The ageing individual: socially constructed, must be understood in an externalised context

Doron has labelled elder law, as initially developed in the US setting, ‘the Positivist-Professional Approach’, defined by the clients served and the areas relevant to legal ‘counselling’. He then pictures the theoretical development of elder law in terms of ‘Monist Jurisprudential Approaches to Law and Aging’ and the ‘Pluralist Approach’ (Doron, Reference Doron, Bengtson, Gans, Putney and Silverstein2009a). Whereas Monist Approaches – such as ‘Law and Economics’ and ‘the Feminist Ethic Care Approach’ – often are not elder-specific, but instead are characterised by the use of a single (monist) conceptual lens, the Pluralist Approach is in fact elder-specific. Doron developed the Pluralist Approach ‘in response to the argument that the only way to fully grasp the richness and diversity of elder law is through a multidimensional model that connects the different functions and targets that law wishes to achieve’ (Doron, Reference Doron, Bengtson, Gans, Putney and Silverstein2009a: 650). These later theoretical approaches within elder law have long since abandoned narrow positivism. A recurring theme, moreover, has been the need to expand elder law into ‘geriatric jurisprudence’ (Cohen, Reference Cohen1978: 229; Doron and Meenan, Reference Doron and Meenan2012: 194). Notwithstanding, the special needs and ‘rights’ of older persons are frequently the starting point. Today, the need for an international convention on human rights of older people is at the centre of discussion – its justification and whether such rights are any different than other human rights (Dabove, Reference Dabove2013, Reference Dabove2016; Doron and Georgantzi, Reference Doron and Georgantzi2018).

I will argue here for a broad and multifaceted approach to the relationship between law and ageing. Just as ‘essentialism’ is long gone in feminist studies (Spelman, Reference Spelman1991), having been replaced by a multifaceted gender approach, the same ought to be true in the present subject area. There is no such thing, namely, as a one-dimensional ageing individual. In line with the theory of social constructivism – part of a broad tradition of sociological theories of knowledge (Berger and Luckman, Reference Berger and Luckman1966) – the ageing individual should not be seen as a given from natural or biological facts, but rather as the result of human interactions at a given moment in history. While this is widely recognised, studies taking their starting point in a more general problem-area, disclosing eventual ageing issues as they appear, are less frequent. So, the 2015 Inter-American Convention on Protecting the Human Rights of Older Persons defines ‘old age’ as a social construct of the last stage of the lifecourse and ‘ageing’ as a gradual process that develops over the course of life associated with dynamic interactions between the individual and their environment – yet an ‘older person’ is said to be a person aged 60 or more, fixing such rights to old age itself (Article 2).

A deepened understanding of the social construct of ageing in a particular area, in my opinion, requires in-depth study and far-reaching knowledge of the area in question. A historically contextualised perspective yields the insight that present arrangements are not necessarily natural or universally determined, but rather are the consequence of past and present circumstances and constraints. Applying a contextualised approach also means complementing a special analysis in terms of ageing with the particular theoretical approaches applicable to the matter at hand. The latter is what I have chosen to name ‘externalisation’. One thing is the contextualisation of ageing in particular in a certain field, another thing is to understand this in terms of a larger story, in itself external to or only indirectly related to ageing. In the example of compulsory retirement as discussed below, such particular theories regard labour law as analysed by Freedland and Lazear, and discrimination law in the terms of Somek and others. The concept of contextualisation can, of course, be understood so as to cover both a narrower contextualisation of the ageing issues proper and an externalisation, taking in the broader understanding of a particular field and its theorisation, in order to ‘unwind’ its specific implications for ageing.

From the perspective of elder law, I find it crucial – as Joanne Conaghan does with regard to gender – (a) to explore the aged nature of accounts of this social world; (b) to focus on disadvantage and change; and (c) to place the ageing individual at the centre of analysis (Conaghan, Reference Conaghan, Morris and O'Donnell1999: 17). An analysis becomes truly multi-disciplinary first when it is carried out both within and outside the legal setting, and in terms of both old age and other realities. Also – as Conaghan notes in relation to gender – the application of an age-specific lens to labour law helps to expose the conceptual and normative architecture supporting that field of knowledge in the externalised perspective, thereby inviting its (critical) scrutiny (Conaghan, Reference Conaghan2017: 107). In this article, I confine my discussion to the area of labour law and to that of age-discrimination regulation.

Thus, by defining the legal field and the age groups studied on the basis of such an external perspective on law, we take some pressure off the issue of ‘old age’ itself as a legal construct. As noted above, the latter is not necessarily a matter of a fixed age or of a certain span of years. The object here is to identify a set of problems at the societal level that relate to old age in much less precise and much more differentiated ways. The boundaries of the relevant age group(s) must be contextualised in different ways depending on the issues studied – including an externalised understanding of the subject area. When we consider labour law and the functioning of labour markets, it becomes particularly clear that an ‘ageing individual’ in this field of study may be remarkably young.

It is true that, in the case of elder law, there is bound to be a connection with the somewhat more mature stages of the lifespan, i.e. with age in a chronological sense, and towards the latter part of the scale. But viewed in a life-long perspective, one's age is simply a point on a continuous scale. There are different stages of age, and they are fluid. Contextualisation opens the field up rather than delineating it. Nevertheless, old-age issues have been found to be rooted – at least to some extent – in physical materiality (Fudge and Zbyszewska, Reference Fudge, Zbyszewska, Numhauser-Henning and Rönnmar2015: 142). In the field of elder law, ‘old’ age is the conceptual starting point, ageism is the historical context (and the history may vary from one issue to the next!) and inequality – or simply injustice – is the most pressing current reality (Conaghan, Reference Conaghan2009: 307; Doron, Reference Doron, Ruebner, Do and Taylor2015; Harding, Reference Harding, Doron and Georgantzi2018). Ageism and inequality are also what make age-discrimination regulation of particular interest in an elder-law perspective – a point on which I elaborate in the next section.

The conceptualisation of an ageing individual, then, must be broadly contextualised in the proper social setting. So too must ‘solutions’ to the problems of old age. Yet we do need conceptual tools for analysis in relation to precisely the ageing individual. Elder law, it has often been said, contains two main dichotomies: that between autonomy and dependency/paternalism, on the one hand, and that between the individual and society, on the other (Doron, Reference Doron and Doron2009b: 70; Numhauser-Henning, Reference Numhauser-Henning and Numhauser-Henning2017b). On the one hand, the needs of the ageing individual perceived through the lens of the legal professional have been the focus of attention, as have the protective and supportive functions of legal intervention. On the other hand, principles of autonomy and independence can also be said to inform elder law, forming the basis for the legal counselling so characteristic of traditional elder law. As I see it, these dichotomies and concepts often yield a fruitful approach to discussions of elder law. However, these dichotomies also ‘overlap’ considerably, making them hard to separate completely. Nor may such separation be necessary. The autonomy versus paternalism dichotomy relates mainly to the values or attitudes behind a given legal solution; by contrast, the individual versus society dichotomy addresses the more organisational level of interests. Doron, in his multi-dimensional model of elder law, thus proposes a more genuinely pluralistic approach to elder-law analysis proper, notwithstanding his inclusion of the four key concepts of the two dichotomies (Doron, Reference Doron and Doron2009b: 60).

Against this background, some scholars have argued that elder law itself – and age-discrimination bans as well – is inherently paternalistic. In this view, both elder law and anti-discrimination regulation make ‘old’ age a ‘problem’, and portray older persons as needing support and protection. In the next section, I examine the problem of ageism, and anti-discrimination regulation as a crucial legal response to it. Also, I contend, a non-paternalistic, autonomous and individual approach is not necessarily the best solution to real-world problems (Numhauser-Henning, Reference Numhauser-Henning and Numhauser-Henning2017b).

The ‘old’ worker, age discrimination and labour law

Hitherto studies on the ban on age discrimination, and its relationship to labour law generally, point up the importance of a broadly contextualised old-age concept for achieving a deepened understanding of the rules of ‘elder law’ in this area. ‘Old’ age is but part of a larger – or externalised – story here. The story itself has its special impact on what is perceived as ‘old’.

The larger story is shaped by assumptions about what constitutes a typical working-life trajectory and about what ‘ideal’ workers – those closely fitted to the needs of production over time – require. An ageing population therefore challenges the functioning of labour markets, of labour law and of social-security systems. At the same time, any ‘solution’ requires a deep understanding of these systems.

Wage work has long been the dominant distributive order in our society. Social-security arrangements – including pension systems – have developed successively as a social order complementary to wage work, accompanied by labour-law developments reflective of ‘the standard employment contract’. According to Freedland (Reference Freedland, Stone and Authors2013), the standard employment contract entails continuous full-time employment with a single employer, with accompanying expectations of remuneration and benefits. It is also grounded in a male-breadwinner model, with a wife and children economically dependent on the male breadwinner (Strauss, Reference Strauss, Countouris and Freedland2013: 337). However, the standard employment contract has never been a one-dimensional or altogether standardised concept or pattern of employment; rather, it has taken somewhat different legal forms in different national contexts. One of its characteristics is seniority wage-setting. That is, workers’ wages do not fluctuate throughout their careers in response to actual productivity; rather, wages tend to rise successively until retirement. This is what led the influential American economist Edward Lazear to launch his lifecycle theory of mandatory (compulsory) retirement as an implicit part of the standard employment contract. Employers pay their employees a wage premium towards the end of their career on the assumption that the employment relationship will cease at a predictable, fixed point in time specified in a bargained or statutory pension scheme (Lazear, Reference Lazear1979). This means employees receive a wage premium at both the beginning and the end of their career, but are paid less than their marginal productivity in mid-career. The wage premium at career's end is a kind of deferred compensation. The implicit contract, as Lazear describes it, depends on employment-protection devices to maintain the inherent promise of a wage premium.

Already early on in industrialised society, production requirements typically led – despite the implicit contract identified by Lazear – to a marginalisation of elderly and less-productive workers. Simply put, employers sought to get rid of workers who had become old and less productive. To some extent, it is true, the implicit contract could be maintained by means of severance-pay arrangements and bargained early retirement schemes. In more recent decades, technological evolution, the growth of service society and globalisation have led to what is generally known as the flexibilisation of work. Flexibilisation has been pictured as a new trade-off between the pronounced need for flexibility, on the one hand, and traditional standard employment and pension solutions, on the other. The introduction and growing importance of anti-discrimination regulation is held to be related to this trend – offering protection in an era when employment standards are being replaced by individualisation and equal treatment. The deregulation of employment protection is being accompanied, namely, by an increase in labour-market segmentation and by the spread of part-time and short-term employment (Numhauser-Henning, Reference Numhauser-Henning, Numhauser-Henning and Rönnmar2015).

Whereas mandatory retirement and the standard employment contract can be said to have influenced general labour-law developments worldwide, the same is not necessarily true of employment-protection devices. In the EU, despite the absence as yet of any harmonisation proper regarding employment protection, there is a partly harmonised EU-law framework, as well as a ‘constitutionalisation’ of employment protection through Article 30 of the Charter of Fundamental Rights, which typically offers fairly extensive employment protection.Footnote ¹ In the USA, by contrast, anti-discrimination regulation developed early on as an important substitute for employment protection, which is minimal in that country (Fineman, Reference Fineman2013). Whereas the employment-at-will doctrine has made prohibitions against discrimination a central feature of the regulation of working life in the USA, the increasing flexibilisation of labour markets has attracted somewhat less attention (there being less of any standard protection of workers to ‘deregulate’). Notwithstanding this, the trend towards individualisation in working life – and precarisation as well – is noticeable in the USA too (Lester, Reference Lester, Numhauser-Henning and Rönnmar2015). Japan may be the country where standard employment and its implicit contract as described above were taken to their extreme, with practices such as life-long employment and seniority wage-setting (Araki, Reference Araki, Numhauser-Henning and Rönnmar2015).

Age-discrimination regulation is thus an important tool within elder law – not least when it comes to working life. A ban on age discrimination was first introduced in the USA through the Age Discrimination in Employment Act (ADEA), in 1967. It was modelled on title VII of the 1964 Civil Rights Act. The ADEA only prohibits differential treatment on grounds of ‘old’ age – now set at 40 years of age or above. The range of age groups covered has changed somewhat over the years, however. Basically, it has been set in accordance with labour-market realities and the need for protective rules in relation to certain age groups.Footnote ² When the EU adopted its ban on age discrimination, it made the regulation age-neutral. Thus, the ban introduced by Directive 2000/78/ECFootnote ³ – hereafter the Employment Equality Directive – applies to all ages. However, policy documents adopted by the EU on active ageing in the wake of the ageing of the population refer to workers in their fifties as a special concern.Footnote ⁴ In a labour-market context, older persons are generally understood to be those approaching their post-employment years (Boudiny, Reference Boudiny2013: 1077); and, from the standpoint of EU policy, ‘the critical period with regard to old age and employment begins in the early end of a person's working life, which mainly concerns employees from the age just above fifty years of age and older’ (Julén Votinius, Reference Julén Votinius, Manfredi and Vickers2016).

Age discrimination is often approached through the concept of ageism. This concept relates to old age, and makes discriminatory behaviour an integral part of its definition. Thus, it covers ‘negative or positive stereotypes, prejudices and/or discrimination against (or to the advantage of) elder people on the basis of their chronological age or on the basis of a perception of them as being “old” or “elderly”’ (Iversen et al., Reference Iversen, Larsen and Solem2009: 4). Schiek (Reference Schiek2011: 777) too cites age-related stereotypes that undermine the individual's dignity as an important rationale for age-discrimination bans – an individual-oriented or human-rights approach. Against this background, anti-discrimination regulation has traditionally been designed as the lodging of complaints-led individual claims in the liberal tradition. However, there is also another important rationale behind the age-discrimination ban. Apart from the individual dignity approach, the differential treatment of older workers involves more material concerns; after all, and as already indicated, age is to some extent rooted in physical materiality (Fudge and Zbyszewska, Reference Fudge, Zbyszewska, Numhauser-Henning and Rönnmar2015: 142). The associated stereotyping evolves within a certain context, reflecting the need to legitimise differential treatment by age in terms of the ‘requirements’ of real life. A collective-interest rationale of a more instrumental kind appears, at a societal level and in terms of age as a traditional social stratifier. We are talking here about economic/distributive concerns and a collective or economic approach. These two main rationales behind the EU's ban on age discrimination – the individual approach and the collective approach – have been referred to as ‘the double bind’ of age discrimination (Hendrickx, Reference Hendrickx and Hendrickx2012; Numhauser-Henning, Reference Numhauser-Henning, Numhauser-Henning and Rönnmar2015; Alon Shenker, Reference Alon Shenker and Langille2019: 279).

Against this background, the EU's ban on age discrimination has been designed in ‘a weaker format’ than its bans on other kinds of discrimination. We find the relevant regulation in the Employment Equality Directive prohibiting harassment, direct and indirect discrimination, and instructions to discriminate on grounds of age (among other grounds) in a working-life context. It also grants some general exemptions from the ban on discrimination in terms of genuine and determining occupational requirements (Article 4), legislative measures necessary for public security and public order (Article 2.5), and for the armed forces in relation to age and disability (Article 3.4). Unlike other discrimination bans, however, the Employment Equality Directive opens up – in Article 6.1 – for the justification of direct age discrimination; thus:

The scope for such justification is ultimately decided by the CJEU, in its case law.

A comprehensive bulk of case law has now been built up by the CJEU. Compulsory retirement is the issue that best reflects the important links mentioned above between labour law, employment protection, standard employment contracts and pension systems.Footnote ⁵

The continued need for collectively imposed norms regarding the role of work in a person's lifecycle, as perceived in the European setting (Suk, Reference Suk2012: 75), is reflected in two features of the Employment Equality Directive: (a) it does not cover provisions on retirement age in public pension schemes; and (b) it grants an exemption – in Article 6(2) – for ‘the fixing for occupational social security schemes of ages for admission or entitlement to retirement or invalidity benefits’ (which are well within the realm of labour law proper).

There is no doubt, however, that the question of compulsory retirement – putting an end to the employment contract at a fixed age – is covered by the Employment Equality Directive, and indeed that in principle it constitutes direct discrimination under the Directive as stated by the CJEU in, among other cases, Palacios de la Villa (CJEU, 2007). The continued acceptance of compulsory retirement has come about through an interpretation of the justification rule in Article 6.1 of the Directive. The CJEU has applied a particularly ‘loose’ proportionality test in relation to compulsory retirement (compare Schlachter, Reference Schlachter2011; Kilpatrick, Reference Kilpatrick2011; Dewhurst, Reference Dewhurst2013).

An early case was Palacios de la Villa (CJEU, 2007). It concerned a Spanish worker forced to retire at age 65, in accordance with the collective agreement in place. The CJEU found ‘promoting employment’ – through the termination of older workers at a certain age – to be a public interest, and considered ‘the legitimate claims of workers’ not to be prejudiced unduly thereby. After all, ‘the relevant legislation is not based only on a specific age, but also takes account the fact that the persons concerned are entitled to financial compensation by way of a retirement pension … the level of which cannot be regarded as unreasonable’ (judgement: CJEU, 2007: 73). The CJEU has confirmed this judgement in a number of subsequent cases such as Age Concern England (CJEU, 2009), Rosenbladt (CJEU, 2010a) and Hörnfeldt (CJEU, 2012). Among the legitimate aims cited by the CJEU so far are intergenerational fairness in terms of access to employment, the prevention of humiliating forms of employment termination, and a reasonable balance between labour-market and budgetary concerns – all of which are expressions for age as a traditional social stratifier. The CJEU has furthermore shown much consideration to the traditions of the various Member States in this area. In the Hörnfeldt case, for example, it noted that

The Hörnfeldt case also made clear that compulsory retirement at a set age of about 65 years is compatible with the requirements of the above-mentioned Employment Equality Directive, provided there is a reasonable system of pensions in place (and even if the pension in case is quite low).

It is thus clear from the CJEU's case law that differential treatment on grounds of age is often justifiable, and the practice of compulsory retirement especially so.Footnote ⁶ At the same time, inducing people to work beyond pensionable age is a core concern of active ageing policies. If people are to be persuaded to do this, however, they must have both the practical and the legal possibility of doing so. Here, of course, the acceptance of compulsory retirement is a key concern. How can this continued practice be understood?

First, compulsory retirement may be regarded as part and parcel of the standard employment contract and standard pension arrangement described earlier. Taken together, these elements form a scheme implying a long-term trade-off in wage society, as developed – at least in Europe – during the main part of the 20th century. Accepting compulsory retirement in line with the findings of the CJEU basically implies a defence of the traditional standard employment contract. The CJEU's reasoning can also be cited in support of the view that upholding the ban may lead to the humiliating dismissal of older workers in terms of presumed ‘incapacity’. To uphold the ban on age discrimination would even threaten to weaken employment protection altogether (Numhauser-Henning, Reference Numhauser-Henning2013, Reference Numhauser-Henning, Numhauser-Henning and Rönnmar2015). A stronger stress on ‘capability’ as an employment requirement would have the effect of undermining not only active ageing strategies but also employment protection well before retirement age. Thus, one result of the abolition of the statutory compulsory-retirement scheme in the UK is assumed to be an increased use of performance management, i.e. regular review, consultation and documentation for monitoring the performance of the employee, and to build up a case either for voluntary retirement or for a dismissal that will hold up against the UK's unfair-dismissal legislation (Barnard and Deakin, Reference Barnard, Deakin, Numhauser-Henning and Rönnmar2015).

Nonetheless, defending the standard employment contract can seem rather senseless, given the fact that the battle has apparently already been lost to flexibilisation. Anti-discrimination regulation as such has been seen, as already indicated, as a natural companion to flexible work and the decline of the standard employment contract. Somek (Reference Somek2011) has gone so far as to label anti-discrimination regulation in general ‘a neoliberal forerunner’. Upholding the ban on age discrimination – and thus outlawing compulsory retirement – does imply a threat per se to traditional employment protection, and so would go hand in hand with the dominant trend towards the flexibilisation of work and the weakening of employment protection and of standard employment conditions. To the extent active ageing comes about in this perspective, it is likely to be in the form of even more flexible employment. Dismissing an elderly worker from a permanent position can be expected to be a difficult process, making a fixed-term contract even more attractive from the perspective of the employer. Also, it should be said here, accepting compulsory employment seems as well to support ‘active ageing’ beyond pension age precisely in flexible contractual forms (see the cases Rosenbladt (CJEU, 2010a), Georgiev (CJEU, 2010b) and Fuchs and Köhler (CJEU, 2011)). The use of fixed-term employment is often more generous beyond ‘normal’ pensionable age, i.e. following compulsory retirement.Footnote ⁷ The conclusion is that both moves – accepting compulsory retirement and upholding the ban on age discrimination – basically enhance the flexibilisation of work. The ban thus works as a kind of ‘Trojan horse’ for flexibilisation, as employment protection grows weaker and standard employment contracts become less common.

I have argued, then, that (a) age is still an acceptable – and maybe even necessary – social stratifier in terms of compulsory retirement; that (b) flexibilisation is a dominant labour-market trend; and that (c) anti-discrimination regulation in the EU, with its weak template and its basic function of promoting further flexibilisation, fits both the traditional model and the push for increased flexibility.

However, not all older workers are among those benefiting from the traditional trade-off in relation to the standard employment contract. Flexibilisation has since long affected the conditions under which large groups of workers (older ones too) carry out their labour; and pension-scheme reforms have ‘flexibilised’ pensions, making them increasingly precarious (Strauss, Reference Strauss, Countouris and Freedland2013). Policies allowing for direct age discrimination in terms of compulsory retirement may continue to work best for certain groups, but they do not necessarily match the social and economic needs of everyone – there is no ‘one solution fits all’ alternative. For the growing number who have never had a standard employment contract, continued employment beyond pensionable age may well be necessary to compensate for reduced pension benefits. As Schiek has pointed out, moreover, it is in situations where age discrimination intersects with differential treatment on other grounds – particularly gender and ethnicity – that the CJEU's deference to national policies gives rise to the most unfair results (Schiek, Reference Schiek2011). This brings us to the theme of the next section: age and intersectionality.

The ‘old’ worker in intersectional perspective

Old age relates to but a part of life. This is why, in the lead-up to the ban on age discrimination, there were not really any social movements for older persons comparable to those for women or for certain ethnic groups. The design of the EU's ban – as an ‘age-neutral’ ban applicable to all ages – is another important explanation here. Of course, organisations based on such groups as retired people may qualify as social movements. Nonetheless, the identities that people bear prior to the onset of old age – centred, for instance, on disability or on a certain gender, religion, ethnicity or sexual orientation – may well override any specific age-related identity (Kohn, Reference Kohn2012: 325). This differentiates age from many other grounds of discrimination. It also requires a deepened understanding of the relationship between age and other grounds of discrimination, in terms of multiple and/or intersectional discrimination.

Multiple discrimination is the broader of the two concepts. It covers any situation of discriminatory behaviour on two or more grounds. Intersectionality is a narrower notion. Coined by Kimberlé Crenshaw, it refers to situations where two or more grounds of discrimination intersect in such a way as to result not merely in the sum of two or more inequalities, but in a unique and complex experience of discrimination (Crenshaw, Reference Crenshaw1989). Intersectional analysis began with a focus on the systematically ignored claims of black women, and was elaborated at first mostly in gendered terms and within feminist analysis (Fudge, Reference Fudge, Davies and Munro2013). As it has shifted, however, from viewing law as mainly instrumental to perceiving it as both ideological and institutional – central in constituting social entities/subjects – intersectionality analysis has broadened to include the examination of many such categories.Footnote ⁸ Schiek, who has criticised later sociological-discourse developments within intersectionality theory for being overly differentiated, argues that a sufficiently precise and non-exclusive definition of grounds of discrimination for the purpose of covering intersectional situations can be achieved if EU discrimination law is refocused on the three nodes of sex, race and disability (with age, together with illness, addressed under the last) (Schiek, Reference Schiek2016a, Reference Schiek2016b). She thus argues – in relation to discrimination law – for a rather more restrictive approach to intersectionality in tune with the function of discrimination law, in which the focus is on market inclusion and equal opportunities (Schiek, Reference Schiek2016b).

EU discrimination law had its start in 1957, with prohibitions against discrimination on grounds of nationality and sex. Broadening its scope in the late 1990s – to cover ethnicity, religion and other belief, disability, sexual orientation and age – it seemingly invited the development of multi-dimensional and/or intersectional approaches. Women, as the group most frequently affected by discrimination on more than one ground, were expressly mentioned in the new directives of 2000.Footnote ⁹ Notwithstanding this, EU discrimination law has been seen as mainly compartmentalising the field into discrete grounds of discrimination, while failing to address intersectional discrimination. This legal approach – reflected also in case lawFootnote ¹⁰ – is promoted by the structure of EU discrimination law, and of the Employment Equality Directive not least. In terms of their justifications and scope of application, namely, these provisions treat different grounds differently. This has the effect of inviting strategic choice in cases of multiple or intersecting grounds, since raising a case under one ground but not another may imply a stricter level of scrutiny (Schiek, Reference Schiek2011: 779, Tryfonidou, Reference Tryfonidou2017: 93).

Despite prohibiting discrimination on a range of grounds, then, EU law still includes no explicit ban on multiple (and still less intersectional) discrimination. Yet many observers see intersectional analysis as ‘particularly useful in revealing the complex issues of dignity and distribution associated with age discrimination in the labour market’ (Fudge and Zbyszewska, Reference Fudge, Zbyszewska, Numhauser-Henning and Rönnmar2015: 142).

The recent Parris case (CJEU, 2016) is the first actually dealing with intersectionality. Discrimination was alleged on the combined grounds of sexual orientation and age. Mr Parris had worked as a lecturer at Trinity College, Dublin. Upon taking early retirement in 2010, he requested that his civil partner be granted a survivor's pension upon his death. This request was denied. Mr Parris, namely, had not entered into marriage or civil partnership before reaching the age of 60, as required by the pension scheme under which he was covered. Mr Parris had been living for over 30 years in a stable relationship with his same-sex partner, and had entered into a civil partnership with him in the UK under the 2004 Civil Partnership Act. Mr Parris was then 63 years old. It was only in 2010 that a Civil Partnership Act was enacted in Ireland, entering into force on 1 January 2011. Eventually, in 2015, same-sex marriages were allowed too.

In its ruling on this case, the CJEU does not refer explicitly to intersectional discrimination, but instead uses the term ‘combined’ grounds of discrimination (Judgement: CJEU, 2016: 29(3)). Nor does Advocate General Kokott use the term intersectionality when discussing a combination of discriminatory factors, but rather ‘multiple discrimination’: ‘(–) exist where a measure does not constitute discrimination on the basis of age alone or on the basis of sexual orientation alone, but does so on the basis of a combination of both grounds for a difference of treatment’ (opinion: CJEU, 2016: 147). The CJEU ruled that the requirement at issue in the pension scheme was not capable of creating discrimination on the combined grounds of sexual orientation and age, inasmuch as ‘that rule does not constitute discrimination either on the ground of sexual orientation or on the ground of age taken in isolation’ (judgement: CJEU, 2016: 82). The CJEU had first tried the case against the ban on discrimination on grounds of sexual orientation. Since the rule in the pension scheme was neutrally worded, the situation concerned possible indirect discrimination. With reference to recital 22 of Directive 2000/78 – which declares the Directive to be without prejudice to national laws on marital status and the benefits dependent thereupon – as well as with reference to the fact that EU law did not require Ireland to provide for marriage or a form of civil partnership for same-sex partners (prior to 2011, when it actually did so), the CJEU found that the rule in question did not constitute discrimination on grounds of sexual orientation (judgement: CJEU, 2016: 51–62). Then there was the question of age discrimination, in connection with the requirement to enter into marriage or partnership before 60 years of age. Despite constituting directly age-related differential treatment, this rule was found to fall within the scope of the express exemption, in Article 6(2) of Directive 2000/78, ‘fixing an age for access to the survivor's benefit’ under the pension scheme in question.

This judgement provides, it would appear, for quite restricted future protections against intersectional discrimination, and it has been widely questioned (compare Tryfonidou, Reference Tryfonidou2017). Take the following statement by the CJEU:

This seems to give little scope for recognising – or taking measures against – intersectional discrimination. Such a conclusion seems all the more warranted in view of the CJEU's declaration that

This implies there must be a separate finding of discrimination on the single grounds at issue before combined discrimination can be said to have obtained. This meshes poorly with the definition of intersectional discrimination, which involves precisely a third ‘unique’ category of discrimination.

This case is a bit special, though, in that both discrimination on grounds of sexual orientation and discrimination on grounds of age were found not to be present, due to the explicit exemptions set out in Directive 2000/78. Establishing discrimination, however, is about proof, and here the rule on the reversed burden of proof comes into play. For a prima facie case of discrimination to be at hand, it is enough that the alleged victim ‘establish facts from which it may be presumed that there has been direct or indirect discrimination’. The respondent is thereupon obliged to prove there has been no breach of the principle of equal treatment. In cases less clear-cut than Parris, situations of intersectionality may well come under this rule. This is especially true when the law – such as the Swedish Discrimination Act (2008: 567), which implements this part of EU law – requires only that the differential treatment be ‘related to’ one or more grounds of discrimination. In the end, it is all a question of interpretation and proof.Footnote ¹¹ The third category here – the intersectional category – is created by the finding that both grounds of discrimination are (presumably) at hand! However, a real recognition of intersectional discrimination – involving the recognition of a new sub-group – seems to require an ‘open-list’ design of anti-discrimination law.

It is noteworthy in this context that Article 21 of the Charter of Fundamental Rights includes precisely such an open list of grounds of discrimination. Perhaps the CJEU will invoke this Article at some point in the future, in a somewhat more permissive interpretation of the possibility of intersectional discrimination than in Parris. So far, however, the CJEU has been reluctant to expand by way of interpretation the grounds on which discrimination is banned as demonstrated in the cases Kaltoft (CJEU, 2014) and Chacón Navas (CJEU, 2006). This was explained in Advocate General Jäskinnen's opinion in Kaltoft: while the list of grounds of discrimination in Article 21 of the Charter is open-ended, and so may include other grounds than the ones actually mentioned there, the Charter as such must not expand the EU's competences.Footnote ¹² The prerogatives of the Member States in social policy matters are to be safeguarded, and the competences bestowed on EU institutions by Article 19 of the Treaty on the Functioning of the European Union are not enough to broaden the substantive scope of binding EU law in these matters. As Schiek (Reference Schiek2016a: 43) puts it: ‘In so far, the case law has come full circle.’

The Parris case has been cited as an indication ‘that the right to equal treatment for same-sex couples remains contingent upon a prior decision of the national legislature to create a legal framework for the recognition of such relationships’ (Bell, Reference Bell2012: 141). As such, the case refers mainly to ‘another story’ than that about age discrimination. But what do we learn about the significance of age and age discrimination here? Understanding an individual's ageing experience requires an appreciation of the relationship between age and other forms of identity, in this case sexual orientation – relating the issue not only to identity but to social contextualisation as well. It is common knowledge that acceptance of homosexuals and LGBT persons (and still more their new family constellations) has come late in time. The specific history of social relations connected with sexual orientation interacts with age in ways that make old-age LGBT persons especially vulnerable.Footnote ¹³ This so-called cohort effect is well known to demographers and was also discussed by Doron and colleagues in relation to the European Court of Human Rights (2017) case Carvalho Pinto de Sousa Morais v. Portugal (Mantovani et al., Reference Mantovani, Spanier and Doron2018: 9).

In Parris, same-sex couples above a certain age were disproportionally disadvantaged by a pension scheme age requirement found to be typically proportionate, in terms of the economic interests of employers in relation to those of employees in general (Article 6.2 in the Directive). Acknowledging the age-related interests of the specific group in question is necessary in order to appreciate the specific historical embeddedness of developments relating to discrimination on grounds of sexual orientation. I do not argue, however, that such a contextualising is necessarily enough to change the law. An intra-legal analysis is bound to follow the legal dogmatics rules of interpretation of the legal sources. Notwithstanding, a better understanding of the consequences for the implicated victim may well help to inform the legal argumentation. In Parris, Advocate General Kokott had found indirect discrimination on grounds of sexual orientation to prevail, ‘as the 60 year age limit affects a large number of homosexual employees in Ireland more severely and more deleteriously than their heterosexual colleagues’ (opinion: CJEU, 2016: 57). Age discrimination was also present since the 60-year age limit was not permitted under Article 6(2) of the Directive, in her view. In another case, E.B. v Versicherungsanstalt öffentlich Bediensteter BVA, the CJEU (2018) has lent itself to a ‘bold’ interpretation related to same-sex ‘indecency’ in the light of later legal developments regarding such relationships. Whereas new bans on discrimination do not apply to legal situations that have arisen and become final under an old law, they do apply to their future effects (judgement: CJEU, 2018: 50). An earlier decision on early retirement of a civil servant for an attempted offence of same-sex ‘indecency’ committed in 1974 was still standing, whereas the Equal Treatment Directive was interpreted as requiring a recalculation of the now unlawful reduction in his pension entitlement for the future – ‘in order to calculate the amount he would have received in the absence of any discrimination on the ground of sexual orientation’ following the entering into force of said Directive.

It has been suggested that a proportionality assessment strictu sensu would be appropriate in cases of intersecting grounds of discrimination (Tryfonidou, Reference Tryfonidou2017: 93). According to Fudge and Zbyszewska (Reference Fudge, Zbyszewska, Numhauser-Henning and Rönnmar2015: 149), a proportionality test of this kind is important not only for recognising ‘individual dignity (–) just for its own sake but also … [for illuminating] and address[ing] the inherent flaw in the particular system of distribution’. Such a test points, namely, ‘to distributional asymmetries and structural arrangements that (re)produce the marginalisation of groups of workers’ (Fudge and Zbyszewska, Reference Fudge, Zbyszewska, Numhauser-Henning and Rönnmar2015: 149). As it is now, replacing intersectional analysis of these cases with parallel but separate analyses of each distinct ground of discrimination risks diluting protection against discrimination, rather than strengthening it. This is especially true when age discrimination is involved, due to the weaker design of the ban in that area.

Concluding discussion

In Europe, elder law has developed first and foremost in relation to the ageing society and its corresponding policies. Here it may suffice to recall an EU declaration from 2012: ‘The Year of Active Ageing and Solidarity Between Generations’, with its accompanying ‘Guiding Principles for Active Ageing and Solidarity Between Generations’Footnote ¹⁴ and the recurring Ageing reports.Footnote ¹⁵ It is also against this background that the Norma Research Programme took the initiative in 2011 to create a new research environment in elder law at the Law Faculty of Lund University, Sweden. The legal challenges implied by the ageing society have constituted our core area of studies in the 2012–2016 period, as reported in the anthology Elder Law. Evolving European Perspectives (Numhauser-Henning, Reference Numhauser-Henning2017a). The concept of ‘law and ageing’ – thus also familiar to elder law in a European setting more generally (Doron and Georgantzi, Reference Doron and Georgantzi2018) – may be said to reflect the perspectives contextualisation and externalisation argued here in a good way! They also come naturally to the way legal research – and a ‘social science approach’ to legal studies (Numhauser-Henning, Reference Numhauser-Henning2010; Numhauser-Henning and Rönnmar, Reference Numhauser-Henning2013) – has been carried out within the Norma Research Programme since its start in 1996, now the host of our Elder Law Research Environment.

If we accept that the ageing individual is an essentially contextualised and socially constructed concept, we must accept too that law itself – as an important institutionalisation of society – is part of the ‘ageing’ process. Law is one of the practices through which old age is acquired; it is a process of which ageing is an effect (compare Fudge, Reference Fudge, Davies and Munro2013). Understanding this process requires in-depth study and far-reaching knowledge of the particular problem area studied. I have thus been arguing here, an understanding of elder law in terms of a broader contextualisation of the ageing process, including what I have labelled the externalisation of the problem-area studied. It is about understanding the ageing process in terms of a larger story, in itself external to or only indirectly related to ageing as such. The concept of contextualisation itself can thus be understood so as to cover both a narrower contextualisation of the ageing issues proper and an externalisation – taking in the broader understanding of a particular field and its theorisation in order to ‘unwind’ its specific implications for ageing. To look at the law from the lens of the ageing individual, i.e. elder law, is in itself applying an external view on law. Talking about externalisation as a method to study ageing in a particular area, we are instead applying an external view to the process of ageing. The externalised perspective yields an argument against elder law-specific theories as the solution. Such theories and concepts must be complemented, namely, by a deepened understanding of the broader research area at stake. This makes elder law a truly multi-disciplinary field, both within and outside the legal setting.

My own studies have centred on labour law and labour-market functionality in relation to ageing workers. As it turns out, we have to include remarkably young workers (40+) in this area of elder law. We also need to consider how regulations interact with pensionable ages (65+). This underlines the risk of merging elder law with gerontology and a focus on the oldest old, health, autonomy, abuse and the like. For an ageing society, after all, relatively young workers and labour-market functionality are of utmost importance. From the perspective of traditional elder law, moreover, not-so-old clients are important too.

The intersectional perspective illustrates how a certain age constraint – in the Parris case an age limit of 60 years for entering marriage or civil partnership if certain pension rights are to be enjoyed – can take on a special meaning in relation to a certain group (such as LGBT persons), due to this group's specific historical embeddedness, by demographers referred to as the cohort effect. ‘Two cohorts have the same age but lived, live or are going to live in different cultural, economic, legal [my addition] and social worlds’ (Mantovani et al., Reference Mantovani, Spanier and Doron2018: 10).

Here – as in most cases – mere contextualisation does not do the trick, however. Legal argumentation regarding positive law is due to take place within certain methodological restraints related to the theory on legal sources and their interrelations. Notwithstanding, law is known to make room for considerable interpretation and a deep understanding of the realities certainly helps to support legal argumentation, whether de lege lata or de lege ferenda. Here, one can draw attention to the legal method called ‘rewriting judgments’ (Mantovani et al., Reference Mantovani, Spanier and Doron2018) as well as to the search for legal principles underpinning certain rights of older persons (Dabove, Reference Dabove2013, Reference Dabove2016).

Contextualisation, in the broad sense I am advocating, can also be questioned from the viewpoint of ‘politics of identity’ – in this case the recognition of older people as a distinct social group in the position to form a civil rights movement (Kohn, Reference Kohn2012). I have already argued that old age to a lesser extent than other grounds protected by bans on discrimination has made way for social (or political) movements. No doubt, emancipation or ‘voice’ go hand in hand with today's debate on special rights for older persons, preferably in terms of an international Human Rights convention. Does not contextualisation as a method per se – implying a risk of fragmentation of the indicated group – threaten to weaken the political movement behind such claims?

This article is about the ‘boundary building’ of elder law. Law-making is a way to institutionalise society in all its complexity. Defining elder law as a profoundly contextualised ageing process, rather than concerning a specific span of old age itself, makes this endeavour equally complex. The level of support to the ‘identity building’ of old people as a distinct social group may, applying such a broad contextualised perspective, vary depending on the issue studied. In my view, a deepened understanding of a societal issue – now in relation to the ageing process – always helps dealing with it. I reckon it is with old age as with women: early ‘essentialism’ may well have seemed a more straightforward way of fighting women's ‘interests’. Not because of that, have we not embraced later developments in terms of different women-groups’ interests and ever more varied views/theories on social gender. There is no ‘one distinct social group’ to defend, but a variety of interests. True, it is about emancipation, but also about understanding the system, its potentials and also shortcomings in many different fields. Over-simplification is not a recommendable way forward. This view is, in my opinion, well reflected in theory-building concerning precisely the human rights of old persons. Arguments have been made about Sen's and Nussbaum's Capability Approach as a way to not only inform the argument for a special convention on the rights of old persons but also, more specifically, age discrimination regulation (Harding, Reference Harding, Doron and Georgantzi2018; Alon Shenker, Reference Alon Shenker and Langille2019). Such an approach requires precisely a careful evaluation of the capabilities offered to ageing individuals in variable situations, including an elaborated discourse on the adequate level of such capabilities and their substantive realisation. Doron's application to ageing of Nancy Fraser's theory on social justice in terms of both redistribution and recognition can also be mentioned. Transformative remedies concerning both types of social injustice are said to ‘blur group differentiation’ (Doron, Reference Doron, Ruebner, Do and Taylor2015: 24).

In my view, elder law deserves recognition as a separate area or field of legal studies, thereby making ‘old age’ a meaningful (but far from fixed) socio-legal category to be analysed on the basis of a contextualised view of law as institutionalised within a certain social setting. To this end, we should see elder law as very much a field in process – a developing body of intellectual and political activity known as elder law (or law and ageing, if you prefer) (compare Hunter and Fletcher, Reference Hunter and Fletcher2009). Within such a field in process, which is ‘moving and changing rather than [being limited to] a fixed, marked-out territory or state of being’ (Hunter and Fletcher, Reference Hunter and Fletcher2009: 292), there is a need for pluralism. For the foreseeable period, it may perhaps suffice to say – to paraphrase what a well-known economist of the 1930s, Jacob Viner, said about economics – that ‘elder law is what elder law researchers do’ (Schwab, Reference Schwab2017: 117).