Social constructionism: social epistemology and social ontology

Put at its most general, the thesis behind Berger and Luckmann's book is the ‘constructed’ nature of our social worlds and the social – inter-subjective – production of its ‘facticity’. Our inter-subjective understandings of the world become internalised in the form of the life-world (a Schutzian term, derived from the philosopher, Edmund Husserl) in which we and subsequent generations are immersed from birth (Seidman, Reference Seidman1997: 129). As a framework for a social epistemology of the social world, social constructionism can at times seem to treat everything as if it were all a ‘social construction’ with the consequence of dissolving the term into little more than ‘an empty signifier’ (Lynch, Reference Lynch2001) whose initial value has become exhausted, a sociological shibboleth of our times.

At the heart of social constructionist thinking, however, are important issues of social ontology and social epistemology – concerning what social things are and how they should be understood. It draws attention to the tension that exists between what John Searle (Reference Searle1995) once called the ‘brute facts’ about the world as external reality contrasted with the ‘institutional facts’ making up society. These latter, Searle argued, are continually being created through and by human action, though not necessarily through individual human intent and often as the unintended consequences of social, political and economic change directed towards other ends.Footnote ² Institutional facts, according to Searle, serve as ‘placeholders’ for possible ongoing activities, determining what can be done and how it can be done but not what is. Brute facts, however they may be interpreted by society, and however they may be represented, remain constant in their facticity (Searle, Reference Searle1995: 57). Hacking (Reference Hacking1999) makes a similar distinction between what he calls ‘human kinds’ and what he calls ‘natural kinds’. While the latter are unaffected by their becoming objects of study, institutional or human kinds are, leading him to describe them as ‘interactive kinds’ since their form is contingent upon human actions interpretations and representations (Hacking, Reference Hacking1999).

Further distinctions have been made between the social construction of ideas (theories, beliefs, representations) and the social construction of entities in the world. While few would dispute that the former (theories, beliefs, representations) are socially constructed, the status of the latter (socially constructed entities or objects) is both more problematic – and more radical. As Searle noted, social objects may depend upon a material existence – but those brute facts may serve only a place-holding role in structuring social relations, so that it is their meaning rather than their materiality that determines their status as social objects. Social philosophers such as Diaz-Leon, Haslanger, Mallon and Marquez have considered two distinct ways in which such socially constructed entities may exist, entities that are socially constitutive and entities that are socially caused (Haslanger, Reference Haslanger1995; Mallon, Reference Mallon2007; Marquez, Reference Marquez2017; Diaz-Leon, Reference Diaz-Leon and Garavaso2018). In the first case, socially constitutive phenomena are entirely dependent upon some aspect of the social. They are constituted by and within the social world. Without a social world, they could not exist. In the second case, something may be causally linked to or implicated by social factors playing a significant part in its emergence, even if its place is held, in part or as a whole, by its status as a ‘brute fact’ existing within the world however it may be socially organised.

Socially constituted phenomena can have no independent existence outside the social; thus buildings, artistic and literary products, planes, trains and automobiles are material entities, but what they are – as books, churches, forms of transport, etc. – only makes sense – has meaning – in the context of human society. The same cannot be said of the latter. These phenomena (socially causative entities) can be considered to have materially real and materially significant properties, even if (a) they have socially important significance and (b) that social change can alter both their symbolic significance and their material properties. Examples that spring to mind might include human diseases, domestic animals, systems of irrigation and so forth. Their existence, their being, constitutes a brute fact that can be studied and understood without reference to their social signification. Even so, this does not mean that they cannot be more thoroughly studied and understood through other means, including through the social sciences and through the arts and humanities. Osteoarthritis can be understood whether in a human being or in non-human animals; the same goes for the shape, size, constitution and development of cows, pigs and sheep. That does not prevent studying the effects of social interventions (or already-existing social conditions) on the manifestation of the disease or the nutritional properties of their meat. Equally, their material facticity does not pre-determiner the symbolic significance they may acquire in a society.

Such ontological and epistemological complexities make social constructionism a potentially controversial theme, particularly in relation to various body-based (corporeal) but socially significant phenomena such as age, disability or gender. While it might be the case that height or weight constitute ‘biological realities’, whose variation may well be affected by social practices, habits and ‘conditions of living’, such socially constituted causal variation may have in itself little or no social significance. Other equally physical features, though no more or even less biologically salient, on the other hand, may have considerable social significance across a wide range of settings. While the existence of a biological reality may be necessary for such features to exist, their biological existence may be more a place holder than a substantive social fact when it comes to determining their part in the social world, whether they are strongly or weakly subject to social causes and influences. Although issues of sex and gender are currently among the more controversial topics when determining their biological or social constitution, the social construction of age can be said to be equally controversial, and whether it constitutes merely ‘a brute fact’, a socially constitutive entity or a socially causal phenomenon.

The social construction of age: developments within social gerontology

As Lynott and Lynott (Reference Lynott and Lynott1996) observed, the social construction of age first emerged as a theme within social gerontology during the late 1970s and early 1980s. At that time, it was particularly associated with the political economy approach, adopted by such academics as Carol Estes, John Myles and Laura Olson in the United States of America, by Chris Phillipson, Peter Townsend and Alan Walker in the United Kingdom, and Martin Kohli, Karl Mayer and Walter Müller in Germany. These writers promulgated the notion that old age ‘is a social rather than a biologically constructed status’ determined by the institutions of the state and the structures of welfare (Phillipson, Reference Phillipson1991: 404; Phillipson and Thomson, Reference Phillipson, Thomson and Bland1996). Other less structurally framed trends supported the questioning of the ‘brute facticity’ of ageing, notably those who, like Jaber Gubrium, have adopted a more interpretive approach to the sociology of ageing (Gubrium and Wallace, Reference Gubrium and Wallace1990). Within this latter tradition, the emphasis has been less on the social construction of old age and ageing as socially structured phenomena but on the social constructed ideas, interpretations and understandings of ageing and old age – what Hazan (Reference Hazan1996: 91) has referred to as ‘the construction of social knowledge and its contextual conditioning’. This issue of the social representation of ageing and old age will be explored in a subsequent section. For now, it is sufficient to note that social constructionism, in all its forms, has been critical in shaping developments both in social gerontology and in the sociology of ageing. Whether adopting what the Lynotts have called a social phenomenological position, or a more structural account of the institutionalisation of old age the social constructivists have chosen to ‘bracket’, or set aside, belief in the brute facticity of age and age-related concepts ‘and focus upon the process or processes by which they are socially constructed’ (Lynott and Lynott, Reference Lynott and Lynott1996: 753). This, the Lynotts claimed, has ‘liberated’ social gerontology and the sociology of ageing from the historical dominance of biological and medical approaches to the study of age and ageing.

Insofar as a theoretical basis can be discerned in such empirical work,Footnote ³ several sociologists of ageing have proceeded along the lines of a social constructionism that takes for granted that ‘social structure [is responsible for] forming, constraining, directing, and empowering individual development’ (Kohli and Meyer, Reference Kohli and Meyer1986: 147–148). Working under the general remit of a ‘political economy’ of ageing, they have begun with the general premise that old age is a socially marginalised position or status, realised less by biology or chronology than through enforced retirement, limited benefits, poorly resourced forms of support and an underlying constraining structure embedded within the legacy of the poor laws (Townsend, Reference Townsend1981; Olson, Reference Olson1982). Pensioner or retired households are characterised as marginal to the central dynamic of post-war economies, maintained through a constrained transfer income and deprived of the means to earn more. Limited alike as producers and consumers, older people are seen as having little scope to contribute to the development of the economy. Old age, in these authors' views, is determined less by the biological status of their bodies than by their marginal economic status. Old age, in short, was constituted by, and within, the social and economic structures of the time.

Towards the latter decades of the 20th century, the economic status of retired households ceased to be confined by such predetermined positions of economic and social marginality. With a steady rise in disposable income, many of those moving into later life began occupying a more central position in society and in the economy. The marginality that had been attributed to later life was not abandoned, however, but slowly became reframed around its cultural and social representation. Age, it was said, became a sign marked by its cultural invisibility, reflecting the more widespread political turn towards issues of recognition over those of redistribution (Fraser, Reference Fraser1995). Identities rather than classes acquired prominence, and the lack of recognition given to older, retired people was seen as continuing to marginalise old age, despite the changing economic circumstances of such households. A parallel could be drawn in relation to gender inequality, much as Naomi Woolf argued that women remained constrained by the ‘beauty myth’ despite their improved economic position (Woolf, Reference Woolf1990).

With this shift came a changing focus in accounts of the social construction of age. How age was represented – its social representation (to use Moscovici's term) – became key, not how financially constrained older people were. Studies focused upon the representations of age in policy documents, in social media and technology, in film and television, in advertising and in literature (Coupland, Reference Coupland2000, Reference Coupland2007; Walz, Reference Walz2002; Zhang et al., Reference Zhang, Harwood, Williams, Ylänne-McEwen, Wadleigh and Thimm2006; Williams et al., Reference Williams, Ylänne and Wadleigh2007; Levy et al., Reference Levy, Chung, Bedford and Navrazhina2014; Brooks Bouson, Reference Brooks Bouson2016; Elmersjö, Reference Elmersjö2020; Burema, Reference Burema2022; Rasi, Reference Rasi2022). The concept of ageism became an increasingly popular area of research in social gerontology, sociology and psychology (Levy and Macdonald, Reference Levy and Macdonald2016; Nelson, Reference Nelson2016). The work of Becca Levy in particular has proved seminal in providing wide-ranging empirical evidence illustrating how the ‘aged gaze’, whether arising from the psychological within or the social without, ‘ages’ those subjected to it (Levy et al., Reference Levy, Slade, Kunkel and Kasl2002, Reference Levy, Slade, Chang, Kannoth and Wang2020; Levy, Reference Levy2009). In short, the social construction of age and agedness has evolved from being framed as a position or status that is structured by the political economy to one that extends into the symbolic domain. This has seen age viewed more as a culturally and socially represented status, positioned variously within binary divisions, between the ‘young’ and the ‘old’, or the ‘old’ versus the ‘not old’, or as ‘boomers’ versus ‘generation Xers’ and so forth. But whether framed largely in generational terms or as age groups, age has become an actively contested identity and in Bourdieu's terms a site of ‘symbolic struggle’ over both its meaning and its representation (Bourdieu, Reference Bourdieu1992: 242–243).

Ageing: realisation and representation

Before reaching any overall evaluation of the social constructionist approach, it is helpful to consider just how hegemonic this approach has come to be within the fields of social gerontology and the sociology of ageing. Arguing for the critical importance of ‘the sociological construction of ageing’, Powell and Hendricks (Reference Powell and Hendricks2009: 85) see the social construction of ageing as ‘an important theoretical alloy’ highlighting the point that ‘the meaning of ageing derives not from innate biological processes but is socially determined’. Whether this is understood in ‘social structural’ or in ‘social representational’ terms, the point these authors stress is that thus conceived, ‘ageing has no existence independent of social interaction and power relationships in society’ (Powell and Hendricks, Reference Powell and Hendricks2009: 85). As alluded to in the earlier part of the paper, what kinds of things that are subsumed by social constructionism encompass a range of phenomena, from entities that make sense only as socially constitutive, to entities whose realisation is socially causal, to entities that are primarily socially representational – matters not of external structure but of meaning and understanding. While the strongly structural approach outlined by Powell and Hendricks emphasises that the very existence of ageing is the product of ‘interaction and power relationships’, the more phenomenological tradition associated with Berger and Luckmann focuses upon the representational aspects of age and ageing. Viewed in this light, one must acknowledge that age and ageing have long been viewed as conceptually problematic – as things thought about as much as materialised entities.

Thus Galen, writing in the third century CE, confessed his own uncertainty as to whether ageing should be considered a ‘natural’ kind or not (Theoharides, Reference Theoharides1971).Footnote ⁴ Ageing, framed as a term for a process, rather than a status like old age, does not seem to have existed, in English, until the late 18th century (Yallop, Reference Yallop2016: 2). Moreover, at least until quite recently, most people in the world had only the loosest notion of their own chronological age, in large part because there was very little cause to use it (Roebuck, Reference Roebuck1979). Brute fact or not, the place-holding function of the term was generally quite limited. Chronological age was, as Christine Fry (Reference Fry, Settersten and Hendricks2018: 275) put it, ‘a remarkable cultural innovation’. For long periods of time, age and ageing impinged little upon society. It was only as the outcome of modern social policies, designed to calculate society's human capital, by counting, enumerating and segmenting all its population that chronological age and chronological agedness began to play a significant role in the social organisation of the state (Roebuck, Reference Roebuck1979; Gilleard, Reference Gilleard2002). Even in some of the early national population censuses, age was either omitted altogether, or used as a grouping measure, such as ‘those 16 years and over’ or as decade-based age groups, before eventually settling on recording everybody's precise chronological age (A'Hearn et al., Reference A'Hearn, Baten and Crayen2009: 794; Whitby, Reference Whitby2020).

In short, while the social representational approach to the ‘social constructionist’ model of ageing can be understood within a socio-historical framework, thus supporting the view that ageing is a rather ‘modern’ social fact (in the sense of its ‘collective’ (Durkheim, Reference Durkheim1898) or social (Moscovici, Reference Moscovici1988) representation), its existence as an observable fact is not to be undermined by insisting upon what Beck (Reference Beck, Brown and Held2011: 233) has called a ‘nothing-but-society sociology’ such as that espoused by some in social gerontology. Human understandings of old age and agedness are, as Hazan (Reference Hazan1996: 91) notes, ‘contextually conditioned’. For well over a thousand years, explanations have been proposed for how ageing happens, most of which have centred upon intrinsic processes of ‘withering’ through the loss of living matter's ‘essential heat’ and its ‘radical moisture’ rather than the mere passage of time (Gilleard, Reference Gilleard2015). Exactly what those processes are, how they should be understood and how if at all they might be altered remains a problem, but not for the sociology of ageing. On the other hand, how age should be represented, what meanings it confers, how it should be morally valued and how far it can, or should be, deployed as an identity, class or category are important matters both of symbolic contestation and shifting sensibilities.

These representational problems have become more pervasive, both as the space that later life occupies in society continues to expand and as the shift taking place towards a more cultural turn in the social sciences (Gilleard and Higgs, Reference Gilleard, Higgs, Twigg and Martin2015). Considered as a site of symbolic struggle, there is a contrast between those adopting a ‘lifecourse’ perspective emphasising that all lives are lived through time and that the past helps shape the future, and what may be termed a ‘life-stages’ approach which considers that there are definable periods or stages in life, each differentially defined by growth, maturity and ageing. Within the former camp, it is often claimed that from the moment of our arrival in the world we are ageing. In the absence of anything more substantive than that all lives are lived in and through time, the further delineation into distinct divisions within the lifecourse is considered to be structured mainly as representational – constituted entirely by the social and cultural discourses and practices of the time.Footnote ⁵ Shorn of the artifices of culture science and society, such divisions as ‘old age’ are thus arbitrary and institutionally contingent constructions: age is fundamentally socially constitutive. This seems an increasingly hegemonic position, both within the humanities and the social sciences, as leading writers such as Margaret Gullette seek to eliminate the very idea of ageing as inherently ‘ageist’ (Gullette, Reference Gullette2018, Reference Gullette2022).

For those still attached to a ‘stages of life’ representation, age continues to be granted an existence outside society and its institutions, albeit one that is socially overlaid and instrumentalised by society.Footnote ⁶ Within this approach, the question then becomes one of teasing out whether these socially representational divisions affect age's brute facticity. Do they constitute more than a struggle conducted within the symbolic order of culture and society, beyond mere social representations, affecting old age's actualisation? If the latter position is adopted, are age divisions like old age examples of entities of an ‘interactive’ kind, to use Hacking's term: potentially socially causal but not constitutive. For those espousing this perspective, the rate, style and extent to which the biological processes of ageing manifest themselves cannot be considered in isolation, i.e. bracketed out from the social milieux within which they are shaped and ordered.Footnote ⁷ As two American social gerontologists have aptly put it, not only does the social get under the skin – it also does work there (Ferraro and Shippee, Reference Ferraro and Shippee2009).

In this more representational framing, social constructionism makes two rather different claims regarding old age. For those who take what I have called the lifecycle approach, ageing can be viewed as little more than a time-dependent variable, whose social representations, moral valuings and cultural depictions appear in a variety of symbolic forms, whose differentiation into life stages ‘has been an historically emergent property of modern societies’ (Settersten and Mayer, Reference Settersten and Mayer1997: 248). Until the modern era, ageing as a form of representational ordering has taken place largely under the radar, outside what Habermas calls the system world (Baxter, Reference Baxter1987). While various biomedical theories have been articulated to define and explain ageing, such pre-modern cultural representations have scarcely touched the organisation of society or indeed the framing of individuals' life-worlds.

The continuing dominance of these cultural representations of ageing and old age have effectively masked the formative influences of social structure that, with the onset of modernity, began to shape how persons are aged. By drawing attention to those structures and their impact on all that we attribute to ‘ageing’ and ‘old age’, the advocates of social constructivism proposed a new direction leading scholars to rethink how they – we – are aged as much by our society as by our biology. As a result of this development, we have begun to better understand how ‘old age is a social rather than a biologically structured status’ (Phillipson and Thomson, Reference Phillipson, Thomson and Bland1996: 14). This shift – as noted by the Lynotts – has proved of seminal influence in how the social sciences have examined age and ageing, as ‘facts’ achieved as much through the social organisation of lives in time as by ageing and old age's cultural representations. While the more extreme form of social constructionism treats ageing as having ‘no existence’ outside social interaction and power, a social fact masquerading as a natural fact, the constructionism derived from Berger and Luckmann's social phenomenology emphasises the social importance of ageing's social representation and its capacity for social causation. The hegemonic position that social constructionism has taken, in both social gerontology and the sociology of ageing, risks concealing these different perspectives.

Old age and its facticity: costs and benefits of a social constructivist approach

While not ignoring the very real value in opening up less biologically deterministic accounts of the lifecourse and its directedness, social constructionist understandings of ageing contrast significantly with the biological framing of ageing as a process or set of interdependent physiological and molecular processes exponentially increasing the risk of death. This view of the biological reality of age and ageing is itself open to criticism, locating ageing as a material process whose existence, form and process can be studied and understood independently of the socially mediated institutions, processes or relations attached to it. The recently emerged ‘geroscience’ offers an even starker contrast, insisting upon the biological reality of ageing as a sui generis process ‘controlled by molecular and physiological fundamental processes, such as macromolecular damage, metabolism, proteostasis, cellular senescence, chronic inflammation, epigenetic factors, and stem-cell regeneration’ (Moffitt, Reference Moffitt2020: 1). Within this paradigm, both ‘ageing’ and all age-associated diseases that render us mortal are equally judged the products of the ‘inner’ brute fact of ageing, irrespective of the social context in which those processes arise (Gems, Reference Gems2022).

In the same issue of the review journal where Moffitt's paper appeared, Elaine Crimmins (Reference Crimmins2020) attempted to establish a degree of balance between the social and bio-medical influences on ageing. Drawing on data from the large US Health and Retirement Study (HRS), she argued that when biological and social hallmarks are allowed to compete on an equal footing with each other (in the statistical sense of each controlling for the influence of the other): ‘Social variables (SES [socio-economic status], childhood health and hardship and adult trauma, psychological, and behaviors) explain 1.8 times the variance in mortality when compared to the … biological variables’ (Crimmins, Reference Crimmins2020: 7).

Do such findings support the claim that age and ageing are as much the products of social as they are of biological processes? That ageing is neither pure brute fact nor socially constituted entity? If the phenomenological concept of bracketing out phenomena is to be considered akin to the statistical partialling out of variance from other variables, might the findings reported by Crimmins imply that the social influences upon ageing are every bit as explanatory and hence at least equally meaningful in shaping the realisation of ageing as biological indicators? And so, for the purposes of social gerontology and the sociology of later life, can the biological be effectively bracketed out, and the social representational and structural organisational aspects of ageing and old age studied in their own right and as matters of equal importance?

The problem with making such an assumption is that human ageing seems a more directly observable outcome – a visible entity – than a relational quality dependent upon its positioning as a part of the socially constructed lifecourse. Looked at as a purely individual phenomenon, a quality of a person rather than a characteristic of a group, ageing certainly appears to be more a material than a social phenomenon; it privileges a person's body – the corporeal – as its ‘real’ site and the source of its realisation, irrespective of the machinations of social life and ‘the power relations in society’ (Powell and Hendricks, Reference Powell and Hendricks2009: 85). Research that seeks to represent the biological and social reality of age can be understood as addressing different aspects of the externality of ageing and old age, each helping to frame what ageing is. In this sense, both are equally and inescapably representations realised within the context of particular historical contingences. This contingency may be evidenced, in the case of social studies, in the shift from the early modern concerns over economic marginality to the late modern concern with cultural invisibility. It can also be observed, in the case of biological studies, in the shift from the early modern concern with distinguishing pathological from normal ageing to the dissolution of such distinctions in the geroscience of late modernity.

Does this mean that ageing and old age are at bottom inescapably contingent, necessarily social constructions, and best studied as social entities? If treated purely as objects of study – epistemologically if not ontologically – one could argue they are. This does not mean, however, that ageing and old age can be reduced to mere symbols, social kinds constituted through the institutions, language and social practices of a particular time and place. This surely was Husserl's point when he sought to distinguish between the reality of the life-world and that of science: the structuring by science of experience is agentic, collective and deliberate – and must be learned. The structuring of our life-worlds, on the other hand, is incidental, operating in part through our acquired habitus and in part through experiential learning. The social construction of ageing helps focus the study of ageing as a set of structured discourses and social practices, but nevertheless inescapably associated with the underlying brute facticity of ageing. It does not deny a place for ageing framed by the biological sciences but recognises that this constitutes an epistemological rather than an ontological choice. In short, the value of social constructionism lies not in any assumed dominance, but in enabling the study of ageing and old age to be pursued as social facts, important in themselves, bracketing out without thereby denying the reality of the corporeal.

For those who feel themselves ageing, no such structuring, whether biological or sociological, may quite capture their experience. If one grants, as de Beauvoir and Sartre have suggested, the basic unrealisability of ageing as a distinct subjectivity, a ‘for itself’ sense of me-ness, then ageing understood will remain a somewhat formless, illusive experience, rendered meaningful, at times, by recourse to one or the other of the competing symbolic frameworks operating within society at a particular time. While age may seem a structuring principle, evident in all living matter and in most human societies, it is by no means clear how firmly it serves as a structuring principle in shaping our lived experience. Arguably it is this ambiguity of experience, not between the biological and the social, but between the subjective and the objective, that remains unresolved, whatever version of social constructionism is applied in the social study of ageing and old age. This, more than anything, seems to ensure that the ongoing symbolic contestations over this particular social space will continue.