Categorising the causes of violence using the integrated model

This blog aims to summarise the literature relating to the causes of aggression and violence drawing on a number of key documents that have synthesised the research literature in a range of domains. It first discusses the social context and prevalence of violence before reviewing the risk factors associated with it. The blog goes on to review historical and socio-cultural views of the causes of violence and its long-term decline, perspectives from biology and neuroscience and then psychological theories. Finally, the blog attempts to bring together the different theories and perspectives using the integrated model of behaviour.

The social context and prevalence of violence

Aggression and violence are complex behaviours and social phenomenon that have many different forms and happen for a wide range of reasons. In some cases, violence and aggression are seen as socially acceptable, for example in contact and combat sports, and they form the mainstay of popular entertainment from depictions of violence in films and other media such as video games to the live action performances of WWE. In addition, violence and aggression by the state, though highly regulated in most countries in the world, is accepted as a necessary element of maintaining law and order. At the same time, violence outside of those parameters is at the top of many people’s concerns and its impact, particularly on young people, can be devastating. According to the World Health Organisation (WHO), youth violence can lead to ‘mental health problems and increased health-risk behaviours… [and] results in greatly increased health, welfare and criminal justice costs’ (WHO, 2015).

In recent years, there has been a mixed picture in terms of trends relating to crime and violence. According to the YEF’s latest review of the data, although there were reductions in levels of crime during the pandemic, since the end of lockdown restrictions violent crime has returned to and in fact exceeded levels previously recorded with homicides increasing by 2%, violence with injury increasing by 4%, and violence without injury increasing by 11% (YEF, 2022). Worryingly the data also indicates that there was a disproportionate rise in potential child victims in 2021 (an increase of 9% compared to 2020). This reinforces the importance of developing and evaluating interventions aimed at reducing the number of children and young people involved in violence either as victims or perpetrators.

Risk factors associated with violence

In order to intervene and prevent youth violence it is important to understand its causes, as this can enable policy-makers, intervention developers and those working on the ground to target the most powerful levers at both a social and individual level. However, the complexity and diversity of the phenomenon makes it difficult to provide simple, coherent explanations for why individual violent acts occur or for their prevalence in particular societies at specific periods of time. The WHO take a risk-based approach, using an ecological framework to identify risk factors for youth violence at individual, family and community levels (WHO, 2015), mapping them against different developmental stages, from conception and early infancy to early adulthood. At an individual level, risk factors range from those associated with an individual’s characteristics (such as sex, cognitive functioning, and psychological and behavioural disorders), behavioural patterns (illicit drug use, harmful use of alcohol, involvement in crime and delinquency), and individual experiences (including child maltreatment, parental drug use and unemployment). In addition to these factors, age itself is a risk factor, with a large proportion of violence being perpetrated by individuals below the age of 25.

Some of these individual factors overlap with those categorised as part of the family and close relationship level, which include poor parental supervision, on the one hand, and harsh and inconsistent discipline by parents on the other. Other factors at this level include parental divorce, and depression, familial unemployment and a history of antisocial behaviour, and among wider social relationships, peer delinquency and gang membership. At the highest ecological level, social and community risk factors include access to alcohol, drugs and firearms and other weapons, and poverty and inequality.

While a risk-based approach to understanding the causes of violence provides a useful framework for assessing broad policy-based decisions about violence prevention, using it to design interventions that aim to work at an individual level can be challenging. This is because identifying factors associated with violence does not distinguish between the causes of violence and its effect nor does it identify the underlying mechanisms involved. In addition, the risk factor model does not clearly distinguish between aspects of an individual’s life or context that can potentially be influenced (such as their drug or alcohol use or who they choose to be their peer group) and those that cannot (such as their sex or other genetic factors). Therefore, other conceptual frameworks need to be drawn on to provide a more powerful understanding of the causes and mechanisms related to violent behaviour.

Historical context and socio-cultural drivers

In his book, The Better Angels of our Nature, Steven Pinker provides an overview of historical trends in violence and an explanatory framework at a social level for the decline in interpersonal and state-sponsored violence that has occurred over the last two thousand years (Pinker, 2011). Though not without controversy, much of Pinker’s analysis is regarded as providing a coherent explanatory framework at a macro-level for some of the key trends in violence. The broad premise is that two historic social trends radically changed the costs and benefits of violence and aggression within society, which led to a dramatic reduction in violent acts. The first, associated with the agrarian revolution and the birth of large-scale farming, was the aggregation of small social groups into large scale political states, whose rulers were strongly motivated to monopolize violence as a means to maintain law and order and maximise economic efficiency. This was not necessarily associated with a lessening of the ferocity of violence by the state but did lead to significant reductions in violence by private individuals, both due to policing and the imposition of punishment and subsequently the internalisation of the new social rules and norms and reliance on their effectiveness. Evidence has indicated that individuals with low socio-economic status and from marginalised ethnicities are treated differently by police and the criminal justice system, undermining the belief among those groups that the system can be relied on to respond to acts of violence or law breaking. This is seen as one explanation for higher rates of violence in those communities.

The second trend was the transition to a modern economic system that used a currency rather than barter and was based on much greater labour specialisation. Both these economic characteristics, it is argued, rely on significant levels of trust, both in individual transactions and in the maintenance of a stable society that means current investment will be rewarded in the future. These structural changes are seen as being reinforced and enhanced by changes in the intellectual discourse, firstly by the Enlightenment’s focus on reason and science (mitigating against superstition and violence associated with religious conflict) and secondly by the creation of a language of human rights, which first came to prominence in the declarations after the French and American revolutions and subsequently gaining greater prominence in the post-World War 2 Declaration of Human Rights in 1948.

Pinker’s framework provides a helpful overview and explanation of historical trends, though suffers from a number of limitations. One major issue is the fact that, although the thesis is supported by empirical analysis, the causes of large-scale historical trends by their nature are difficult to pinpoint definitively. For example, although the rise of a modern economic system is credited in part for reducing violence, some economic historians point to the decline in violence as enabling the development of trade and commerce, meaning the direction of causation is difficult to untangle. In addition, while the historical analysis is useful in understanding long term trends, it is perhaps less helpful in considering what interventions may or may not be effective at a particular timepoint. Increasing the strength and legitimacy of a society’s system of crime and justice, widespread changes in the cultural discourse around violence and further developments of its economic system are outside of the control of the developers of interventions aimed at the individual level.

Biology, neuroscience and evolutionary perspectives

Robert Sapolsky’s work aims to draw together the wide array of biological perspectives on violence, beginning conceptually with the theory of evolution, within which violence is seen from the perspective of patterns of behaviour that promote the passing on of genes (Sapolsky, 2017). Historically, evolutionary perspectives have been associated with right-wing political ideology that in some circumstances has either condoned violence against particular groups or contradictorily seen violence as more ‘natural’ among some groups, thus providing a rationale for exercising control. Modern interpretations of evolutionary theory recognise that evolutionary pressures can act to both encourage violence and competition and promote restraint and altruism. Individual selection and kin selection predict the use of violence for the purposes of sexual access of males to women, where the potential of violence increases as the degree of relatedness decreases. These patterns that have been observed in non-human species and in humans, for example analysis of the likelihood of violence of fathers towards their natural versus their step-children. However, co-operation within kin groups and the development of pseudo-kinship (as happens among warrior classes in traditional societies and in modern armies) shows the potential for evolutionary pressures to discourage violence. This effect is also seen in rituals aimed at creating the shared fiction of common ancestors, such as among peoples who might naturally compete for water rights. The flip side of this is that the creation or emphasis of in-group out-group distinctions can promote violence, reaching an extreme with the phenomenon of ‘despeciation’, where an out-group is conceptualised as ‘non-human’ and therefore not worthy of the same level of sympathy or respect.

At a genetic and biological level, genes associated with a high or low level of the neurotransmitter serotonin have been found to interact with environmental experiences, and particularly with the occurrence of adverse childhood events, to predict higher levels of aggression, though this is a highly complex field that in many ways is still in its infancy. The higher prevalence of violence among men is well established and there is evidence for the role of testosterone being part of the causal pathway. However, research indicates that it is the presence of testosterone in normal levels rather than the amount of testosterone that is most influential and increases in testosterone amplifies existing behaviour rather radically changing it. However, levels of testosterone outside of the normal range, as can happen due to the misuse of steroids, can lead to substantially increased levels of aggression not restricted to existing social patterns. 

At the level of neuroscience, brain areas thought to be important in influencing violence include the amygdala, which is part of the limbic system, which is associated with emotions and the perception of them and us dichotomies. Evidence indicates that post-traumatic stress disorder is associated with increases in the size of the amygdala, leading individuals to be more responsive to stress and fear and therefore react more aggressively and to a heightened sense of ‘us and them’ social cognitions. As individuals with low socio-economic status and of marginalised ethnicities are more likely to experience traumatic events, this provides a potential explanation for the increased levels of violence among those groups. Another important area of the brain in relation to violence is the prefrontal cortex, which is important in ‘executive function’ and decision-making. The prefrontal cortex is seen as important in regulating the amygdala, including promoting delayed gratification, which is an important component in resisting the urge to act violently. The prefrontal cortex does not fully develop in humans until around the age of 25, which helps explain the fact that the age group 16-25 is responsible for much of violence in modern societies. Other areas of the brain that play a role in violence or its absence include the anterior cingulate cortex, associated with empathy. and the insula, associated with physical and moral disgust (which can play a role in reinforcing the sense of them and us and potentially justifying acts of retributive violence). The importance of neurobiology in explaining the causes of violence is illustrated by the statistic that up to 75% of violent criminals have experienced a serious head injury and also by the fact that 7-28% of the fall in homicide rates in the US is estimated to be due to the introduction of lead-free fuel.

Psychological theories

Psychological models have been criticised for placing an overemphasis on the role of an individual in the causes of violence, while underplaying the importance of context and systemic or structural factors. However, no matter how significant those wider factors are they must in the end be channelled through the individual, as it is the individual who ultimately commits the act of violence. Therefore, a psychological perspective on violence will also be indispensable to any discussion of the issue and is of particularly relevant to developers of interventions that are intended to work at an individual level (rather than those aimed at changing the wider social or economic context). One component of the psychological approach is related to ‘nature’, where the capacity and inclination for violence is ‘hard-wired’ into humans. This overlaps with biological perspectives discussed above and is focused on biological and dispositional vulnerabilities and dysfunctions. The other core component of the psychological approach is related to ‘nurture’, those elements of behaviour that are the result of experiences, including an individual’s particular life history and the experiences shared with a social group. This second component of the psychological model may be of particular interest to intervention developers as it is natural to think that tendencies that have developed as the result of environmental influences as opposed to biological ones could potentially be more easily counteracted or reversed.

From a developmental perspective, a disrupted or traumatic attachment relationship between a child and their caregiver can result in the underdevelopment an individual’s ability to self-regulate, leading them to a more extreme response to perceived threat or frustration, including increased levels of aggression and violence. The attachment relationship is likely to be negatively affected by hostile and punitive parenting styles, particularly if combined with inconsistency, leading to maladaptive coping mechanisms. Issues with attachment and the potential for shame and anxiety are also seen as potentially leading to violence if an individual feels abandoned or feels at risk of abandoned, one explanation given for intimate partner violence. More generally shame, which may be the result disrupted attachment relationships or other experiences, is seen as motivating violence through the need of individuals to regain a sense of power and self-esteem (though generally self-esteem is as the result of behaviour rather than a driver of it).

The other major area of theory regarding the cause of violence relates to cognitive frameworks, defined as sets of information or beliefs that unconsciously influence how someone interprets new information and responds to it. Within this realm, ‘cognitive distortions’ and ‘pathological belief systems’ lead people to either mis-interpret situations (for example perceiving threats where they don’t actually exist) or to believe that situationally inappropriate responses (such as aggression and violence) are in fact justified and reasonable. These ‘social-information-processing’ structures can be the result of adverse childhood experiences, such as disrupted attachment patterns or abuse.

A sociological and historical perspective would link cognitive patterns that promote aggression and violence to cultural context. For example, aggressive responses have been shown to be more likely in cultures of ‘honour’, ie where a high value is placed on an individual’s reputation and overt indicators of respect (which in evolutionary and game theory terms is seen as creating a ‘credible threat’). One experiment, for example, tested the response of people who were involved in a minor altercation (deliberately instigated by the research team) found that people from the southern American states had a far greater physiological reaction to the incident than those from the north. Other cultural factors that have been associated with aggression and violence have been ‘myths of victimisation’, where a particular individual or social group is seen as been treated unfairly by another social group.

Contextual factors linked with violence and aggression that are mediated by psychological processes include the presence of alcohol. Alcohol is seen as having a disinhibitory effect, meaning that it primarily does not cause people to be aggressive but amplifies existing tendencies to aggression where they exist. This disinhibitory effect has a basis in physiology but also can also act in a social permissive way, with both the individual and onlookers feeling that aggression is more likely and more acceptable in the presence of alcohol. More immediate contextual factors that could be seen as triggers for aggression and violence include stress, frustration and pain, which can lead to an immediate response or retaliation but is also seen as been expressed as ‘displacement’ aggression, with the aggression being aimed at someone other than the individual that has caused the frustration or injury.

Sitting across all the different psychological explanations for violence are typologies of the immediate causes or goals of violence. One typology (the Four Roots of Evil) identifies four goals of violence: instrumental violence to achieve a particular material outcome; in response to threats to an individual’s ego or reputation; to punish individuals perceived to have committed a crime or morally erred; for sadistic pleasure or some intrinsic reward from the violent behaviour itself. Related to this is the concept of the ‘algebra of aggression’, which breaks down aggression into component parts that form an unconscious cost-benefit analysis that determines if an individual is aggressive and to what degree. The components are: the instigation to aggression (angry vs instrumental); habit strength (influencing the nature of the response); inhibitions to aggression; and, stimulus factors.

An integrated model of the causes of violence

The literature on the causes of aggression and violence reflects the wider literature on behaviour and behaviour change in that explanations and theories have been developed within different disciplines and there has been limited attempts to provide a holistic understanding. Indeed, the different theories have sometimes been positioned as ‘rival’ or ‘alternative’ theories. Using the integrated model of behaviour, the causes of aggression and violence discuss have been categorized in table 1 below. The intention is that these causes can be linked to the mechanisms involved in violence prevention programmes and help developers create effective interventions and evaluators explore the barriers and facilitators to the impact through quantitative and qualitative approaches.

Table 1: Causes of violence categorized using the integrated model

This text is also available as an appendix to Barnard, M. 2023. The Integrated Model of Behaviour (pdf). London: ICF [Available at: https://www.researchgate.net/publication/374675385_The_Integrated_Model_of_Behaviour]

References

King, B. 2012. Psychological Theories of Violence. Journal of Human Behaviour in the Social Environment, 22:553-571

Moser, C. and Shrader, E. 1999. A Conceptual Framework for Violence Reduction. [pdf] Washington: The World Bank [Available at: https://documents.worldbank.org/en/publication/documents-reports/documentdetail/774791468782146689/a-conceptual-]

Pinker, S. 2011. The Better Angels of our Nature. New York: Viking

Sapolsky, R.M., 2017. Behave: The Biology of Humans at Our Best and Worst. New York: Penguin Press

World Health Organisation. 2015. Preventing youth violence: an overview of the evidence. [pdf] Geneva: World Health Organisation [Available at: https://www.who.int/publications/i/item/preventing-youth-violence-an-overview-of-the-evidence]

Youth Endowment Fund. 2022. Statistics update (May 2022): The latest data on police recorded crime, youth victims of homicide and modern slavery. [pdf] London: Youth Endowment Foundation [Available at: https://youthendowmentfund.org.uk/reports/statistics-update-may-2022-the-latest-data-on-police-recorded-crime-youth-victims-of-homicide-and-modern-slavery/]