Elan-Cane, R (on the application of) v Secretary of State for the Home Department [2021] UKSC 56 (15 December 2021)

LORD REED: (with whom Lord Lloyd-Jones, Lady Arden, Lord Sales and Lady Rose agree)

This appeal raises two questions:

(i) Does article 8 of the European Convention on Human Rights (“the Convention”), either taken in isolation or read together with article 14, impose an obligation on a contracting state, when it issues passports, to respect the private lives of individuals who identify as non-gendered, by including a non-gendered (“X”) marker for the passport-holder’s gender, as an alternative to the markers for male and female?

(ii) If not, is such an obligation nevertheless imposed on the Home Secretary by the Human Rights Act 1998?

For the reasons explained below, I answer both questions in the negative.

1. A preliminary note on language

The choice of language in the discussion of gender identification often signifies allegiance to a particular point of view. I should therefore explain that the language used in this judgment is intended to reflect the impartiality which is incumbent upon the court. Although the appellant is categorised for legal and administrative purposes as female, the present appeal involves a challenge to that categorisation, so far as it affects the application of a particular administrative policy. I shall not, therefore, use words which would normally follow from that categorisation. I will refer to the appellant as identifying as non-gendered, in the sense that that is how the appellant’s gender is self-described. It will become apparent that the word “gender” is used in that context in a different sense from that in which it has been used by the Government and other public authorities, including Parliament. I will give a brief explanation of both usages which should suffice for present purposes.

2. The factual background

(1) The parties

The appellant was born female. However, growing up, and particularly during and after puberty, the appellant felt revulsion at having a female body. In 1989 the appellant underwent a bilateral mastectomy, and in 1990 a hysterectomy, in order to alleviate those feelings. The first of those procedures was carried out privately; the second, as publicly-funded treatment under the National Health Service. The appellant now identifies as non-gendered, and has become a campaigner for the legal and social recognition of a non-gendered category of individuals. The term “gender” is used in this context to describe an individual’s feelings or choice of sexual identity, in distinction to the concept of “sex”, associated with the idea of biological differences which are generally binary and immutable. According to a witness statement provided by the appellant, “X” passports (that is to say, passports in which an individual’s gender may be described not only as “M” or “F” but also as “X”) are a focal point of this campaign.

Her Majesty’s Passport Office (“HMPO”) is an agency of the Home Office which deals with the issuing of passports and related matters. It is the sole issuer of United Kingdom passports to citizens of the United Kingdom. Passports are issued by HMPO at the discretion of the respondent, the Home Secretary, in the exercise of the Royal Prerogative. Under the law of the United Kingdom, the Royal Prerogative is a lawful basis for decision-making of this nature.

In 1995 the appellant contacted the United Kingdom Passport Authority, a predecessor of HMPO, to inquire whether it was possible for a passport to be issued without making a declaration of being male or female. The appellant was informed that this was not possible because a declaration of gender as either male or female was a mandatory requirement. In this context, as in others, public agencies generally use the terms “gender” and “sex” interchangeably, to refer to the biological categories of male and female, subject to the inclusion of transgender persons within the category of their acquired gender. The appellant accordingly applied for, and was issued with, a passport in which the relevant gender was recorded as female. The appellant made similar inquiries in 2005 and in correspondence between 2010 and 2016, with similar results.

(2) The policy

HMPO continues to operate a policy that an applicant for a passport must state on the application form whether their gender is male or female. If no gender is stated, the gender shown on the applicant’s supporting documents is selected. The passport is issued recording the passport-holder’s gender as male (“M”) or female (“F”). Transgender people, in the sense of people who have acquired a different gender from the one recorded at birth, can obtain passports showing their acquired gender.

It is important to make the latter point clear in order to avoid confusion. The term “transgender” can be used in a wider sense, as it is by the intervener, Human Rights Watch, so as to include persons in the position of the appellant. The term is used in this judgment in the narrower sense in which it has been used in the European case law and in most of the documentation to which I shall refer. So used, it describes those individuals who have acquired a gender, either male or female, which is different from the one recorded at birth. Such persons are not non-gendered. In the United Kingdom, they can obtain a passport which conforms to their acquired gender on the production of a gender recognition certificate, a re-registered birth certificate showing their acquired gender, or a doctor’s letter confirming that their orientation to their acquired gender is likely to be permanent.

The appellant contends that the policy operated by HMPO contravenes the Convention rights of individuals who do not identify as either male or female.

(3) Reviews of the policy

In 2014 HMPO completed an internal review of gender marking in passports. The findings are set out in a document entitled “Gender Marking in Passports - Internal Review of Existing Arrangements and Possible Future Options”. It stated at para 1.4:

“There is no provision in the passport or on the passport application form for a person to transition from one gender to no gender or to state that they do not identify in either gender. This is in line with UK legislation that recognises only the genders Male and Female.”

It noted at para 1.6 that the record of a person’s gender in their passport is used for a variety of purposes. One is to assist in verifying the identity of applicants for passports. The review explained that applicants for a passport must also provide documents which record their gender, such as a birth or adoption certificate or a gender recognition certificate. When the passport application is considered by HMPO, a check is made that the gender shown on the application form matches the gender shown on the documents supplied, in order to prevent fraud. Interviews are carried out in cases of doubt, with gender being one of the details investigated. Checks, including checks relating to gender, can also be carried out with other Government departments.

A second purpose is to assist in verifying the identity of passport users. Border staff check that the gender recorded on the passport appears to match the gender of the person using the passport. This is particularly valuable in the case of persons with names which may not be indicative of gender to a border officer who is unfamiliar with the traveller’s language or culture. Other bodies which are required to check identity, such as banks carrying out their obligations under money laundering legislation, and employers carrying out their duties under immigration legislation, can also use passports for that purpose (passports, like driving licences and birth certificates, being one of the types of document used as evidence of identity in the United Kingdom when such evidence is required).

A third purpose is to enable officials to deal appropriately with members of the public in passport-related matters, for example by addressing them in appropriate terms, and by arranging for physical checks at borders to be carried out by officers of the appropriate gender, without their having to ask embarrassing questions about the passport-holder’s gender.

The review acknowledged that two groups might be negatively affected by the current policy. One comprised individuals in the process of transitioning from one gender to the other (an issue which was subsequently resolved, as explained in paras 6 and 7 above). The other comprised individuals who did not consider themselves to be either male or female. There had however been very few requests for an “X” provision, other than from the appellant. There were no calls for change from gender representative groups or civil liberties groups (para 2.6).

The review noted that United Kingdom legislation, including discrimination and equality legislation, is based on the categorisation of all individuals as either male or female (and, if they are parents, as either mothers or fathers). There is no legislative provision for the recognition of individuals as non-gendered. There were no plans across government to introduce a third gender category (para 4.5). The norm across government was for gender to form a key part of the personal information gathered in respect of individuals. The review stated that “HMPO could introduce recognition of a third gender but it would be in isolation from the rest of government and society. There are likely to be so few applications for such a passport, but we would need to avoid issuing a document that was not recognised by other parts of government or wider UK society” (para 4.7). It was noted that introducing an “X” gender marker in passports would also result in administrative costs of about £2m being incurred.

The Government considered the issue again in July 2016, when responding to a recommendation by a Parliamentary committee that the United Kingdom should follow Australia’s example. In its response, “Government Response to the Women and Equalities Committee Report on Transgender Equality”, the Government repeated the points which had been made in the 2014 review, and stated: “We would not see the passport as being used in the UK to recognise a third gender marking in isolation from other areas of government.”

During the same year, the Government began an internal review of gender markers in official documents generally. The Government Equalities Office (“GEO”) also sought information during 2017 from other countries in relation to the legal recognition of a third gender and its inclusion on identity documents. In 2018 the Government consulted on amendments to the Gender Recognition Act 2004, and included a question relating to individuals who identify as non-binary (ie who identify their gender outside the categories of male and female). In its response to the consultation, the Government did not propose any changes in that regard. In 2019 the Government appointed the National Institute of Economic and Social Research (“NIESR”) to undertake a review with the aim of improving understanding of individuals who identify as non-binary, and also began a review of gender markers by the GEO. The NIESR project was paused in December 2019 following the formation of a new Government, and subsequently cancelled when the Government decided to focus specifically on gender markers. The GEO review was suspended in 2020 as the officials engaged on it were re-deployed to work on matters arising from the Covid pandemic. In March 2021 the Government decided that work should resume, focusing on the issue of introducing gender-neutral markers in identity documents (including passports), and the potential impact of such a change on the provision of public services and on systems across government.

(4) International practice

At the time of the hearing of this appeal, there were agreed to be six contracting states of the Council of Europe which, in some circumstances, allow passports to include markers other than male and female. It appears that Denmark (since 2014), Malta (since 2015), and Iceland (since 2021) permit “X” markers in passports on application. It appears that the Netherlands (since 2018), Austria (since 2018) and Germany (since 2019) permit “X” to be entered on the passports of persons born with ambiguous sexual characteristics (“intersex”), subject in some cases to a court order or in others to the production of satisfactory evidence, such as an “X” birth certificate. The other 41 contracting states issue passports only with male or female markers.

Other countries which permit passports to bear an indicator other than male or female are New Zealand (since 2005), Australia (since 2011), India (since 2014), Nepal (since 2015), Pakistan (since 2017, in the case of transgender people) and Canada (since 2017). India, Nepal and Pakistan are states with a cultural history of treating hijras or khawaja sara (traditionally, but somewhat misleadingly, translated into English as “eunuchs”) as a distinct gender category.

On 22 April 2015 the Parliamentary Assembly of the Council of Europe adopted Resolution 2048 (2015), “Discrimination against transgender people in Europe”, which, among other matters, called on member states of the Council of Europe to “consider including a third gender option in identity documents for those who seek it” (para 6.2.4). On 12 October 2017 the Assembly adopted Resolution 2191 (2017), “Promoting the human rights of and eliminating discrimination against intersex people”, which, among other matters, called on member states to “ensure, wherever gender classifications are in use by public authorities, that a range of options are available for all people, including those intersex people who do not identify as either male or female” (para 7.3.3).

The International Civil Aviation Organisation (“the ICAO”), a United Nations agency established under the Convention on International Civil Aviation of 1 March 1947, has issued international standards for machine-readable passports which require information as to the sex of the passport-holder to be included. Since at least 2006, the ICAO standard (ICAO 9303) has permitted countries to issue passports with “M”, “F” or “X” (denoting “unspecified”) in the section dealing with sex or gender.

Both New Zealand and the United Kingdom have raised within the ICAO the need for gender to be a required field on machine-readable travel documents. The ICAO concluded (at its 21st Meeting of the Technical Advisory Group on Machine Readable Travel Documents, held between 10 and 12 December 2012), in agreement with New Zealand’s recommendation, that “[t]he costs of removing the requirement to display the holder’s gender on travel documents outweigh the benefits at this stage”. It also accepted that there would be “adverse effects on the operations of border authorities and … potential inconvenience for passengers”.

In 2016 the ICAO agreed to the United Kingdom’s leading a review of the necessity of sex or gender markers in passports. Working with five other countries, HMPO sent a questionnaire to ICAO member states in October 2016. 21 countries responded. All of them considered gender markers in passports to be important in order to confirm identity, ensure security and provide accurate statistics, for example in relation to immigration and settlement trends. In relation to security, it was said that when checking passport holders against security databases, such as terrorist watchlists, gender operates as a filter which enables a large proportion of records and individuals to be immediately eliminated. If that filter could not be operated, passport holders would take longer to process at frontiers. One respondent supported the removal of gender markers.

3. The proceedings below

(1) The Administrative Court

In 2017 the appellant issued an application for judicial review of HMPO’s policy on the grounds that it (1) breached the appellant’s right to respect for private life under article 8 of the Convention, (2) breached the appellant’s right not to be discriminated against on the basis of sex or gender under article 14 of the Convention taken together with article 8, (3) was irrational and (4) failed to take account of relevant considerations, while taking account of irrelevant considerations.

On 22 June 2018 Jeremy Baker J dismissed the claim: [2018] EWHC 1530 (Admin); [2018] 1 WLR 5119. He accepted that “private life” within the meaning of article 8 included an individual’s identification as being non-gendered, with the consequence that a person who identified as non-gendered had a right under article 8 to respect for that identification. However, he held that the pre-eminent consideration when determining whether article 8 imposed a positive obligation on the state to ensure that an individual’s non-gendered identification was respected and, if so, the scope of that obligation, was the striking of a fair balance between the competing interests of the individual and the community as a whole. In making that determination, the state’s margin of appreciation was a relevant consideration. On the facts of the present case, it was relevant to the width of the margin of appreciation that (1) although the appellant had a strong personal interest in gaining legal recognition as being non-gendered, the judge was “less convinced that such strong emotions are justified by the current HMPO policy of not permitting the [appellant] to enter ‘X’ in gender/sex field on the passport” (para 115), (2) there was no consensus among the contracting states of the Council of Europe to recognise the gender identity of those who did not identify as being either male or female, nor was there a trend of sufficient strength to affect the matter, (3) there were security-related justifications for refusing to issue non-gendered passports, and (4) the Government was entitled to pursue the legitimate aim of “maintaining an administratively coherent system of gender recognition across all government areas and legislation” (para 117) and should be able to consider a change to the policy as “part of a more fundamental review of policy in relation to these issues across government” (para 121). It followed that, in relation to the issue raised by the appellant, the margin of appreciation was relatively wide.

Accordingly, article 8 did not impose a positive obligation on the Government to permit the claimant to apply for and be issued with a passport with an indicator in the sex field signifying that the gender was unspecified. The answer would be the same if the issue were looked at in terms of a negative interference with the article 8 right.

In relation to the complaint under article 14 taken together with article 8, although the appellant’s non-gendered identification fell within the ambit of article 8, it was arguable that there was no difference in treatment in respect of that right between the appellant and others put forward for comparison (such as transsexuals who identify within the categories of male and female), since neither the appellant nor the comparators could obtain a passport which did not specify their gender. In any event, any difference in treatment would be objectively justifiable for the same reasons as those governing the existence and scope of the Government’s positive obligations towards the appellant under article 8.

(2) The Court of Appeal

That decision was upheld by the Court of Appeal (King, Irwin and Henderson LJJ): [2020] EWCA Civ 363; [2020] QB 929, although for different reasons so far as the complaint under article 14 was concerned. King LJ, in a judgment with which the other members of the court agreed, accepted that the appellant’s identification as non-gendered was an aspect of private life within the meaning of article 8, but agreed with the judge that there was no positive obligation on the part of the state to issue a passport with an “X” marker. Nor did HMPO’s policy amount to an unlawful interference with the appellant’s article 8 rights. The question whether any difference in treatment was objectively justified under article 14 would receive the same answer.

In reaching that conclusion, King LJ accepted that there was nothing approaching a European consensus in relation to either the broader issue of the recognition of non-gendered status, or the narrower issue of the use of “X” markers on passports. States therefore enjoyed a relatively wide margin of appreciation. However, King LJ rejected the Secretary of State’s submissions in relation to national security, stating at para 73:

“In my judgment, issues of security do not affect the fair balance in circumstances where the ICAO has, for many years, been content for passports to carry an ‘X’ marker. Additionally, people from countries that already have such provision have been entering the UK for many years without there being any security issues articulated before us.”

On the other hand, King LJ accepted that it was important for the Government to pursue a coherent approach to the issue of gender recognition across all the areas where the issue arose. The passport issue could not reasonably be considered in isolation (partly because, as the appellant’s counsel recognised, the introduction of “X” markers in passports would inevitably lead to legal challenges to binary gender classification in other contexts). The current policy fell within the margin of appreciation and did not amount to a breach of the appellant’s article 8 rights.

In relation to article 14, King LJ accepted that there was a difference in treatment between individuals who identify as non-gendered and those who identify as gendered. The latter group could obtain a passport which reflected their gender identification, whereas the former group could not. Nevertheless, the outcome under article 14 was the same as under article 8. The complaint under article 14 was essentially the same as that under article 8, albeit seen from a different angle.

4. The legal issues arising under the Convention

(1) The argument based on article 8

(i) General principles

Article 8 provides:

“1. Everyone has the right to respect for his private and family life, his home and his correspondence.

2. There shall be no interference by a public authority with the exercise of this right except such as is in accordance with the law and is necessary in a democratic society in the interests of national security, public safety or the economic well-being of the country, for the prevention of disorder or crime, for the protection of health or morals, or for the protection of the rights and freedoms of others.”

There is no judgment of the European Court of Human Rights which establishes a positive obligation to recognise a gender category other than male or female, and none which would require the Secretary of State to issue passports without any indication of gender. Indeed, there does not appear to have been any case before the European court concerned with the application of the Convention to individuals who identify as non-gendered. However, in the light of the court’s case law concerning transgender individuals, it is no longer in dispute between the parties that the appellant’s identification as non-gendered is an aspect of private life within the meaning of article 8.

The question then arises whether the issue arising under article 8 should be analysed by asking whether article 8 imposes a positive obligation upon the Secretary of State to provide the appellant with an “X” passport, in order to secure the appellant’s right to respect for private life, or whether the Secretary of State’s unwillingness to do so constitutes an unjustified interference with the appellant’s right to such respect. The courts below favoured the former analysis, and the authorities support that approach.

The European court has generally analysed cases concerned with gender identity from the perspective of whether the respondent state has failed to comply with its positive obligation to secure to the persons concerned their right to respect for private life under article 8: see, for recent examples, the Grand Chamber judgment in Hämäläinen v Finland [2014] 37 BHRC 55; Reports of Judgments and Decisions 2014-IV, p 369 (“Hämäläinen”), para 64, and the Chamber judgments in AP, Garçon and Nicot v France (Applications Nos 79885/12, 52471/13 and 52596/13, [2017] ECHR 338) (unreported) given 6 April 2017 (“AP”), paras 99-100, X v Former Yugoslav Republic of Macedonia (Application No 29683/16, [2019] ECHR 55) (unreported) given 17 January 2019, paras 63-65, and Fedotova v Russia (Application No 40792/10, [2021] ECHR 636) (unreported) given 13 July 2021, paras 44-47. In the light of this body of case law, both parties approached the present appeal as one concerning the existence of a positive obligation.

In Hämäläinen, the Grand Chamber gave guidance at paras 65-67, in similar terms to that given in several earlier judgments, as to the general principles applicable to assessing a state’s positive obligations in a context of this kind. First, it explained at para 65 that the principles applicable to assessing a state’s positive and negative obligations are similar. In both contexts, it said, “[r]egard must be had to the fair balance that has to be struck between the competing interests of the individual and of the community as a whole, the aims in the second paragraph of article 8 being of a certain relevance”.

Next, it considered at para 66 the concept of “respect”. In that regard, it observed that the notion of respect “is not clear cut especially as far as positive obligations are concerned”, and that “having regard to the diversity of the practices followed and the situations obtaining in the contracting states, the notion’s requirements will vary considerably from case to case.” The court noted that, nonetheless, certain factors had been considered relevant for the assessment of the content of those positive obligations. Some of them related to the applicant. They concerned “the importance of the interest at stake and whether ‘fundamental values’ or ‘essential aspects’ of private life are in issue or the impact on an applicant of a discordance between the social reality and the law, the coherence of the administrative and legal practices within the domestic system being regarded as an important factor in the assessment carried out under article 8.” Other factors related to the impact of the alleged positive obligation at stake on the state concerned. In that regard, the question was “whether the alleged obligation is narrow and precise or broad and indeterminate or about the extent of any burden the obligation would impose on the state”.

Finally, the court considered the state’s margin of appreciation at para 67. It accepted that in implementing their positive obligations under article 8 the states enjoy a certain margin of appreciation. A number of factors had to be taken into account when determining the breadth of that margin. Where a particularly important facet of an individual’s existence or identity was at stake, the margin allowed to the state would be restricted. Where, however, there was no consensus within the member states of the Council of Europe, either as to the relative importance of the interest at stake or as to the best means of protecting it, particularly where the case raised sensitive moral or ethical issues, the margin would be wider. There would also usually be a wide margin if the state was required to strike a balance between competing private and public interests or Convention rights.

(ii) The present case

Considering the present case in the light of that guidance, it is necessary to examine in the first place the factors which concern the appellant. As the courts below recognised, the appellant’s identification as non-gendered is an aspect of the appellant’s sense of self: that is why article 8 is considered to be engaged. It is therefore understandable that the appellant wishes to obtain legal recognition as a non-gendered individual. But that is not the issue in these proceedings, as the appellant’s counsel emphasised. Regardless of the outcome of these proceedings, the appellant will continue to be treated as female for legal purposes. These proceedings are concerned solely with HMPO’s current policy relating to the issuing of passports. The interest at stake in these proceedings, as far as the appellant is concerned, is therefore the appellant’s interest in obtaining an “X” passport.

As explained earlier, in the absence of identity cards in the United Kingdom, a passport is recognised as an identification document. In a witness statement, the appellant maintains that it is demeaning and distressing to use a passport as an identification document, for example when opening a bank account, when it does not reflect the appellant’s identification as a non-gendered person. It is also said that applicants for passports who identify as non-gendered are forced to make a false declaration of their identity. There is also said to be a risk of difficulties or harassment when a person who identifies as non-gendered uses a passport at borders, although the appellant has never personally been subject to harassment. In relation to these matters, counsel argued that the appellant’s situation was comparable to that of the applicant in B v France (1992) 16 EHRR 1.

So far as these points are concerned, it needs to be borne in mind, in the first place, that in the United Kingdom the need to establish one’s identity arises only occasionally: most commonly, when accessing certain financial services, such as opening a bank account. In that regard, the United Kingdom differs from many other contracting states. Secondly, when the need to establish one’s identity arises, there is no obligation to use a passport for that purpose. As the appellant recognises, other documents, such as a birth certificate or a driving licence, can be, and commonly are, used instead. It is not alleged in these proceedings that the use of those documents for that purpose involves any violation of article 8. It may be worth adding that a United Kingdom driving licence bears no obvious marker of gender. Counsel explained to the court that a multi-digit number which appears on the licence includes one digit which is an encoded reference to gender (which can be changed from one gender to the other on request), but the significance of the digit is not apparent on the face of the licence, and is not something of which most people are aware.

In relation to the process of applying for a passport, it is true that applicants, including those who identify as non-gendered, have to select either “male” or “female” on the application form, and make a declaration as to the accuracy of the information provided. But the purpose of providing that information is not to inform HMPO as to the applicants’ feelings about their sexual identity, and the applicants are not being forced to lie about those feelings. As was explained at para 9 above, the form is concerned with the applicants’ gender as a biographical detail which can be used to confirm their identity by checking it against the birth, adoption or gender recognition certificates provided and other official records. It is therefore the gender recognised for legal purposes and recorded in those documents which is relevant. The gender recorded on the passport can also be used for the other purposes mentioned in paras 10-11 above: purposes which are associated with the passport-holder’s appearance and physiology rather than their innermost thoughts. In relation to the use of the passport at borders, it is acknowledged that the appellant has no personal experience of harassment.

The other difficulty mentioned is the appellant’s feeling that the use of a passport bearing a female gender marker is demeaning and distressing. In relation to that matter, there is a significant difference, at least in degree, from the situation of the applicant in B v France. In that case, the applicant was a post-operative male to female transsexual who lived as a woman and had a female appearance. The European Commission on Human Rights found that “the applicant indisputably suffers particularly trying ordeals in her daily life because of the discrepancy between her appearance and the entries concerning both gender and forename on the documents relating to her, whether official or not, and more particularly on her identity papers. This discrepancy forces her to reveal to third persons details of her private life, whether in performing everyday actions, travelling, having even minor treatment, finding work, etc” (para 69 of its opinion, set out at para 30 of the judgment of the European court). Taking together the applicant’s inability to obtain a birth certificate and identity papers which reflected her acquired gender, and her inability to adopt a female forename, the Commission concluded that “[t]he combination of all those factors gives rise to disruption in the applicant’s daily life of … a serious nature” (para 71 of the opinion). The court agreed, and concluded that “the inconveniences complained of by the applicant in this field reach a sufficient degree of seriousness to be taken into account for the purposes of article 8” (para 62).

By comparison, the appellant is free to use any forename, and in fact uses a name which is in use for both men and women. The appellant makes no complaint about the law governing birth certificates. In relation to proof of identity, it has been explained that such evidence is not frequently required, and that the appellant is under no obligation to use a passport for that purpose. Perhaps most importantly, there is not the obvious discrepancy between the appellant’s physical appearance and the “F” marker in the appellant’s passport that there was between the feminine appearance of the applicant in B v France and her male identity papers.

Having regard to all these circumstances, the degree of prejudice to the appellant which is attributable to the unavailability of an “X” passport does not appear to be comparable to that suffered by the applicants in the cases before the European court on which counsel relied, such as B v France and Goodwin v United Kingdom (2002) 35 EHRR 18.

In relation to the coherence of the administrative and legal practices within the domestic system, which is regarded as an important factor in the assessment carried out under article 8 (para 34 above), counsel for the appellant emphasised that the NHS had treated the appellant’s gender dysphoria by providing the appellant with a hysterectomy. It was said to be incoherent for the Secretary of State then to decline to provide the appellant with an “X” passport. Reference was made to the European court’s case law concerning transsexuals, where it was regarded as illogical for the state to provide gender reassignment surgery, on the one hand, but to decline to give legal recognition to the acquired gender, on the other hand.

One needs to approach this argument with some care. In cases such as B v France, para 63, and Goodwin v United Kingdom, paras 77-78, the European court considered it illogical that the long and difficult process of transformation which was undergone by transsexuals as a result of gender reassignment surgery was not met with full recognition in the law. By contrast, the appellant has not undergone gender reassignment surgery, but a procedure which is undergone by many women, without any alteration to their gender.

Nevertheless, it can fairly be said that the appellant is in a different position from women who undergo hysterectomies for the usual medical reasons, since the appellant was provided with a hysterectomy, at public expense, in order to alleviate the psychological distress resulting from identifying as non-gendered while possessing a woman’s reproductive physiology. In those circumstances, it might be argued that it would be logical for the Government to accord recognition to the applicant’s non-gendered identity in the context of passports, as the NHS did in the context of medical treatment. There appear to me to be two flaws in the argument. First, the NHS did not recognise the applicant as being a non-gendered person: what it recognised was that the applicant was suffering from the medical condition of gender dysphoria, and it provided medical treatment to alleviate that condition. Secondly, the fact that the Government, through its funding of the NHS, bore the cost of the appellant’s medical treatment does not logically entail that it should in addition bear the far greater costs which would be involved in introducing “X” passports, or accept the other disadvantages described at paras 46-53 below.

So far as concerns the extent of the burden which an obligation to issue “X” passports would impose on the state, three principal points are made on behalf of the Secretary of State. The first relates to the public interest in security. In relation to this issue, the courts below were provided with evidence from senior officials in the Government unit with the relevant responsibilities. Their evidence was consistent with the position adopted in the 2014 review and the 2016 response to the Parliamentary committee, summarised at paras 9-11 and 14 above. One of the officials, Mr Woodhouse, addressed the contention, advanced on behalf of the appellant, that advances in biometric technology and facial recognition had rendered gender markings in passports redundant. He explained that such technologies were not universally employed. Even where employed, they did not eliminate human checking of passports in cases where the automated gates refused admittance, whether because of mistakes by the passport holder or faults affecting the machines, or because of warnings triggered by security databases.

As explained at paras 23 and 25 above, the judge attached weight to the issue of national security, but it was dismissed by the Court of Appeal on the basis, first, that “the ICAO has, for many years, been content for passports to carry an ‘X’ marker”, and, secondly, that “people from countries that already have such provision have been entering the UK for many years without there being any security issues articulated before us”: para 73. The court should not have dismissed the issue so easily, particularly bearing in mind the limitations to its competence, both institutional and constitutional, in relation to questions of national security.

In the first place, the ICAO is not responsible for the security of this country, or any other: the Secretary of State is. The ICAO’s willingness to permit countries to use an “X” marker, if they choose to do so, does not imply that such a course of action is without security implications. Secondly, the relevant security issues, relating to the issuing of passports and to the use of passports, particularly but not only at borders, were fully articulated in the evidence. The Court of Appeal did not address that evidence. It appears to have thought that the evidence was undermined, to the extent that it could be dismissed, because of the absence of evidence relating specifically to visitors to the United Kingdom from countries which have permitted “X” markers for many years. It presumably had in mind New Zealand and Australia, which were the only countries mentioned, and the only countries where “X” markers had existed for more than a few years. But the absence of evidence is not evidence of absence. There was no evidence before the Court of Appeal which contradicted, or even questioned, the evidence of the officials with responsibility for security.

It is, of course, evident that the governments of countries which permit “X” markers (at least, beyond the very restricted use of the kind permitted in the Netherlands, Austria, Germany and Pakistan: paras 17 and 18 above) have made a different judgment from the United Kingdom and the great majority of other countries as to how the competing considerations should be balanced. That seems likely to reflect differences in the circumstances of different countries. For example, some countries have a cultural tradition of recognising a third gender category, as was noted at para 18 above. In some countries, terrorism may be seen as a less serious threat, or unlawful immigration as a less serious problem, than in others. In some countries, campaigners for the use of “X” markers may have greater political influence than in others. Be that as it may, the United Kingdom is not the only country whose government consider that gender markers in passports play a significant part in relation to security: see para 21 above. On the evidence before the court, that is not an unreasonable view to take. That being so, weight should be attached to the Secretary of State’s assessment that the introduction of “X” markers would have a detrimental impact on security.

The second point made by the Secretary of State is that to allow “X” markers to be used in passports would result in significant costs being incurred, as the documents and the application process would have to be altered. The costs were estimated in 2014 at about £2m (para 13 above), and there is evidence that they are likely to have increased since then. This is a relevant factor, although not conclusive.

The third point made by the Secretary of State, and the one to which the courts below attached the greatest weight, returns to the issue of the coherence of the administrative and legal practices within the domestic system. It is argued that to recognise a non-gendered category of individuals in the context of passports would be anomalous, in the context of United Kingdom law and administrative practice, and would therefore undermine the coherence of that system of law and practice. In relation to this point, it should be said at the outset that, although the appellant’s challenge is only to the absence of an “X” marker in passports, that does not mean that the Secretary of State is required to treat that matter as a discrete issue, which must be considered separately from the remainder of the legal and administrative system of the United Kingdom.

As was explained in evidence, there is no legislation in the United Kingdom which recognises a non-gendered category of individuals. On the contrary, legislation across the statute book assumes that all individuals can be categorised as belonging to one of two sexes or genders (terms which have been used interchangeably). Some rights differ according to whether a person is a man or a woman: for example, rights of succession to hereditary titles. There are criminal offences that can only be committed against persons of a particular gender: for example, female genital mutilation. There is a raft of legislation which assumes that only a woman can give birth to, or be the mother of, a child, including legislation relating to maternity rights and benefits, health provision and fertility treatment, and nationality. The legislation governing the registration of births requires the sex of children to be recorded. Legislation relating to marriage and civil partnership (including legislation permitting same sex marriages) assumes that everyone is either a man or a woman. The Gender Recognition Act 2004, enacted following the judgment of the European court in Goodwin v United Kingdom, likewise assumes that all individuals belong to one of two genders, albeit not necessarily the gender recorded at birth. Equality legislation protects people from discrimination if it arises from their being a man or a woman.

A binary approach to gender also forms the basis of the provision of a wide variety of public services. The prison estate, for example, is divided into male and female prisons. Hospitals have wards where patients can only be of a single sex. Local authorities may fund rape crisis centres or domestic abuse refuges which offer their services only to women. Many schools only admit pupils of a particular sex. Much of this is underpinned by, or permitted by, legislation.

Against this background, it is apparent that the questions whether other gendered categories should be recognised beyond male and female, including a non-gendered category, and if so, on what basis such recognition should be given, raise complex issues with wide implications. Counsel for the appellant argued that the courts below had erred in treating the coherent treatment of individuals in the appellant’s position as a significant consideration. On the contrary, the courts were right to conclude that the need for a legally and administratively coherent system for the recognition of gender was an important factor.

Turning to consider the margin of appreciation, this is a concept of particular significance in relation to positive obligations. That is because the imposition of such obligations requires contracting states to modify their laws and practices, and possibly (as in the present case) to incur public expenditure, in order to advance social policies which they may not wholly support, or which they may not regard as priorities, without the imposition of the obligation being supported by any democratic mandate or accountability. While not a conclusive objection, those characteristics of positive obligations indicate the importance of exercising caution before they are imposed. An important conceptual mechanism by which the European court exercises such caution is by interpreting and applying the Convention in a way which allows contracting states a margin of appreciation.

As explained in para 35 above, the width of the margin of appreciation varies according to the circumstances. In that regard, two particularly important factors are, first, whether a particularly important facet of an individual’s existence or identity is at stake, and secondly, whether there is a consensus within the member states of the Council of Europe, either as to the relative importance of the interest at stake or as to the best means of protecting it, particularly where the case raises sensitive moral or ethical issues.

In relation to the first of these issues, notwithstanding the centrality of a non-gendered identification to the appellant’s private life, it is difficult to accept that a particularly important facet of the appellant’s existence or identity is at stake in the present proceedings. That is because it is only the designation of the appellant’s gender in a passport which is in issue. This was explained at para 36 above.

In relation to the second issue, the importance of a consensus within the Council of Europe is readily understood. Courts, including the European court, are expert in adjudication. They do not, on the other hand, possess the capacity, the resources, or the democratic credentials to be well-suited to social policy-making. When adjudication by the European court requires it to consider questions of social policy, it accordingly finds guidance in a consensus on the part of the contracting states, and is cautious before embarking on such policy-making in the absence of a consensus.

In the present case, there is no consensus among the member states of the Council of Europe that passports should be available with an “X” marker, whether it is taken as signifying membership of a non-gendered category or of an unspecified gender. Nor is there any consensus, even among those member states which issue “X” passports, as to the circumstances in which they should be issued (whether, for example, they should be confined to individuals who are biologically intersex, or should be available to any person who identifies as non-gendered), or as to the nature of any evidence or procedure which may be required (whether, for example, a court order should be required, or any form of medical or other evidence). This was explained at para 16 above.

It was argued by counsel for the appellant that the courts below had erred in attaching disproportionate weight to the lack of consensus among the member states of the Council of Europe, while overlooking a wider international trend. That criticism is ill-founded. In considering the margin of appreciation available under the Convention, the focus in relation to issues of consensus must be, and is, on whether there is a consensus among Council of Europe member states. In any event, there is no more consensus internationally than there is among the contracting states on the question whether “X” markers should be provided on passports: see para 17 above.

In addition, the question in this case raises sensitive moral and ethical issues, especially in so far as it impinges on the broader question of gender determination on the basis of an individual’s feelings or choice, regardless of biological sex and physiology, and unconfined by the categories of male and female.

Drawing these various considerations together, the courts below were right to conclude that the considerations relating to the appellant’s interest in being issued with an “X” passport were outweighed by the considerations relating to the public interest put forward by the Secretary of State. In that regard, the importance of a coherent approach across government to the question whether, and if so in what circumstances and on what basis, any gender categories beyond male and female should be recognised, is of particular importance. In addition, it is clear that the matter is one in relation to which the member states should be permitted a wide margin of appreciation, having regard to the absence of any consensus within the member states, the complexity and sensitivity of the issue, and the need for a balance to be struck between competing private and public interests.

In addition, to hold that the appellant’s Convention right to respect for private life requires the Secretary of State to issue the appellant with a passport reflecting the appellant’s identification as non-gendered would go well beyond the case law of the European court. As was explained in R (AB) v Secretary of State for Justice (Equality and Human Rights Commission intervening) [2021] UKSC 28; [2021] 3 WLR 494 (“AB”), para 59, it is open to domestic courts to develop the law in relation to Convention rights beyond the limits of the Strasbourg case law, on the basis of the principles established in that law. They should not, however, go further than they can be confident that the European court would go: AB, para 57. In the present case, that test is far from being satisfied. On the contrary, the application of the principles set out by the European court in cases such as Hämäläinen supports the conclusion that the appellant’s case, so far as based on article 8, was rightly rejected by the courts below.

(2) The argument based on article 14 taken together with article 8

Article 14 provides:

“The enjoyment of the rights and freedoms set forth in this Convention shall be secured without discrimination on any ground such as sex, race, colour, language, religion, political or other opinion, national or social origin, association with a national minority, property, birth or other status.”

The nub of the complaint under article 14 taken together with article 8 is that, whereas individuals who identify as male or female can obtain passports with markers which correspond to that gendered identification, the appellant cannot obtain a passport with a marker which corresponds to a non-gendered identification. That state of affairs is argued to constitute a difference in treatment, in relation to a matter falling within the ambit of article 8, which is unjustifiable: that is to say, either the difference in treatment does not pursue a legitimate aim, or there is not a reasonable relationship of proportionality between the means employed and the aim sought to be realised.

In its essentials, this amounts to the same complaint as that made under article 8, albeit viewed from a different perspective. For the reasons explained at paras 45 and 50-54 above, the aims underlying the Government’s unwillingness to treat the appellant as an exception to the general rule that passport gender markers are either male or female, including the objective of maintaining a coherent approach to issues of gender across the law and administration, are legitimate. In addition, for the reasons explained at paras 56-61 above, this is an area where the contracting states enjoy a wide margin of appreciation.

In these circumstances, the appellant’s case based on article 14 taken together with article 8 was rightly rejected by the courts below.

5. The argument based on Re G

(1) Introduction

In addition to the foregoing arguments, counsel for the appellant argued that, even if the European court would permit a wide margin of appreciation and conclude that there was no violation of the Convention, the domestic courts should nevertheless hold that the Secretary of State was obliged under the Human Rights Act to issue the appellant with an “X” passport, and that her failure to do so was a breach of the appellant’s Convention rights. Counsel relied on dicta in the case of In re G (Adoption: Unmarried Couples) [2008] UKHL 38; [2009] AC 173 (“Re G”) in support of this submission.

(2) Re G

Re G was a decision of the House of Lords on appeal from Northern Ireland. It concerned the question whether legislation which excluded unmarried couples from being eligible to adopt children was compatible with Convention rights. The lower courts concluded that it was, by reference to the case law of the European court. The argument of counsel in the House of Lords, as reported, was directed solely to the interpretation and application of the European jurisprudence: see [2009] AC 173 at pp 175-179.

The majority in the House of Lords decided that an absolute bar on adoption by unmarried couples, regardless of the circumstances of any individual case, was irrational and lacked an objective justification: see paras 16 and 18-20 (Lord Hoffmann), 53-54 (Lord Hope of Craighead), 110-112 (Baroness Hale of Richmond) and 134-135 and 143-144 (Lord Mance). Lord Hoffmann considered that the European jurisprudence had developed to the point that, if the case were heard in Strasbourg, the European court was “not at all unlikely” to hold that the legislation violated article 14 taken together with article 8, with the consequence that the House should not be inhibited from so declaring by the thought that it might be going further than the European court (paras 27 and 29). Lord Hope agreed that it was “not at all unlikely” that the European court would hold that there was a violation of article 14 (para 53), while Lord Mance considered that it was “likely” (para 125).

In the alternative, a majority also indicated their view that even if the European court would consider that the legislation fell within the margin of appreciation allowed to the contracting states, it was nevertheless open to the domestic courts to hold that the legislation violated the Convention rights guaranteed by the Human Rights Act: see paras 30-31 and 36-37 (Lord Hoffmann), 50 (Lord Hope), 119-120 (Lady Hale), and 126-130 (Lord Mance). It was said that where the European court declared a question to be within the national margin of appreciation, it was for the courts in the United Kingdom to interpret the relevant articles of the Convention and to apply the division between the decision-making powers of the courts and Parliament in the way which appeared appropriate for the United Kingdom. On that basis, even if the European court would have found the legislation to be compatible with the Convention, the domestic court could nevertheless hold that the legislation violated Convention rights.

It is reasonably clear that these dicta in the speeches of Lord Hoffmann, Lord Hope and Lord Mance were obiter. Lord Hoffmann considered it not at all unlikely that the European court would find a violation: see para 70 above. Lord Hope, in the critical passage in his speech, stated (para 50):

“Would it [ie, the European court], as my noble and learned friend, Baroness Hale of Richmond, suggests at the outset of her opinion, regard this matter as within the margin of appreciation allowed to member states? If it would, we would indeed be presented with a dilemma. If so, I would be inclined to resolve it in favour of exercising the judgment which that margin of appreciation permits by removing the discrimination against unmarried couples. … But I consider, in agreement with Lord Hoffmann and my noble and learned friend, Lord Mance, that the dilemma is less acute than Baroness Hale’s opening remarks might suggest.”

It appears from that passage that Lord Hope considered his alternative reasoning to be obiter, and understood Lord Hoffmann and Lord Mance to be in the same position. Lord Mance considered it “likely” that the European court would find a violation (para 125), but added that he also agreed with “the alternative route identified by Lord Hoffmann” by which he “would, if necessary”, be prepared to find a breach of Convention rights (para 126). It appears therefore that Lord Mance also regarded this part of his reasoning as obiter.

The view that these dicta were obiter is supported by the absence of any reported argument on the point: it is unlikely that the House would have decided such an important point, as part of the ratio decidendi, without its having been argued by counsel, or at least raised with them. It is also supported by the absence of any discussion of the point in the dissenting speech of the remaining member of the committee, Lord Walker of Gestingthorpe. It is unlikely that he would have ignored the point if he had understood it to be part of the ratio of the majority decision.

(3) Analysis

It is necessary, first, to examine these dicta in Re G, so as to identify the essential elements of the reasoning and consider them at the level of general principle; next, to examine the consistency of the dicta with the statutory scheme created by the Human Rights Act, and with the intention of Parliament as inferred from the terms of the Act and the purpose which it was designed to achieve; and finally, to examine the consistency of the dicta with other decisions of the House of Lords and of this court.

(i) The dicta in Re G

The question raised by the dicta in Re G is whether an act which does not result in a violation of the Convention can nevertheless be incompatible with Convention rights, within the meaning of the Human Rights Act. Clearly, there are differences between the Convention and the Human Rights Act. The Convention is an international treaty which imposes obligations upon the contracting states of the Council of Europe under international law, whereas the Human Rights Act is a statute which imposes obligations upon public authorities under the law of the United Kingdom. The question can therefore be given a more precise formulation: it is whether an act which does not violate the international law obligations of the United Kingdom under the Convention can nevertheless be incompatible with a public authority’s obligations under the Human Rights Act. The answer to that question depends upon a comparison between the content of the obligations imposed by the Convention and the content of the obligations imposed by the Human Rights Act. Since both instruments define the obligations which they impose by reference to the same texts, as will appear from an examination of the Act, the answer must depend upon whether domestic courts can interpret those texts as having a different meaning in the context of the Human Rights Act from the meaning which they have in the context of the Convention.

By this stage of the analysis, one can conclude that the fact that Convention rights within the meaning of the Human Rights Act are domestic and not international rights does not in itself entail that domestic courts can interpret and apply Convention rights in a more expansive way than the European court. Nor did the House of Lords suggest in In re McKerr [2004] UKHL 12; [2004] 1 WLR 807 that they should: that decision, cited by Lord Hoffmann and Lord Mance in Re G in support of their alternative reasoning (at paras 33 and 128), was concerned with the different question of when the domestic Convention rights came into force: a question which evidently fell to be answered by reference to domestic law. The critical factor, according to the dicta in Re G, is whether the reason for the European court’s finding (or likely finding) that an act does not violate an article of the Convention at the international level - or, otherwise put, for finding that the relevant article of the Convention does not impose an obligation to refrain from the act in question - is the contracting states’ margin of appreciation.

The margin of appreciation doctrine is a principle of interpretation of the Convention, based on the need for judicial restraint on the part of the European court. By applying the doctrine, the European court sets the boundaries of compliance with the Convention rights correspondingly wide, and so allows the contracting states a degree of latitude or discretion in relation to their domestic law and practice. The width or narrowness of that latitude can reflect a number of factors, such as whether there is a consensus among the contracting states in relation to a particular issue, whether the national authorities are better placed to make the relevant assessment, and the importance of the right affected. Where the margin of appreciation doctrine is applied in deciding whether there has been a violation of one of the substantive articles of the Convention, the answer therefore depends upon whether the act complained of was within the margin of appreciation allowed to the contracting states.

Accordingly, where the European court applies the margin of appreciation doctrine so as to conclude that there has been no violation of the Convention, it does so by adopting a correspondingly restrained interpretation of the relevant article of the Convention. For example, the concept of family life, within the meaning of article 8, was interpreted in 2001, having regard to the margin of appreciation, as not applying to the relationships of same-sex couples: Mata Estevez v Spain (Application No 56501/00, [2001] ECHR 896) Reports of Judgments and Decisions 2001-VI, p 311 (more recently, the case law has developed beyond that position as a European consensus has emerged). The right to respect for private life, within the meaning of the same article, has been interpreted, on the basis of the margin of appreciation doctrine, as not conferring a right to an abortion (A, B and C v Ireland (2011) 53 EHRR 13), or a right to assisted suicide (Pretty v United Kingdom 35 EHRR 1 (“Pretty”)). Countless other examples could be given. They illustrate the centrality of the margin of appreciation to the interpretation of the Convention.

Of course, it remains open to the contracting states to go beyond the limits of the Convention right, in the exercise of their own national sovereignty, as the Convention itself recognises in article 53. For example, the House of Lords held in 1999, two years before the decision in Mata Estevez, that the same-sex partner of a statutory tenant was a member of the tenant’s “family” within the meaning of the relevant provision: Fitzpatrick v Sterling Housing Association Ltd [2001] 1 AC 27. In doing so, the House was not giving effect to the Convention, or to the Human Rights Act, the material provisions of which were not yet in force, but was construing legislation in a way which went beyond the United Kingdom’s obligations under the Convention as they had been interpreted up to that point. Parliament has chosen to allow abortions within Great Britain, although the Convention imposed no obligation upon it to do so. In doing so, Parliament has created rights which go beyond the ambit of the rights arising under the Convention.

It is necessary to examine the dicta in Re G against that background. Lord Hoffmann, with whose reasoning Lord Hope and Lord Mance expressed agreement (at paras 56 and 126 respectively), stated at para 31 that in “a case in which the Strasbourg court has declared a question to be within the national margin of appreciation … [t]hat means that the question is one for the national authorities to decide for themselves”. “The question” raised in a case before the European court of the kind which Lord Hoffmann had in mind is whether the applicant’s rights under the Convention have been violated. The European court answers that question itself: it does not declare it to be within the national margin of appreciation. Its answer to the question may be influenced, possibly decisively, by its application of the margin of appreciation doctrine. But that does not mean that it is declining to interpret the Convention, or declaring that the interpretation of the Convention is something which it is leaving to the national authorities to decide. The question whether there has been a violation of the Convention is answered by the European court.

The misunderstanding in Re G appears to arise from the European court’s acceptance that, where a margin of appreciation exists, it is open to the national authorities to decide on their own approach to the issue in question, provided they remain within the margin allowed to them. So, for example, if the European court decides that the statutory prohibition of broadcast political advertising is compatible with the right to freedom of expression guaranteed by article 10 of the Convention, having regard to the state’s margin of appreciation, as it did in Animal Defenders International v United Kingdom (2013) 57 EHRR 21, then it is open to Parliament to maintain that prohibition in force without placing the United Kingdom in breach of the Convention. The fact that the margin of appreciation doctrine was applied does not mean that the question whether the prohibition violates article 10 of the Convention has been left to the national authorities to decide, with the consequence that the domestic courts might in principle decide under the Human Rights Act that the ban was indeed a violation of article 10. On the contrary, the question of compatibility with the Convention has been decided.

Lord Hoffmann continued at para 32 by saying that when the European court “says that a question is within the margin of appreciation of a member state, it is not saying that the decision must be made by the legislature, the executive or the judiciary. That is a matter for the member state”. Lord Hoffmann again appears to be proceeding on the basis that the application of the margin of appreciation doctrine means that the question whether the act or legislation in issue is compatible with the Convention is not decided by the European court but is left to the decision of the contracting state. For the reasons I have explained, I consider that to be a misunderstanding. Of course, where the European court decides that there has been no violation of the Convention, applying the margin of appreciation doctrine, it remains open to the contracting state to recognise the right in issue under its domestic law, although it is under no obligation to do so by virtue of the Convention. For example, it is open to the United Kingdom to recognise a right to assisted suicide, although there is no such right under the Convention. The European court evidently has no role in deciding which institution in the United Kingdom should determine whether there should be such a right. That question depends on the United Kingdom’s constitution, under which legislative power is the preserve of the legislature, and the courts’ role in law-making is confined to the development of the common law.

Lord Hoffmann accepted at para 35 that domestic courts applying the Human Rights Act should generally follow the interpretation of the Convention adopted by the European court, for the reasons explained by Lord Bingham of Cornhill in R (Ullah) v Special Adjudicator [2004] UKHL 26; [2004] 2 AC 323 (“Ullah”), para 20: that is to say, because the rights given effect by the Human Rights Act reflect the United Kingdom’s international obligations, and the scope of those obligations is in principle determined by the European court. However, Lord Hoffmann said at para 36 of Re G that “none of these considerations can apply in a case in which Strasbourg has deliberately declined to lay down an interpretation for all member states, as it does when it says that the question is within the margin of appreciation”. In other words, where the European court decides that there has been no violation of the Convention, and its reasoning relies upon the margin of appreciation doctrine, it is to be regarded as having declined to interpret the relevant article of the Convention. As I have explained, I respectfully disagree with that contention. If, for example, the question is whether article 8 should be interpreted as conferring a right to assisted suicide, then the European court decided that question in Pretty, applying the margin of appreciation doctrine. The margin of appreciation doctrine is not an abdication of the task of interpretation, but an important aspect of that task.

On the premise that the application of the margin of appreciation doctrine entails a failure by the European court to interpret the Convention, Lord Hoffmann reached his conclusion at para 37:

“In such a case, it is for the court in the United Kingdom to interpret articles 8 and 14 and to apply the division between the decision-making powers of courts and Parliament in the way which appears appropriate for the United Kingdom. The margin of appreciation is there for division between the three branches of government according to our principles of the separation of powers. There is no principle by which it is automatically appropriated by the legislative branch.”

This passage proceeds, in the context of that case, on the hypothesis that the European court would decide that a law preventing unmarried couples from being adoptive parents did not violate their rights under articles 8 and 14, having regard to the margin of appreciation. The first step in the reasoning, as we have seen, is then to characterise such a decision as a failure to interpret those articles. For the reasons I have explained, I respectfully disagree with that characterisation, and with the consequent inference that it is the function of the domestic court to remedy that supposed failure. It follows that I also disagree with the next step in the argument, which treats the domestic court as consequently having the function of effecting a “division” of the margin of appreciation between Parliament, the executive and the courts, so that the supposed failure can then be remedied by whichever of those institutions the court considers appropriate. I accordingly also reject the concluding step in the argument: that, where the court considers that it should itself remedy the supposed failure, it can consequently interpret the relevant articles more expansively than the European court. As has been explained, the margin of appreciation is itself a principle of interpretation. When the European court finds that the contracting states should be permitted a margin of appreciation, it does not cede the function of interpreting the Convention to the contracting states, or enable their domestic courts to divide that function between their domestic institutions. Contracting states can of course create rights going beyond those protected by the Convention, but that power exists independently of the Convention and the Human Rights Act, is not dependent on the margin of appreciation doctrine, and is exercisable in accordance with long-established constitutional principles, under which law-making is generally the function of the legislature.

(ii) The statutory scheme and intention of the Human Rights Act

The drafting and structure of the Human Rights Act also point away from the alternative approach favoured in Re G. Section 1(1) defines “the Convention rights” as “the rights and fundamental freedoms set out in - (a) articles 2 to 12 and 14 of the Convention, (b) articles 1 to 3 of the First Protocol, and (c) article 1 of the Thirteenth Protocol, as read with articles 16 to 18 of the Convention”. In terms of section 1(2), those articles are to have effect for the purposes of the Act subject to any designated derogation or reservation: that is to say, derogations or reservations by the United Kingdom at the international level. Section 21(1) defines “the Convention” as “the Convention for the Protection of Human Rights and Fundamental Freedoms, agreed by the Council of Europe at Rome on 4 November 1950 as it has effect for the time being in relation to the United Kingdom”. The same section defines “the First Protocol” as “the protocol to the Convention agreed at Paris on 20 March 1952”. The relevant articles of the Convention and its protocols are set out in Schedule 1 to the Act.

The Convention rights which the Act introduced into the domestic law of the United Kingdom are thus defined as the rights set out in the relevant articles of the Convention and its protocols, as those instruments have effect for the time being in relation to the United Kingdom. The Convention and its protocols have effect in relation to the United Kingdom by imposing obligations on the United Kingdom under international law to act compatibly with the rights which they guarantee. The Act therefore defines the Convention rights to which it gives effect in domestic law as the rights which are enforceable against the United Kingdom under international law. It follows that the rights given effect in domestic law have the same content as those which are given effect under international law, although they are enforceable before domestic courts rather than the European court, and against public authorities rather than the United Kingdom as a state. Since the rights have the same content at the domestic level as at the international level, it follows that the relevant articles of the Convention should in principle receive the same interpretation in both contexts. That is not to say that domestic courts are bound to follow every decision of the European court, but there should in principle be an alignment between interpretation at the international and domestic levels.

This identity of content at the domestic and the international levels is the norm in the Council of Europe. The majority of contracting states apply monist theories of international law, whereby the Convention has immediate effect in domestic law without implementing legislation. Under systems of that kind, the articles of the Convention and its protocols are directly enforceable before domestic courts against domestic institutions, without any need for the creation of equivalent rights in domestic law. In the dualist state of Ireland, the European Convention on Human Rights Act 2003 explicitly defines the domestic obligations which it imposes by reference to compatibility with the state’s obligations under the Convention.

Section 2(1) of the Human Rights Act confirms the importance of the case law of the European court to the interpretation of Convention rights, by stipulating that a court or tribunal determining a question which has arisen in connection with a Convention right must take into account any relevant judgment of the European court. The issue which lies at the heart of many of those judgments is the application of the margin of appreciation doctrine. As has been explained, that doctrine is central to the European court’s interpretation of the rights set out in the Convention and its protocols, since it sets limits to the ambit of those rights. The European court’s treatment of the margin of appreciation cannot therefore be severed from its interpretation of the Convention rights: it is part and parcel of the exercise of interpretation.

It is also necessary to consider section 3(1) of the Act, which requires that “[s]o far as it is possible to do so, primary legislation and subordinate legislation must be read and given effect in a way which is compatible with the Convention rights”. This provision, as interpreted by the House of Lords in Ghaidan v Godin-Mendoza [2004] UKHL 30; [2004] 2 AC 557, has remarkable consequences if the alternative reasoning in Re G is adopted. Following that approach, it is open to domestic courts to modify unambiguous legislation under section 3(1) so as to bring about a result which departs from Parliament’s intention in enacting that legislation, where they consider that Parliament’s approach fails to comply with Convention rights, even though the European court would itself accept that Parliament’s assessment was legitimate. That would constitute a significant encroachment on the principle of Parliamentary sovereignty: a principle which, it has long been recognised, the Human Rights Act is careful to protect (see, for example, R v Director of Public Prosecutions, Ex p Kebilene [2000] 2 AC 326, 367, and Wilson v First County Trust Ltd (No 2) [2003] UKHL 40; [2004] 1 AC 816, para 120). It is true that section 3 on any view extends the common law principle under which international conventions can influence the interpretation of legislation, but its effect is confined (but for the dicta in Re G) to cases where there would otherwise be a breach of international law. The dicta in Re G would go considerably further, by permitting judges to override the intention of Parliament on the basis of their individual assessments of the requirements of the Convention rights, notwithstanding Parliament’s compliance with the requirements imposed by the Convention under international law. That would represent a substantial expansion of the constitutional powers of the judiciary, at the expense of Parliament.

Further consequences would also flow from the dicta in Re G. For example, it would be open to the courts to make declarations of incompatibility under section 4 of the Act in relation to Acts of Parliament, notwithstanding that the European court would find the legislation to be compatible with the Convention. The courts could also find (subject to section 6(2)) that public authorities were acting unlawfully in implementing Acts of Parliament, contrary to section 6(1) of the Act, and grant remedies including injunctions and damages under section 8, notwithstanding that the European court would consider the legislation to be compatible with the Convention.

It is also apparent that the dicta in Re G would depart from the intention “to make more directly accessible the rights which the British people already enjoy under the Convention … [i]n other words, to bring those rights home”, as it was put in the White Paper which preceded the Act (Rights Brought Home: The Human Rights Bill, Cm 3782 (1997), para 1.19). It is true that the same White Paper, referring to the “living instrument” doctrine of the European court, also expressed the intention that “[i]n future, our judges will be able to contribute to this dynamic and evolving interpretation of the Convention” (para 2.5), but as I shall explain, they are able to do so without following the dicta in Re G. More importantly, those dicta would undermine the constitutional principle of Parliamentary sovereignty. They would also undermine legal certainty, since the meaning of legislation would not depend on the ordinary meaning of the language used, or on the legal principles established in the case law of the European court, but on domestic judges’ interpretations of Convention rights. It might be argued that, even on an orthodox approach to the interpretation of Convention rights, there is a degree of tension between those principles and the aims of the Act, but the tension would be significantly greater if the dicta in Re G were followed, and domestic courts were consequently at liberty to create Convention rights of their own devising, unmoored from the case law of the European court.

These issues were not discussed in Re G, but they appear to me to be destructive of the dicta in that case.

(iii) Other decisions of the House of Lords and the Supreme Court

The dicta in Re G are also difficult to reconcile with a large body of authority at the highest level, both prior and subsequent to that decision. As this court explained in AB at paras 54-59, the Human Rights Act has long been construed as having been intended to enable the rights and remedies available in the European court to be asserted and enforced by domestic courts, rather than as being intended to provide a basis for the development of rights of the domestic courts’ own creation. In that regard, the court cited in AB the well-known passage in the speech of Lord Bingham in Ullah, para 20, as well as Lord Bingham’s statement in R (SB) v Governors of Denbigh High School [2006] UKHL 15; [2007] 1 AC 100, para 29 that “the purpose of the Human Rights Act 1998 was not to enlarge the rights or remedies of those in the United Kingdom whose Convention rights have been violated but to enable those rights and remedies to be asserted and enforced by the domestic courts of this country and not only by recourse to Strasbourg”.

Similar statements preceding the decision in Re G can be found in Aston Cantlow and Wilmcote with Billesley Parochial Church Council v Wallbank [2003] UKHL 37; [2004] 1 AC 546, paras 6-7 (Lord Nicholls of Birkenhead) and 44 (Lord Hope), R (Greenfield) v Secretary of State for the Home Department [2005] UKHL 14; [2005] 1 WLR 673, para 19 (Lord Bingham), N v Secretary of State for the Home Department (Terrence Higgins Trust intervening) [2005] UKHL 31; [2005] 2 AC 296, paras 23-25 (Lord Nicholls), R (Quark Fishing Ltd) v Secretary of State for Foreign and Commonwealth Affairs [2005] UKHL 57; [2006] 1 AC 529, paras 25 (Lord Bingham), 32-34 and 44 (Lord Nicholls), and 87-88 and 92 (Lord Hope), R (Al Skeini) v Secretary of State for Defence (The Redress Trust intervening) [2007] UKHL 26; [2008] 1 AC 153, paras 56-58 and 81 (Lord Rodger of Earlsferry), 90-91 (Lady Hale) and 105-107 (Lord Brown), R (Countryside Alliance) v Attorney General [2007] UKHL 52; [2008] 1 AC 719 (“Countryside Alliance”), para 126 (Lady Hale), and R (Animal Defenders International) v Secretary of State for Culture, Media and Sport [2008] UKHL 15; [2008] AC 1312 (“Animal Defenders International”), para 53 (Lady Hale).

With the exception of Ullah, none of these authorities appears to have been mentioned in the speeches, or cited in argument, in Re G. As explained earlier, Lord Hoffmann accepted the general correctness of Lord Bingham’s dictum in Ullah, but distinguished that case on the basis that it was not a case in which the European court had applied the doctrine of the national margin of appreciation, and was therefore not a case in which the European court had declined to interpret the Convention and had instead left that task to the domestic authorities. For the reasons explained earlier, I do not find that a convincing distinction. One might also add that other cases decided prior to Re G could not be distinguished on that basis. In Countryside Alliance, for example, Lady Hale said at para 126 that “when we can reasonably predict that Strasbourg would regard the matter as within the margin of appreciation left to the member states, it seems to me that this House should not attempt to second guess the conclusion which Parliament has reached”. This, she said, was not to do with the subject matter of the issue, whether it be moral, social, economic or libertarian; “it has to do with keeping pace with the Strasbourg jurisprudence as it develops over time, neither more nor less”, citing Lord Bingham’s dictum in Ullah, para 20.

In Animal Defenders International, Lady Hale returned to the question of the court’s role in a context where the European court would permit a margin of appreciation, and stated at para 53:

“The Human Rights Act 1998 gives effect to the Convention rights in our domestic law. To that extent they are domestic rights for which domestic remedies are prescribed: In re McKerr [2004] 1 WLR 807. But the rights are those defined in the Convention, the correct interpretation of which lies ultimately with Strasbourg: R (Ullah) v Special Adjudicator [2004] 2 AC 323, para 20. Our task is to keep pace with the Strasbourg jurisprudence as it develops over time, no more and no less: R (Al-Skeini) v Secretary of State for Defence (The Redress Trust intervening) [2008] 1 AC 153, para 106.”

As Lady Hale noted, there was nothing to stop Parliament from legislating to protect human rights to a greater extent than the Convention and its jurisprudence currently required. Nor was there anything to prevent the courts from developing the common law in that direction. Nevertheless, she continued:

“But we are here concerned with whether an Act of the United Kingdom Parliament is compatible with the Convention rights: if prima facie it is not, Parliament has given us the duty, if possible, to interpret it compatibly with those rights (Human Rights Act 1998, section 3(1)); and if that is not possible, the power to declare it incompatible: section 4. I do not believe that, when Parliament gave us those novel and important powers, it was giving us the power to leap ahead of Strasbourg in our interpretation of the Convention rights. Nor do I believe that it was expecting us to lag behind. The purpose, in my view, of a declaration of incompatibility is to warn Government and Parliament that, in our view, the United Kingdom is in breach of its international obligations. It is then for them to decide what, if anything, to do about it.”

That reasoning is consistent with the present judgment. The penultimate sentence, in particular, is premised on the view that the obligations created by the Human Rights Act have the same content as those imposed at the international level by the Convention.

Other cases which I have cited also indicate that the alignment between Convention rights at the domestic and international levels described in Ullah was not as narrow in scope as was suggested in Re G. For example, in N v Secretary of State for the Home Department, para 25, Lord Nicholls stated as a general proposition, in relation to the courts’ role under the Human Rights Act: “It is not for us to search for a solution to [the appellant’s] problem which is not to be found in the Strasbourg case law”. In R (Quark Fishing Ltd) v Secretary of State for Foreign and Commonwealth Affairs, para 25, Lord Bingham stated unequivocally: “A party unable to mount a successful claim in Strasbourg can never mount a successful claim under sections 6 and 7 of the 1998 Act”. In the same case, Lord Nicholls stated at para 34 that “the Act was not intended to provide a domestic remedy where a remedy would not have been available in Strasbourg”.

Later cases have reaffirmed the linkage between Convention rights at the domestic and international levels. An illustration is the case of Ambrose v Harris (Procurator Fiscal Oban) [2011] UKSC 43; [2011] 1 WLR 2453, in which Lord Kerr, in a dissenting judgment, challenged the correctness of the Ullah approach as a matter of principle. The other members of the court responded by reasserting the principle: see paras 16-20, 68, 70 and 72 (Lord Hope, with whom Lord Brown, Lord Dyson and Lord Matthew Clarke agreed), 86-87 (Lord Brown) and 105 (Lord Dyson). In Smith v Ministry of Defence (JUSTICE intervening) [2013] UKSC 41; [2014] AC 52, Lord Hope said at para 43, with the agreement (on this issue) of all the members of a court of seven judges:

“Lord Bingham’s point [in Ullah] was that Parliament never intended by enacting the Human Rights Act 1998 to give the courts of this country the power to give a more generous scope to the Convention rights than that which was to be found in the jurisprudence of the Strasbourg court. To do so would have the effect of changing them from Convention rights, based on the Treaty obligation, into free-standing rights of the court’s own creation.”

At the same time, the later cases, including most recently AB at paras 54-59, have also made it clear that Parliament’s intention to implement domestically the rights which are available under the Convention at the European level does not mean that domestic courts are bound to follow and apply the jurisprudence of the European court slavishly or unquestioningly: on the contrary, they can and do decline to follow Strasbourg judgments where there is a good reason to do so, as was explained by an enlarged court of nine judges in Manchester City Council v Pinnock (Secretary of State for Communities and Local Government intervening) [2010] UKSC 45; [2011] 2 AC 104, para 48. Nor are domestic courts prevented from developing the law in relation to Convention rights beyond the limits of the European case law. As was explained in AB at para 59, the protection of Convention rights under the Human Rights Act can go beyond the situations previously considered by the European court, where the principles established by that court enable such a step to be taken. In these ways, domestic courts can and do engage in dialogue with the European court and influence the development of its case law. It is also important to remember that, quite apart from the Human Rights Act, the Convention can influence statutory interpretation and the development of the common law. But it has generally been accepted that the domestic application of Convention rights under the Human Rights Act has to be based on the principles established in the case law of the European court, rather than upon a distinct domestic interpretation of Convention rights.

The dicta in Re G have been cited, in the later case law, mostly in dissenting or concurring judgments (as in D v Comr of Police of the Metropolis (Liberty and others intervening) [2018] UKSC 11; [2019] AC 196, para 153, per Lord Mance, and R (DA) v Secretary of State for Work and Pensions (Shelter Children’s Legal Services and others intervening) [2019] UKSC 21; [2019] 1 WLR 3289, paras 149 and 167, per Lady Hale and Lord Kerr respectively), or in obiter dicta (as in In re Recovery of Medical Costs for Asbestos Diseases (Wales) Bill [2015] UKSC 3; [2015] AC 1016, para 54, per Lord Mance, In re Northern Ireland Human Rights Commission’s Application for Judicial Review [2018] UKSC 27; [2019] 1 All ER 173, paras 38 and 115, per Lady Hale and Lord Mance respectively, and In re McLaughlin [2018] UKSC 48; [2018] 1 WLR 4250, para 34, per Lady Hale).

It may be worth mentioning the case of Rabone v Pennine Care NHS Foundation Trust (INQUEST intervening) [2012] UKSC 2; [2012] 2 AC 72, para 112, where Lord Brown (with whom Lord Walker agreed) affirmed that it is open to the domestic courts, consistently with Ullah, to reach a conclusion which flows “naturally from existing Strasbourg case law”, even if the existing case law does not deal with the precise issue at hand. That, Lord Brown said, was the position in Re G. The implication is that the dicta in Re G, to the effect that domestic courts can depart from the Ullah principle when the European court would permit a margin of appreciation, were not considered to form part of the ratio of the decision.

The same approach was adopted in Moohan v Lord Advocate (Advocate General for Scotland intervening) [2014] UKSC 67; [2015] AC 901, where the court held that the statutory provisions which prevented convicted prisoners from voting in the referendum on Scottish independence did not breach Convention rights. Lord Hodge (with whom Lord Neuberger, Lady Hale, Lord Clarke and Lord Reed agreed; Lord Kerr and Lord Wilson dissenting) cited Re G and para 112 of Lord Brown’s judgment in Rabone as authority for the proposition that “[o]n occasion our domestic courts may choose to go further in the interpretation and application of the ECHR than Strasbourg has done where they reach a conclusion which flows naturally from Strasbourg’s existing case law” (para 13). On the facts of the case, however, the Strasbourg court had rejected the suggestion that the relevant article applied to referendums, with the consequence that the petitioners’ claim therefore failed (para 18). The reasoning is consistent with the approach described in para 101 above. The passages in Rabone and Moohan which I have cited were also cited with approval in AB, para 59.

One of the few cases in which the dicta in Re G have been cited in judgments contributing to the majority decision is R (Nicklinson) v Ministry of Justice (CNK Alliance Ltd and others intervening) [2014] UKSC 38; [2015] AC 657. In that highly unusual case, this court decided by a majority (Lord Neuberger, Lord Mance, Lord Clarke, Lord Wilson, Lord Sumption, Lord Reed and Lord Hughes; Lady Hale and Lord Kerr dissenting) not to grant a declaration that the statutory provision which criminalises assisted suicide was incompatible with article 8 of the Convention. Although it was noted that the European court would have concluded that there was no violation, applying the margin of appreciation doctrine, that was not treated as necessarily conclusive. The ratio of the decision, if it has one, is elusive. Four members of the majority (Lord Clarke, Lord Sumption, Lord Reed and Lord Hughes) considered that the court should respect Parliament’s assessment that the prohibition was justified. The other members of the majority (Lord Neuberger, with whom Lord Wilson agreed, at paras 71-75, and Lord Mance at para 163) cited and endorsed the dicta in Re G. However, all three of those judges also concluded that the appellants had failed to show that the prohibition was a disproportionate interference with article 8 rights: see Lord Neuberger at paras 115 and 126-128, Lord Mance at paras 175-177, 182 and 187-190, and Lord Wilson at paras 197-201. The court did not hear argument on the correctness of the dicta in Re G, and the issues discussed above were not considered.

In R (Steinfeld) v Secretary of State for International Development [2018] UKSC 32; [2020] AC 1, the question for the court was whether statutory provisions which precluded couples of different sexes from entering into civil partnerships, which the Secretary of State accepted fell within the ambit of article 8 and gave rise to a difference in treatment on a ground proscribed by article 14, were justified. The court unanimously held that they were not. Lord Kerr, in a judgment with which the other members of the court expressed agreement, rejected the Secretary of State’s argument based on the assertion that the European court would allow a wide margin of appreciation, referring in that regard to the dicta in Re G. Nonetheless, Lord Kerr’s approach was aligned with that of the European court. He observed that, on the supranational plane, a narrow margin of appreciation applied in cases concerned with discrimination on the ground of sexual orientation (para 32). Referring to a relevant decision of the European court, he concluded that “[t]he margin of discretion available to the Government and Parliament in this instance, if it exists at all, must be commensurately narrow” (para 32).

In summary, while Nicklinson, in particular, might be regarded as providing support for the dicta in Re G, albeit without any discussion of the issues raised above, there is a substantial and consistent body of case law in which the previously prevailing approach has continued to be followed. The line of decisions from Ullah to AB demonstrates the general alignment between the interpretation of Convention rights at the domestic and the international levels, whether or not the European court would apply the margin of appreciation doctrine. A final illustration of the point, in a context where a wide margin of appreciation is generally permitted by the European court, is the line of domestic cases concerned with discrimination in relation to social security benefits, recently considered by this court in the case of SC. In Humphreys v Revenue and Customs Comrs [2012] UKSC 18; [2012] 1 WLR 1545, Lady Hale (with whom the other members of the court agreed) stated at para 15 that the proper approach to justification in cases involving discrimination in relation to state benefits was to be found in a judgment of the European court which allowed a wide margin of appreciation, according to which the court would generally respect the legislature’s policy choice unless it was “manifestly without reasonable foundation”. The same test, she explained, had been applied by the House of Lords in the earlier case of R (RJM) v Secretary of State for Work and Pensions (Equality and Human Rights Commission intervening) [2008] UKHL 63; [2009] 1 AC 311, and Lady Hale went on to apply it in the case before the court. That approach, whereby the margin of appreciation permitted by the European court in that field is reflected in the domestic application of Convention rights, has been followed in numerous decisions of this court ever since, as was explained in SC at paras 143-156.

(iv) Conclusions in relation to the argument based on Re G

In the light of the foregoing discussion, the dicta in Re G (1) are best understood to have been obiter, (2) are based upon a misunderstanding of the nature of the margin of appreciation doctrine, (3) are difficult to reconcile with the structure and purpose of the Human Rights Act, (4) result in an encroachment upon Parliamentary sovereignty which Parliament is unlikely to have intended, (5) undermine legal certainty, and (6) are inconsistent with the prevailing approach of the House of Lords and this court both prior and subsequent to that decision. In these circumstances, the dicta in Re G should be disapproved.

It follows that those dicta do not provide a basis for allowing the present appeal.

6. Conclusion

Although the European court does not appear to have considered the issue raised in these proceedings, this court, and the courts below, have endeavoured to assess whether, applying the principles established in the European case law, the Secretary of State is under an obligation, by virtue of article 8 of the Convention, considered either alone or together with article 14, to provide the appellant with an “X” passport. The conclusion is that the Convention imposes no such obligation, at least at the present time. There is no reason why that assessment of the position at the European level should not be followed at the domestic level in the application of the Human Rights Act.

In these circumstances, the appeal should be dismissed.

United Kingdom Supreme Court