Brazil's football warriors: Social bonding and inter-group violence

https://doi.org/10.1016/j.evolhumbehav.2018.06.010Get rights and content

Abstract

Football-related violence (hooliganism) is a global problem. Previous work has proposed that hooliganism is an expression of social maladjustment. Here we test an alternative hypothesis, that hooliganism is typically motivated by a parochial form of prosociality, the evolutionary origins of which may lie in intergroup raiding and warfare. In a survey of Brazilian football fans (N = 465), results suggest that fan violence is fostered by intense social cohesion (identity fusion) combined with perceptions of chronic outgroup threats. In contrast, maladjustment is unrelated to indices of past acts of football-related violence or endorsement of future violence. Our results suggest that to reduce hooliganism and other forms of inter-group violence, efforts could be made to harness the extreme pro-group sentiments associated with identity fusion in more peaceful ways.

Introduction

Football (or soccer) related violence (or ‘hooliganism’) is a persistent, global problem. Negative outcomes include injury, destruction of property, state security costs, and even death (Raspaud & Da Cunha Bastos, 2013; Sekulic, Kühl, Connert, Krastl, & Filippi, 2015; Spaaij, 2005; Stott, Adang, Livingstone, & Schreiber, 2008). To address the problem, we need to understand what motivates it in the first place.

For decades, hooliganism has been attributed to social ‘maladjustment’ or ‘abnormal social functioning’ (Lawther, 1972; Wakefield & Wann, 2006), sometimes associated with masculinity and class (Robson, 2000; Spaaij, 2008). By contrast, we argue here that football-related violence stems from a particularly potent form of pro-group commitment dubbed “identity fusion” (Swann, Gómez, Seyle, Morales, & Huici, 2009; Swann, Jetten, Gómez, Whitehouse, & Bastian, 2012). People strongly fused to a group experience a profound sense of ‘oneness’ between their personal identity, that is who they are as an individual, and their social identity, e.g., family, religious affiliation, football team, etc. This synergy motivates a wide range of personally costly, pro-group actions because for fused persons, threats to the group are interpreted also as personal threats (Buhrmester, Fraser, Lanman, Whitehouse, & Swann, 2015; Swann et al., 2014; Swann, Gómez, Huici, Morales, & Hixon, 2010; Whitehouse, Mcquinn, Buhrmester, & Swann, 2014). This ‘warrior’ psychology is consistent with evolutionary explanations of coalitional conflict and warfare (McDonald, Navarrete, & van Vugt, 2012; Whitehouse et al., 2017). We present a warrior psychology hypothesis of football hooligans, i.e., that these fans are motived by identity fusion to defend their fellow fans in the face of perceived outgroup threats, using membership to Brazilian superfan groups (torcidas organizadas) as a proxy to assess threat. Here we attempt to adjudicate between ‘social maladjustment’ and evolved ‘warrior psychology’ hypotheses at the levels of both proximate and ultimate explanation.

Football-related violence has often been attributed to social maladjustment, characterized by abnormal or impaired social functioning (Lawther, 1972; Wakefield & Wann, 2006; Zani & Kirchler, 1991). The ‘hardened hooligan’ is frequently depicted in mainstream media as a violent thug, gangster or other form of social deviant. This hooligan stereotype is also widely evident in popular books and films in the UK over the last two decades (Poulton, 2014).

Academic representations of football hooliganism emphasizing class (Dunning, Murphy, & Williams, 1986; Robson, 2000) and masculinity (Spaaij, 2008; Taylor, 1987) contribute to this popular caricature. According to this view, the typical hooligan is young, working-class, troubled, and male (Wakefield & Wann, 2006; Zani & Kirchler, 1991). However, causal evidence in support of the social maladjustment hypothesis is scant. Contrary to popular parodies, research shows that football violence is not limited to the youth and the working classes (though they are perhaps most visible in the media) (Pearson & Stott, 2016; Treadwell, 2006).

Some football fans experience a profound sense of ‘oneness’ between personal and social selves known as ‘identity fusion’ (Newson, 2017; Swann et al., 2009; Swann et al., 2012; Swann & Buhrmester, 2015). Dozens of studies focusing on a range of group alignments (e.g., national, political, religious) show that strongly ‘fused’ persons are willing to sacrifice themselves for fellow group members (Buhrmester et al., 2015; Swann et al., 2014; Whitehouse et al., 2014). This is due to the porous boundary between the personal and social selves of highly fused individuals and the fact that both personal and social selves of highly fused individuals are activated in tandem (Swann, Gómez, Huici, et al., 2010). Likewise, fused individuals are emboldened by the sense that the group's collective strength is invested in their personal agency (Swann et al., 2012). As a result, highly fused people are extraordinarily committed to the group, as any attack on the group feels like an attack on oneself.

Fused individuals are not just co-operators; they regard one another as connected by family-like ties (Buhrmester et al., 2015; Whitehouse & Lanman, 2014). The extreme cooperation caused by fusion is thought to result in part from individuals recognising other group members as ‘psychological kin’ (or even closer than kin) due to sharing intense, self-transformative experiences and evolved tendencies for sacrificial behaviour to protect them (Jong, Whitehouse, Kavanagh, & Lane, 2015; Newson, Buhrmester, & Whitehouse, 2016; Swann et al., 2012; Whitehouse et al., 2017; Whitehouse & Lanman, 2014). Psychological kinship thus refers to fictive ties (e.g., as seen among ‘fraternities’ or ‘brothers in arms’), as opposed to actual kin detection (e.g., via cohabiting with siblings or exposure to mother-infant dyads (Lieberman, Tooby, & Cosmides, 2007)).

While past work has shown that fusion with a group is sufficient to produce pro-group action, other group factors may interact with fusion to produce the particularly strong pro-group, violent acts that are found in certain highly fused groups (Sheikh, Gómez, & Atran, 2016; Swann, Gómez, Dovidio, Hart, & Jetten, 2010; Whitehouse et al., 2014). For instance, recent work with a sample of Israeli residents has also shown that individuals with high fusion scores are particularly hostile toward out-groups when they perceive high levels of threat to the ingroup (Fredman, Bastian, & Swann Jr, 2017). The warrior psychology hypothesis is that football hooligans are motived by identity fusion to defend their fellow fans in the face of perceived outgroup threats, which are a continual feature of life for a member of a superfan group.

In trying to understand the root of football-related violence, scholarly attention has focused largely on proximate physiological mechanisms, such as post-match testosterone release (Bernhardt, Dabbs, Fielden, & Lutter, 1998) or lower basal-cortisol levels in aggressive fans (van der Meij et al., 2015). Such explanations, though controversial, are not inconsistent with the social maladjustment hypothesis. It is less obvious, however, how a social maladjustment hypothesis might fit into a broader evolutionary account. By contrast, it is relatively easy to provide a rationale for the warrior psychology hypothesis at the level of ultimate functions in intergroup competition.

Research on the relationships between group membership, individual differences in violent behaviour, and fitness, have increasingly contributed to an understanding of intergroup conflict and warfare (Choi & Bowles, 2007; Kurzban, Tooby, & Cosmides, 2001; McDonald et al., 2012; Navarrete, McDonald, Molina, & Sidanius, 2010; Whitehouse et al., 2017; Wood, 2010). In particular, evolutionary approaches to the study of coalitional psychology have made considerable progress in understanding the phenomenon of fandom (Fessler, Holbrook, & Dashoff, 2016; Winegard & Deaner, 2010).

Coalitionalism is a universal human trait. Examples include states, ethnic conflict, gang rivalries, male social-clubs, competitive team sports, and online video gaming. Coalitional psychology and intergroup violence are also documented in chimpanzee society (Goodall, 1986; Wilson, Wallauer, & Pusey, 2004), with coalitional aggression improving reproductive output (Gilby et al., 2013). Evolutionarily, we might expect to see higher levels of violence among male and masculinised groups due to a combination of male competition for access to females and the additional aggression required for inter-group conflict (McDonald et al., 2012; Pinker, 2011; Wrangham & Peterson, 1996). In phylogenetic terms, humans share the trait of male violence with other great apes (Pinker, 2011).

Football violence or ‘hooliganism’, in its many cultural settings, is best understood as a product of our evolved coalitional psychology, featuring aggression as a tool for group gain, hate and anger toward threatening out-groups (that could reduce fitness), and the ability to co-ordinate with the group (and potentially displaying this ability to increase the group's formidability; Tooby & Cosmides, 2010). Furthermore, in football, teams exhibit a number of non-sport-related traits that make them comparable to warring groups - players tend to: be connected with a particular territory; wear highly distinctive group uniforms, and; gain ‘spoils’ after the game, e.g., material wealth, status, and desirable mates (Winegard & Deaner, 2010).

Coalitional conflict has been explained in ultimate terms using evidence from game-theoretic analysis and agent-based simulations (Choi & Bowles, 2007). These analyses suggest that the volatile conditions of the late Pleistocene and early Holocene would have put selective pressures on altruism and parochialism in humans, due to conditions of heightened inter-group conflict. Clearly football has not existed long enough to influence evolution. Nonetheless, fan culture capitalises on our propensity to categorise and distinguish our coalitions. Just wearing a football shirt could be enough to alert others to one's coalitional affiliation, which in turn has been found to increase perceptions that the individual is both more physically formidable and aggressive (Fessler et al., 2016). Thus, an evolutionary framework can shed light on the role of group psychology in football hooliganism.

Recent research focusing specifically on the role of identity fusion in motivating extreme pro-group behaviour has modelled the evolutionary consequences of shared past experience for cooperation (Whitehouse et al., 2017). Specifically, models examined the extent to which groups facing dysphoric experiences (operationalized as fitness-reducing setbacks) are able to solve collective action problems, as compared with groups facing euphoric (fitness-enhancing) experiences (given standard assumptions regarding the transmission of genes that would make cooperation possible, but not inevitable). Following many simulations of this model, the gene effects for cooperation were found to be larger for dysphoric than euphoric experiences.

Empirical support for Whitehouse et al.'s model comes from surveys and experiments showing that dysphoric experiences with others lead to extreme social cohesion (identity fusion) in a diversity of sample populations ranging from military veterans, college fraternity/sorority members, martial arts practitioners, twins (beyond the effects of genetic relatedness), and football fans (Whitehouse et al., 2017). Of the 725 British football fans sampled by Whitehouse et al., those who were fans of loosing (dysphoria-inducing) teams were more likely to sacrifice themselves in a classic trolley dilemma, compared to fans of winning (euphoria-inducing) teams, an effect mediated by fusion. Furthermore, evolutionary advantages to the individual of engaging in personally costly, pro-group behaviours in inter-group conflict may be an outcome of direct reciprocity (Choi & Bowles, 2007) and indirect reciprocity and reputation enhancement (Fessler et al., 2016). Not all these possibilities have yet been integrated into fusion theory.

A limitation of much previous research on issues of group membership and outgroup hostility is that studies have predominantly sampled participants from Western, Educated, Industrialized, Rich, and Democratic (i.e. ‘WEIRD’) populations, and non-WEIRD populations have been under-investigated (Henrich, Heine, & Norenzayan, 2010). Here we address this limitation by applying theories of coalitional conflict discussed above to a less WEIRD sample: Brazilian football fans and torcidas organizada members (groups of extreme fans; hooligans). Brazil, a developing nation, is culturally rich, ethnically diverse, and famed for its love of football, making it the ideal site for this research. Case studies report tight-knit networks of supporters and tales of camaraderie among hooligans (Francis & Walsh, 1997; Harvey & Piotrowska, 2013; Robson, 2000). These networks, rife with kin terminology, seem far removed from the social misfits and deviants implicated in maladjustment explanations of hooliganism. An alternative to the maladjustment hypothesis is the theory that shared dysphoria is a pathway to fusion.

In Brazil clashes between rival ‘super fan’ groups known as torcidas organizadas have become increasingly violent, resulting in injury, stadia bans, and imprisonments (Raspaud & Da Cunha Bastos, 2013) as well as escalating loss of life (Murad, 2013; Raspaud & Da Cunha Bastos, 2013). Membership to a torcida organizada entails persistent reminders of rivalries and the threats that one's rivals pose to the group's material successes, reputation, and security (de Toledo, 1996; Pimenta, 2000); such threats are both real and imagined and constitute an essential aspect of a torcida organizada's identity (Newson, 2017). For strongly fused persons, involvement in torcidas organizadas may embolden acts of hostility and self-sacrifice as efforts to defend one's psychological brothers-in-arms (Fredman et al., 2017; Whitehouse et al., 2014). In contrast, for persons strongly fused to their club's fellow fans but not involved in a super fan group, violence may be less prevalent because they tend to not perceive strong outgroup threats, relative to highly fused members of super fan groups. Here we examine the extent to which engaging in football violence is the result of the combination of membership in a torcida organizada and fusion to one's fellow fans.

The purpose of this research was to test the following hypotheses, using a sample of football fans in Brazil (N = 465):

  • (1)

    Social maladjustment is unrelated to past reports of football-related violence, willingness to fight/die, and membership in a torcida organizada;

  • (2)

    Instead, hooligan acts reflect an interaction between torcida organizada membership and fusion, such that highly fused torcida organizada members are especially more likely to be more violent and endorse fighting and dying for fellow fans;

  • (3)

    Fused torcida organizada members are especially likely to report violence against rival team fans compared to general fans, but there will not be significant differences between torcida organizada members and general fans concerning other targets (e.g., the police), suggesting that ‘brothers in arms’ are primarily battling opposing ‘warriors’.

Section snippets

Experimental design

For this study, we focused on Brazilian football fans of all major leagues. Using a snowball technique, our Brazilian partners distributed a 20–40 min online questionnaire to affiliates of the Minerão stadium, surrounding fan groups, two subject pools, and posted on social media (e.g. Facebook, Twitter etc. and inviting people to ‘share’/‘retweet’ the study). Participants were told they could win one of three football jerseys of their club. This prize complied with Brazilian ethical regulations

Prevalence of violence

Of the 439 participants, almost half (45.1%) reported membership to a torcida organizada (see Table 2 for descriptives and Fig. 1 for a comparison of general fans and torcidas organizadas). A quarter of the total sample reported having engaged in football-related physical violence but this was significantly higher for those in a torcida organizada than general fans, χ2(1) = 49.30, p < .001 (Fig. 1). Willingness to fight and die for other fans was also significantly higher for members of these

Discussion

Our results indicate that both membership to a super-fan group, such as a torcida organizada, and identity fusion may play key roles in motivating violent behaviours among football fans. Our data suggest that rather than being socially maladjusted, a deep bond between fans motivates football hooliganism; to the extent that they regard one another as family. Furthermore, we found that it is the interaction between fusion to one's fellow fans and belonging to a super fan group that serves to

Conclusions

We cannot say that either decreasing fusion or membership to extreme groups will stop football-related violence. Rather than ‘defusing’ the group, which may be unrealistic given the irrevocable nature of fusion (Fredman et al., 2015), another possibility is to encourage fusion to groups that place less emphasis on intergroup rivalries and experience lower levels of outgroup threat. Further, brutal police tactics may only serve to increase hostility by inflating perceived threat levels toward

Funding

This work was supported by a Large Grant from the UK's Economic and Social Research Council (REF RES-060-25-467 0085) (HW, MB, MN, TSB); an award from the Templeton World Charity Foundation (TWCF0164) (HW, MB); an Advanced Grant from the European Research Council (ERC) under the European Union's Horizon 2020 - Research and Innovation Programme (grant agreement No. 694986) (HW, MB, and MN); an award from the John Fell OUP Research Fund (131/072) (HW and MB); a grant from the European Union

References (79)

  • J.-K. Choi et al.

    The coevolution of parochial altruism and war

    Science

    (2007)
  • L.H. de Toledo

    Torcidas organizadas de futebol

    (1996)
  • H. dos Reis et al.

    Eric Dunning's explanations about hooliganism in the Brazilian context: A critical review

    Movimento

    (2015)
  • E. Dunning et al.

    Spectator violence at football matches: Towards a sociological explanation

    The British Journal of Sociology

    (1986)
  • C.M. End et al.

    Sport fan identification in obituaries

    Perceptual and Motor Skills

    (2009)
  • D. Fessler et al.

    The strategy of affect: Emotions in human cooperation 12

  • D.M. Fessler et al.

    Dressed to kill? Visible markers of coalitional affiliation enhance conceptualized formidability

    Aggressive Behavior

    (2016)
  • A.P. Fiske et al.

    Virtuous violence: Hurting and killing to create, sustain, end, and honor social relationships

    (2014)
  • M. Francis et al.

    Guvnors: The autobiography of a football hooligan gang leader

    (1997)
  • L.A. Fredman et al.

    God or country? Fusion with Judaism predicts desire for retaliation following Palestinian stabbing intifada

    Social Psychological and Personality Science

    (2017)
  • L.A. Fredman et al.

    Identity fusion, extreme pro-group behavior, and the path to defusion

    Social and Personality Psychology Compass

    (2015)
  • I.C. Gilby et al.

    Fitness benefits of coalitionary aggression in male chimpanzees

    Behavioral Ecology and Sociobiology

    (2013)
  • Á. Gómez et al.

    On the nature of identity fusion: Insights into the construct and a new measure

    Journal of Personality and Social Psychology

    (2011)
  • Á. Gómez et al.

    The devoted actor's will to fight and the spiritual dimension of human conflict

    Nature Human Behaviour

    (2017)
  • A. Harvey et al.

    Intolerance and joy, violence and love among male football fans: Towards a psychosocial explanation of ‘excessive’ behaviours

    Sport in Society

    (2013)
  • A. Hayes

    Introduction to mediation, moderation, and conditional process analysis: a regression-based approach

  • J. Henrich et al.

    Beyond WEIRD: Towards a broad-based behavioral science

    Behavioral and Brain Sciences

    (2010)
  • M. Hewstone et al.

    Intergroup bias

    Annual Review of Psychology

    (2002)
  • J. Jong et al.

    Shared negative experiences lead to identity fusion via personal reflection

    PLoS One

    (2015)
  • M. Joo et al.

    Effect of identity fusion on decision to make extreme sacrifices in romantic relationships: The moderating role of impulsiveness

    The British Journal of Social Psychology

    (2017)
  • R. Kurzban et al.

    Can race be erased? Coalitional computation and social categorization

    Proceedings of the National Academy of Sciences

    (2001)
  • J.D. Lawther

    Sport psychology

    (1972)
  • P. Lewis et al.

    Reading the riots: Investigating England's summer of disorder

    (2011)
  • D. Lieberman et al.

    The architecture of human kin detection

    Nature

    (2007)
  • M.-K. Lizotte

    Gender differences in support for torture

    Journal of Conflict Resolution

    (2017)
  • M.M. McDonald et al.

    Evolution and the psychology of intergroup conflict: The male warrior hypothesis

    Philosophical Transactions of the Royal Society B

    (2012)
  • M. Murad

    Práticas de violência e mortes de torcedores no futebol brasileiro

    Revista USP

    (2013)
  • C.D. Navarrete et al.

    Prejudice at the nexus of race and gender: An outgroup male target hypothesis

    Journal of Personality and Social Psychology

    (2010)
  • M. Newson

    Football, fan violence, and identity fusion

    International Review for the Sociology of Sport

    (2017)
  • Cited by (50)

    • Child images affect sports fans' prosociality and aggression

      2024, Journal of Behavioral and Experimental Economics
    • Fanning the flames: Transmitting negative word of mouth of rival brands

      2023, Journal of Business Research
      Citation Excerpt :

      With the increasing consumer empowerment and popularity of social network sites, individual hostility may evolve into collective actions, resulting in the development of anti-brand communities (Brandão & Popoli, 2022; Dessart et al., 2020; Popp, Germelmann, & Jung, 2016). When conflict intensifies, this rivalry can turn into inter-group violence, which is common among competing teams in sports (Newson et al., 2018). Previous research has revealed a series of factors that drive consumers’ resistance to competing brands, such as brand community identification (Kuo & Hou, 2017), customer satisfaction (Liang, Guo, & Zhang, 2019), and community commitment (Kuo & Feng, 2013).

    • Sports

      2022, Encyclopedia of Violence, Peace, &amp; Conflict: Volume 1-4, Third Edition
    View all citing articles on Scopus
    View full text