Introduction

Heroes dominate public discourse at the times of crisis. For example, the global COVID-19 pandemic saw the emergence of heavily laden heroic narratives around ‘frontline key workers’ and especially nurses (e.g., Kinsella & Sumner, 2022; Mohammed et al., 2021; Pennella & Ragonese, 2020). The recent spike of interest in heroes goes beyond that. It stems from ongoing appeal to heroes and heroism by politicians (Brown, 2007; Browning & Haigh, 2022; Peabody & Jenkins, 2017), educators (Jones, 2007; Pickett et al., 2010; Power & Smith, 2017), media producers and TV screen writers (Korte & Falkenhayner, 2021), and sport and entertainment personalities (Lines, 2001; Parry, 2009). In short, an enduring ‘allure of heroes’ penetrates the multiple facets of public life. Exploring the appeal of heroism, this paper develops an innovative approach to heroism by shifting the focus from heroic narratives and reputations to situating the public preferences for popular heroes within the context of social change, and, specifically, diversity, and inclusion politics.

Recognising that heroes fulfil the vital ‘cognitive and emotional needs, such as our need for wisdom, meaning, hope, inspiration, and personal growth’ (Allison & Goethals, 2016: 188; Allison et al., 2017; Franco et al., 2018) and accepting that heroes reflect an individual desire for self-change and self-development (e.g. Kinsella et al., 2015), this paper posits that heroes can also function as powerful drivers, and harbingers of socio-political innovations. This approach goes beyond recognising heroes as complex socio-political constructs that ‘deeply imbued with political power’ (Kitchen & Mathers, 2019: 12; see also Cubitt & Warren, 2000; Power & Smith, 2017; Peabody & Jenkins, 2017) which can also channel and foster particular cultural attitudes and values (Jones, 2007; Korte, 2021). Building on research on social change (Inglehart & Flanagan, 1987; Norris & Inglehart, 2011; Inglehart & Welzel, 2005), the proposed approach allows exploring gender- and race/ethnicity-specific dynamics as effecting hero-choices and revealing the overlooked drivers of societies’ openness to inclusion of marginalised groups.

The analysis of heroes in the context of social change offers an exciting and under-used opportunity to examine the levels of acceptance of traditionally disadvantaged and under-represented groups such as women and ethnic minorities and of egalitarianism more broadly. Indeed, it would be logical to assume that choice of heroes is indicative of people’s societal and political aspirations and that a person declaring Benito Mussolini as a hero and a person choosing Pope Francis as such would be quite dissimilar in terms of their socio-political values. Choosing Malcolm X or Rosa Parks as a personal hero can potentially be indicative of a greater acceptance of the contribution of ethnic minorities to public life, while considering Hilary Clinton or Margaret Thatcher inspirational may point at greater acceptance of women in politics.

The emergence of post-materialist values that emphasises tolerance and individual identity over traditional socio-political constructs (i.e., attachment to a particular nation-state) should, in theory, broaden a variability of hero-choices through weaking the pattern of same sex and sex-race hero-choices, particularly among younger people due to their socialisation into the more economically, technologically and educationally sophisticated — post-materialist — environment of Western democracies. However, our comparative analyses of hero-choices in the USA and Britain challenge this expectation of post-materialist theory, highlighting the diverse effects of individual identities such as gender and ethnicity on society’s readiness to embrace popular heroes from diverse backgrounds.

The paper proceeds as follows. The literature review outlines the framework for identifying heroes as harbingers of social change and explores the effects of gender and race on popular heroism. It then outlines the research methodology and discusses the hero-landscape in the USA and UK. Using rigorous exploratory regression analysis, we then evaluate how specific socio-demographic factors influence preferences for women- and ethnic minority hero-figures. Finally, we discuss the implications of exploring popular heroism through the prism of social change and signpost the ways of advancing egalitarianism and inclusion of marginalised groups through the medium of heroes and heroines.

Heroism, Social Change, and Gender/Race Lens

Drawing on findings from a range of disciplines, including political psychology, political science, political and cultural history, sociology of heroism, and media studies (e.g. Franco et al., 2018; Peabody & Jenkins, 2017; Frisk, 2018; Korte & Falkenhayner, 2021), we explore popular heroism is a socio-political phenomenon which channels identifications with important socio-political values and communities. Specifically, this paper explores heroes and heroism as a landscape which is reflective of engagement with social change and, particularly, values of inclusion and diversity.

There are many strands of theorising of the individual, cultural, social, and political importance of heroes and heroism. Despite a huge variation in approaches to heroes and heroism, our conceptual approach draws and expands on two interdisciplinary debates. On the one hand, existing scholarship within heroism studies, political psychology, and political socialisation literature is dominated by discussion of the individual preferences and ‘needs’ for heroes and inspiring personalities. The foundational premise of this literature as well as the main finding is that heroes reflect a deeply personal need for structuring individual behaviour and self-development based on the life of an inspirational hero-figure (e.g. Allison & Goethals, 2011, 7–11; Allison & Goethals, 2016). For instance, Kinsella et al. (2015) distinguishes between three functions of heroes, including self-enhancing, moral modelling and protecting function of heroes. Importantly, although this set of literature recognises that heroes as behavioural anchors and role-modelling structures help to overcome obstacles associated, for instance, with gender-prejudices and other experiences of hardship and discrimination. This approach places much more emphasis on heroes as inspirational life-goals which signpost the ‘final destination’ of individual development. Conversely, political scientists, political historians, sociologists and scholars of culture and media communication treat heroes as both vehicles and manifestations of ‘cultural attitudes and practices’, emotionally charged social and political constructions’ (Cubitt & Warren, 2000; Jones, 2007; Frisk, 2018; Kitchen & Mathers, 2019; Korte, 2021, 2–3) as well as the instruments of ‘political socialisation, institutional trust, and political communication’ (Peabody & Jenkins, 2017, 4). Within this diverse body of literature, heroes emerge as evocative rhetorical devices, complex and ever-evolving socio-cultural and political phenomena, routinely utilised by politicians, cultural elites, educators, and popular opinion-makers. This set of literature generates many important findings on the socio-political functions of heroism in the context of modern societies, yet, surprisingly, most research in this field prioritises the cultural and political texts over changes in the public perceptions of modern-day heroes. Consequently, existing analysis of social, cultural, and political functions of heroism prioritises the analysis of exceptional personalities designed by researchers and elites as heroes while overlooking the drivers which determine the public preferences for heroes and heroines (see discussion in Peabody & Jenkins, 2017; Danilova & Kolpinskaya, 2020).

Building on this rich interdisciplinary debate and filling existing gap in the heroism literature, this paper situates the phenomenon of popular heroism within the context of social change. It engages with Inglehart’s and others’ analysis as a key framework to theorise the socio-political functionality of heroism and, problematise the effect of gender and race-specific dynamics. Inglehart’s theory posits that ‘the rise of post-industrial values… declining respect for authority and growing emphasis on participation and self-expression’ gives rise ‘to a new type of humanistic society that promotes human freedom and autonomy on numerous fronts’ (Norris & Inglehart, 2011: 43), as well as ‘gender equality – along with tolerance of outgroups, such as people of other races, foreigners, and homosexuals’ (2011: 273; see also Inglehart & Abramson, 1999: 668; Inglehart & Flanagan, 1987). Notably, this scholarship does not offer a linear explanation of social change and acknowledges that this process displays a substantial variation across modern societies. For example, recently, many modern democracies have experienced the rounds of ‘cultural backlash’, gender re/de-alignment, and neo-traditionalism which have occurred alongside the insurgence of right-wing political parties and conservative social movements which, in turn, tend to follow the growth of libertarianism and liberalism (Norris & Inglehart, 2011; Inglehart & Norris, 2016; Tilley, 2005). Combining this premise of the disrupted, staggered social change with the recognition that public attitudes to heroes from marginalised backgrounds can be informative of associations with heroic journeys of those who have faced and overcame the experience of discrimination and injustice, this paper explores gender and race dynamics in public preferences for heroes as one of the main arenas for the study of deeply divisive contentious politics.

Gender and Popular Heroism

Although most scholars of heroism identify gender as an important factor of individual and collective preferences for women-heroes and heroes from LGBTQ + communities (e.g., Allison & Goethals, 2011; Rankin & Eagly, 2008; Kinsella et al., 2017; Kitchen, 2019), there has been a lack of academic discussion on the diverse aspects of gender-specific dynamics. This gap has been further complicated by the overall lack of progress in envisioning the ways of moving beyond gender-equal (in)visibility politics while responding to the wave of anti-feminist backlash and a shift towards neoconservatism.

The analysis of gender divides within political psychology, political representation, and political socialisation literatures mostly utilises a binary understanding of sex-difference and prioritises the analysis of the role-model effect of female/gender-non-conforming heroines on women. There is an overwhelming consensus within this diverse body of literature that women continue to be represented and related to as ‘invisible’, ‘forgotten’, ‘marginalised’, and ‘hidden’ or ‘transparent’ hero-figures (Allison et al., 2017; Kinsella et al., 2017; Estrada et al., 2015; Power & Smith, 2017). This finding feeds discussion of the (in)visibility politics and urges for the greater investment in public recognition of women’s contribution in society, citing multiple factors, including positive role-model effect of women-heroes on improving educational, professional attainment of most women along with their political engagement (e.g., Beaman et al., 2012; Wolbrecht & Campbell, 2007).

Although the campaign for the greater visibility of a few publicly recognised heroines reiterates the necessity of the gender diversification of the hero-landscape, this scholarly debate prioritise the presuming positive, same-sex, role-modelling effect and pays little attention to the manifestations and, specially, problematic implications of this trend For instance, recent research has revealed that same-sex choice of women-heroes develops alongside privatisation of heroines (Danilova & Kolpinskaya, 2020; see also Rankin & Eagly, 2008; Estrada et al., 2015). This evidence of women choosing heroes from a cycle of female friends, mothers, aunts, and daughters hints at the reproduction of a public/private divide, a central question within the feminist analysis of patriarchy. This finding suggests that gender parity in heroes cannot be dealt with through the ‘quick fixes’ in gender-equal visibility politics, due to the compound effect of a systematic under- and misrepresentation of successful women’s careers and life choices in mass media and popular culture (Lines, 2001; Kian et al., 2013; Korte & Falkenhayner, 2021), gender-equality in education and pedagogical practices, high-cost of women’s public careers due to remaining barriers in employment and childcare, and an unavoidable gap between publicly honoured heroines, and the majority of women due to differences in class, education and experience (Durose et al., 2011; Kanthak & Woon, 2015).

The novelty of our analysis lies in the approach which does not only treats the same sex-choice of women-heroes subject to the empirical testing, we situate this choice within discussion of social change. Reframing gender-dynamics in heroes through the lens of social change theory suggests that a move towards egalitarianism, and wider inclusion of women in public spaces can only be successful if preferences for in-group (women)heroes and role model effects lessen (Norris & Inglehart, 2011: 18), as society at large becomes more open to ‘out-group’ heroes and role models who not like us. Indeed, for men having such heroes as Florence Nightingale or Ada Lovelace would be indicative of society’s awareness and acceptance of these women’s achievements and their challenging life/career trajectories, reminiscent of stereotypes and injustices faced by contemporary women in their attempts to secure their place in politics and public life.

Race and Heroism

Surprisingly, race and ethnicityFootnote 1 have been one of the most under-researched aspects within heroism literature. Most empirical studies on American heroism prioritise generational, individual and gender differences over racial divisions (Allison & Goethals, 2011; Kinsella et al., 2015; Senkbeil, 2016; Peabody & Jenkins, 2017; Allison et al., 2017; Franco et al., 2018). Equally, analysis of racial hierarchies is marginalised in the studies of heroic reputations of British colonial and imperial heroes, military heroes, heroes of polar exploration, mountaineering, and everyday heroes (e.g. Cubitt & Warren, 2000; Dawson, 1994; Jones et al., 2014; Sebe, 2015) as well as research on contemporary British sport heroes and heroines, heroes of television, and military heroes (Lines, 2001; Parry, 2009; Kian et al., 2013; Korte & Falkenhayner, 2021; Browning & Haigh, 2022).

Like with gender effects, existing empirical research suggests same race/ethnicity preference for heroes among British and US children and young adults (Anderson & Cavallaro, 2002; Power & Smith, 2017; Wineburg & Monte-Sano, 2008). The same race choice may relate to racial polarisation existing within the wider context (Hutchings & Valentino, 2004), or to a systematic under-representation of ethnic minority heroes in history textbooks, popular culture, and education curricula (Grewal, 2003; Hutchins, 2011; Jones, 2007). This coincides with the white hero-celebrities emerging as bearers of cosmopolitan values (see discussion of Angelina Jolie in Kitchen and Mathers (2019) and Matt Damon in del Mar Azcona-Montoliu (2019)), while non-white heroes are depicted as ambivalent hero-figures and/or hero-villains (van Deburg, 2013; Noel & Joel, 2017; Blaine & Brenchley, 2017). This racially polarised representation of heroes builds on the association of ‘whiteness’ with moral goodness and ‘blackness’ with moral ambivalence. For instance, the image of a ‘dark’ hero-villain from so-called ‘ghetto action film cycle’ films and ‘hood’ films of the 1990–2000s, which made connections between ethnic minorities and socially deviant behaviour thereby ‘connecting race to the culture of violence’ (Watkins, 1998: 8; Park, 1996: 497).

Beyond the silver screen, the political activism of non-white communities in the USA and Britain generated a polarised representation of the non-white hero. On the one hand, campaigners for equal rights, including leaders of the 1960–1980s urban riots and the Black Power Movement, through their willingness to respond to violence with violence, challenged social norms and values of a predominantly white society that ‘curbed and compromised the black demand for freedom’ (Schuman et al., 1997: 34). On the other hand, such prominent equal rights campaigners as Nelson Mandela, Martin Luther King Jr, and Rosa Parks rejected violence as a means of working against racial discrimination, which transformed them into global human rights heroes.

Like previous studies on gender and popular heroism, the literature on racial dynamics of hero-choice suggests a compound effect of same race choice and role model effect. Both trends contradict the expectation of post-materialist theory that egalitarianism and tolerance (in this case manifested through having non-white heroes) flourish with the emergence of post-materialist values in a society.

Methodology

In this paper, we move beyond traditional single-case studies of hero-figures and offer a systematic comparative analysis of popular heroism in the USA and Britain. This approach allows challenging a contextually blind approach which dominates scholarly debates in heroism studies (e.g., Allison & Goethals, 2011; Franco et al., 2018). The assumption of comparability of Britain and the USA is determined by three ‘similar’ factors. First, these societies are similar in terms of having comparable levels of economic development, modernisation, and values and belief structures (Norris & Inglehart, 2011; Inglehart & Welzel, 2005). Second, they have political systems, both structured around two dominant political parties, and facing a resurgence of populism and antifeminist backlash, thereby to a different degree, stronger in the USA and less manifestly in the UK (Gest et al., 2018; Wilson, 2017). Finally, Inglehart and many others point towards the similarity of English-speaking value-structure and foreign policy trajectories (Dumbrell, 2009), including the commitment of both countries to the values of liberal humanitarianism.

Data

Our analyses use original datasets representative of public opinion of British adults (n = 1683, excluding Northern Ireland) and American adults (n = 1000) conducted by YouGov in 2015.

This study makes the following contributions to the existing literature.

First, it develops an innovative conceptual framework by situating popular hero-choices within debates on social change and inclusion of marginalised groups. Second, this study, for the first time, examines hero choice in Britain and the USA using comparative analysis of nationally representative survey data. Previous studies of heroes and heroism draw on data from unrepresentative samples (e.g., Allison & Goethals, 2011; Peabody & Jenkins, 2017). The substantial proportion of empirical studies focuses on hero-figures chosen by children and adolescents, overlooking adults, or, for instance, elderly groups of the population (e.g., Estrada et al., 2015; Gibbons & Stiles, 2004). Most research on the effect of sex-based difference is based on small group purposive and/or convenience samples, often of students, and makes very limited efforts to account for the effect of a wider socio-political context which again substantially limits the generalisability of the findings (e.g., Campbell & Wolbrecht, 2006; Kinsella et al., 2017; Rankin & Eagly, 2008). Finally, most studies guide participants by pre-selecting types of heroes or their character traits or introducing restrictions on the ‘appropriate’ types of heroes by excluding parents and personal heroes from the analysis (e.g., Power & Smith, 2017). Moving away from such prompting and the resulting bias, we use a free-text, type-in question as a means of gauging a range of popular associations with heroes.

The core dependent variable for the analysis is based on the unprompted question, ‘Who is your biggest personal hero?’. However, the US respondents had an opportunity to choose more than one hero, whereas the UK respondents were choosing only one hero.Footnote 2 Answers to this question were manually re-coded by the authors to feed into the following variables in both sets of data:

  • Categorical variables classifying heroes into professional/relational groups,

  • Dichotomous variables recording the sex of respondents’ heroes,

  • Dichotomous variables recording the ethnic background of respondents’ heroes.

Method

Considering that this study uses hero-choice as an indicator of the levels of egalitarianism, which a novel approach, we do not formulate formal statistical hypotheses. We do, however, expect that the trends identified in the previous research, particularly in terms of same sex and same race choices will stand.

We start from descriptive analysis of the hero-landscapes in Britain and the USA, where we reflect on the distributions of heroes from different professional and relational backgrounds in both countries.

For further analysis, we use sex and ethnicity of heroes as independent variables in multivariate logistic regression models. Predicted probabilities facilitate interpretation of the results and improve visualisation on the outputs.

Although the hero-related variables for the UK and US datasets are the same, there are differences in terms of independent and control variables. In both sets of analyses, we use gender and age as the main independent variables. In the US case, we also model the effects of respondents’ ethnicity, which is unavailable for the UK data. We also speculate about the effect of age on hero-choices, drawing on the concepts of generational replacement (Abramson & Inglehart, 1992), life cycle (Cornelis et al., 2009) and period effects (Wolbrecht & Campbell, 2007). Although our data do not allow detailed age-period-cohort analysis, we discuss the direction of age-linked patterns.

Finally, we control for region of residence and party affiliation to account for regional differences and the effects of political predispositions (Johnston et al., 2019). Additionally, we use several measures of social class, including college education and income for the USA, and social grade and income for the UK (see discussion in the Appendix). Although social grade and education are not interchangeable characteristics, a high degree of association between the two, as well as the fact that social grade is not determined by income, allows increasing the explanatory potential of the UK set by including social grade (in the absence of education). However, this choice also urges caution when drawing comparison between the two countries.

Regression coefficients reflect the direction and statistical significance of the relationship, whereas we use predicted probabilities to illustrate the magnitude of the effects. Considering small sample sizes, we report statistical significance at both 95% and 90% confidence levels.

Analysis: Hero-landscape in Britain and the USA

An implicit assumption that the facets of modern heroism are similar across both countries find partial support in our data (Table 1). First, in both countries personal, ‘everyday’ heroes occupy the most popular position among all hero-categories, accounting for one in three choices of heroes in British sample and 41% of the US ones. This corresponds with findings from previous research (Allison & Goethals, 2011; Peabody & Jenkins, 2017) confirming that despite a different support for heroism as national self-identification, both societies share a great deal of respect for individualistic values (Inglehart & Welzel, 2005).

Table 1 Hero types and examples (three most popular choices of hero figures by type and country)
Full size table

The striking difference between the two countries emerges in support for religious figures and celebrities as heroes. The British respondents chose hero-celebrities, actors, and TV presenters as the second popular category of hero, with only 1.2% choosing religious figures, thereby demonstrating the affinity of British society with secular heroic values. In contrast, religious figures were the second most popular type of hero in the USA, reaching almost 7% of all reported heroes. This supports the view of the USA as a deviant-outlier case, exhibiting ‘more traditional and religious values than other rich societies’ (Inglehart & Welzel, 2005: 47; Braunstein, 2021). Moreover, British respondents tend to choose living religious leaders (e.g., bishops, the Pope), compared to the US respondents’ affinity with Biblical figures — God, Virgin Mary, and Jesus Christ.

Secondly, in both countries, politicians are popular as heroes, with former US presidents (e.g., Ronald Reagan, Abraham Lincoln, and Barak Obama) and British Prime Ministers (e.g., Winston Churchill and Margaret Thatcher) dominating this category for the US and the British data, respectively. These preferences reflect the country-specific political and historical context such as the prominent place of Abraham Lincoln in the American public life and popular memory as ‘a prophet of American civil religion’ (Schwarz & Schuman, 2005: 186; see also Hutchins, 2011). Moreover, recent research highlighted the use of heroic narratives by Barak Obama, which is congruent with Obama’s legacy for racial politics (Tesler, 2016). In Britain, prominence of Winston Churchill and Margaret Thatcher in the education curriculum, and cultural discourse can explain their standing as popular political heroes. Churchill also emerges as the most popular British hero, potentially, because of his twofold heroic reputation as a prominent political leader as well as a national war hero through his contribution to the Allied victory in World War II.

Following on the relatively high support for politicians in both samples, the US and British respondents identify military heroes as the fifth most popular category, with the US ‘pool’ of military heroes having a higher level of public support. This corresponds with twice as much support for ‘strong defence forces’ in the US in comparison with Britain (Inglehart et al., 2014). However, the military in both countries is not the most popular hero, a finding which challenges a popular premise within heroism studies and wider debates on heroes, both of which consider ‘martial heroism as a basic pattern of hero-perception’ (Korte & Falkenhayner, 2021, 137; see also Franco et al., 2018) and also the dominant model of male-centric martial heroism (e.g. Bacevich, 2013; McCartney, 2011; Browning & Haigh, 2022). Notably, the ‘pool’ of military heroes in both countries includes few historical personalities, mostly from the context of World War II, with most respondents declare serving military personnel or the British/US Armed Forces as contemporary collective/institutional military heroes. This suggests that the public recognition of modern soldiers is either substituted with historical associations or martial heroism is linked to the military institution by by-passing individual soldiers, both dynamics suggests the enduring perception of the military as heroic (Frisk, 2018) thereby challenging the premise that martial heroism in the countries with all-volunteer forces has moved towards the ‘post-heroic stage’ and personalisation of military heroes (e.g. Scheipers, 2014).

The final similarity between two countries refers to human rights activists and campaigners, which is the sixth most popular category in both countries. Both samples feature Martin Luther King Jr, whose popularity competes with Nelson Mandela and Malala Yousafzai in the UK sample, and Mahatma Gandhi and Malcolm X in the USA. Although in both countries human rights driven, liberal regimes align with each other in terms of humanistic values (Inglehart & Welzel, 2005), these values manifest in choosing hero-figures who mostly reside outside these societies. Particularly, in Britain, human rights hero-identifications derive solely from non-British contexts and hons on recognisable human rights icons, including into the heroic landscape of liberal humanitarian interventionalism. This finding corresponds with the analysis of media representations of Malala Yousafzai in the UK newspapers, which reveals a hierarchy between the representation of Malala as a global human rights activist, who is simultaneously recognised and othered (Walters, 2016).

Searching for a Hero (1): Women-Heroes in Britain and the USA

Despite the expectations of a shift towards relative gender equality across post-industrial societies (Norris & Inglehart, 2011: 88–89), our analysis reveals a (white) male-dominated landscape of contemporary heroism, with only one in four Britons and one in five Americans having a heroine (See Tables S.8 and S.9 in the Appendix).Footnote 3 The percentage of women-heroes is far from being representative of the population structure of both countries and of the level of women’s presence in public life.

While women are present among hero-figures, we observe strong gender polarisation through the prominence of same sex choice, with women from both British and American samples being more likely than men to have women-heroes. The results of logistic regression analyses (see the full Output S.1 in the Appendix and predicted probabilities in Fig. 1) show that this effect is strong, positive, and statistically significant at 99.9% confidence level. American and British men alike are around 4 times less likely to have a woman-hero than women, i.e., 9 to 34% in the US and 9 to 40% in Britain. This is in line with previous studies conducted on adolescents and children (e.g., Estrada et al., 2015; Yancey et al., 2011) and strongly supportive of the claim that women ‘are proving more supportive of gender equality, especially in post-industrial societies’ (Norris & Inglehart, 2011: 18–19).

Fig. 1
figure 1

Predicted probabilities of having a woman-hero by country

Full size image

The analysis of preferences for public and personal heroes based on respondents’ sex reveals the widening gap between the USA and Britain. Once we exclude family members from the analysis, only 11% of Americans choose a women public figure hero, while one in five Britons do (see Tables S.10 and S.11 in the Appendix). In other words, although privatisation of heroes is prominent in both countries, the presence of women among public hero-figures is stronger in Britain than in the US.

Our analysis of the relationship between age and preferences for women-heroes challenges the linear pattern of the generational replacement across post-industrial societies (Norris & Inglehart, 2011: 19). If in the USA, the under 25-year-olds have almost a 50% chance to have a woman-hero, which declines gradually for each subsequent age group, yet spikes to 21% for 55 to 64-year-olds before declining for the oldest age group to 9%. In the UK, the under 25-year-olds have only a 25% chance of having a woman-hero, similar to the oldest age group. However, middle aged Britons (between the ages of 40 to 59) are the least likely to have women-heroes, which is the only statistically significant effect of age identified in the British data. Thus, although there is a multitude of empirical evidence demonstrating a linear trend of increasing support for gender equality over the last 50 years (Norris & Inglehart, 2011: 39–40; see also Childs & Webb, 2011), this support does not converge in the linear age-based pattern of support for women-heroes.

Finally, our analysis of support for political parties and women-heroes demonstrates the divergent dynamics in Britain and the USA, challenging the argument of a similar pattern of gender re-alignment (Norris & Inglehart, 2011). For instance, in our study, support for the Conservative Party in the UK and the Republican Party in the USA has the opposite effect on the probability of having a woman-hero. In Britain, supporting the Conservatives increases the probability of having a woman-hero expressed as a 27% chance of having a woman-hero compared to 13% for Labour, 22% for Liberal Democrats and 14% for UKIP. This outcome represents a ‘Thatcher effect’ — with Margaret Thatcher accounting for 18% of all women-heroes listed by Conservative supporters, which is second only to female family members (Table S.2 of the Appendix). This interpretation resonates with the view expressed by some Conservative Party members arguing that Thatcher ‘opened the door’ for women into high politics and was ‘a real role model’ for them (Childs & Webb, 2011: 94).

Supporting the Republican Party in the USA, by contrast, substantively reduces chances of having a woman-hero (significant 90% confidence level), with the Republicans having a 13% chance of having a woman-hero compared to the Democrats’ 25%. In other words, the American version of conservative ideology impedes the shift towards recognition of heroines.

Searching for a Hero (2): Ethnic Minority Heroes in Britain and the USA

When comparing the predictors for an ethnic minority hero (see the full Output S.2 in the Appendix and predicted probabilities in Fig. 2), there are more differences than similarities. This stems from a wider range and greater presence of ethnic minority heroes in the US ‘pool’, while they are barely visible among British heroes.

Fig. 2
figure 2

Predicted probabilities of having an ethnic minority hero by country

Full size image

In Britain, non-white minority heroes account for only 15% of all heroes. Chosen by only 47 respondents, this illustrates extremely low probabilities of having an ethnic minority hero at all among all groups of Britons. They barely reach 21% at a maximum (for women), but mostly, for all income groups, social grades, political parties, and regions hover around 10%, thus, indicating that there is a 90% probability that the chosen hero is white. Furthermore, ethnic minority hero-figures tend to include non-British political activists such as Nelson Mandela, Martin Luther King Jr., Malala Yousafzai, Mahatma Gandhi, and Muhammad Ali, while only three Britons made it to the list, including boxer Lennox Lewis, Formula 1 driver Lewis Hamilton, and a Victoria Cross recipient Johnson Beharry.Footnote 4 Such limited range and low numbers of ethnic minority heroes, as well as a lack of UK-born heroes among them (see Table S.6 in the Appendix), gives weight to the argument that heroism in Britain has not only been male-centric (Danilova & Kolpinskaya, 2020; Jones, 2007), but it is White- and Euro-centric. This trend reflects the othering of the racial/ethnic difference while also signposting a limited progress in both visibility and critical engagement with experiences of non-white communities in the UK.

In the meantime, despite the ongoing struggle to overcome racial discrimination, there is a much wider presence of non-white Americans in the ‘pool’ of the US heroes. They account for one in three heroes listed by the US respondents and include African Americans, Hispanics, Asians, Native Americans, and those of mixed ethnic background (Table S.3 of the Appendix). Although they are present in greater numbers, ethnic minority hero-figures tend to include prominent public figures such as Barack Obama and Martin Luther King Jr. among African Americans and Pope Francis among Hispanics.

The analysis of the US sample generates strong support for the same race/ethnicity choice in line with the findings of previous studies conducted on children and adolescents (e.g., Anderson & Cavallaro, 2002). Particularly, we identify the strikingly high probability of having a non-white hero among African Americans (88%), Hispanics (70%) and Americans belonging to other ethnic minority groups (89%) compared to white Americans (6.5%). The effects are statistically significant at 99.9% confidence level for all non-white backgrounds, but particularly strong for African Americans and people with mixed ethnic background. Unfortunately, the limitations of our data, i.e., the absence of the ethnicity of respondent variable for the British sample, do not allow establishing if this pattern holds for Britain. However, the extremely low presence of ethnic minority heroes (only 47 cases) and a lack of variation among them (only 18 unique heroes) (Table S.4 in the Appendix) would have yielded few meaningful results even if this variable were available.

Overall, of all socio-demographic characteristics, only sex of respondents has a statistically significant effect on the probability of have an ethnic minority hero for Britain. This effect supports the argument that women in post-industrial societies appear to be more tolerant towards minorities (Norris & Inglehart, 2011) — with 30 out of 47 ethnic minority heroes declared by women (Table S.4 in the Appendix). For instance, British women identified nine out of 14 preferences for Nelson Mandela, four out of seven — for Martin Luther King Jr. and all preferences for Malala Yousafzai and Oprah Winfrey (Table S.5 of the Appendix).

The US model, by contrast, has a decent explanatory power (with the Pseudo R-square of 0.493) and identifies several statistically significant predictors of preference for an ethnic minority hero, in addition to sharing the hero’s non-white ethnic background. As in the case of predicting preference for women-heroes, age has a non-linear effect on the probability of having an ethnic minority hero (Fig. 2). It hovers around 6% — for the under 25 s, spikes up to 52% for the 25- to 34-year-olds and then settles at one in four chance (23–25%) for the middle age group up to the age of 64. Although the probability of having an ethnic minority hero drops slightly for the over 65 s, it is still almost three times higher than that of the under 25 s, thus, casting considerable doubt that acceptance of differences is associated with younger age (De Graaf & Evans, 1996; Cornelis et al., 2009; Norris & Inglehart, 2011: 87). It might be also a result of under-representation of African Americans, Hispanics, Native Americans, and Asians within the educational curricula (Banks, 1991; Hutchins, 2011) and their continuous misrepresentation in popular culture. For the 25- to 34-year-olds who spent most of their formative years during the Obama presidency, which has affected aspirations of African Americans, this effect may be lower (Marx et al., 2009; Tesler, 2016).

As in Britain, American women are disproportionately — and statistically significantly — more likely to have ethnic minority heroes with one in three women declaring an ethnic minority figures as a private hero (family and friends). Based on these findings, our analysis does not only show the positive effect of gender on tolerance towards minorities (Norris & Inglehart, 2011), it suggests the transfer of gender differences into preferences for certain types of heroes. For instance, in both samples, women demonstrated the weaker attachment to traditional types of white public hero-figures and affinity with female family members and prominent civil rights activists of non-white backgrounds such as Nelson Mandel, Maya Angelou, and Martin Luther King Jr.

Our data does not support an assumption that non-white hero-figures are disproportionately hero-villains (Schuman et al., 1997; van Deburg, 2013). Indeed, both in Britain and the USA, the pool of non-white heroes includes prominent civil rights activists, writers, television personalities, politicians, military figures, singers, models, and sport personalities, and as such it replicates the structure of white-male-dominated heroism.

Conclusion

The extensive scholarly and popular debate on the functions of heroism generates a wide range of explanations of reasons for paying attention to who, how and why is seen as a hero. Some scholars treat heroes and heroines as inspiring and life-affirming figures whose stories allows us to overcome everyday troubles and find the way to move forward (e.g. Allison & Goethals, 2011; Kinsella et al., 2015; Franco et al., 2018). Others see heroism and heroes as an under-utilised and under-explored resource of political socialisation of young people into the foundational values of their respective communities, suggesting that young people need to be taught to recognise and appreciate ‘true’ heroes (e.g. Estrada et al., 2015; Jones et al., 2014; Power & Smith, 2017). Finally, there are those who see heroes as multifaced cultural, emotional devices, and ‘unicorns’, ‘a partly mythical, partly manufactured, but arguably important cultural marker of a nation’s purported union around key values, symbols and icons’ (Peabody & Jenkins, 2017: 210). The main contribution of this paper in this expanding scholarly debate lies in a shifted focus from discussion of heroes as desirable, yet exceptional and rarely achievable role-models to the empirically grounded analysis of popular heroes and heroism as informative of social change and embracement of marginalised groups such as women and ethnic/racially- minorities.

Applying the concept of social change to the analysis of hero-landscape in Britain and the USA, we demonstrate that although popular choices of heroes reflects some degree of diversity and respect for women-heroes and heroes from non-white backgrounds, this support remains limited and strongly determined by the pattern of gender/race sameness. Both dynamics work through the same sex and same race choice, thus, casting doubt in the inevitable rise of egalitarianism and embracement of diversity and inclusion politics argued by post-materialism theory. Both dynamics lead to gender and racial polarisation of heroes from different backgrounds.

Regrettably, the analysis of gender composition shows that women remain largely ‘invisible’ heroines in both countries. This is reinforced through two interlinking dynamics: privatisation of women-heroes and through the prominence of same sex choice pattern among women, which is particularly strong in the case of Britain. In other words, we reveal a persistent gap between the publicly prominent white male hero-figure and a perpetually ‘invisible’, and ‘forgotten’ heroine. Due to the limitations of our dataset, we cannot identify the exact causes of this persistent gender marginalisation of women-heroes, but we can speculate that this trend continues to be enhanced through the systematic underrepresentation, and/or misrepresentation of women-heroes, including the misrepresentation of women’s achievements in mass media (e.g., Kian et al., 2013; Lines, 2001), educational curricula (Pickett et al., 2010; Power & Smith, 2017), an ongoing uphill struggle to improve gender representation in politics alongside a wave of antifeminist backlash and neoconservatism, particularly, in the US. Perhaps, surprisingly, our study demonstrates the possibility of endorsing prominent women (Margaret Thatcher) as heroines, a step which can potentially lead to the modernisation of social conservative parties (e.g., the Conservative Party under David Cameron: Childs & Webb, 2011: 114), but the lasting change seems less likely (Bale, 2010: 20). The analysis of the relationship between party affiliation and preferences for women-heroes in the USA reinforces this doubt showing that the affiliation with Republican Party impedes support for gender equality in heroes due to the overlap between conservative ideology with the high level of support for religious values, which, in turn, forms one of the cornerstones of the American version of popular heroism and leaves little space for embracing gender equality in heroes.

In terms of ethnic diversity of heroes, our study reveals a stark difference between Britain and the USA, with only 15.2% of non-white heroes in Britain to 31% in the USA. Although these figures resonate with the ethnic composition of both countries (ONS, 2012; U.S. Census Bureau, 2017), they are also reflective of the limited visibility and recognition of ethnic minorities in public life of both countries. The legacy of the equal rights movement in the USA brought forward recognisable and well-known figures into the public eye — and into the educational curricula — such as Martin Luther King Jr and Rosa Parks. The political legacy of Barak Obama has further advanced public debate relating to racial politics and discrimination. Having said this, preference for heroes of the same race/ethnicity drives choices of non-white heroes in the USA. This pattern advances racial polarisation and as such it reveals the limitation of racial integration through heroes, particularly between white Americans and Afro-Americans.

In the UK, although racial prejudice is declining and the level of acceptance of ethnic minorities as a group is growing (Ford, 2008), there are few easily recognisable heroes from non-white backgrounds. This is obvious from the fact that only three out of 44 non-white heroes in Britain are British — Lennox Lewis, Lewis Hamilton, Johnson Beharry and — while the rest are prominent African Americans, overseas nationals and icons of liberal humanitarianism. This finding reinforces the ‘whiteness’ of British heroism (Jones et al., 2014; Korte & Falkenhayner, 2021), leading to the systematic lack of public recognition for achievements of Britons from ethnic minority backgrounds (The Telegraph, 2019) alongside difficulties of acknowledging the decades, if not centuries-long experiences of discrimination and injustice.

Overall, our analysis demonstrates both a lack of progress in bringing gender equality and equal recognition of minorities in heroes and heroism of both countries as well as a strong trend towards gender and ethnic polarisation through preference for same sex and same race/ethnicity heroes. Both trends run a risk of compartmentalising women and ethnic minority heroes to their respective social groups., cementing rather than overcoming the existing divisions. This trend highlights the most problematic implication of (in)visibility politics of heroes from marginalised backgrounds. Drawing on the postmaterialist theory and discussion of social change, we argue that it is vital to move beyond signposting difference through the equal visibility of heroes and heroines from marginalised backgrounds, and work towards recognising and engaging with multiple injustices faced and overcome by marginalised groups.

Finally, we tend to agree with the caveat to post-materialism which suggests that ‘cultural change is not sufficient [italic in the original] by itself for gender equality’ as well as for the greater acceptance of ethnic minorities (Norris & Inglehart, 2011: 9). This caveat necessities the expansion in the empirical analysis of public choices of heroes and heroines as a way of gaining insights in the drivers of and barriers to social change and a truly inclusive politics.