The Moral Ecology of Extremism

The question of what drives extremism has generated a wealth of observations,
relating to the characteristics of people who adopt extremist beliefs and engage
in extremist behaviour, and to the features of situations that encourage the
adoption of such beliefs and the engagement in such behaviour.
Concerns about the harmful potential of new technologies have fuelled a
renewed interest in the environmental drivers of extremism. Exposure to
violence-supportive narratives, discrimination, migration, disinformation,
cultural norms, the mobilisation of social identities by social movements – these
are only some of the exogenous contributors hypothesised to play a part.
This multiplication of drivers begs the question of knowledge integration. How
do they fit together? Are all individuals equally exposed and permeable to their
effects or are some of us more ‘at risk’ than others? On what basis should we
assess whether social, political, technological, economic or cultural change
might suppress or support extremism?
This paper does not enumerate all possible contributors to extremism. Rather,
it addresses some conceptual challenges in this problem space; namely, a lack
of clarity as to problem definition, and the limited integration of individual and
exogenous accounts of extremism. Inasmuch as these questions remain
unresolved, they contribute to the persistence of the problem in and of
themselves.
At the outset, the argument is made that the definitional problem is not trivial
and that perpetually moving the goalposts risks turning extremism into an
insoluble policy issue. As such, a case is put forward in favour of a narrower
problem statement. The requirements of an integrative approach are outlined
and an inference framework is used to structure current knowledge about the
drivers of extremism, grounded in a functional account of morality.1
The paper
concludes that an effective counter-extremism strategy must adopt a systemic
perspective to address the emergence of the moral ecologies that foster this
phenomenon.
Beware wicked problems
Problems come in kinds. Simple problems are straightforward. Complicated
problems have many moving parts, but the relationships between them are
THE MORAL ECOLOGY OF EXTREMISM
knowable. These problems can usually be tamed, given an adequate
commitment of resources. What sets complex problems apart are the non-linear
relationships between their constituents.2
They involve concepts that are fuzzily
defined. They are embedded in systems with emergent properties, which
interact in unpredictable ways. Programmes to tackle these problems tend to
have unforeseen effects.3

Many social problems are complex.4
Understandingly, it can be tempting to
reframe them as complicated problems that can be solved if enough money
and effort are spent. When this approach fails, we might choose to move the
goalposts.
9/11, 7/7 were traumatic events. Their consequences were so staggering, while
at the same time the parts implicated in bringing them about seemed so
knowable (men directed by a terrorist organisation known as Al Qaeda), that a
standard response to dealing with complicated problems was implemented:
massive resources were invested to counter terrorism. Yet, framed as a military
matter, the problem was ill-posed. When interventions abroad failed to quell
terrorism, the problem was brought home and reinterpreted as one of homegrown radicalisation. A few iterations of the Prevent Strategy later, however,
terrorism remained a threat and this was seen as evidence that the target had
not been broad enough. The real problem was extremism, the “root cause” of
many social ills, above and beyond terrorism.5

Once more, the goalposts shifted.
A complex problem becomes wicked when it is characterised by piecemeal,
clashing, constantly changing requirements.6
The perception of the problem
itself is contentious and malleable: definitions are disputed or elusive. Because
there are no boundaries, it is impossible to say where the problem stops and
where the solution begins. Interventions have unintended consequences on
other, interlocking problems. The very claim that there is a problem may be
challenged.
This seems a fair description of the problematisation of extremism today.7
At best, extremism is well on its way to becoming a wicked problem. At worst, it is
one already.
THE MORAL ECOLOGY OF EXTREMISM 6
Problematising extremism
Agreeing consensual problem definitions is difficult. Policy, the law, science, civil
society at large each have their own requirements. Consensus may not even be
desirable, because a definition is good in relation to its purpose, a point wellmade by Lord Anderson of Ipswich in his recent Treasurer’s Lecture on
Extremism and the Law.

A good definition from the point of view of the law or
policy may not be of service to science and vice versa.
How, then, to conceive of a (wicked?) problem like extremism for the purpose
of uncovering its drivers, in order to inform a national strategy to counter it?
As a starting point, we can ask ourselves: what are the manifestations of the
phenomenon, which we would like to suppress? We are concerned with
extremism because it has harmful social effects, brought on by the actions of
extremists, actions which contravene certain categories of moral norms and
rules of conduct.9
It is worth returning to Lord Anderson, who made three very important remarks.
The first is that most of the activities we think of as harmful extremist behaviour,
from hate crime to terrorism, already fall under criminal law. The second is that
we should resist policy change that “leaves citizens unclear whether their
actions or projected actions will be judged to be objectionable”, as this “risks
undermining the rule of law.”10 The third remark, in the form of advice to the
Commission for Countering Extremism, is that “[w]hen it comes to
recommending new offences or other coercive measures, [the Commission
should] work with the grain of what is already there: just because extremism is
a word does not mean that it is a useful legal concept.”11
Let us take on board the last remark, first: because a concept gains traction in
some domain does not mean that it designates a phenomenon characterised
by a coherent set of properties or a specific set of causes.
Next, if we are concerned with stopping citizens from engaging in certain
harmful actions, then the first and best mechanism we have in place is clear
moral guidance: explicit rules of conduct that state what is considered right or
wrong to do, attached to unambiguous consequences should they be broken.
In functioning societies, the law is the most effective crime prevention
technology.
Lastly, if most of the extremist actions we wish to prevent are already
criminalised, then reducing extremism to unlawful extremist actions would take
into account much of what we find problematic about the otherwise amorphous
THE MORAL ECOLOGY OF EXTREMISM 7
and ill-defined phenomenon. To establish what is extremist, “one requires a
benchmark.”12 Arguably, the most explicit benchmark we have is the law.
Hence, in the interest of crafting a problem statement that is solvable, this paper
concerns itself with explaining why some individuals come to see committing
acts of unlawful extremist behaviour as morally legitimate and choose to carry
them out.
This problem statement has a number of advantages.
1. It conceptualises extremism in such a way as to be applicable regardless
of time period and jurisdiction: i.e. as the breaking of rules of conduct
set out in the laws of a given society.13

2. It defines its object in terms of what is observable (unlawful actions).
3. It does not refer to any specific set of values (e.g. particular religious or
political beliefs) leading to any specific set of violent or non-violent
actions, which means that the ensuing discussion is relevant to all
unlawful extremist behaviour, regardless of driving ideology. The actions
considered do not have to be circumscribed to violent actions and the
account remains applicable even if the legal definition of extremist
behaviour changes over time. It is left up to society to define what
specific actions are objectionable and should be lawfully proscribed, in
the same way that any society – not its academics – decides what actions
constitute any type of crime.
4. It does not refer to any state of mind required for an act to be considered
extremist. In this way, the problem statement avoids the possible
tautological pitfall of containing both the description of the object and
its explanation. Extremist behaviour may be caused by particular social
dynamics (e.g. perception of threat from an out-group)14, states of mind
(e.g. need for cognitive closure) 15 or motivations (e.g. desire for
existential significance)16, but prior to analysis, we do not know this. It is
what we are trying to find out.
To sum up: extremism is, by definition, a relational term.17 What a society
perceives as extremist is subject to change and may differ widely from what
another society would consider extremist. 18 What remains is that actions
deemed extremist contravene a set of norms, both formal (legal rules of
conduct) and informal (social norms), and that the most objective measure of
the normative benchmark we have at our disposal is the law. Consequently, one
THE MORAL ECOLOGY OF EXTREMISM 8
way to reduce extremism to a useful concept, for the purpose of discovering its
causes and in order to inform a national counter-strategy, is to define extremist
acts as what is deemed to be unlawful extremist behaviour in a given
jurisdiction.19

Does everything matter?
A recent brief by the European Union’s Radicalisation Awareness Network (RAN)
states that violent extremism is best understood as the outcome of a
“kaleidoscope of factors”, which create “infinite individual combinations”. 20
While the document refers specifically to violent extremism, it is fairly reflective
of an approach to the explanation of extremist behaviour, whereby factors at
various levels of explanation associated with the problem are described, but the
question of their combined interplay is not addressed in detail.
In her review of research on home-grown radicalisation in Europe 21 , Anja
Dalgaard-Nielsen observes that explanations for violent extremism fall into
three categories: sociological accounts, social movement and social network
theories, and largely atheoretical, empirical individual-level accounts. She
concludes that, while each category of accounts addresses important elements
of the radicalisation process, all fail to tackle the problem of specificity.22

The Commission for Countering Extremism identifies this same problem in its
Terms of Reference, when it states:
There is […] widespread academic disagreement over the drivers
and causes of extremism. It is also the case that the majority of
people who share what are assumed to be drivers do not go on to
engage in extremist activities or behaviour. 23 (emphasis added)
Dalgaard-Nielsen suggests that the three types of accounts should be seen as
complementary, rather than competing. In the same vein, Schmid24 contends
that radicalisation studies have put too much emphasis on the micro (individual)
level of analysis, and that full explanations should integrate the meso (group
and community) and macro (structural) levels.
How, then, should we carry out this integrative task, in order to make sense of
the “infinite individual combinations” of factors implicated in the explanation of
extremism?
THE MORAL ECOLOGY OF EXTREMISM 9
Ours is not to reason why, but where
The fragmentation of causal accounts is a familiar problem to criminologists25
and much recent effort has gone into developing empirically-grounded
theories, which articulate the “rules of interaction”26 between personal and
external drivers of crime. A fundamental take-away is that much of the risk
associated with this kind of complex social problem emerges from the
interaction between individual and context. Whether someone will become
involved in crime depends in a large part on their environment – where they
are, as much as who they are.
Indeed, some of the strongest evidence available in criminology relates to the
temporal and spatial concentration of crime, to the socio-physical
characteristics of criminogenic places, and to the causal effects of place-based
mechanisms on crime and criminal development.27 It has also been observed
that where crime concentrates, so do other kinds of disadvantage, like health
problems and psychological distress.28 The very notion of Prevent priority areas
implies that radicalisation, too, is thought to concentrates geographically, and
early research findings support this.29
If there is a way to address the problem of specificity with regards to extremist
behaviour, it will require that we understand what kinds of individual drivers
matter in what kinds of contexts, as well as understanding how these contexts
emerge.
The inference framework
In 2010, ahead of the revision of the Prevent Strategy, the author and her
colleague, Professor Per-Olöf Wikström, were commissioned by the Office of
Security and Counter Terrorism to synthesise the knowledge-base on the
causes of Al-Qaeda-influenced radicalisation.30 As a point of departure, they
took a theory of moral action and crime causation known as Situational Action
Theory (SAT). 31 Since then, this work has been the foundation of several
sophisticated studies on the individual and ecological causes of extremism,32
and of a number of research projects on lone-actor terrorism risk33 and the
emergence of radicalising environments.34

To enable the integration of knowledge from other behavioural science
domains and to support the operational and strategic analysis of risk associated
with violent extremism, the original theory was developed into the S5
inference
THE MORAL ECOLOGY OF EXTREMISM 10
framework (Figure 1). Rather than enumerate all known drivers or possible risk
indicators of extremist behaviour (a theoretically infinite, ever-changeable list),
S
5
sets out how five key categories of determinants interact to generate or
suppress the risk of extremist propensity development and extremist action.
Extremist propensity as moral propensity
To explain how an individual acquires an extremist propensity is to explain 1)
how they come to see extremist actions as morally legitimate and 2) how they
fail to develop, or do not make use of, their capacity for self-regulation, leaving
them more likely to choose to carry out an extremist act, in certain situations.
Propensity is first and foremost a question of one’s commitment to actionrelevant moral rules of conduct, because much of human behaviour is guided
by rules about what is right or wrong to do – acceptable or unacceptable – in a
given context.35 Evolutionary biologists hold that we developed systems of
moral values in large part because of the need to choose the most
advantageous between alternative courses of action, by anticipating their
consequences, and because of the need to predict the behaviour of others to
enable social cooperation.36

This functional perspective seems congruent with a number of observations,
such as the fact that social identities appear rooted in specific sets of moral
guidelines, values and judgements;37 that individuals are more sensitive – to a
stressful, physiological degree – to in-group moral disapproval compared to
outgroup moral disapproval; 38 that groups act as “moral anchors” in
competitive and conflicting moral environments, and that contravening group
moral rules can have highly negative social consequences for their members.39
Also congruent is research suggesting that claims of moral superiority in
reaction to perceived threats to group social advantage are central
characteristics of extremist belief systems.40 This hyper-salience of morality may
also explain why societies react to antinomian crimes – that not only breach,
but aim to overthrow normative consensus (like violent extremism; terrorism;
certain forms of paedophilia) – with acute disapproval and severity.41

At the individual level, recent experiments in human cognition indicate that
morality constrains people’s representations of what they think of as possible
to do, to the extent that people tend to default to treating immoral events as
impossible.42 That is why understanding how people come to see extremist
action as morally justified is so crucial for an account of extremist behaviour. In
THE MORAL ECOLOGY OF EXTREMISM 11
a nutshell: to see extremist action as morally legitimate is to see it as something
that one could possibly do.
But while extremist moral change, commonly known as radicalisation, raises the
potential for future action by making extremist actions appear as possible, it
does not necessitate it. To explain how an individual actually engages in
extremist action is to explain how they find themselves in a situation that 1)
motivates them to act; 2) leads them to perceive extremist action as a legitimate
and advantageous alternative in the circumstances; and 3) leads them to choose
to carry it out then and there.43

Both processes – radicalisation and action – should be explained independently,
because it is possible (in fact, likely) for a radicalised individual never to choose
to carry out an extremist action, and it is possible for an individual who is not
radicalised to carry out an act of extremism in certain situations, such as, for
example, intense peer pressure or coercion.
People and contexts
Four of the five levels of analysis of the S5 framework are concerned with context,
meaning that exogenous drivers are seen as fundamental contributors to the
explanation of extremist behaviour. Even at the individual level, it is those
factors that impact the degree to which an individual is susceptible (or not) to
salient contextual influences, which are theorised to matter most.
To explain why some people and not others, who can appear similar in many
ways, acquire an extremist propensity or engage in extremist behaviour is
effectively to explain why certain kinds of people find themselves in certain
kinds of contexts at certain times.44
The main purpose of S5
is to guide the formulation of inferences about what
kinds of people in what kinds of contexts at what times should be considered ‘at
risk’.
In the remainder of this paper, the role of each key category of determinants is
outlined and plausible contributions for some of the factors and processes
reportedly associated with extremist behaviour are inferred on that basis. As
stated, the aim is not to enumerate everything, but to structure thinking about
the role of suspected contributors.45

THE MORAL ECOLOGY OF EXTREMISM 12
Figure 1. The S5 inference framework
A multilevel analysis of extremism
Susceptibility
At the individual level, the key determinant is susceptibility to moral change.
Evidence of differential susceptibility – the notion that some individuals are
more susceptible to environmental influence than others 46 – has been
documented in several domains. Most importantly for us, there is reason to
think that individuals differ in their susceptibility to 1) the extremist socialising
influences in their environment; and 2) situational features that support the
emergence of the motivation and choice to engage in extremist behaviour.
Differential susceptibility to crime has been studied at length. The main
determinants have been identified as weak commitment to law-relevant
morality, inappropriate moral emotions, poor self-control, and several
neuropsychological characteristics, such as impulsivity, thrill-seeking, poor
affect regulation, cognitive rigidity, attention deficit and other evidence of poor
executive functioning.47

Given difficulties associated with research on extremism, findings are somewhat
less robust, but they suggest similar processes may be involved in individual
susceptibility to extremism. In a large survey of young Belgian adults, low selfcontrol had the strongest direct effect on self-reports of political aggression,
THE MORAL ECOLOGY OF EXTREMISM 13
regardless of ideological leaning.48 Cross-sectional research looking at the link
between extremism and common mental illness finds that past criminal
convictions are a key risk factor among White British and Pakistani populations
in England, alongside comorbid anxiety and depression. The authors interpret
this to mean that receptivity to extreme ideologies is related to poor selfregulation. 49 Most recently, psychological experiments have suggested that
cognitive inflexibility is a reliable predictor of extremist attitudes, including
endorsement of violence and willingness to die for a group.50

Regarding the common roots of criminality and extremism, analysis of data
from a large longitudinal study suggests that people who already see violence
and rule-breaking as justifiable are more vulnerable to involvement in violent
extremism, especially when they live in environments characterised by conflict,
repression, or social and economic strains.51 No other measure, whether related
to generalised trust, parental involvement, deviant peer group or violent media
consumption proved a better predictor, with the exception of gender. Further
longitudinal research has found that both non-violent support for a right-wing
extremist ideology and the potential for violent extremism are associated with
weak commitment to law-relevant morality and a poor ability to exercise selfcontrol.52 Additionally, a recent examination of men convicted of violent hate
crime has found that these individuals have extensive criminal careers involving
a wide variety of criminal offences and tend to be even more criminally prolific
than other (non-hate crime) violent offenders.53

These findings are not surprising, if we consider that many extremist actions are
criminal in nature. Much has been made of a new nexus between crime and
violent extremism, because a significant number of individuals implicated in
terrorist offences have criminal antecedents.54 Given the above, more than a
crime-terror nexus we might speak of a crime-crime nexus, inasmuch as the
same susceptibility appears to be a driver for both types of offending.55

The existence of a differential susceptibility to unlawful extremist behaviour,
rooted in weak commitment to context-appropriate rule-guidance, and
compounded by poor executive functioning, 56 makes sense of conflicting
findings as to the importance (or lack thereof) of a host of exogenous factors,
such as exposure to extremist content online and offline, individual and
collective strains (e.g. those related to loss of personal significance or to
perceived threats to sacred group values), influence of extremist peers, social
disintegration, community polarisation, declining perceptions of legitimacy,
collective anomy, migration experiences, and so on.57 Their effect, if any, are
likely to be conditioned on the susceptibility of the individuals exposed.
THE MORAL ECOLOGY OF EXTREMISM 14
Crucially, however, susceptibility is a dimension, not a category. People who
have exhibited heightened susceptibility in the past (e.g. criminals) might be, in
a manner of speaking, the ‘canaries in the coal mine’: people especially reactive
to moral strain, attracted to groups claiming that commitment to their moral
system will reduce the cognitive pressures these individuals are often illequipped to handle, and will lead to more advantageous action choices than
these individuals have made in the past (and which, on occasion, have landed
them in prison). That is not to say that no one else can be susceptible.
In order to explain why some susceptible individuals end up adopting extremist
beliefs and engaging in unlawful extremist actions, while other susceptible
individuals do not, or why some individuals, who, on the face of it, do not seem
as susceptible are nevertheless drawn into extremism, we must turn to the role
of context, which is where the other four levels of explanation of S5
come in.
Selection
When we think of vulnerability to extremism, we might think first of all about
the sort of susceptibility previously discussed: characteristics which make some
people more susceptible to extremist influence. Yet, to be truly vulnerable to
something, one needs to be at risk of coming into contact with it. In other words,
one needs to be at risk of exposure.
58 Much as we need to understand why
people vary in their susceptibility to extremism, we need to understand why
they vary in their risk of exposure to extremism-enabling environments. We
need to understand selection.
Briefly put, social selection operates on the basis of social forces that encourage
or compel, or discourage or bar, certain categories of people from taking part
in particular kinds of place-based activities.59 Residence and socio-economic
status are some of the most common factors of social selection. Living in a
particular neighbourhood or belonging to a particular social group (ethnic
group, religious, professional, and so on) affects the chance of exposure to
certain places and the participation in certain activities. If these place-based
activities lead to contact with extremism-supportive influences, then social
selection becomes a factor in the explanation of why some people rather than
others become extremists.60

If social selection sets the stage for exposure, people also choose to spend time
in particular settings as a result of personal preferences acquired over their
lifetime. This we can call self-selection.
61 If I am barred from taking part in a
political rally because I am a woman, I am subject to social selection on the basis
THE MORAL ECOLOGY OF EXTREMISM 15
of gender discrimination; if I choose to attend the meetings of a local women’s
group because I prefer the company of other women, I am enacting selfselection.
Selection is the bridge between individual and environmental levels of
explanation. Given the way social life is organised (its routines; its structures; its
norms) and given where extremism-enabling settings are found, some
categories of people will be more at risk of exposure compared to the rest of
the population. This may be why, at different times and in certain places,
individuals who belong to certain group – young people, residents of particular
neighborhoods, criminals, students, the socially disadvantaged, inmates, certain
social media users, and so on – have been (reportedly) over-represented among
extremists.62

Selection is likely the key process which explains why it is possible to find
statistical profiles of extremists in particular places at particular times, yet a
stable, general profile remains elusive.63 In any given environment, extremismenabling, place-based activities will select for similar kinds of people. But these
activity settings differ across environments and can displace over time, so the
kinds of people selected for exposure and the kinds of moral systems they are
exposed to will also vary.
Selection is also likely to be part of the explanation why individuals who do not
seem especially susceptible to moral change end up adopting extremist beliefs.
They may possess characteristics that select them for habitual exposure to
extremism-enabling settings. Even if their susceptibility threshold is relatively
high, sustained, effective exposure could result in moral change over time.64
Vulnerability to extremism, then, is context-dependent, because it combines
individual susceptibility characteristics, which may be relatively stable, and
selection susceptibility characteristics, which are likely to vary across place and
time.65

Settings
People are socialised and act in places, even if these places are virtual, so what
makes an extremism-enabling place? Observations would suggest that settings
which enable the acquisition of a morality supportive of extremism are
characterised by certain socialising affordances.66
THE MORAL ECOLOGY OF EXTREMISM 16
Cognitive affordances
Reportedly, extremism-enabling settings have features that can bring about
cognitive states which make people more amenable to the adoption of new
moral beliefs, such as features that mobilise their attention and foster cognitive
receptivity. 67 Various types of perceptions have been associated with this
process, such as loss of effective control, feelings of insignificance, fear, and
more generally any features, which provoke experiences that threaten one’s
confidence in their own rule-system as the best guide for advantageous action
– such as perceived injustice, alienation, loss of relative social status, or threat
to survival.68

Cognitive affordances do not have to foster only negative perceptions.
Cognitive ease, 69 if associated with extremist discourse, may support
extremism-enabling learning. Also relevant here are those features that can
bring about habituation by encouraging repeated exposure to the setting (such
as addictive features),70 or otherwise heighten the intensity and effectiveness of
exposure (such as discouraging exposure to counter-opinions).71
Moral affordances
By definition, these settings allow for exposure to extremism-supportive moral
norms. They offer discursive opportunities to promote ideas, which characterise
extremist behaviour as morally legitimate72 and extremist moral systems as the
best possible guide for action. Depicting particular group values and
commitments as morally superior is one way of achieving these effects.73 This
may be conveyed through narrative devices, which can be broadly characterised
as transcendental (about ‘meaning of life’ stuff), categorical (‘black-and-white’)
and prescriptive (action-orientated).74
The prescriptive, action-orientated dimension of extremist narratives (giving
people ‘things to do’, not just ‘things to believe’) is likely a key feature of
successful extremist socialising discourse. 75 Together, these characteristics
contribute to the entativity of moral systems promoted by extremist groups,
making the extremist ideology seem clear and actionable for individuals who
might otherwise struggle with the cognitive cost of uncertainty.76 Repeated
exposure to such settings may result, over time, in normalisation – the
perception that a moral commitment, which may have been considered ‘outside
of norms’, is now the acceptable standard. Algorythmic technologies, to the
extent that they contribute to the emergence of moral “filter bubbles” have
been hypothesised to be one such normalising feature.77

THE MORAL ECOLOGY OF EXTREMISM 17
Attachment affordances
Socialisation operates within a web of micro-interactions. For most people, the
agents of socialisation with the greatest influence are their parents or guardians.
Effectiveness of family socialisation practices is in large part conditioned by the
strength of the child’s attachment to his guardians. That attachment, in turn, is
a function of the caregiving relationship between child and guardians. Humans
tend to get attached to the people who provide for their physical and emotional
well-being. Eventually, people form other attachments with peers, teachers and
spouses, who care for them and who come to have their own socialising
influence.78

Hence, extremism-enabling settings are likely to be effective to the extent that
they foster attachment to individuals who already hold extremism-enabling
moral beliefs.79, 80

Social control affordances
Lastly, settings that support extremist socialisation are characterised by
ineffective law-relevant social control. It may be that individuals with regulatory
authority over the setting lack the means or the willingness to enforce pro-legal
norms, or that extremism-supportive individuals have asserted their own
control over the space.81 In that case, formal and informal mechanisms for
behavioural monitoring support the extremism-enabling moral context, rather
than suppress it.
Discussions about Internet governance and extremism revolve increasingly
around the idea that certain online settings, given what new technologies afford,
are beyond the reach of state enforcement and deterrence, or that they foster
a breakdown of informal social control on the part of private citizens,
contributing to the normalisation of extremism-supportive discourse.82, 83
Social Ecology
Extremism-enabling settings are not equally distributed in space and time.
Online, they are associated with particular platforms, forums or websites. Offline,
some streets, neighbourhoods, boroughs, cities, societies, even some countries
have more of these kinds of settings at particular times.84 Claims that there are
‘hotbeds’ or ‘hotspots’ of extremist activity are a mainstay of the conversation
on extremism.85 Concern about prison radicalisation flares up periodically, but
THE MORAL ECOLOGY OF EXTREMISM 18
even then the problem is not evenly spread out across the penitentiary
system.86

The observation that extremist settings concentrate suggests that there are
processes at work in certain contexts, which encourage (or fail to suppress) the
emergence of such settings.
If we hold that:
 extremism is characterised in relation to other moral norms and rules of
conduct;
 unlawful extremist behaviour is an instance of moral action;
 individuals with a weak commitment to law-relevant moral rules of
conduct are more susceptible than others to developing an extremist
propensity;
 extremist moral systems are attractive to individuals experiencing moral
friction and perceived loss of advantageous moral rule-guidance (i.e.
moral strain);
 perception of moral superiority is a key feature of extremist normative
systems and groups;
 particular moral contexts are determinant characteristics of extremismenabling settings
then, to understand the conditions from which extremism emerges and upon
which it thrives, we should focus on those factors that shape moral ecologies.
In thinking about these factors, the following rule-of-thumb applies: to
anticipate or evaluate the impact of any given social ecological (broadly
speaking, community-level) feature, we can ask ourselves how that feature
could support or suppress the emergence, convergence and maintenance of
the aforementioned cognitive, moral, attachment, and social control affordances.
Research on social ecological processes of social disorder, as well as work on
extremist ‘hotspots’, points to several plausible processes. 87 At this level,
attention should be paid especially to changes – including technological
innovations – which affect social segregation (keeping people apart on the basis
of ethnicity, religion, culture, and so on) and foster perceived group
competition; increase social disorganisation and disadvantage; undermine
levels of trust in legitimate authorities; undermine trust between pro-legal
community members and impair community resourcing; compromise law
enforcement effectiveness, as well as the effectiveness of informal deterrence;
damage perceptions of procedural justice; afford unchallenged propagation of
THE MORAL ECOLOGY OF EXTREMISM 19
extremist group norms and unmonitored exposure to radicalising agents;
determine the selection of certain (susceptible) populations for exposure to
criminogenic settings; and lead to collective and individual experiences of
hardship, absent effective social support.
These factors, especially to the extent that they occur in combination, are likely
to contribute to the emergence of extremism-enabling social ecologies.
System
Logically, at the whole-system level, the focus should be on factors that
promote the emergence of extremism-supportive moral ecologies. Once again,
the most effective approach is to identify the key processes that systemic drivers
might shape, rather than try and enumerate all possible drivers.88

Norms
Unsurprisingly, given the functional importance of morality, systemic processes
which influence moral norm-making, promotion or suppression will play a key
role in the emergence of extremism-supportive moral ecologies. Returning to
Lord Anderson’s recent remarks, the law is democracies’ most effective and
consensual normative instrument. Hence, the importance of explicitness about
what a society considers unacceptable extremist actions and what
condemnation it associates with these actions cannot be understated, as is the
importance of the fair application of the law and of effective procedural justice
to citizens’ perception that these rules are legitimate.89

On the flipside, mechanisms that contribute to the normalisation of extremist
values, norms and behaviours are particularly salient.90 To the extent that they
are not systematically associated with clear and immediate negative social
consequences (i.e., that they are depicted as ignorable or tolerable; that they
are not received with explicit sanction and disapproval, especially from moral
authorities), habitual representations of extremist beliefs and behaviours
permeating people’s social ecologies – through mainstream and social media,
political discourse, cultural products, and so on – may result, over time, in the
normalisation of extremist moral systems.
In sum, the role of norm formulation, diffusion and endorsement at this level
cannot be underestimated, as it sets the conditions for the emergence of moral
contexts for whole populations.
THE MORAL ECOLOGY OF EXTREMISM 20
Governance
The moral context as it is perceived by citizens is made up of moral norms and
rules of conduct, and of the mechanisms of their formal (e.g. courts and police)
and informal (social control by other citizens) enforcement. Any systemic
change which impairs the effectiveness of formal and informal governance can
be determinant, in that it will shape, very concretely, the micro situations
individuals experience.91 Changes which affect the organisation of social life in
general, or trust between citizens, and between citizens and public authorities
in particular, can impair governance and through it the controls that suppress
extremism-supportive moral contexts. For example, economic hardship can reorder citizens’ priorities and deter them from participating in the life of their
community, which includes demonstrating their disapproval of others’
behaviour. Perception of loss of political representativeness can damage
citizens’ perception of the legitimacy of state representatives, to such an extent
that citizens undermine the representatives’ authority or fail to report breaches
of conduct. Indeed, anything that brings on a loss of trust in moral authorities
(public servants, but also experts, religious elites or other ‘thought-leaders’)
may have the same effect.
Perhaps one of the most significant changes in governance experienced in our
societies in recent years is the outsourcing of the regulation of public social
settings to private owners of online platforms on the one end,92 and to the users
to which these owners have delegated some of their governing responsibility
by providing powerful (and poorly evaluated) mechanisms of informal social
control on the other.93 It is reasonable to think that we do not yet understand
the full impact of these changes on the perception of behaviourally-relevant
moral contexts, on the perceived legitimacy of public and private authorities to
regulate these settings, on the exposure to extremist moral systems, and on the
emergence of extremism-enabling moral ecologies in general.
It is also with regards to governance that system interconnectedness and
permeability may be the most salient. Authorities are limited in the extent to
which they can affect the governance of other systems and their own efforts (or
lack thereof) to suppress the emergence of extremist moral ecologies.
Segregation
As posited, moral rule-systems are core features of social groups that enable
cooperation between members and advantageous selection of action
alternatives. Perceived threats to the group’s status – notably, in terms of
relative social, existential, economic, cultural or normative advantage – can
THE MORAL ECOLOGY OF EXTREMISM 21
result in the tightening of norms of conduct and a heightened commitment to
the group’s values, up to and including absolute claims of moral superiority. To
remain attractive, groups need to assert that their system of moral ruleguidance will lead to optimal outcomes, offering their members greater control
over their lives in general and the achievement of certain goals in particular.94
If it is so, then processes which induce effective or symbolic segregation
between social groups are likely to contribute to the emergence of an
extremism-supportive moral ecology.95 This is true, in particular, of any process
of segregation which heightens perception of social injustice in the form of
group discrimination, favouritism, competition, polarisation on the basis of ingroup and out-group, and so on.96

At this level, information technologies may do more than facilitate the
propagation of extremist discourse: they expose groups of people to
information about the existence and treatment of other groups, creating more
opportunities for frictions (which weaken bonds) and perceptions of relative
disadvantage (which heighten competition). By mediating the activities of
certain moral entrepreneurs and by popularising certain frames, they can also
entrench the association between specific social identity markers (e.g. ethnicity;
nationality; sexual orientation) and given moral systems, reshaping and
delocalising the groups that people choose as moral anchors.
In the past, friction between groups may have been tied to land and
neighbourhood – the most salient moral competitor was the one next door,
especially if they were noticeably other. With global awareness of other groups
in distant places, this may no longer be the case, multiplying the sources of
friction far beyond perceived competition for place-based resources, and
fostering ties between ideologically proximate, yet geographically distant moral
allies.97
Likewise, systemic processes that contribute to discrimination (exclusion from
collective social identity and its material advantages) and social ostracism (real
or perceived loss of group membership) may be determinant to the extent that
they constrain some individuals without sufficient human or social capital to
seek out new group membership and to display evidence of heightened
commitment (e.g. willingness to participate in extremist action) to prevent the
recurrence of ostracism. 98 In other words, such processes can affect
mechanisms of selection for exposure to extremism-supportive settings,
through individual segregation and lack of social integration.
THE MORAL ECOLOGY OF EXTREMISM 22
Strains
Given the salience of group-threat perception and friction, and the nature of
individual susceptibility to extremism-supportive moral change, systemic
processes that result in (the perception of) either collective or individual strains
are the last key category of systemic contributors, inasmuch as they have the
potential to impact normative, governance and segregation processes, and to
shape the experiences of individuals embedded in particular situations and
social ecologies.
Processes which contribute to strains in the form of perceived or effective value
violation, the unjust administration of justice and power imbalance have been
more specifically associated with political violence, as well as crime.99 Normative
strain 100 has also been implicated in violent extremism. Some systemic
processes in particular – such as migration or economic inequality – can bring
about the conditions for these experiences. Feelings of unfitness and
normlessness (marginalisation and insignificance) of some second and third
generation immigrants have been documented.101 These might be attributable,
in part, to the perception that their parents’ culturally-appropriate moral
systems haven’t provided adequate guidance in their present context (i.e. moral
commitment hasn’t resulted in positive outcomes and in some negative ones,
such as discrimination), while barriers to social integration have made local
moral systems inaccessible or unattractive. Experience of increasing, relative
social disadvantage for some (non-migrant) social groups could have similar
effects, leading some of their members to seek new moral anchors.
Cultures regularly exposed to acute exogenous strains have been shown to
display tighter commitment to norms and to punish rule-breaking more
severely, compared to looser, more liberal cultures, which have traded social
order for adaptability and creativity, suggesting that normatively looser systems
may struggle to suppress unlawful extremist behaviour without losing some of
the benefits of liberalism.102 It may be that ‘tightening’ the moral context to
suppress extremism has a systemic cost for liberal democracies.
At the same time, sustained exposure to systemic strains (including, but in no
way limited to, violent extremism), which have assailed a largely insulated
middle social class, might contribute to the weakening of the heretofore
relatively stable normative centre of liberal democracies. Commitment to
historically-advantageous moral systems (liberalism, communism, socialism) is
no longer perceived as delivering satisfactory outcomes. Trust in moral norms
(traditional ideologies), their enforcers (the justice system) and their promoters
(mainstream politicians) is eroding. 103 Social systems characterised by the
THE MORAL ECOLOGY OF EXTREMISM 23
contradictory forces of hyperchoice and uncertainty are perceived as cognitively
taxing, even for socially-advantaged citizens.104

To date such experiences of strain may not have resulted in a general shift to
political or religious extremism for middle social categories, though changes in
voting patterns towards extreme ends of the spectrum are seen as a growing
trend. Yet, it may be that very similar processes of vulnerability, exposure and
emergence taking place in different social ecological contexts manifest in
moves towards other kinds of extreme moral anchors. One may think, for
example, of the growing opposition to the previously consensual belief in the
good of vaccination, with potentially dire systemic consequences.105

Conclusion
Extremism is, rightly, a central concern for our democratic social systems.
However, when addressing it we should be careful not to create a wicked policy
problem or to unwittingly contribute to the conditions that support the
emergence of extremist moral ecologies, by feeding into, for instance, the
perception of competition between moral groups.
Anticipating the unintended consequences of our own actions, as well as
devising strategies to control the actions of others, requires a systemic
approach. The evidence suggests that some individuals are more susceptible to
the attraction of extremist moral systems and to engagement in unlawful
extremist behaviour than others, but because there are ‘canaries in the coal
mine’ does not mean others will forever remain impervious to the toxic effects
of prolonged exposure. Supporting those we perceive to be most at risk is not
enough. Changing contexts, rather than changing people, is the more effective
strategy, because vulnerability is inherently context-dependent.
The word ‘systemic’ might strike fear in the heart of strategy designers. It may
imply that anything and everything is involved in creating the problem, and that
everything and everyone is a target or a potential lever. That is not the case.
Defining extremism as a relational moral concept; extremist behaviour as moral
action; extremist susceptibility as susceptibility to moral change; the main
appeal of extremist moral systems as one of moral entativity; extremist contexts
as a matter of the emergence of extremism-enabling settings – all of this means
narrowing our focus down to a set of specific mechanisms relevant to the
emergence and maintenance of extremist moral ecologies.
THE MORAL ECOLOGY OF EXTREMISM 24
While it makes no claim as to the problem being any less complex than it is, the
proposed functional approach offers a systematic framework for the
identification and organisation of counter-extremism strategic goals and the
actions needed to achieve them. It roots the conceptual (extremist moral
systems) in the material (places) and conceives of resilience to extremism not
as a transitory, individualistic trait, but as an emergent social feature of
functioning societies.