This CREST report explores the intra-group mechanisms through which members of extremist groups contribute to, establish, and maintain parameters on their own violence.

Why do some ‘extremists’ or ‘extremist groups’ choose not to engage in violence, or only in particular forms of violence? Why is it that even in deeply violent groups there are often thresholds of violence that members rarely if ever cross? 

Part of the answer will lie in the counter-measures deployed by state and non-state actors, and in other external constraints that inhibit the opportunities, capabilities and motivation of such groups to deploy violence. Yet the fact that few if any groups carry out as much violence as they are capable of indicates that in most cases external constraints comprise only part of the answer. 

The basic premise of this report is that another part of the answer lies in what we call the ‘internal brakes’ on violent escalation: the intra-group mechanisms through which group members contribute to, establish, and maintain parameters on their own violence. 

Such internal brakes are often evident in detailed accounts of decision-making within groups that use or flirt with violence, yet they are rarely examined systematically.

This report presents a descriptive typology of the internal brakes on violent escalation that could provide a basis for more systematic analysis of such brakes.

We explore three very different case studies to construct, test, and refine the typology: the transnational and UK jihadi scene from 2005 to 2016; the British extreme right during the 1990s, and the animal liberation movement in the UK from the mid-1970s until the early 2000s (see Annexes A–C).

Drawing across the literature on social movements and contentious politics, terrorism studies, peace studies, and the sociology, psychology and anthropology of violence, we also undertake a review of existing research into the inhibition or non-emergence of violence, how violence ends or declines, and intra-group dynamics during processes of escalation, de-escalation and non-escalation.

The typology

The typology is based around five underlying logics on which the internal brakes identified in this project operate:

  1. Strategic logic
  2. Moral logic
  3. The logic of ego maintenance
  4. The logic of out-group definition
  5. Organisational logic

While these logics sometimes coincide with one another, at other times they operate independently and sometimes appear even to contradict one another. Each of these underlying logics underpins a brake, which in turn is associated with a series of ‘sub-brakes’. 

Strategic logic

Identification of non- or less violent strategies of action as being as or more effective than more violent alternatives.

  • Expressions of scepticism about their ability to beat their opponents in a violent struggle, including concerns that greater militancy will increase backlash or repression from opponents or the state towards them and their supporters.
  • Expressions of concern that violent escalation will undermine support for the group.
  • Attempts to build or maintain ties with strategically useful allies who are not supportive of violent escalation.
  • Identification of political opportunities that favour (re)adoption of non- or less violent strategies of action.
  • Identification of non- or less violent strategies of action that are perceived to be effective, including the identification of ‘sufficient’ levels of violence.

Moral logic

Construction of moral norms and evaluations that inhibit certain forms of violence and the emotional impulses towards violence (e.g. revenge).

  • Articulation and performance of general moral norms and principles that problematise certain forms of violence, require violence to be justified or enable activists to forestall on entering the ‘tunnel of violence’ (e.g. the conception of violence as a tactic of last resort; positioning non-retaliation as a virtue; emphasising values such as mercy and compassion).
  • Identification of some groups of actors as illegitimate targets for violence.

The logic of ego maintenance

Self-identification as a group that is either nonviolent or uses only limited forms of violence.

  • Production of group narratives that emphasise non-violence or the limited use of violence either by themselves or by those they claim have inspired their movement.
  • Disassociation from more violent groups or factions and/or association with less violent groups or factions.
  • (The threat of) sanctions for activists who advocate or undertake violence beyond the established parameters of the group’s action repertoire, and/or opportunities to achieve intra-group respect and prestige without undertaking or encouraging the use of violence at or beyond the parameters of the group’s action repertoire.
  • Circulation of limited expectations that they will be involved in greater levels of violence.

The logic of out-group definition

Boundary softening in relation to putative out-groups (e.g. opponents, opponents’ perceived supporters, the general public or state actors).

  • Resistance to generalizations about their opponents.
  • Identification of segments of the public beyond their previously-imagined support base as potential converts to their cause.
  • Limited intra-movement pressure to ‘burn-bridges’ with social contacts outside of the movement or outside of the radical flank of the movement.
  • Expressions of reluctance to conceive of the state security forces as ‘the enemy’.

Organisational logic

Organisational developments that either (a) alter the moral and strategic equations in favour of non- or limited violence, (b) institutionalise less violent collective identities and/or processes of boundary softening, and/or (c) reduce the likelihood of unplanned violence.

  • Limited investment in capabilities to escalate violence, and/or development of capabilities to undertake strategies of action that either entail non- or limited violence or more controlled violence.
  • Foregrounding more modest or intermediate objectives and de-prioritising revolutionary goals.
  • Construction and maintenance of spaces in which a range of activists that includes and extends beyond the radical flank are able to freely discuss tactics and movement objectives.
  • Concerns among some group members that violent escalation will compromise their ability to shape the movement’s direction and/or negatively affect their position within it.
  • The concentration of energy on targeting movement rivals, leading to reduced capability to prosecute campaigns of violence against their external enemies.

Discussion and conclusions

While there was considerable cross-case variation in terms of the distribution and effectiveness of the internal brakes – as might be expected given the case studies selected – we were able to develop a typology that, across the three case studies and the wider reference literature, enabled us to describe and categorise the practices though which group members sought to establish and maintain the parameters of their own violence.

There are important limitations with regards to how this typology might be used. Foremost among these, it should be emphasised that the typology cannot be deployed as a straightforward ‘checklist’ with which to make inferences about the risk of violence.

This in part is because the (increased) presence of internal brakes within any given case might be open to a number of possible interpretations: it might indicate a limited risk of violent escalation due to extensive intra-movement opposition to such escalation, but it might also indicate that there are increasingly active attempts within the movement to escalate violence, or simply that there are growing intra-movement tensions.

We believe however that this descriptive typology can provide a vocabulary with which researchers and analysts can begin to investigate and better understand important questions about how members of extremist groups contribute to establish and maintain the limits on their own violence. Furthermore, by organising such analysis around the underlying logics on which these brakes work, it can help us to understand how the brakes work, and when and how they might be more likely to fail. 

Proposed applications of this research

  • For academic researchers interested in the dynamics of conflict and political violence, we propose that this descriptive typology can be used to advance and stimulate research into processes of non- or limited escalation. While escalation and de-escalation have received considerable academic attention in recent years, non- or limited escalation has not. 
  • For security, intelligence and law enforcement practitioners working in areas of risk assessment, we propose that this typology can provide a further tool with which to identify indicators of the propensity towards and away from particular forms of violence by specific groups or sub-groups.
  • For security, intelligence and law enforcement practitioners undertaking interventions with extremist groups, we propose that this typology can be used to inform assessments how externally applied counter-measures might interact with and sometimes undermine internal brakes.

Building on this typology, we believe that particularly productive avenues for further research and analysis are likely to relate to:

  • The conditions under which certain brakes, or configurations of brakes, are more or less likely to be effective.
  • How the patterns and functioning of internal brakes are affected by wider conflict dynamics and vice versa e.g. how they affect and how they are affected by interactions between group members and state security services, opposition groups etc.
  • How and why the distribution of brakes varies across groups and what, if anything, this tells us about their propensity for violence.

How the internal brakes on violent escalation operate at different points within waves or cycles of conflict and what the implications of this are in terms of how we might use analysis of the internal brakes on violent escalation to assess risk and guide interventions with extremist groups.