Changing the narrative: Loneliness as a social justice issue

The understandings of loneliness that dominate in western societies focus primarily on how it emerges from individual shortcomings, such as too little engagement in social activities or sub-optimal social skills. In this article, we characterize and contribute to changing this dominant narrative by arguing that it is both unhelpful and incomplete, in that it contributes to further setting those who feel lonely apart from the rest of society and encourages interventions that


| 3
bs_banner _bs_banner LONELINESS AS SOCIAL JUSTICE Ickes, 1989).The prioritization of chronic loneliness as the problem to be examined (as opposed to more transient or situational loneliness; Vasileiou et al., 2017), has contributed to this interest in relatively stable characteristics, which are more easily associated with long-term experiences than the ever-changing (social) environment.This body of research characterizes those who report loneliness as a particular type of person, one who tends to be introverted and neurotic (Buecker et al., 2020;Mund et al., 2020;Teppers et al., 2013), has a negative outlook on life and insecure attachment (Cacioppo et al., 2000), has low self-esteem (Vanhalst et al., 2013), displays depressive tendencies (Van As et al., 2022), and lacks trust in others (Qualter et al., 2013) (for a recent review, see Dahlberg et al., 2022).These characteristics, in turn, are associated with interpersonal inadequacies such as difficulties initiating interactions, low self-disclosure, and aggressive tendencies (Lasgaard et al., 2016;Lim et al., 2020;Qualter et al., 2015).Underlying these behavioral patterns is the existence of cognitive biases that lead lonely individuals to fail to make the best of the social opportunities provided to them.Specifically, lonely individuals prioritize attention to socially threatening stimuli, make negative attributions for others' behavior that assume hostile intentions, expect rejection, evaluate themselves and others negatively, avoid social risks, and have low self-efficacy regarding social interactions (Bangee et al., 2014;Spithoven et al., 2017;Vanhalst et al., 2015Vanhalst et al., , 2017)).These individual differences and behavioral patterns predict loneliness, but to make matters worse, they are also enhanced by loneliness experiences, in a cycle that perpetuates social withdrawal and poor-quality interactions (Cacioppo et al., 2006;Hawkley & Cacioppo, 2010;Mund & Neyer, 2019).Given this situation, researchers have concluded that loneliness is "all in the mind" of those who report it (Spithoven et al., 2017).
The development of the evolutionary theory of loneliness in the early 2000s further encouraged the examination of relatively stable individual-level predictors, in particular the biological underpinnings of loneliness, from the genetic to the molecular, hormonal, and neural (for an overview, see Cacioppo & Cacioppo, 2018).Particularly influential has been the suggestion that loneliness is at least in part genetically heritable (Boomsma et al., 2005;Gao et al., 2017;Matthews et al., 2016Matthews et al., , 2022;;Spithoven et al., 2019), which implies it stems from an individual's biological make-up, but also that it is fairly immutable.In this scenario, interventions might at best encourage individuals to compensate for this risky genetic disposition.
Another focus of loneliness research, especially among young people, has been on the role of interpersonal experiences, such as parental negativity and child maltreatment (de Heer et al., 2022;Qualter et al., 2013), cybervictimization (Olenik-Shemesh et al., 2012;Varghese & Pistole, 2017), lack of peer acceptance (Qualter et al., 2013(Qualter et al., , 2015)), and school victimization and bullying (Matthews et al., 2022;Vanhalst et al., 2014).Although these interpersonal experiences point to factors outside the individual, they are most often explained in this literature by reference to characteristics of the victim, such as their social competence.
A range of demographic factors have also been considered in relation to loneliness, such as age, gender, marital status, migration history, race and ethnicity, disability, socioeconomic status, employment status, and sexual orientation (e.g., Altschul et al., 2021;Beutel et al., 2021;Dahlberg et al., 2022;Qualter et al., 2021).However, insights into the mechanisms through which these demographic variables are associated with increased loneliness are scarce (e.g., Hymas et al., 2022 for autistic individuals;Elmer et al., 2022 for sexual minorities; Refaeli & Achdut, 2022 for socioeconomic status).Instead, demographic variables are most often mentioned in passing as "risk factors" that increase an individual's vulnerability to loneliness and are frequently statistically "controlled for" when examining psychological mechanisms of interest.
If loneliness is largely understood as a problem with the individual, then it is not surprising that the majority of evidence-based interventions aim to fix the lonely individual, for example, by correcting maladaptive social cognitions through psychological therapy or social skills training or by increasing opportunities for social interaction and social support (Eccles & BARRETO et al. Qualter, 2021;Masi et al., 2011).Individuals are advised by policymakers to seek these corrective measures, get out more, and join group activities (see, for example, NHS, n.d.; NIA, n.d.).As Wilkinson (2022) points out, the dominant idea is that "we must constantly work to avoid loneliness" (p.24).
When it comes to understanding the effects of loneliness, the individual level has also been the primary focus.Indeed, research on loneliness became more urgent when evidence started revealing its detrimental effects on individual health.Specifically, a vast body of evidence shows that loneliness leads to poor physical (e.g., Hawkley et al., 2022;Hawkley & Capitanio, 2015), mental (e.g., Cacioppo & Cacioppo, 2015;Wang et al., 2018), and brain health (Karelina et al., 2009), with impressive evidence pointing to how loneliness increases the risk of early mortality (Holt-Lunstad et al., 2015;Rico-Uribe et al., 2018).The magnitude and consistency of the associations between health and loneliness have led to calls for loneliness to be considered a public health priority (Cacioppo & Cacioppo, 2018;Holt-Lunstad et al., 2017; National Academies of Sciences, Engineering, and Medicine (U.S.) et al., 2020).This characterization of loneliness, together with concerns about the costs of loneliness to society (Meisters, Westra, et al., 2021), has raised the profile of loneliness and garnered the attention of policymakers (for a recent example, see Office of the Surgeon General, 2023).However, what this has most often been taken to mean is that loneliness is experienced by a lot of people, is a threat to population health, and is costly to governments and health services-rather than that loneliness is socially patterned, unequally distributed, and needs structural solutions.In addition, the association of loneliness with public health has also contributed to pathologizing and stigmatizing loneliness as reflected in metaphors of loneliness as "contagious" (Cacioppo et al., 2009), "like a bad cold" (Willey, 2009), or the "leprosy of the 21st century" (Fergusson, 2018).
All in all, the dominant narrative offers a bleak picture of the individual reporting loneliness: They are mostly portrayed as being lonely (rather than feeling so) because they behave in undesirable ways.Loneliness is most often attributed to individual factors-from genes to personality, mood, and cognitive biases-and the role of the social contexts in which these individual factors and associated loneliness emerge is significantly underappreciated.

PROBL E M S W I T H T H E DOM I NA N T NA R R AT I V E
Although the dominant narrative stems from valid research that sheds light on aspects of the loneliness experience, we claim it is significantly limited and in need of extension to highlight the role of social inequalities.Below we expand on some of the limitations of this narrative, which in a nutshell are that (1) it increases loneliness and makes people feel worse about this experience; (2) it is incomplete and does not account for important predictors of loneliness; (3) it guides us to interventions that do not produce sufficiently effective or sustainable change; and (4) it hinders broader understandings of the societal impact of loneliness.

The narrative itself increases loneliness
The dominant narrative around loneliness, outlined above, largely stems from psychology's long-standing adoption of the natural science paradigm, which is itself dominated by neoliberal and individualistic lenses that downplay the role of context (Arnett, 2008).Individualism celebrates autonomy and self-reliance.Neoliberalism is defined around increased competition, which intensifies protectionism and isolation, and a reduced role for the state, which deprioritizes collective care (Ostry et al., 2016).The foci in neoliberalism are personal responsibility, individual growth, and emotional self-regulation.Although neoliberalism emerged, as an ideology, with a focus on economic growth, it affects both discourses about loneliness and | 5 bs_banner _bs_banner LONELINESS AS SOCIAL JUSTICE policies to address it-directing attention to the individual for the identification of both causes and solutions (Wilkinson, 2022).
Scholars have previously criticized the dominance of these specific lenses in psychology as neither representing the experiences of most humans nor adequately accounting for contextual factors (Arnett, 2008;Bhatia, 2018;Gergen et al., 1996;Sugarman, 2015).Arnett (2008), for example, refers to the dominance of psychological research by American authors on American samples and explains this by reference to the superiorly financed U.S. system and its predominant "philosophy of science that emphasizes fundamental process and ignores or strips away cultural context" (p.602).This philosophy of science has enabled the rise and dominance of research in areas such as cognition, neuroscience, and behavioral genetics, which in turn perpetuate it.Bhatia (2018) additionally describes how this natural science approach contributes to the ideology of the neoliberal self, the idea that each person is a set of assets that they need to manage and grow, autonomously and by choice.From this point of view, individual struggles, like loneliness, are perceived as stemming from a failure to make the right life choices.
Interestingly, the origins of this narrative match conditions under which loneliness becomes most prevalent.Historian Fay Bound Alberti, for example, contends that loneliness became understood as it is today due to the development of modern individualism and neoliberal thinking, associated with the growth of the consumer economy, the decline of the importance of religion, and the popularity of evolutionary biology, with its focus on interindividual competition for the survival of the fittest (Alberti, 2019).The role of individualism is also stressed by Vivek Murthy (2021) who attributes loneliness to the fact that "the values that dominate modern culture … elevate the narrative of the rugged individualist and the pursuit of selfdetermination" (p.xxi).
Empirical research has also demonstrated that loneliness is enhanced in individualistic contexts (Barreto et al., 2021), although some studies find the opposite pattern-that it is in fact higher in collectivist settings (Heu et al., 2019).While evidence regarding the role of countrylevel variations in individualism is inconsistent, the causal role of neoliberalism is less ambiguous.For example, experimental research has revealed that participants who imagined living in a neoliberalist society reported more loneliness than participants who did not (Becker et al., 2021).In addition to this direct causal effect on loneliness, neoliberalism is known to increase social inequalities (Ostry et al., 2016;Piketty, 2015), which are "recast as virtuous" within this ideology, in that they are claimed to merely reflect different levels of deservingness (Mobiot, 2016).Social inequality, in turn, has been associated with more violence, less trust, less happiness, and less social cohesion (Oishi et al., 2011;Wilkinson & Pickett, 2010), none of which are grounds on which a sense of connection can flourish.That is, the ideologies that underlie the focus on individual deficits as the cause for loneliness are self-fulfilling, increasing loneliness, which in turn can be associated with social withdrawal, rendering individual-level explanations for loneliness seemingly sensible and plausible.In short, the dominant narrative feeds itself.
Neoliberalist ideologies further encourage the stigmatization of loneliness, which in turn worsens the loneliness experience.The stigma associated with loneliness has been defined as "a constellation of beliefs that derogate and devalue those who feel lonely" (Barreto et al., 2022, p. 2), which includes attributing loneliness to the individual's traits or behavior.Neoliberalism values individual's abilities to thrive and take control of their lives and encourages individuallevel attributions as a step towards achieving control.It follows that those who are not thriving-for example, because they feel lonely-can be blamed for not making the choices that would allow them to prevent or address their relational difficulties (Barreto et al., 2022;Ko et al., 2022).Consequently, those who feel lonely are often derogated by others and seen as less likely to be good friends (Kerr & Stanley, 2021;Lau & Gruen, 1992).Importantly, these stigmatizing perceptions can be held not only by others but also by lonely individuals themselves (Barreto et al., 2022;Ko et al., 2022).As a result of this stigma, those who experience loneliness BARRETO et al.
often do not disclose this feeling to others or seek help to overcome this experience, both aggravating and perpetuating the loneliness they feel.
In sum, the dominant narrative that portrays loneliness as the result of individual deficits originates from neoliberal ideologies that can both facilitate loneliness and worsen its experience, by adding the burden of social stigma to the already aversive feeling of loneliness.

Individual factors do not explain everything
Another problem with the dominant narrative is that it largely neglects the role of the social contexts in which loneliness emerges and is perpetuated.Research on the role of life transitions shows, for example, that loneliness increases after separation, job loss, or when moving away from home; this research convincingly demonstrates that life contexts can overrule individual differences to elicit loneliness among even those not so predisposed (Buecker et al., 2021).In addition, individual-level variables explain only part of the interindividual variance in loneliness.For example, regarding genes, some have claimed loneliness has 48% heritability (Boomsma et al., 2005)-which leaves at least 52% to noninherited factors-but others have reported more modest percentages (Gao et al., 2017).The relationship with loneliness seems stronger for other variables, such as personality traits (Buecker et al., 2020) and self-esteem (Geukens et al., 2022).However, although individual difference variables undeniably constitute important proximal determinants of loneliness, they are often themselves triggered by contextual factors that are its root causes, and which will continue to operate if only proximal factors are attended to through interventions.For example, while poor self-esteem and sensitivity to social threats predict loneliness, as already mentioned, these are often caused by more distal factors, such as childhood adverse experiences or bullying (Bellmore et al., 2004;Cappadocia et al., 2012;Cook et al., 2010;Hawker & Boulton, 2000;Hoglund & Hosan, 2013).One could argue, then, that interventions like psychological therapy or social skills training can only hope to treat symptoms; they do nothing to eliminate the causes of those symptoms (Akhter-Kahn & Au, 2020).
In addition, some factors considered to be individual risk factors are not associated with loneliness by themselves, but through the intervention of social processes that take place in specific contexts.This is often the case for demographic variables.For example, there is nothing inherent to sexual minorities that might cause their heightened levels of loneliness-indeed, when they feel safe, members of sexual minority groups engage in high levels of social activity (e.g., Croff et al., 2017).Instead, and as we will discuss in greater detail below, what often leads to heightened loneliness among sexual minorities (versus heterosexuals) are mechanisms of social exclusion that operate outside of the lonely individual, in between the individual and society (e.g., Elmer et al., 2022).

Individual interventions do not help much
Recent reviews (Christiansen et al., 2021;DCMS, 2023;Lasgaard et al., 2023) show that most interventions designed to reduce loneliness, listed in the gray and academic literature, fit into five categories: (1) social support via befriending or mentoring programmes, which in some cases are socially prescribed, (2) training of social skills and emotional competency, (3) social connection opportunities to increase an individual's social network, (4) psychological therapies or mindfulness training, and ( 5) increased knowledge about what loneliness is.There are many criticisms of such approaches, including the fact that befriending, provision of social support, and social connection mainly target social isolation and not loneliness (Motta, 2021).Importantly, there are currently no complex community interventions for loneliness (Lasgaard et al., 2023) or documented attempts to address systemic marginalization mechanisms that drive loneliness.Unsurprisingly, then, existing interventions have found to have only small, short-term effects (Christiansen et al., 2021;Lasgaard et al., 2023).Indeed, ignoring the distal factors and important societal drivers of loneliness in intervention strategies means the potential for reducing loneliness is significantly limited.
Although not yet implemented, pharmacological interventions are also being seriously considered to reduce the behavioral effects of social isolation, or to dampen the fear and stress responses that contribute to the emergence and perpetuation of loneliness (Cacioppo et al., 2014(Cacioppo et al., , 2015;;Cacioppo & Cacioppo, 2015;DeWall et al., 2010;Entis, 2019).Aside from also failing to address the root causes of loneliness, this approach neglects the possibility that social anxiety and hypervigilance are appropriate responses when the stressor is a significant and reasonable threat, such as intergroup hostility and discrimination.As such, one might question how ethical it is to intervene by providing treatments that can dampen such protective responses.
Although it is important to identify and mitigate the role of proximal factors in loneliness, those often reflect the operation of exclusion processes at the interpersonal and structural levels.As such, addressing only individual factors in interventions is not optimally effective, and it does not eliminate the source of the problem.In addition, many existing social campaigns and interventions, because they highlight what individuals can do to address their loneliness, exacerbate the idea that loneliness is a product of faulty individuals.Those strategies communicate that loneliness is something we can each address ourselves, but only if we want to.Such strategies fail to recognize that those who feel lonely often do not have the confidence and motivation to engage in the ways being highlighted; if those who feel lonely cannot engage in the available solutions, or if those solutions do not address their problem, they might end up feeling even more alienated.
In addition, strategies aimed at the individual do not recognize the obstacles that those who are stigmatized might face precisely when trying to engage in social activities because the spaces on offer are also spaces where further exclusion and devaluation can easily take place.Although well intentioned and potentially effective for those whose loneliness is driven by lack of knowledge about or motivation to engage with social activities, such interventions likely do little to reduce marginalization or to ensure that those who are marginalized have their needs met.

It hinders understanding of consequences of loneliness beyond the individual
Reflecting the dominant narrative, research on the consequences of loneliness has also most often focused on its impact on the individual.However, research has begun to show that loneliness has broader societal impacts.For example, loneliness experienced at work has a high cost to employers, partly by reducing commitment to the employer, increasing work absence, and reducing company productivity (Co-op and the New Economics Foundation, 2017; Ozcelik & Barsade, 2018).Loneliness is also costly to society, more generally.For example, loneliness reported at one time increases the probability of being unemployed later by 17.5% (Morrish et al., 2022).Loneliness has also been documented to reduce economic productivity and to increase use of health services (Gerst-Emerson & Jayawardhana, 2015;Meisters, Westra, et al., 2021;Mihalopoulos et al., 2020).
In addition, loneliness has been associated with reduced political participation.Specifically, those who report loneliness are less likely to vote, in part because they feel less of a sense of duty towards society (Langenkamp, 2021b).Besides being less likely to vote, individuals who feel lonely are also less inclined to sign a petition or contact a politician (Langenkamp, 2021a).In addition, loneliness might be associated with greater support for right-wing (but not left-wing) populist parties (Langenkamp & Bienstman, 2022).As such, loneliness impairs normative BARRETO et al.
political participation and encourages polarization, which might be seen to threaten democracy, disenfranchise individuals, and perpetuate marginalization (see also Arendt, 1973).
Crucially, research has yet to examine whether the relationship between loneliness and these costs to society is not in fact another vicious cycle, in that loneliness might both cause these problems and be enhanced by them.Indeed, by focusing so heavily on individual deficits and solutions, dominant understandings of loneliness draw attention away from these collective disenfranchisement and marginalization processes, which, as a result, remain poorly understood.However, failing to understand those mechanisms limits appreciation of the role of loneliness in social cohesion and impairs our ability to persuade policymakers and appropriate institutions of the importance of this problem and the need to address it.

C H A NGI NG T H E NA R R AT I V E: LON E L I N E S S A S A SOC I A L J UST IC E I S SU E
Increasing evidence suggests that factors beyond the individual play a role in the emergence and perpetuation of loneliness.To start with, loneliness is geographically patterned, which is hard to explain without a consideration of context (e.g., Fokkema et al., 2012).In the United Kingdom, for example, the highest rates of loneliness were reported in areas of the country with the highest concentration of young people and in areas with higher rates of unemployment (Office of Statistics Regulation, 2021).
Making note of the importance of social and structural factors in loneliness, scholars from various disciplines have been contributing to changing the narrative.Across psychology, many have called for the consideration of more social factors when analyzing loneliness (e.g., de Jong Gierveld & Tesch-Römer, 2012;Yang, 2019).Feminist critiques of existing approaches have pointed out the role of power differentials and oppression and advocated for the need to "reframe loneliness as a structural condition, rather than as a personal failure" (Wilkinson, 2022, p. 24; see also Magnet & Orr, 2022).Likewise, Stauffer (2015) highlighted how loneliness in members of persecuted groups (which she labeled "ethical loneliness") stemmed from neglect and "the experience of not being heard" by those with the power to help them (p.1).The view of loneliness we wish to outline here is consistent with and complements these perspectives.Our analysis extends and consolidates prior calls for the consideration of the social determinants of loneliness by specifically focusing on the existence of social inequalities in loneliness as well as the social disadvantage it stems from and which it perpetuates.

What is social justice and why loneliness qualifies
Social justice is often defined by reference to the idea of equality of opportunity in access to the means that allow one to make effective use of individual freedoms (Rawls, 2003).A just society is "a fair and equal society in which each individual matters, their rights are recognized and protected" (Park & Allaby, 2013, p. 397).A just society enables equal participation of all citizens and groups in its activities, including in the shaping of the society itself (Bell, 2007).Some go further to advocate that true social justice is not just having equality of access to means to satisfy one's needs, but to actually having one's needs satisfied in an equal way (Faden & Powers, 2008).In addition, while many authors include only basic needs in this definition, others have argued for the importance of considering multiple aspects of well-being, including social needs (Faden & Powers, 2008).Consistent with this view, scholars have advocated for individuals to have the right not to be lonely (Lederman, 2023;Meisters, Putrik, et al., 2021) with some defending loneliness as a moral injustice (Brownlee, 2020) and a breach of human rights (Brownlee, 2013).Social justice is | 9 bs_banner _bs_banner LONELINESS AS SOCIAL JUSTICE not merely a philosophical or theoretical principle operating at a high level.A vast body of research has demonstrated that individuals care about justice and that perceiving the systems they inhabit as unjust has a variety of consequences for their well-being and ability to contribute to society (Tyler, 2001).Understanding loneliness as an unequally distributed burden, heavily caused by an unjust allocation of resources, can help us understand how systems and structures need to change to address this problem at large scale and in a sustainable fashion.To provide empirical ground for the notion that loneliness needs to be considered a social justice issue and addressed through structural change, in what follows we review evidence for the existence of social inequalities in loneliness and the role of social exclusion and marginalization in these.

Loneliness inequalities
Although many have stated that loneliness does not discriminate in order to convey the idea that it can affect anyone (DCMS, 2018), we have previously summarized evidence that revealed that higher rates of loneliness are reported by those who belong to social groups that are marginalized by society (Barreto et al., 2023).For example, overall, more loneliness is reported by migrants than by natives of the host society (Buecker et al., 2021;Lim et al., 2020).Crucially, migrants report more loneliness than members of the same cultural group who remained in the native country (Polo & López, 2009).Within the migrant group, migrants report less loneliness when they have been longer in the host country (Wu & Penning, 2015), were born there (Madsen, Trab Damsgaard, et al., 2016), or immigrated at a younger age (Albert, 2021), all factors that increase similarity with the host society.Racial and ethnic minorities consistently report more loneliness than members of groups that are not minoritized on the basis of their race or ethnicity (Franssen et al., 2020;Lasgaard et al., 2016).Individuals who identify as LGBTQ+ report considerably higher levels of loneliness than cisgender heterosexual individuals (Fish & Weis, 2019;Gorczynski & Fasoli, 2022).Although little work explores the social experiences of transgender and gender diverse individuals, the evidence suggests that they undergo significant relational difficulties (Lewis et al., 2023) and that loneliness is particularly prevalent in this group (Anderssen et al., 2020).Disabled individuals, including those who are autistic (Hymas et al., 2022), visually impaired (Sorokowska et al., 2022), deaf (Shukla et al., 2020), or have chronic illnesses (Dahlberg et al., 2022), all report more loneliness than those who are not disabled.Finally, higher levels of loneliness are reported by those with low socioeconomic status, measured objectively (Buecker et al., 2021) or subjectively (Qualter et al., 2021), by children whose parents have lower social status (Madsen et al., 2019), and by those who are unemployed (Morrish & Medina-Lara, 2021).

The role of exclusion
The very fact that loneliness is more frequently experienced by individuals who belong to social groups that are marginalized in society already suggests that social exclusion, operating at multiple levels, is likely to be an important mechanism underlying the loneliness experience (see Figure 1; see Barreto et al., 2023 for a more detailed overview).Evidence directly supports this idea.

Interpersonal exclusion
At the interpersonal level, loneliness can directly emerge from exclusion from play groups in childhood, peer groups in adolescence, or coworkers in adulthood-indeed, one way in which people express loneliness is by stating they feel left out (Russell, 1996).Although not all members of minoritized social groups are excluded (or feel lonely), this exclusion, in the form of bullying, victimization, or plain ostracism, more frequently targets members of minoritized social groups (Bouldin et al., 2021;Killen & Rutland, 2022), providing a convincing explanation for why they tend to feel more lonely.
Prejudice and discrimination are often involved in interpersonal exclusion and reported experiences with discrimination predict loneliness among sexual minorities (Doyle & Molix, 2014), migrants (Priest et al., 2014), international students (Zhao et al., in prep), ethnic (Visser & El Fakiri, 2016) and racial minorities (Juang & Alvarez, 2010).Some studies show that experiences with discrimination at one time are associated with increased loneliness at a later time, but not the other way around (Jackson et al., 2019;Priest et al., 2017).Moreover, experiences with discrimination explain loneliness inequalities between members of dominant and marginalized groups (Doyle et al., in prep;Doyle & Molix, 2016;Visser & El Fakiri, 2016).Stigmatization has also been associated with increased loneliness when stigma is experienced by association, for example, among parents of autistic children (Jiao et al., in prep-a, Jiao et al., in prep-b).
Although interpersonal exclusion can cause loneliness fairly directly, it often does so as a distal factor, through its effect on health, well-being, and poverty.These, in turn, function as proximal factors that more directly affect loneliness, but which should not be understood as its primary cause.For example, experiences with prejudice and discrimination often reduce selfesteem and lead to mental health difficulties, which in turn are associated with relational difficulties and loneliness (Doyle & Barreto, 2023;Doyle & Molix, 2014).In addition, loneliness disparities between minoritized and majority groups are at least partially explained by group differences in psychological well-being (Visser & El Fakiri, 2016) and in financial comfort (Fokkema & Naderi, 2013;Visser & El Fakiri, 2016).The effects of physical health, psychological well-being, and poverty, in turn, can reduce the opportunity or the motivation to socialize and increase relational strain (for example, via increased vigilance to social threats or rejection sensitivity; see Barreto et al., 2023, for a more detailed account of these mechanisms).
In sum, interpersonal mechanisms exclude individuals from valued and valuable social interactions, eliciting loneliness.This most often emerges among members of minoritized social groups and through the operation of stigmatization processes.Importantly, these create psycho-social states that are strikingly similar to those most commonly described in the dominant narrative as individual traits, such as low self-esteem, social anxiety, or depression.In this way, a consideration of the context of loneliness highlights where these individual differences might originate and raises the possibility that the relative stability in what is often considered an individual difference trait might in fact reflect the stability in contexts of disadvantage.

Cultural hegemony
Marginalization does not always operate in obvious ways.One subtle way through which it functions is through what Gramsci designated as "cultural hegemony" (Forgacs, 1988).The term "cultural hegemony" refers to how dominant groups not only hold power over the distribution of resources, but in fact determine what values, norms, and behaviors are perceived as appropriate in a given society.We have previously explained how this applies to social interactions too (Barreto et al., 2023).In a nutshell, norms regarding social interactions are set by the dominant group and reflect how it prefers to operate.As a result, these norms exclude other ways of being social, which might be valued by minoritized groups, leaving the latter with the option to either assimilate or be left out (Antrobus et al., 2014).This is often the case with cultural minorities, who frequently attribute their loneliness to clashing cultural norms and misunderstandings (Byrne et al., 2015;Koehn et al., 2022;Sawir et al., 2008).Another example is that of autistic individuals, whose social needs and behaviors tend to differ from those of neurotypical individuals (Sedgewick et al., 2022).Other examples are those with visual impairments (Sorokowska et al., 2022) or those using sign language (Batten et al., 2014).Assimilating to cultural norms, or masking difference, is not only hard and sometimes impossible, but it can also result in feeling devalued and behaving in awkward ways that do not increase acceptance (Newheiser & Barreto, 2014).Importantly, when migrants interact with others from the same cultural background, or autistic individuals with another autistic individual, or even a neurotypical one they know well, interactions can be highly satisfying for both parties (Kim & Bottema-Beutel, 2019), suggesting the problem is a property of interactions between minoritized and dominant group members, rather than the characteristics of these individuals.This also highlights the importance of not romanticizing communities, as is often done when speaking about loneliness, but acknowledging that they are always dominated by particular views of who they are, what they are for, and how things must be done (Joseph, 2002).

Exclusive communities
Consistent with tenets of socioecological models in psychology (Bronfenbrenner, 1979;Oishi, 2014), individual experiences are inevitably shaped by the communities in which one is located.Community can refer to a geographical place (e.g., neighborhood, classroom) as well as a relational network (e.g., interest group, voluntary organization), or a combination of these (Obst et al., 2002).Geographical neighborhoods tend to play an important role in individuals' experiences of community (Chavis & Pretty, 1999).This is influenced to a large extent by life stage, with emerging adults often deriving sense of community from institutions such as schools and universities (Slaten et al., 2016).Importantly, communities do not automatically enable and encourage belonging for all people.For example, within the context of universities, studies have reported lower levels of belonging for Black compared to White Americans (Walton & Cohen, 2011), sexual minorities compared to heterosexuals (Wilson & Liss, 2022), and those with disabilities compared to those without (Barnes et al., 2021).Notably, some research has also established that disparities in self-reported sense of community belonging between marginalized and dominant groups are in fact due to true differences between groups rather than measurement issues, such as differential item functioning on measures of community belonging (Coffman & BeLue, 2009;Fisher et al., 2020).
In an intriguing example, one recent study (Trawalter et al., 2021) demonstrated that students from lower SES backgrounds at elite universities in the United States tended to use public spaces on campus-particularly iconic public spaces, such as campus monuments-less often than their higher SES peers.The authors argued that this disparity in use of public space was due to lower SES students perceiving that these spaces were not "for them."In turn, lower SES students reported lower levels of belonging to the university compared to higher SES students, and this gap could be explained by differences in use of public space.This dovetails with research demonstrating that the lack of social group identification with a community hinders belonging and increases loneliness (Cruwys et al., 2022;McNamara et al., 2021).Importantly, much past research has demonstrated the powerful role of belonging in shaping experiences of loneliness.Belonging is a fundamental human motivation which, when thwarted, can lead to substantial distress (Baumeister & Leary, 1995).Indeed, unmet belongingness needs have been shown to predict greater loneliness (Mellor et al., 2008), including for deficits in neighborhood belonging specifically (Marquez et al., 2022).Undeniably, communities bear a responsibility for ensuring that all members located therein feel included and can forge a collective identity.
Why then might communities be experienced so differently by members of different social groups?From an evolutionary perspective, social stigma has historically functioned as a protective mechanism, keeping communities safe by marking out those who might carry disease or be likely to betray the group (Kurzban & Leary, 2001).That is, stigma has been used to push people out to protect the larger collective from harm (Phelan et al., 2008).Although such signals may have served an important protective function for communities in our ancestral past, they most often lead to unjust social exclusion in modern societies, which should be capable of caring for the ill and rehabilitating those who might engage in acts of betrayal.Supporting this notion of the function of stigma, evidence confirms that experiences of prejudice and discrimination do result in reduced feelings of belonging within communities for member of marginalized groups (Froehlich et al., 2023).In turn, reduced belonging mediates an increased risk of loneliness resulting from prejudice and discrimination (Lattanner & Hatzenbuehler, 2023;Liu et al., 2014).However, as with interventions aimed at reducing gaps in loneliness, interventions aimed at improving disparities in belonging (e.g., Walton & Cohen, 2011) have been critiqued as too individualized-potentially pathologizing members of marginalized groups by locating the source of the problem within their own psychology rather than the structural environment (Johnson, 2022).Thus, disparities in belonging, as with disparities in loneliness, must be understood as products of the communities and the resulting structures in which people are embedded.

Discriminatory policies
One key way in which communities fail to address disparities in belonging, or even exacerbate interpersonal stigma and exclusion, is through public policy.Structural stigma has been defined as "societal-level conditions, cultural norms, and institutional policies that constrain the opportunities, resources, and wellbeing of the stigmatized" (Hatzenbuehler & Link, 2014 , p. 2).Public policy and institutional practice are amenable to change, but they are also likely to be strategically wielded by those in power to enable maintenance of social hierarchies within communities, keeping those from marginalized groups either down or out.For example, segregation and redlining, policies aimed at excluding members of ethnic minority groups from obtaining housing in specific neighborhoods, have been shown to be deleterious for health and life chances above and beyond effects of interpersonal experiences of prejudice and discrimination (Gee, 2008;Williams & Collins, 2001).Even policies that are not explicitly about keeping people out of communities can signal exclusion to members of marginalized groups.One study conducted in Australia during the samesex marriage plebiscite in 2017 found that sexual minorities living in communities with a greater proportion of people voting against same-sex marriage reported greater levels of loneliness (Perales & Todd, 2018).More generally, further research among sexual minorities has confirmed that on average, greater structural stigma at neighborhood (Lattanner & Hatzenbuehler, 2023), state (Doyle & Molix, 2015b), and national levels (Doyle et al., in prep) are each associated with increased loneliness.Because structural stigma manifests at the level of institutions and cultures, this happens in a covert way that absolves particular individuals of personal responsibility for the loneliness that they ultimately create within their own communities, unlike most forms of interpersonal exclusion.
Public policy can also add to the burden of loneliness for members of marginalized groups by failing to adequately manage diversity in a way that ensures that members of all social groups feel included and represented within a community.Absolute diversity in a community (i.e., the likelihood that any two people randomly chosen from a given community or organization will belong to the same social group) could be an indicator of a more tolerant social environment, but it does not automatically lead to feelings of representation and inclusion for members of all marginalized groups (Purdie Greenaway & Turetsky, 2020).In general, ethnic minority adults tend to experience less loneliness in neighborhoods in which they are in closer proximity to others of the same ethnic background (i.e., the "ethnic density effect"; Bécares et al., 2009;Tseng et al., 2021;Visser & El Fakiri, 2016).Ethnic minority youth also tend to report lower levels of loneliness in schools in which more peers in their classroom share their ethnic identity (Madsen, Damsgaard, et al., 2016).However, greater diversity is also not inherently linked to greater loneliness for members of marginalized groups, and communities can implement policies and practices to manage diversity in supportive ways.Diversity policies within organizations that prioritize support for members of marginalized groups and intolerance of prejudice and discrimination have been shown to improve social relationship outcomes at work (Ciftci et al., 2020).Conversely, poor or ineffective diversity policies within organizations can reduce belonging among members of marginalized groups, thereby increasing loneliness at work (Wright & Silard, 2021).While communities may create loneliness by pushing people out intentionally, they may also inadvertently promote loneliness through deprivation and lack of infrastructure to support socialization, particularly for members of marginalized groups.

Deprived neighborhoods
Structural factors in deprived neighborhoods can severely limit opportunities for socialization and connection between community members.Indeed, evidence is accumulating that communities with fewer green spaces (Astell-Burt et al., 2022), lower home ownership rates (Morris & Verdasco, 2021), greater residential density (Lai et al., 2021), higher crime rates (Portacolone et al., 2018), and less accessible transportation (e.g., Hille & Gather, 2022) tend to have higher rates of loneliness among residents (see also van den Berg et al., 2016, for an analysis of the role of neighborhood characteristics in loneliness).Building on this research, scholars have proposed the term "lonelygenic environments" to refer to complex structural factors (e.g., lack of green space, poor public transportation infrastructure) that cause or exacerbate loneliness within communities (Feng & Astell-Burt, 2022).Moreover, they highlighted that lonelygenic environments are socially patterned, with greater likelihood in deprived communities and areas shaped by structural stigma.
A growing body of research, primarily from human geography and urban design, points towards the important role of "third places," that is, places other than work and home in which members of a community can socialize, in community well-being (Oldenburg & Brissett, 1982).Notably, third places may be particularly protective against loneliness in deprived neighborhoods that suffer from other structural disadvantages (Hickman, 2013).Living in a deprived neighborhood is especially likely to increase loneliness for members of marginalized groups.For example, there are independent as well as interactive effects of indicators of personal and neighborhood SES on loneliness, with the worst outcomes for the most socioeconomically disadvantaged in the most deprived neighborhoods (Algren et al., 2020;Wu et al., 2022).Some authors have argued that objective neighborhood features do not impact loneliness, with neighborhood effects being primarily driven by subjective appraisals, which are themselves shaped by characteristics such as mental health and preexisting loneliness (Matthews et al., 2019;Wen et al., 2006).However, this research does not consider that neighborhoods are experienced differently by different people or, in particular, examine the experiences of those who are marginalized.Moreover, it is not consistent with the growing evidence base to state that structural factors only matter for loneliness insofar as they are perceived as negative by the individual.Real change can reduce loneliness among members of marginalized groups if communities invest in infrastructure that supports social health and implements policy that protects against stigma and discrimination.

I M PL ICAT IONS OF T H E C H A NGE I N NA R R AT I V E
The dominant narrative around loneliness can be classified as a type of "psychopolitics" (Han, 2017;Yılmaz, 2021), the process through which neoliberalist systems exert power by isolating individuals from each other, ensuring that blame for failure is individualized and social change is undermined.Blaming individuals for their loneliness gets governments off the hook to enact true and sustainable change.In counterpoint to this, our view of loneliness as a social justice issue encourages us to examine the role of power and identity in loneliness (Hammack, 2017), in so far as social inequalities are closely linked to group identities and power differences.This approach aligns with feminist theories that shed light on how knowledge production is often a result of "othering" and highlights the importance of situational and contextual factors (Harding, 2004;Hartsock, 2004).Although psychologists have been often criticized for feeding such reductionist, pathologizing, and stigmatizing views of phenomena like loneliness, it is precisely those of us who are alert to the social and political contexts where individuals operate who are best placed to shed light on how these different domains come together to influence social connection and loneliness and therefore to provide a more complete understanding of how loneliness is experienced and can be addressed.
It is important to acknowledge that not all members of minoritized social groups feel lonely, as well as that loneliness is not exclusive to these social groups.Our argument, instead, is that mechanisms of marginalization increase the likelihood that members of minoritized groups feel lonely.In addition, often multiple marginalizing mechanisms come together to further increase these odds.For example, those identifying as a sexual minority are more likely to feel lonely when they have personal experience with harassment and live in states or countries where sexual minorities are not supported by egalitarian policies (Doyle & Molix, 2015a).
Our desire to change the narrative around loneliness stems from the knowledge that narratives have consequences, shaping experiences, as well as what is done to address them.This change in the narrative around loneliness has the potential to reduce the stigma associated with loneliness, which is likely to remove some of its burden, but it is primarily important because it points to avenues for intervention that might reduce loneliness more effectively and for more people at a time.This article highlights that these solutions need to be rooted in the understanding that loneliness is unequally distributed in the population in the same way as many other resources are unequally distributed in the population-this, in turn, neatly describes loneliness as a social justice issue that converges with other sources and outcomes of marginalization.As such, solutions to loneliness need to address structural sources of marginalization and in doing so they will address other problems as well.Studies demonstrating the role of welfare states in loneliness (Nyqvist et al., 2019), for example, point the way towards structural change in the provision of welfare services that can have multiple benefits by reducing social inequalities across a range of outcomes, including loneliness.Research demonstrating the importance of well-maintained and safe leisure areas in residential neighborhoods (Hong et al., 2018), green spaces (Coley et al., 1997), third spaces (Hickman, 2013), and good transport links (Rachele et al., 2017) also directly point to concrete steps that can be taken to distribute social connection more fairly.Evidence for the role of nondiscriminatory and inclusive policies that celebrate diversity highlights the benefits these can have for social health in environments as varied as schools, workplaces, and broader societies (Doyle et al., in prep;Jefferson et al., 2023;Wright & Silard, 2021).More generally, structural solutions acknowledge systemic causes, which reduces stigmatization, and validate individual experiences of devaluation, which can in itself increase felt understanding and decrease feelings of disconnection, as well as one's confidence that episodes of interpersonal exclusion will be addressed.Finally, structural solutions can benefit those who feel the most stigma associated with loneliness, and who as such would be unlikely to seek other forms of support.
Interestingly, it is possible that approaching loneliness as rooted in structural inequalities is in fact cheaper than the more dominant individual approaches.In fact, while neoliberal approaches are believed to be cheaper, due to the limited role of the state, when it comes to loneliness it might be useful to measure the cost of interventions against the cost of loneliness to society, evidence of which is only beginning to emerge.More widely, scholars have increasingly argued that loneliness results from economic, political, and social systems, structures, and decisions around what is prioritized and funded in any given society (Taylor, 2020;Wilkinson, 2022).Political decisions around what is funded and what is cut affect people's ability to connect and address loneliness: Libraries, youth centers, public transport, and meeting spaces in residential areas are all crucial in facilitating meeting and belonging.Lack of street safety, poor accessibility of public spaces, cuts in housing benefits and social care budgets, are all examples of how unequally distributed the impacts of these cuts/decisions can be.As Taylor (2020) argued, "the hypocrisy of governments that talk about loneliness while systematically destroying key sources of social connectedness is breathtaking".

CONC LUSION
Loneliness is unequally distributed in the population, being higher among marginalized groups and linked to mechanisms of marginalization at the interpersonal and structural levels.This contradicts the dominant narrative that prioritizes attention to individual factors and interventions that fix faulty individuals but does not account for systematic variation between dominant and marginalized social groups.In addition to considering individual sources of variation in loneliness, and based on an analysis of existing evidence, we advocate for considering loneliness as a social justice issue that needs addressing by tackling its structural causes.

F
I G U R E 1 How distal and proximal determinants of loneliness contribute to loneliness inequalities.14679221, 0, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/pops.12965 by University Of Exeter, Wiley Online Library on [15/03/2024].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License