belief) ‘A’ treats ‘B’ less favourably than he treats or would treat other persons.”
On the one hand this makes it plain that the discrimination against “B” can be because of the belief of some third party – for example, a spouse or partner – or because they refuse to obey an instruction not to serve, say, Muslims or Jews.
Arguably, however, it has narrowed the protection in this way. The original definition, by concentrating on “grounds of religion or belief” was clearly wide enough to include perceived, even if false, belief. However, by focusing on the religion or belief of “B”, it is at least arguable that this means the actual religion or belief. If that is right, then the employer who discriminates against someone he thinks is a Muslim or Christian is not acting unlawfully if in fact the employee is not.
The second and perhaps more important change is that the protection formerly afforded only to those with a religious or similar philosophical belief has now been extended to include a religious belief or any philosophical belief. This amendment is designed to reflect the language of the Directive which covers “religion or belief” without any requirement that the belief be religious or similar to a religious belief. So there is a free-standing protection for any philosophical belief.
Meaning of religion or belief
None of these concepts – religion, or religious or philosophical beliefs – are further defined. It is left to the courts – and that ultimately means the ECJ – to determine how these concepts should be construed. There are not even guidelines, nor a list of criteria that ought to be taken into consideration. It is assumed that, like the elephant, the court will know it when it sees it.
The courts have had to give meaning to the concept of a religion in the context of charities’ law. They have taken the view that it at least requires a belief in a supreme being, although the nature of that being may vary from religion to religion.
Presumably a religious belief is something that in some way arises out of the religion. It may flow from a particular interpretation of scripture or from particular religious customs or practices. It is, however, a purely subjective matter. It is not for the courts to test the belief against some objective standard. Is it enough that the individual genuinely claims that the belief is rooted in his or her religion?
Again some observations of Lord Nicholls in the Williamson case are to the point. That was a case where some Christian parents contended that the law forbidding smacking of children in schools infringed their art. 9 rights. They relied upon such scriptural authority as “spare the rod and spoil the child.” They perceived it as “loving corporal correction”. Their Lordships thought that the art. 9 rights had been engaged but that the law was justified. An argument that the belief could be tested objectively was roundly rejected in the following terms: “When the genuineness of a claimant’s professed belief is an issue in the proceedings, the court will inquire into and decide this issue as a question of fact. This is a limited inquiry. The court is concerned to ensure an assertion of religious belief is made in good faith: ‘neither fictitious, nor capricious, and that it is not an artifice’, to adopt the felicitous phrase of Iacobucci J in the decision of the Supreme Court of Canada in Syndicat Northcrest v Amselem (2004) 241 DLR (4th) 1, 27, para 52. But, emphatically, it is not for the court to embark on an inquiry into the asserted belief and judge its ‘validity’ by some objective standard, such as the source material upon which the claimant founds his belief or the orthodox teaching of the religion in question or the extent to which the claimant’s belief conforms to or differs from the views of others professing the same religion. Freedom of religion protects the subjective belief of an individual. As Iacobucci J also noted, at page 28, para 54, religious belief is intensely personal and can easily vary from one individual to another. Each individual is at liberty to hold his own religious beliefs, however irrational or inconsistent they may seem to some, however surprising ... The relevance of objective factors such as source material is, at most, that they may throw light on whether the professed belief is genuinely held.”
Exceptionally, the courts will be entitled to conclude that the belief relied on is not genuinely held. In Kosteski v The Former Yugoslav Republic of Macedonia [2006] ECHR 403 the law gave certain rights to Muslims to take time off for religious purposes. The claimant alleged that his rights had been infringed because he was disciplined for not going to work. The Court disbelieved him, not least because before then he had taken all the Christian holidays, did not follow the Muslim diet and appeared to know nothing about the religion! But such cases will be exceptional and presumably there will need to be some basis for doubting the claimant’s assertion.
Will the scrutiny of the courts be different where the alleged discrimination covers a philosophical belief? I think it might, at least to the extent that there may be a fuller review involved in the determination whether a belief can properly be labelled “philosophical” or not. This is no easy question.
The European Convention jurisprudence, in connection with the freedom to manifest beliefs, includes veganism, pacifism and Druidism among the protected beliefs. As Lord Walker pointed out in the Williamson case, the tendency of the courts has been to adopt a very expansive view of these issues. But there is a limit. In Campbell and Cosans v UK (1982) 4 EHRR 293, the ECHR observed that mere opinions or deeply held feelings were not enough. The views must attain a certain level of “cogency, seriousness and importance.”
This issue was also raised by Lord Nicholls in Williamson . He made the important point, which potentially sets an important limitation on the range of protected beliefs, that they: “must be consistent with basic standards of human dignity or integrity. Manifestation of a religious belief, for instance, which involved subjecting others to torture or inhuman punishment would not qualify for protection. The belief must relate to matters more than merely trivial. It must possess an adequate degree of seriousness and importance. As has been said, it must be a belief on a fundamental problem.”
This classification of beliefs is likely to raise highly controversial issues. However, it is perhaps worth emphasising the obvious point that if the real nature of someone’s belief is religious or philosophical, then that should be made plain to the employer or other party. They can hardly be discriminating on grounds of philosophical or religious belief if they know nothing about it.
The failure to clarify precisely the nature of the belief caused difficulties for the claimant in McClintock v Department for Constitutional Affairs [2008] IRLR 29. In that case, a Justice of the Peace objected to placing foster or adopted children with gay parents. He was a practising Christian with strong principles. But the employment tribunal found that he had expressed his objections not in terms of any fundamental belief but rather on the basis that he felt that there was a real risk that such placings would not be in the interests of the children and that further research was necessary before it was right to take risks with their welfare. The tribunal found that this objection did not engage the Regulations and in the EAT we agreed. In any event, any discrimination was indirect – because the objection was directed to anyone who sought to qualify adherence to the judicial oath irrespective of religion – and in the circumstances it was justified.
Let me give two examples of cases where difficulties might arise in classifying beliefs. First, take the case of doctors who refuse to carry out abortions and are disciplined or dismissed for adopting that stance. The first holds a strong Roman Catholic faith and believes that the sanctity of life is absolutely sacrosanct. Plainly that is a religious belief.
The second doctor has the same belief in the importance of life but does not have any religious beliefs. If he or she is to be protected, his or her belief must amount to a “philosophical belief”. I would have thought that it does, but the position is not perhaps quite so clear.
A third doctor will not carry out abortions as a protest because the other two have been disciplined and he thinks they have been treated unfairly. That is plainly not a philosophical belief; it is simply a strong feeling. Would it be different if his protest was because he thought that conscience should always be respected? Is that a philosophical belief? I doubt it, but I am not at all sure. The potential for sophisticated argument and fine distinctions is obvious. (The example is not a good one because of section 4 of the Abortion Act, which provides a conscience clause in any event.)
A second example concerns the potential discrimination against persons because of their political beliefs. They were specifically intended not to be covered by the original Regulations, and were excluded because the philosophical belief had to be akin to a religious belief. But with that limitation removed, why should some political beliefs not also constitute philosophical beliefs?
Democrats have what I suspect all would agree is a philosophical belief in the inherent equality of man. Certain racist or fascist organisations believe that man was not born equal. That is surely equally a philosophical view, however repellent to many people it might be. It may be that they would fail the test of being consistent with basic standards of human dignity, but I doubt it. There is a substantial difference in believing that the races should live separately and that persons from other races should be subject to torture. The protection is not just for those who hold beliefs of which we approve.
I would have thought that there is a very good case for saying that if I am dismissed for being a fascist or a communist that would be because of my philosophical beliefs. An ideology reflects an underlying philosophy. Such persons hold views about how society should be organised and which values in society should be given priority.
At the other end of the spectrum, beliefs about political issues would not engage the protection. For example, a belief that hunting is or is not a good thing would surely not of itself be a philosophical belief, no matter how strongly the issue is felt. If, however, it is part of a wider belief in animal rights, that would be different. That is surely a philosophical belief. So a huntsman who refuses to employ someone who is opposed to hunting might be within his rights – or, more accurately, would not be infringing any rights of the objector. There is still room for prejudice in the field of recruitment!
The protection afforded to political beliefs could raise some difficult issues. Would this mean that discrimination against someone in the British National Party or the Socialist Workers Party could be unlawful under these Regulations? As I shall shortly indicate, I think that there are circumstances when it could be, although such discrimination may be indirect rather than direct.
Paradoxically, it is more difficult to see how discrimination against someone because he or she is a member of the Labour or Conservative Party would fall within the scope of the Regulations. I doubt whether adherence to those parties could be said to reflect any strong philosophical belief at all. No doubt most members of both parties believe in the virtues of democratic capitalism, and that may be said to be a philosophical belief. But they are not being discriminated against for that reason. It is because they display the wrong tribal loyalty.
When is discrimination direct and when indirect?
Again, this raises a problem of some difficulty. Are there circumstances where the manifestation of the belief must be protected as part of the belief itself? In R (MSF) v Secretary of State for Trade and Industry and others [2004] IRLR 430 Mr Justice Richards had to consider whether the Sexual Orientation Regulations were compatible with the Framework Directive. The argument was that some of the exceptions which permitted religious groups to discriminate against homosexuals were too widely drafted. The claims were rejected but the judge made a number of important observations on the proper construction of the exceptions. In particular, he said – contrary to the guidance that had been issued by the DTI – that: “the protection against discrimination on the grounds of sexual orientation relates as much to the manifestation of that orientation in the form of sexual behaviour as it does to sexuality as such.”
It follows that in his view discriminating on grounds of sexual behaviour with the same sex would be direct discrimination on grounds of sexual orientation.
If one translates this into belief, does this mean that any manifestation of the belief must be treated as commensurate with the belief itself? I do not think that it does; certainly under art. 9 there is a clear distinction between the belief itself and its manifestation, and I think that runs through the Regulations also. How then does one explain the dictum of Richards J?
Plainly anyone who is homosexual will, unless remaining celibate, engage in homosexual relations. Of course, it is possible to discriminate against practising homosexuals but not non-practising; that is the stance of the Anglican Church – or at least part of it – towards gay priests. Nonetheless, discrimination against practising homosexuals inevitably involves discrimination against homosexuals, even though not all homosexuals are targeted. The fact that it is only a subset of homosexuals is irrelevant. I might only discriminate against blacks from Africa and not the West Indies but that does not prevent my conduct being direct race discrimination.
The issue was debated in Azmi v Kirklees Metropolitan Borough Council [2007] IRLR 484. This was the case concerning the Muslim woman who insisted on wearing the veil when teaching children in the presence of male colleagues. She was suspended for refusing to remove it. The college thought that it interfered with effective communication with the children, to conceal the face and mouth, and that this had to override her religious beliefs. The employment tribunal found that there was no direct discrimination; the objection was not to the veil in particular but rather to the wearing of any garment that covered the face. Plainly there was indirect discrimination; the rule had a disparate impact on female Muslims since many of them habitually wear the veil. However, it was in the circumstances considered to be a justified and proportionate response. The EAT held that there was no error of law.
On appeal, Wilkie J observed that there had been much debate about whether the belief of the claimant that she should wear the veil was itself a religious belief. It was apparently perceived that if the answer was that it was a religious belief in its own right that would establish direct discrimination. Wilkie J pointed out that this was an error. That issue was whether there was less favourable treatment on forbidden grounds. Here the reason for the less favourable treatment was wearing something that covered the face.
The tribunal had found that anyone doing that would have been treated the same way. Whether that was properly analysed as a belief or the manifestation of the belief was immaterial. Of course, it may be far more likely that Muslims rather than members of other religions will be affected by this rule, and to that extent it is indirect discrimination. However, that allows the defence of justification to be run.
The difference between this and the example of practising homosexuals is that the insistence on wearing the veil is not always the consequence of religious beliefs; it is not, as it were, a badge exclusive to the religion. So directing sanctions to all in that category is not inevitably discriminating on grounds of religious belief. It will no doubt be an interference with religious belief and would, for example, engage art. 9. But not all forms of interference will constitute discrimination.
I would suggest that in considering whether a manifestation of a religion or belief constitutes direct discrimination, the issue is not whether it can properly be treated as a belief in its own right. Rather the key issue is whether the reason for the treatment is a reason which necessarily and exclusively causes discrimination on a forbidden ground. You necessarily and exclusively discriminate against some gays if you discriminate against those who are practising homosexuals.
This could give rise to difficulties with the British National Party. Does discrimination against members of that party inevitably involve discriminating against the philosophical views which the party espouses? Does such discrimination necessarily involve discrimination against a subset of those holding racist views in the same way that discriminating against practising homosexuals necessarily involves discriminating against a subset of homosexuals? Logic would tend to suggest that there is a case to say that it does.
No doubt most of the members of the BNP could not tell the difference between a philosophical view and a wet fish, and they may be surprised to be told that their racist attitudes reveal anything philosophical at all. Alternatively, if they are required to sign up to the party’s beliefs and do so, is it permissible for the court to question whether they actually hold or understand the philosophical ideology that the party represents? No doubt the employer can genuinely say that he is wholly indifferent to the beliefs, philosophical or otherwise, of these members and simply wants to avoid the problems in the workplace that such membership brings. But it is trite law that the employer’s intention or purpose is irrelevant. If this is direct discrimination then it cannot be justified.
If it is not direct discrimination, may it be indirect discrimination on the grounds that any criterion that bars a member of the National Front from holding a post or office will have a disparate impact on persons who share their fascist views? I would have thought that if they can establish that the beliefs fall within the scope of a philosophical belief, there is a cogent case for saying that they do. The only issue then will be whether it can be justified.
The potential for political organisations of this nature to gain protection under the Religion or Belief Regulations could have a significant impact. In Redfearn v Serco [2006] IRLR 263, the Court of Appeal had to deal with a case where a member of the BNP, a candidate in the election for a local councillor, was dismissed. He did not have a claim for unfair dismissal and sought to bring his case under the Race Relations Act but was unsuccessful. The Court further held that even if there were indirect discrimination, it was justified on health and safety grounds. The employers were under pressure from third parties and feared unlawful violence from members of the public. He might have better luck now, under the Religion or Belief Regulations. However, if it is only indirect belief discrimination, then the justification defence that was found to be a defence in the race case would presumably succeed here also.
Am I the only person who has reservations about justification succeeding in that case? It was because of pressure from third parties and the fear of unlawful violence from members of the public. I suspect that there would be other situations where the courts would be most unhappy to think that discrimination laws could be effectively side-stepped because of the response of third parties. Would it be acceptable to say that I don’t serve blacks in my pub because the white racist clientele will riot? But that is by the way.
The wearing of jewellery provides some difficult issues. What of those who wear a cross? This would not be direct discrimination, since discriminating against those wearing the cross is not inevitably discriminating against Christians, as is the case with discriminating against practising homosexuals. Not all Christians inevitably and exclusively wear a cross. (The position would, in my view, be different if it could be shown that the intention was to victimise Christians. In those circumstances, even if some non-Christians were also affected, that would be unlawful direct discrimination. It would in form be indirect discrimination, but the discriminatory intention would render it direct.)
There are some obiter observations of Mummery LJ to the contrary in Secretary of State for Defence v Elias [2006] IRLR 934, but the issue was not argued there, and I think with respect that any intention to discriminate renders that discrimination direct whatever form it takes. Moreover, if the ban is only on crosses and not on jewellery of other religions, that might establish direct discrimination, at least if the intention is to treat Christians differently. Where, however, the ban is on all jewellery of a religious nature, that will be indirect rather than direct discrimination and the question of justification will arise. But if there were simply a general prohibition on jewellery applying to all persons, it is difficult to see how any discrimination arises. It might well be an interference with the right to freedom of religion conferred by art. 9 of all those who wished to wear a piece of religious jewellery as a manifestation of their religious beliefs. That would depend on the reason for the ban. But it would obviously not be direct discrimination, and it is hard to see how it would be indirect.
To be indirect, it would be necessary to show that the proportion of religious persons, or persons of a particular religion, who wore jewellery was greater than the proportion of non-religious people. I doubt whether any such adverse impact could be established. There is no a priori reason to assume that such a rule would have any such impact and, furthermore, I very much doubt whether it would be possible even to find any relevant statistics. That is a particular problem with indirect discrimination in this field, as I will indicate later.
I have drawn a distinction between cases where the form of discrimination is on grounds which necessarily and inevitably will lead to discrimination on forbidden grounds – as with practising homosexuals – and where it does not, which is in almost all cases of manifestation of belief. The former is direct discrimination and the latter is not.
But I can see an argument to the contrary along the following lines. Take the case of the woman wearing the veil over her face. The issue is whether that might constitute direct discrimination. The answer given by Mr Justice Wilkie, with which I agree, is essentially that if the employer is indifferent to the fact that the conduct or behaviour is a manifestation of the belief, and would act in exactly the same way whether it was or not, then there is no direct discrimination.
A converse argument may be put along the following lines. The reality is that only Muslim women will in practice wear the niqab. If one adopts the “but for” test, which was adumbrated in James v Eastleigh Borough Council [1990] IRLR 288 (the swimming pool case), and link it to what happens in practice rather than what could happen in theory, it is surely the case that, but for her religion, she would have been treated differently. Accordingly, it might be said that whatever the employer’s intention, in reality he has discriminated against a subset of Muslim women who wear the veil; the fact that he would have treated them the same way even if they were not exclusively Muslim women is irrelevant, just as in James v Eastleigh it was irrelevant that men would have been treated the same way had they had the same retiring age.
There is some merit in the argument and perhaps three potential answers to this. The first is to apply the “but for” test only where the distinction on the forbidden grounds necessarily and in all cases identifies those who receive more and less favourable treatment respectively. That was the position in James because the discriminatory retirement age was fixed by statute. On this basis, others apart from Muslims could theoretically choose to wear the veil.
The second answer is to adopt the “reason why” approach to direct discrimination rather than the “but for” test laid down in the James case. Adopting that analysis, the reason why was the wearing of the veil over the face, not the fact of her religion, and whilst this of course raised an issue of indirect discrimination, it was then capable of justification.
The third answer, which is not, however, permitted under the current law, is to allow justification in an appropriate case for direct discrimination. It does not then matter whether a “but for” or “reason why” test is adopted. I have always felt that there is a powerful argument for this. Certainly the failure to allow it means that undue emphasis is placed on how precisely one analyses the nature of direct discrimination rather than looking at the substantive issue.
I have no doubt that in principle the question of manifestation of religion should be tested by a justification test, whether it is exclusive to the religion or not. Take Muslim staff who go on a haj. Only Muslims do that. But it would surely be impossible to contend that it was direct discrimination and therefore unlawful to refuse this whatever the circumstances of the employer. Either the concept of direct discrimination has to be limited, or justification has to apply to direct discrimination.
Accommodating differences
One of the most important areas where these Regulations have the potential to impact on working practices is in a case such as Copsey v WWB Devon Clays Ltd [2005] IRLR 811, where the employee wishes to be allowed to observe religious practices in ways that require a modification of the usual rules adopted by the employer.
The Acas guidance on the Regulations is a mine of information on different religions and the kinds of steps that might be taken to accommodate religious practices, including providing flexible hours, different days off depending on the religious festival in issue, allowing a departure from certain clothing rules and so forth. Indeed, the requirement can go further and involve a claim that the employee should not be required to carry out certain jobs at all.
The example I have given of the doctors refusing to carry out abortions is relevant here. Must an employer always be required to weigh up the potentially conflicting arguments and to justify any refusal to accommodate?
The real analogy here is surely with disability discrimination. The employee has special needs – in this case arising from his beliefs rather than from his disability – which he wishes the employer to accommodate. The duty of reasonable adjustment, an important part of disability discrimination law, is designed to achieve just that.
Unfortunately, the Framework Directive has not been drafted in that way and the Regulations merely follow the Directive. So any duty of accommodation, whether to make it possible for the employee to do the job in the first place or to modify the contract once he or she is employed, will only arise if the rules operate in an indirectly discriminatory way. That will often be difficult to establish, particularly if the employee is in a minority religion.
It is true that because reasonable adjustment operates on an individual basis it imposes less of a burden on the employer to identify and remove discrimination in advance. The employer is responding to the particular needs of the individual, and these must be identified.
By contrast, the principles of indirect discrimination require the employer to consider whether in principle a rule operates in a potentially discriminatory way and, if it does, to consider the issue of justification. But given the variety of religions and beliefs this can be a complex exercise, and the problems may never arise. It would no doubt be possible for an employer to anticipate the fact that Muslims need time to pray, for example, or that there are a number of religions in which the members will need to be home before sunset on a Friday. But other practices will be far less well known.
It seems to me that there is a real risk that by linking the need to adjust to findings of indirect discrimination, the law will simply fail properly to protect some religious beliefs at all.
Moreover, even where indirect discrimination can be established, it will frequently provide protection in relatively arbitrary ways. I return to the example of the doctors. It will no doubt be relatively easy for the Catholic doctor to show that imposing a requirement to carry out abortions will impinge adversely on those of the Catholic faith. A considerably smaller proportion of those doctors who are Catholic as opposed to those who are not can meet the requirement consistent with their religious beliefs. Since the requirement operates to his detriment, he can establish prima facie indirect discrimination – i.e., discrimination subject to the need objectively to justify. (Although even that may be difficult if the pool adopted is the particular workplace.)
But what of the doctor who has a philosophical objection – if that is what it is – based on his belief that it undermines the sanctity of life to carry out abortions. How does he or she establish indirect discrimination in that situation? There would have to be statistics about how many doctors hold such a belief and what proportion they constitute of doctors who could carry out operations of that nature.
It may be that this information is available, but in principle I would have thought that there must be many situations where the statistics required to establish the requisite adverse impact are simply not available. It is of course true that statistics are not always necessary to establish the necessary disparate impact, but often they will be the only effective way of doing so. A duty of reasonable adjustment would eliminate the difficulties which bedevil the field of indirect discrimination.
I suspect that the proportionality test as applied to indirect discrimination will be identical to that where the issue of reasonable adjustment applies. In that context, it may well be that the issue raised by Rix LJ in the Copsey case, namely whether it is the employer or the employee who is seeking to vary the terms, could be highly relevant to the application of the proportionality test. (Although perhaps even that analysis will have to be nuanced where there is a genuine change of religion.)
Conclusions
Discrimination law is an extremely complex field, and these Regulations pose greater problems than most. This is in part because certain key concepts are left undefined. The difficulties of defining them are indeed formidable, but when cases are determined on a case-by-case basis this inevitably encourages litigation and creates uncertainty.
I would make three final observations arising out of this analysis:
First, I think that European law has failed to provide adequate protection for religious beliefs by making the assumption that issues of religion and belief are like other forms of discrimination. Where the discrimination is race or colour, the key point is to ensure that persons are not treated less favourably because of that wholly irrelevant feature. (I am not here concerned with the possibility of positive discrimination to remedy social or economic disadvantage, which raises quite different issues.) The problem is not that black or white have different needs or requirements from each other that need to be recognised and respected.
However, with disability and religious discrimination, that is the principal problem. Of course, there will be some direct discrimination on these grounds. But that is not widespread, at least not in Great Britain. The greater problem arises because of a failure to accommodate the needs or requirements that are an aspect of the religion or disability, as the case may be. That is recognised in the disability context by the duty of reasonable adjustment. But it is not recognised in the area of religious discrimination, and in my view it ought to be.
That raises the second and related observation. While it will often be possible to get to the same position by identifying indirect discrimination and then focusing on justification, that is not in my view a satisfactory alternative. It may operate in arbitrary ways, and establishing the relevant disparate impact – which I intend to be a shorthand for putting people of that religion at a particular disadvantage – can be extremely difficult. Potentially, two employees treated in exactly the same way and feeling equally aggrieved will end up with one having a remedy and the other not. That is always a risk with indirect discrimination, but it is, I suspect, likely to be particularly striking here.
Equally importantly, the law does not engage directly with the real complaint that the employee has, namely that the employer should take steps to accommodate the particular religious requirements.
The final point is that an analysis of these provisions shows yet again that even the concept of direct discrimination is not as clear as it might be. The distinction between the “but for” test and the “reason why” test may continue to cause real difficulties in this field.
So far, there have been relatively few cases in this area. But I have no doubt they will increase. Many of them will raise difficult issues of justification. It is not an easy job for judges – and particularly the employment judges who have to make the initial decisions in the employment tribunals – to strike the balance fairly in these highly charged cases. Genuine respect for philosophical views very different from our own will surely be the starting point for careful judicial analysis.
