2011. Vol.2, No.4, 393-404
Copyright © 2011 SciRes. DOI:10.4236/psych.2011.24062
Evolution and the Prevention of Violent Crime
Jason Roach1, Ken Pease2
1Huddersfield University, Huddersfield, UK;
2Loughborough University, Loug hborough, UK.
Received April 28th, 2011; revised June 2nd, 2011; accepted July 3rd, 2011.
This paper suggests how violence prevention can be better informed by embracing an evolutionary approach to
understanding and preventing violent crime. Here, ethical crime control through an evolutionary lens is consid-
ered and speculation is offered as to what an evolution-evidenced crime reduction programme might look like.
The paper begins with an outline of the current landscape of crime prevention scholarship within criminology
and presents some possible points of contact with actual or possible violence reduction practice, including child
homicide and violence against women. The paper concludes with suggestions for an ethical research agenda for
reducing violence, whereby it is hoped that an audience of open-minded criminologists and diverse students of
evolution may lend a hand in in c reasing the sophistication of the criminological study o f violence prevention.
Keywords: Violence, Evolu t i o n , Child Homicide, Prevent i on
Criminology generally is justly criticized for its theoretic in-
sularity, and in particular its general hostility towards or neglect
of approaches other than that of sociological determinism
(Pease, 2010; Roach and Pease, in press). Its journals tend to be
titled according to geography or methodological focus (e.g.
European Journal of Criminology, Journal of Quantitative
Criminology). There are few cross-discipline titles characteris-
tic of faster-advancing disciplines, and cited references tend to
be older and more often book based than one finds in livelier
branches of scholarship. When one of the present authors told a
distinguished professor of mainstream criminology that he was
co-writing the present paper, the reply was a derisive “You’re
better than that”.
Insofar as systematic attempts at crime reduction take their
theory from academic criminology, evolutionary perspectives
are therefore predictably marginal or absent. In the near eight
hundred pages of Nick Tilley’s authoritative edited “Handbook
of Crime Prevention and Community Safety” (Tilley, 2005),
evolution is mentioned once, in a chapter by the second author
(Pease, 2005). It is not mentioned at all elsewhere, even in the
chapter on violent and sexual crime (Maguire & Brookman,
2005). It would be tedious to rehearse the absence of evolution
as a topic from all other leading recent texts on crime preven-
tion, such as Welsh and Farrington (2006) and Tilley (2010).
Attempts to translate evolutionary insights into social policy
have an unhappy history. Many social scientists may thus feel it
prudent to avoid such contributions as it may make. For exam-
ple, in 2008 when the authors asked attendees at the British
Society of Criminology annual conference to complete a short
questionnaire asking for their opinion on the utility of evolu-
tionary thinking in criminology, only a derisory fifteen per cent
condescended to do so, and most of those who did, considered
it to be irrelevant to criminology.
Tooby and Cosmides (1990) define evolutionary psychology
as, psychology which centres on the fact that the structure of
the human mind, which has been inherited, and represents the
product of evolutionary processes (i.e. natural and sexual selec-
tion). Put more simply, what we can do today is a direct result
of what was needed to be done in order to increase our ances-
tors’ survival and reproductive chances in the past. What as
humans we are able to do today, therefore, is a direct result of
the evolution of our species—namely how our ancestors
adapted to their environment. Genes being simply the method
of transmission between generations.
Evolutionary psychologists seek “functional” explanations
for behaviour. For example, why a particular human behaviour,
ability or characteristic has evolved, and not causal explana-
tions, focused on immediate precursors or causes of that be-
haviour, in the environment in which it occurs (Smith & Ste-
vens, 2002). This is a rapid departure from, for example, be-
haviourist psychology and environmental criminology, where it
is the immediate environment which is of primary focus. Evo-
lutionary psychologists focus instead on the distant past, at-
tempting to locate the function of certain behaviours in our
evolutionary past by looking at our ancestors. Smith and Ste-
vens (2002), for example, pose the question, why is it human
beings have evolved spoken language when other close pri-
mates, such as chimpanzees, have not? Many other aspects of
chimpanzee physiology (including a 95% DNA match) are
similar, so what were the evolutionary pressures that supported
the ability to develop and use language in humans only? Evolu-
tionary psychologists adopt reverse engineering in seeking
answers to questions such as this (Tooby & Cosmides, 1990).
An evolutionary informed criminology of the sort called for
in this article, should consider how distal factors (such as dis-
plays of aggression related to sexual competition and fitness
among young men) interact with the more proximal (e.g.
crowded, noisy bars full of young men) and how this can in-
form violence prevention strategies and interventions. We call
for an “Evolutionary informed criminological” approach to
violence, which, therefore, considers how distal and proximal
factors contribute and interact when explaining (and preventing)
J. ROACH ET AL.
crimi nal be haviou r, refe rred t o as Gene × Environment (G × E)
effects. Caspi and colleagues’ wonderful example of the inter-
action between expression of the MAOA gene and violent be-
haviour is presented later in the article (Caspi et al., 2002).
Eugenics and social Darwinism apart, the central objection of
criminologists to accepting evolutionary insights seems to be
that crime, including violent crime, is a social construct, and
explaining a socially constructed phenomenon in biosocial or
evolutionary terms constitutes a category error. This objection
is not a powerful one. To adopt and adapt the Gottfredson and
Hirschi (1990) characterization in their general theory, crime
can be seen as actions seeking to gain advantage over con-
specifics by force or fraud. Minor definitional tweaking is nec-
essary to exclude sanctioned combat sport and warfare, but the
vast bulk of criminal law reflects this, consisting of more or
less tightly worded proscriptions of intra-species advan-
tage-taking by force or fraud. Violent crime represents the first
of these means of securing advantage. This is not of course to
say that most attempts to gain advantage by force or fraud are
treated, still less processed, as breaches of criminal law. More
such attempts probably occur in a day’s playground bullying or
questionable sales techniques than in a century’s processed
crimes, even in societies that would see themselves as governed
by the rule of law. Neither is it the case that law is equally ap-
plied cross-nationally, with violence against women represent-
ing a particularly troubling case in point (see Weldon, 2002).
There are tautologies involved in the linkage of evolutionary
thought and what counts as a crime. Criminal law is enacted for
forms of behaviour which are deemed to be too serious or dis-
ruptive of social order to be dealt with by citizens informally or
via civil law. There is no need to enact criminal laws in respect
of behaviours which no-one wants to perform, or which harm
only the behaver. There is no law against coprophagia, for ex-
ample. Matters are more complex than this, given a religious
underpinning to the criminal law which has in the past forbid-
den suicide, for instance. Nonetheless, it is tenable to assert that
criminal law proscribes those actions which sele ction pressures
of the environment of evolutionary adaptedness (EEA) gave
some or all of us the inclination to perform, as the EEA-derived
equivalent of Original Sin. Insofar as this is the case, crime
reduction constitutes an attempt to curb or sublimate evolu-
tion-derived predatory actions. As Owen Jones (2005) notes,
law is “a tool for moving human animals to behave in ways
they would not otherwise behave if left solely to their own de-
vices” (p. 953). Of course this also applies to education and
regulated sport, as well as to the behaviour of both the offender
and the preventer. Jones framed the “law of law’s leverage” as
“The magnitude of legal intervention necessary to reduce or
to increase the incidence of any human behaviour will correlate
positively or negatively, with the extent to which such a pre-
disposition contributing to that behaviour was adaptive for its
bearers, on average, in past environments” (p. 962)
Save for a quibble with the insertion of legal as the fourth
word in Jones’ law (since most effective interventions seem to
consist in the manipulation of opportunities rather than weight
of legal response) the point is well made. That said, criminal
law has to be seen as an attempt to produce behaviour different
from what it would be were selection pressures to have their
way unimpeded. Effective laws against murder or dangerous
driving, for example, alter selection advantage for those who do
not waste their reproductive years in prison (and careful pedes-
trians). This, of course, represents a crude position. And the
application to problems of violent crime of memetics as a form
of evolutionary thinking would yield a more sophisticated per-
spective. Indeed, a wholly different perspective could be taken
on the issue by concentrating on enhancing cooperation rather
than reducting its antithesis, agonistic behaviour. While this
approach was seriously considered, given the extensive current
work from that perspective (see for example Nowack, 2011) it
was rejected for no better reason that the length of the piece
was already unwieldy and the deadline for its submission past.
Having thus excluded the option of “accentuating the posi-
tive” in favour of seeking to “eliminate the negative”1 what,
then, can the present article hope to contribute? It aspires (very
diffidently) to consider ethical crime control through an evolu-
tionary lens, and thereby to invite the deployment of evolution
insights in criminology journals and crime reduction practice,
and to speculate what an evolution-evidenced crime reduction
programme might look like. To that end, it will
1) Outline the current landscape of crime prevention schol-
arship within criminology, with the hope of eliciting contribu-
tions from those with expertise in evolution science; and
2) Present some possible points of contact with actual or
possible violence reduction practice, by reference to extant
literature in evolutionary psych ology and crimino logy.
The hoped for audiences are open-minded criminologists and
diverse students of evolution who may lend a hand in increas-
ing the sophistication of the criminological study of violence
prevention. Given these aims, it is perhaps inevitable that some
ideas will be seen as banal, others as tendentious. Successful
crime reduction approaches must, arguably, be conistent with
evolutionary thought, otherwise they would not have been suc-
cessful. To anticipate a conclusion, attempts to change those
embarked upon a criminal career based upon presumptions of
pathology enjoy limited success relative to those based upon
common perceptions of crime opportunities. Evolutionary in-
sights are sought primarily in relation to how people understand
and respond to violence-precipitating situations.
In what follows, violent crime prevention refers to the pre-
clusion, by ethically defensible means, of individual violent
offending acts. Harm reduction involves the reduction of
harmful consequences in unprevented violent acts.
Crime reduction: the state of the art.
The conventional classification of crime reduction distin-
guishes primary, secondary and tertiary methods. This best-
known categorization was devised by Brantingham and Faust
(1976). Primary prevention reduces crime opportunities without
reference to characteristics of criminals or potential criminals.
Secondary prevention seeks to change people, typically those at
high risk of embarking upon a criminal career. Tertiary preven-
tion works by the truncation of a criminal career, in length,
seriousness, or frequency of offending. The Brantingham and
Faust classification was later refined by Van Dijk and de Waard
(1991) and remains useful. Primary prevention has focused
upon proximal causes or precipitators of the crime event, and
can be criticized for neglect of distal ‘causes’. However, this is
a matter of convenience and immediate utility, rather than being
1And not “mess ing with Mr. In-Between ”, as enjoined in th e Johnny Mer-
cer-Harold Arlen song.
J. ROACH ET AL. 395
intrinsic in the definition. Primary prevention simply assumes a
supply of motivated potential offenders, and changes the cir-
cumstances which realize the latent criminality. While this has
typically involved place design, hot spot management, and
other such factors, it does not in logic exclude structural factors
of evolutionary interest, such as the gender composition of
populations. Indeed, the origins of routine activity theory,
which underpins much primary prevention, lie in the linkage of
distal factors such as the lengthening of childhood dependency
and general access to goods. Thus primary prevention is here
used to include all approaches which are person-neutral, i.e.
which does not seek or target individuals according to their
present or predicted criminality. Primary prevention also sits
more easily with the public health perspective favoured by the
World Health Organisation (see Krug et al., 2002).
Primary crime prevention has some advantages over rival
approaches in terms of the weight of evidence brought to bear
in its support, and in terms of the slightness of the level of so-
cial engineering often necessary for its implementation in its
proximal form. Underpinning the approach, as noted above, is
the routine activities theory of Marcus Felson (2002), in which
an offender who is ready, willing and able encounters, seeks out
or engineers a crime situation comprising a vulnerable and
attractive target of crime, in a favourable environment and in
the absence of motivated and capable preventers. Routine ac-
tivities theory was initially developed to address the paradox of
increased urban violence at a time when social conditions
deemed relevant improved (Cohen & Felson, 1979). These
authors saw predatory crime as a by-product of routine activi-
ties, in particular the dispersion of activities away from home
If crime, especially violent crime, is simply the natural work
of evil or damaged people, then convicting and incapacitating
evil and damaged people lies at the heart of police work. For
those of the left politically, the emphasis is on the remediation
of damage to the person, deemed to have been occasioned
through deprivation of one kind or another. Marcus Felson
(1994) refers to this as the “pestilence fallacy”, which asserts
that bad things come from other bad things. He writes:
Why was the major period of crime rate increase in the
United States 1963 to 1975, also a period of healthy economic
growth and relatively low unemployment? Why did Sweden’s
crime rates increase greatly as its Social Democratic govern-
ment brought more and more programmes to enhance equality
and protect the poor? (pp. 11-12).
Felson concludes that “crime seems to march to its own
drummer”, and that the richness of crime opportunities is the
crucial factor which determines the beat. This is a conclusion of
fundamental importance for crime reduction, suggesting that
the regulation of the supply of crime opportunities is central to
the reduction enterprise.
The common criticism by scholars in the biological sciences
of the application of evolutionary ideas to behaviour and its
regulation is best known as simply retailing “Just So” stories,
speculative and unconstrained by evidence (see for example
Coyne, 2009). Our riposte is that much of criminology has con-
sisted of Just-So stories of sociological origin, to the point
where the discipline of criminology is fragmented and possibly
in terminal decline (see for example Loader & Sparks, 2010). If
evolutionary concepts turn out to have no more than heuristic
value in designing research and crime-reductive practice, it will
be no worse than what has gone before. If it turns out that suc-
cessful crime reduction can be derived from and is consistent
with such concepts, a new direction for criminology will have
been signposted. Co-evolution has already been used as a simile
for the “arms race” between offenders and purveyors of pri-
mary prevention (Ekblom, 1997, 1999),2 and some primary
prevention techniques have apparently been inspired by animal
defence mechanisms, for example the squid’s deployment of
opaque fluid when under attack.3 Felson (2006) devotes a chap-
ter of his book “Crime and Nature” to “Crime’s First Defences”.
Understanding offender foraging behaviour has informed opti-
mal police deployment (Johnson et al., 2009).
Primary Violent Crime Prevention: Can
Evolutionary Thinking Contribute?
The renowned astrophysicist Stephen Hawking argued in
April 2010 that we should avoid making contact with extrater-
restrial life forms. The media reported this in somewhat xeno-
phobic terms (if that word can be applied to extra-terrestrials).
The more misanthropic truth was revealed by his remark that
“We only have to look at ourselves to see how intelligent life
might develop into something we wouldn't want to meet.”4
As Buss and Shackelford (1997) point out, the ubiquity of
human violence cannot be explained solely by the variables of
usual criminological interest. Although of no little importance,
the question whether violence has roots as an adaptive strategy
that once increased reproductive success, it is not the brief for
this article, it being specifically about prevention. The age
crime curve and the preponderance of male violence are per-
haps the two central facts about the distribution of criminality.
Self-reported and recorded offending, both violent and other,
peak in the late teens and are higher for males at every age. The
claim that the age-crime relationship is invariant across epoch
and culture (Hirschi & Gottfredson, 1983) is roughly true, with
the exceptions being of equal interest as the generality. For
example Zhong (2005) showed that in a period of rapid eco-
nomic development in Taiwan, the age-crime curve peaked
earlier for property but not for violent offending, which Zhong
ascribes to the continuity of Chinese ‘culture’ over the period.
Robert Wright (1994) asserts that the leading cause of vio-
lence is maleness, to which may be added youth and unmarried
status (see Mesquida & Wiener, 1996, 1999). In brief, violence
is often about sexual selection. The classic psychological theory
of violence is the frustration-aggression hypothesis, which
holds that aggression is the response to the expected interfer-
ence with a desired goal (Dollard et al., 1939, see also Berko-
wittz’s 1989 reformulation). Stressing the primacy of sexual
selection as the root of frustration is the distinctive contribution
of evolutionary thinking. If that insight holds, there seem to be
two strategies which fit into the primary prevention category,
two distal, two proximal. The distal tactic concerns establishing
sex ratios or sexual mores which do not deny sexual expression
2Paul Ekblom, shown the section in the text, comments Biological
co-evolution of predator/prey capabilities etc is a simile, but cultural/
technical co-evolution of offending and preventing techniques is the real
3http://www.smokecloak.co.uk/en/ accessed April 17th 20 11.
4http://news.bbc.co.uk/1/hi/uk/8642558.stm accessed April 13th 2011.
J. ROACH ET AL.
among male adolescents and young adults. As Gottschall (2008)
“…a real shortage of female reproductive capacity, relative
to male demand, is endemic to the human condition. In this
shortage lies the answer as to why men fight… sexual selection
has shaped men to compete for women and for concrete mate-
rial resources and for intangible social resources because…
these resources [are] reliably converted to reproductive advan-
tage” (pp. 48-49).
It is entirely unrealistic to suppose that sex ratios at birth
should be manipulated in the interests of the prevention of later
violent crime. Indeed, since the female population defines ef-
fective reproductive channel capacity, the population implica-
tions of such a strategy would be horrendous.
The second distal approach would involve an attempt to re-
duce anticipations of lifetime failure amongst adolescent males.
Daly and Wilson (2005) note the higher rate of future dis-
counting (in effect preferring short over long term rewards)
among males and among the young (paradocically, given the
increasing uncertainty of the future as one gets older). They
note the greater variability of reproductive outcome among
males, and the resulting competition amongst males. After jus-
tifying rate of homicide as a proxy for inter-male competition,
they find homicide to be most frequent amongst those with
least to lose, unemployed and unmarried men, with divorced
and widowed men reverting to higher rates of homicide, Char-
acterising the deployment of reproductive effort as a gamble,
anticipation of the high probability of an early death or other
form of removal from the reproductively active is consistent
with the choice of a strategy of attempting risky and frequent
impregnation. While plausibly argued, it is difficult to think of
viable social policies which would substantially inflate the re-
sidual reproductive value of young men which would not be
justif ied more powerfully by other ega litarian argume n ts.
The impracticality of both distal options makes the proximal
ones more important.
The proximal tactics concern:
1) The nudgeability of human behaviour;
2) Harm reduction (here treated as a form of primary preven-
tion), whereby the affordance or means for translating violent
intent into injury or death are denied those likely to exhibit such
These will be dealt with in turn.
Over the past 25 years or so there has been a gradually in-
creasing recognition of the possibilities offered by focusing on
crime events rather than on offenders (Gilling, 1997; Crawford,
1998).Overstating the case only slightly, the evidence for the
success of intelligently conceived and well-implemented pri-
mary prevention (Clarke, 1992, 1997; Goldblatt & Lewis, 1998;
Sherman et al., 1998; Tilley, 2010) is very substantial. Such
success is consistent with situational rather than dispositional
theories in social psychology. The capacity of slight changes in
setting to evoke substantial changes in behaviour represents a
well-established tradition in social psychology, with Walter
Mischel (1968) a pioneer. Mischel’s analyses revealed that the
individual’s behavior, when closely examined, was highly de-
pendent upon situational cues, rather than expressed consis-
tently across diverse situations that differed in meaning. A ne-
glected feature of Stanley Milgram’s obedience studies was his
manipulation of the proximity of the person apparently admin-
istering electric shocks and the person to whom they were ad-
ministered, which yielded massive changes in the intensity of
shocks they could be induced to administer (see Milgram,
1977). Momentous decisions (such as committing suicide) are
remarkably changeable by removing one means of doing so, by
making domestic gas supplies non-toxic (Clarke & Mayhew,
1988), installing catalytic converters to vehicles (Lester &
Clarke, 1989), and restricting the size of paracetomol and aspi-
rin packets (Hawton et al., 2001). More recently Zimbardo
(2007) develops the point. Thaler and Ornstein (2008) rename
the situational determinant the “nudge” and apply it widely to
social policy. Human nudgeability, and the capacity to afford
multiple uses for objects, it seems embarrassingly obvious to
relate, is a consequence of the human evolutionary route,
whereby a wide range of environments become habitable by
dint of behavioural flexibility and innovation. The longstanding
recognition of the fact of imitative learning, and the more re-
cent exciting research and speculation about mirror neurons as
the basis of such learning, including the imitation of prosocial
behaviour (see Thomson 2010 for an introduction).
Turning to harm reduction, the concept of affordance pro-
vides one way of thinking about the situational specificity of
behaviour alluded to earlier, and permits the elaboration of the
concept of opportunity. The original Gibsonian (1950) meaning
of affordance concerned things that you could actually do with
objects. An object’s affordance comprised what one could do
with it. Don Norman (e.g. 1998) extended this to perceived
affordance, ie actions that the thing suggested you do with them.
Ekblom and Sidebottom (2007) describe it in the crime oppor-
tunity context as an offender’s capacity to see utility in an ob-
ject. While doing participant observation in an ex-prisoner’s
hostel, one of the writers observed a fight break out very
quickly and a heavy glass ashtray was wielded as a weapon. To
the writer, it had just been an ashtray a moment before! Base-
ball bats “afford” thoughts of violence more than cricket bats,
and are sold in all large British sports stores, despite the fact
that baseball is played very little in the UK. The position on
glass as a weapon and on gun availability is more difficult to
summarise. Conventional (annealed) drinking glasses or bottles,
when broken, yield sharp edges capable of inflicting serious
injuries. Such events are generally known as glassings. Tough-
ened glass shatters on impact, and is relatively useless as a
weapon. Yet a randomised controlled trial showed 60% more
injuries from toughened glass than from annealed glass. The
conclusion was reached that glass with lower impact resistance
caused more injuries. “Toughened” glassware had lower impact
resistance. Standards for toughening need to be developed.
(Warburton & Shepherd, 2000, see also Coomaraswamy &
Shepherd, 2003). Subsequent analysis showed that this coun-
terintuitive result was a consequence of the greater resistance to
breakage of annealed glasses, which occasioned the plea for
revised standards for toughening (not developed at the time of
writing). While the results were obviously disappointing, para-
doxically they do illustrate that weapon characteristics were
relevant to the degree of harm inflicted. For instance, assaults
with bottles caused less severe injuries than assaults with
drinking glasses. In a local initiative in Liverpool in 1997, a set
of simple measures preventing glasses and bottles being taken
onto the street reduced the rate of glassings by some 50%.5
accessed April 9th 2011.
J. ROACH ET AL. 397
The media characterise Swiss gun crime as low, despite the
wide availability of weapons by military reservists. However
Killias (1993) found that Switzerland has five times as many
homicides committed with guns as Great Britain versus only a
slightly higher non-gun homicide rate. His explanation for the
lack of gun use in other crime is that the military weapons that
are kept in Swiss households are of little use to ordinary crimi-
nals, because they are heavy and far too long to be concealed
under a coat or in a case. Zimring and Hawkins (1987) contend
that the firearm effect consists in escalating consequences.
They provide data on altercations involving knives far less
often result in death, for example. Lott (2005) argues that the
power-equalising effects of handguns is neglected in statistics.
He uses national victimization survey to assert that women
without a gun resisting attack are some four times greater than
women resisting with a gun. The male difference is smaller but
in the same direction. Grossman (2009) notes the relationship
between remoteness of assailant to target enabled by modern
weaponry and psychological ease of assault. He notes that
changes in the realism of targets used during weapons training
will affect rate of fire in battle. A historical nod in the direction
of Konrad Lorenz is necessary. To him we owe the insight that
animals with the most fearsome natural weapons are those with
the best developed behavioural inhibitions on lethal assault. He
“…all heavily-armed carnivores possess sufficiently reliable
inhibitions which prevent the self-destruction of the species. …
No selection pressures arose in the prehistory of mankind to
breed inhibitory mechanisms preventing the killing of con-
specifics” (Lorenz, 1966, p. 233).
It is perhaps not too much of a stretch to say that the paucity
of natural inhibitions on the expression of violence is consistent
with the primary reduction focus on limiting the precipitating
circumstances of violence and the weapons availability which
set limits on the amount of harm done in violent attacks. Cer-
tainly some of the major examples of successful practice are of
this type. These include place-oriented approaches to gun crime.
Braga (2003) writes, with supporting research, that
“Officers can reduce gun crime by changing the features, fa-
cilities, and management of problem places. For example, if
problem analysis reveals that easy access to common areas in
front of a high school causes youth gun crimes to be clustered
there immediately upon school release, police should experi-
ment with ways to limit access to these areas during problem
times. The practice of problem-oriented policing is still devel-
oping, and additional research is needed on different ap-
proaches to controlling gun violence hot spots”.6
Violence at pubs and bars is reduced (Scott & Dedel, 2006)
“A comfortable and entertaining atmosphere reduces both
frustration and boredom among patrons, which can reduce ag-
gression levels. Lighting should not be so bright that it acts as
an irritant, but also not so dim that it can conceal customers’
activities. An important environmental consideration is the
crowding level. Police in some jurisdictions enforce occupancy
limits (primarily adopted for fire safety) as a means to control
the bar crowding that can lead to fights. Redesigning a bar’s
interior to improve traffic flow and prevent congestion can
reduce the opportunities for accidental bumps and drink spills
that may escalate into fights”.7
While such conclusions may seem self-evident, they do illus-
trate that manipulable aspects of the immediate environment
can nudge behaviour from violence and/or reduce harm. Some
sixty-four design guides on crime reduction, about a third ad-
dressing violence, have been published by the Centre for Prob-
lem-Oriented Policing (http://www.popcenter.org/) and are
worth consulting to gain an impression of the range and types
of problem in which demonstration projects have found place
manipulation to be effective.
Secondary Violent Crime Prevention
It will be recalled that secondary prevention is oriented to-
wards those who may be particularly prone to develop prob-
lematic behaviour. Aggressive behaviour appears very young
and is characterized by high stability coefficients across many
years (Olweus, 1979). Serbin and Karp (2003) reported that
aggression (and depression) in primary school aged girls were
predictive of aggression amongst their children. Of course the
mechanisms may have no evolutionary relevance, but this does
mean that there is reason for secondary intervention, particu-
larly since aggression was linked with a variety of other unhap-
pinesses in the lives of aggressors (Junget et al., 2007).
Many lines of research in evolutionary psychology have clear
implications for secondary violence prevention. The two most
obvious are selected for attention here. One is the evidence for
apparent gene-specific effects of environment on the develop-
ment of violent people. The second concerns the particular risks
to children associated with step-parentage.
In her admirable, unfashionable classic The Nurture As-
sumption, Judith Rich Harris (1998) contends that the contribu-
tions which parents make to their children’s development pri-
marily comprise their genes and where they decide to set up
home. She points out that in the socialization literature, the
effects of parenting styles are confounded with those of shared
genes. She reviews the literature on the relationship between
parent characteristics and child characteristics and demonstrates
that either the same characteristics must have different and
largely unpredictable effects on children, or that those inﬂu-
ences are largely illusory. Is poor parenting a risk factor in a
child’s development, or do parents tending to (for example)
impulsivity, transmit that attribute through the conventional
Familial inheritance is always both the result of genetic en-
dowment and environment, but environments are made, and are
often correlated with the dispositions of those who inhabit them.
Moffitt and Caspi suggest with regard to antisocial behaviour;
“Environments are provoked by a person’s genetically influ-
enced behavior, or a person chooses environments consonant
with her genotype, or a person’s genotype results in him selec-
tively finding him self in certain environments” (2006, p. 127)
The implications of Rich Harris’s work are profound. They
should be no surprise to those familiar with the earlier literature
on criminality and biology (see, for example, Mednick &
Christiansen, 1977). The latter authors demonstrated that crimi-
nality in the biological parent is reﬂected in a higher prevalence
7http://www.popcenter.org/pr oblems/assaultsinbars/accessed April 9th
J. ROACH ET AL.
of criminality in the child, whatever the criminal record status
of the adoptive parent. More recent work under the flag of neu-
rocriminology has looked in detail at brain functioning, condi-
tioning and cri minality (see for example Gao et al., 2010).
Some scholars suggest that searches for answers to questions
such as these have led to a preoccupation with ‘risk factors’ that
often confuse co r re l ates with caus e s;
“Influential reviewers have concluded that the study of anti-
social behaviour has been stuck in the ‘risk-factor’ stage (Far-
rington, 1988, 2003; Hinshaw, 2002; Rutter, 2003a, 2003b)
because so few studies have used designs that are able to
document causality (Rutter et al., 2001). A variable is called a
‘risk factor’ if it has a documented predictive relation with an-
tisocial outcomes, whether or not the association is causal”.
(Moffitt & Caspi, 2006: p. 109).
Arguing whether behaviour is a result of biological charac-
teristics or specific environmental experiences was likely the
result of misdirection (Suomi, 2009). More exciting is the re-
alization by many that both genetic (nature) and environmental
(nurture) factors play crucial roles (e.g. Collins et al., 2000) and
indeed both often interact to shape behaviour and its individual
development (e.g. Rutter, 2001).
One illustration of the importance of understanding behav-
iour (including violence) from a gene x environment interaction
perspective (hereafter, G × E) is provided by studies with
rhesus monkeys. It has been consistently shown that it those
that have been reared by peers and not their biological mothers,
appear much more “anxious in nature” and excessively “fear-
ful” than their “mother-reared” counterparts’ (Suomi, 1991,
1995). Indeed longitudinal studies have shown that male
“peer-reared” monkeys consistently exhibit more extreme be-
haviours (and with increased frequency) such as being overly
aggressive and overly impulsive, than those reared by biologi-
cal mothers (Suomi, 1991, 2009).
So how can adopting evolutionary thought help explain such
G × E interactions? Put briefly, we can surmise that those
reared by peers and not their mothers, do not have their aggres-
sive survival impulses attenuated. Indeed a predisposition to
aggression and violence would be considered vital to survival
in an environment where access to food and reproductive part-
ners is hotly contested with rival male peers. With those fortu-
nate to be raised by their mothers having their early nutritional
and protection needs provided for in their formative years at
So should predictions of such behaviour by rhesus monkeys
be based simply on an identification of early-rearing environ-
ments and not on gene × environment interactions?
The short answer is no as recent research points to such iden-
tifiable differences in behaviour between male “peer-reared”
and “mother reared” monkeys as being more the result of gene
× environment interaction rather than just early-rearing experi-
ence alone (Suomi, 2009).
The neurotransmitter serotonin is posited to be an important
consideration when explaining unusual and extreme behaviour.
More specifically, the occurrence of certain mental health
problems, such as anxiety and depression, alongside behaviours
such as fear, aggression and violence, has long been associated
with decreased serotonergic functioning. The serotonin trans-
porter gene (5-HTT) is generally held responsible for such a
decrease in functioning when it occurs (Lesch et al., 1996),
because it has ‘a length variation in its promoter region that
results in allelic variation in serotonin expression’ (Suomi,
2009: p. 17). Heils et al. (1996) distinguish “long allele” (LL)
from “short allele” (LS), with the latter raising the possibility of
decreased serotonergic functioning and therefore, an increase in
the likelihood of aggressive and violent behaviour.
Barr et al. (2003) found a “buffering effect” which appeared
to mitigate some of the monkeys with the LS allele from the
full effects of decreased serotonergic functioning, according to
their early-rearing experience. Where high levels of aggression
was observed in peer-reared monkeys with the LS allele, for
mother-reared monkeys with the LS allele it was not, demon-
strating an important interaction between genes and environ-
ment. Put simply, the consequences for these monkeys of hav-
ing the notorious LS allele appear to differ dramatically ac-
cording to whether the monkeys were raised by peers or
whether they were buffered from its effects by being raised by
their biological mothers. Suomi (2009) summarises the signifi-
cance of this finding rather nicely;
Indeed, it could be argued on the basis of these findings that
having the LS allele may well lead to psychopathology among
monkeys with poor early rearing histories but might actually be
adaptive for monkeys who develop secure early attachment
relationship with their mothers’ (p. 18).
In sum, compelling evidence is presented that genetic and
early experience factors in interaction affect a monkey’s capac-
ity to regulate expression of fear and aggression (Suomi, 2009),
with important implications for our own species;
“Clearly, the context in which development takes place mat-
ters a great deal for rhesus monkeys. It is hard to imagine how
it could be less so for human development”. (Suomi, 2009: p.
The importance of interactions is lent support by the work of
Terrie Mofﬁtt and her collaborators (see Mofﬁtt, 1993, 1997,
2003) whose distinction of adolescent-limited and life-course
persistent offenders casts the latter as predisposed as a result of
inherited and/or early acquired neuropsychological deﬁcit. Of
particular current interest is the gene variant MOAO which
lowers the activity of the enzyme monoamine oxidase A and
which seems implicated in violence. This relationship is
stronger amongst maltreated children (see Caspi et al., 2002).
Of course, reﬂecting Rich Harris’s insights, it is not necessarily
the case that maltreatment activates the propensity to violence.
Equally plausible is the notion that MOAO and other as yet to
be identiﬁed violence-relevant genes drive both parental vio-
lence (in the form of maltreatment) and child aggression. In
path terms, the message is simply that one ignores at one’s peril
genetic or other early factors disposing to preferences. Kim-
Cohen et al. (2006) conducted a replication and meta-analysis
and concluded “These findings provide the strongest evidence
to date that the MAOA gene influences vulnerability to envi-
ronmental stress, and that this biological process can be initi-
ated early in life” (p. 903). Widom and Brzustowicz (2006)
found the effect in white children only, adding a further interac-
The relevant research has not been slow to make its way into
the courtroom. Bernet et al. (2007) describe their own experi-
ence and anticipate future developments as follows
“It seems possible that both the defence and the prosecution
may be interested in introducing evidence regarding a defen-
dant’s life experiences and genetic make-up to a jury. In a case
of aggravated assault, for instance, the prosecution may say that
J. ROACH ET AL. 399
a defendant has violent tendencies (based on the person’s
genotype conveying low MAOA activity and a history of se-
vere child mistreatment) and should be removed from society
for as long as possible. On the other hand, the defence may say
during the sentencing phase of a first-degree murder case that a
person’s genotype and history of severe abuse during childhood
is mitigating” (p. 8).
It is telling that the work of the Moffitt-Caspi team and its
replication has been applied to the court literature before the
child protection literature (as far as the writers have been able
to determine after questioning relevant domain experts). What
are the implications for child protection? We may discard the
notion of being indifferent to the parental practices of those
children whose genotypes afford them some protection against
the acquisition of violent personalities! Rather it places an extra
premium on ensuring as fa r as possible the quality of child care
generally. Not to do so is effectively to collude in allowing
preventable harm which compounds the effects of the genetic
lottery itself. A parallel may perhaps be drawn with phenylke-
tonuria (PKU), where screening on neonates allows manage-
ment, and avoidance of the progressive mental impairment
which otherwise ensues.
A brief aside can perhaps be permitted. It has been specu-
lated that the mechanism underlying the MAOA-abuse interac-
tion may be epigenetic in nature. Kramer (2005) defines epige-
netics as “the study of stable alterations in gene expression by
nongenetic mechanisms resulting in stable alterations in phe-
notype” (p. 16). In an astonishingly prescient article, Kendler
and Eaves (1986) anticipate the essential facts of epigenetics,
focusing on sensitivity to the environment. This, both in the
sense of responsivity to the cues which make primary crime
reduction effective, and in the particular sense of empathy (or
its lack, see Baron-Cohen, 2011) will be a crucial point of de-
parture for research which seeks to link applied criminology
Speculatively, epigenesis makes evolutionary sense in a wide
number of cognitive and affective contexts. This form of gene
regulation is primarily under matrilineal control and has
evolved partly to co-ordinate in-utero development with mater-
nal resource availability (Keverne & Curley, 2008). A detailed
general introduction to epigenetics is to be found in Allis,
Jenuwein and Reinberg (2007). Szyf et al. (2009) provide a
clear brief account of the methylation and other events under-
pinning epigenesis, and report research on putative epigenetic
effects on maternal care, and preliminary work on epigenesis
and lifetime trajectories of aggressive behaviour.
In respect of violence, epigenetic effects which favour ago-
nistic behaviour only in environments in which parents had
exhibited similar behaviour towards them, makes intuitive
sense, and its neglect culpable. Kramer (2005) contends
“…the future of child and adolescent psychiatry is ethically
connected to the developmental processes designed by natural
selection and inherited by way of evolution. For anything re-
sembling the continuation of normal human interactions and
normal family-centred developmental processes, our work
should be inter-woven with the complex interactions occurring
between the epigenome and the phenotype” (p. 297).
The same could certainly be said of criminology. The defen-
sible next step will be to reanalyze longitudinal studies of
criminality (the Cambridge study now incorporates data from
three generations) to test epigenetic hypotheses.
Preventing Child Homicide?
For most people, the killing of a child is felt to be the most
heinous and distressing of all crimes (Adler & Polk, 2001). It is
mercifully rarer for children to be victims of homicide than
adults. In England and Wales there will be approximately 110
child homicides per year from an average total number of ap-
proximately 700 recorded homicides, roughly equating to 14%
of all homicide victims (Brookman, 2005). Quite understanda-
bly, child homicide provokes the most outrage from the public
(Adler & Polk, 2001). As a category of crime it has also been
identified as an important influence on public confidence in law
enforcement (Innes, 2003)
Gene-centred evolution-based speculation (as no doubt
elaborated elsewhere in this issue) would suggest that kinship
would reduce child killings, and those in a parenting role with-
out a kin relationship would be more induced to kill, This
should be particularly true of de facto “fathers” since effort
spent nurtuting someone else’s children would retract from
available future reproductive attention. Daly and Wilson (1988)
in a ground-breaking study of child homicides in the US con-
cluded that children were approximately 100 times more likely
to be killed by a “non-biological parent” (e.g. step-parent, co-
habitee or boyfriend/girlfriend of a biological parent) than a
biological one. Similar findings have since been found in other
parts of the world including Canada (Daly & Wilson, 1998) and
Australia (Adler & Polk, 2001). In Canada, Daly and Wilson
(1988, 1998) found a co-residing step-parent was approxi-
mately seventy times more likely to kill a child under two years
of age than a co-residing genetic parent.
Creighton (1989), in a report produced for the children’s
charity the NSPCC (the National Society for the Prevention of
Cruelty to Children), found that in England and Wales for the
period 1983-1987 almost a third of those recorded as victims of
intentionally inflicted physical injuries lived with one “natural”
(biological) and one “substitute” (non-biological) parent. A
random sample of children within the same age-range from the
wider population of England and Wales would have comprised
only three percent (Daly & Wilson, 1998).
In support of Daly and Wilson’s (1988, 1998) finding that
young children are disproportionately at risk of homicide by
step-parental males (more than 100 times more so). Adler and
Polk (2001) in their Australian study of child homicide also
identify that when young children are killed it is often by their
mother’s de facto partner (i.e. a young step-father figure) rather
than at the hand of their biological fathers. Further support is
provided by a recent more localized study of 127 child homi-
cides carried out by the first author (Roach & Shepherd, 2010)
where in the case of young child victims the mother’s de facto
partner (i.e. the child’s non-biological father) was most com-
monly charged with their murder. British readers will be re-
minded of the tragic case of “Baby P”, where seventeen month
old Peter Connolly was killed by his mother, her de facto part-
ner (and his brother) after months of sustained physical and
%20march%202009.pdf (accessed 14/04/2011).
J. ROACH ET AL.
Research has uniformly shown that it is more common for
older children to be killed by their biological fathers than by
male step-parental figures. (Daly & Wilson, 1988, 1998; Adler
& Polk, 2001) A common scenario is where the father attempts
filicide-suicide, as response to a marriage break-up, separation
from his children, or as an act of revenge against the mother
(Adler & Polk, 2001). Mothers who kill older children appear
generally to do so “to save” them from a perceived life of mis-
ery, often by successful filicide-suicide. For mothers, revenge
does not appear to be a motivating factor in the killing of their
children. Love does (Adler & Polk, 2001).
Mothers who kill their children appear to overwhelmingly do
so within the child’s first year of life (neonaticide—within the
first 24 hours being most common see Adler & Polk, 2001)
Psychiatric illness is often deemed a mitigating circumstance
(Adler & Polk, 2001; Roach & Shepherd, 2010).
Traditional criminological explanations struggle to explain
why, for example, young children are often more vulnerable to
becoming victims of homicide at the hands of a step-parental
figure than by a biological parent. Often quoted explanations
for crime per se, such as deprivation and poor educational at-
tainment, do not adequately explain, for example, why it is that
step-parental figures from all social strata and educational back-
grounds are consistently over-represented as perpetrators of
child homicide. Indeed the problem of explaining why takes us
back to the old problem of confusing correlates and causes. As
we have endeavoured to show here in this short article, apply-
ing evolutionary thinking fares much better.
Although our desire for British criminologists to incorporate
evolutionary thinking into the current understandings and ex-
planations for crime, we must not lose sight of our purpose with
this article of preventing violence. Theoretical understanding
may be one thing but practical application is another. What we
find most worrying is that the illuminating research on child
homicide discussed has not made its way to practitioners
working in child protection arenas in the UK.
From discussions with child social work colleagues, the au-
thors are quite convinced that neither the work of Daly and
Wilson, Adler and Polk, or any other like-minded research of
child homicide (enlightened by evolutionary thinking) has
made its way to those in the UK charged with the Hercuean
task of working in child protection. Our initial soundings sug-
gest that this most enlightening research on child homicide has
not been translated into prevention practice. The reason may be
its origin, grounded in evolutionary thought dismissed out of
hand. It may also be that there is a trade-off between child pro-
tection and the stigmatization of step-parents As for the first
possible reason, miscomprehensions and misgivings about
genes and behavior have probably prevented the dissemination
and adoption of such vital knowledge being put into practice by
those working in child protection. If this proves to be the case
then the most significant child homicide prevention measure we
can propose is that those working in child protection, such a
social workers and health visitors, be educated about this re-
search and thinking as part of their training, and that the associ-
ated guides and literature incorporate it as a matter of urgency.
Our plea to wider criminology, and most importantly to those
practitioners charged with protecting children, is therefore, the
same. In order to understand and prevent violence evolutionary
thinking needs to be embraced.
Violence against Women
According to the 2009/10 British Crime Survey (BCS) seven
per cent of women aged between 16 and 59 years in England
and Wales were victims of domestic abuse, compared with four
per cent of men (Hall & Innes, 2011). The majority of the vio-
lence was described as “non-sexual” abuse by a partner. Al-
though only fourteen per cent of the total 2,087,000 violent
incidents estimated by the BCS for that year were described as
domestic violence, those that collected the data are careful to
add a caveat proclaiming that equivalent figures have been
found to be up to five times higher where participant “self-
completion” had been used instead of the BCS “face-to-face”
method (Hall & Innes, 2011).
Our point is that it is the consensus that the prevalence and
frequency of domestic violence are far greater than recorded
crime statistics or the BCS suggest. Domestic violence is ubiq-
uitous and as such if we are working to prevent violence than
we can best make inroads by focussing on intimate partner
violence. For example, a study in England and Wales (1995-
2000) showed that 30 per cent of all the homicides were ‘femi-
cides’. Moreover, 57 per cent of these female victims were
killed by an intimate (or ex-intimate) male sexual partner
How can an evolutionary informed violence prevention
strategy help reduce domestic violence? The first step, as al-
ways, should be to try and understand what is going on. Let’s
start with why men might go to war?
Tooby and Cosmides provide an exquisite evolutionary psy-
chology of warfare where they ide n tify four esse ntial conditions
that must be met for adaptations to evolve that allow for initiat-
ing such coalitional aggression as war. They call these condi-
tions “the risk contract of war”—we draw the reader to their
first condition: The average long-term gain in reproductive
resources must be sufficiently large to outweigh the reproduc-
tive costs of engaging in warfare over evolutionary time.
(Tooby & Cosmides, 1990)
The most likely candidate is an increase in copulations as
men have a great deal to gain if the result of engaging in war
results in substantial increase in reproductive access to women.
As Buss phrases it,
“Although few wars are initiated solely with the stated intent
of capturing women, gaining more copulations is almost always
viewed as a desired benefit of successfully vanquishing an en-
emy” (2004: p. 298).
The work of Chagnon in his study of the warfare between the
Yanomamö villager tribes lends support to this argument,
where it was found to be nearly always linked to reproductive
access to women (see Chagnon, 1983 for a very interesting
anthropological study). We cannot resign the use of violence by
men to secure female reproductive access to “less developed”
societies. The mass rape of females by the victors has been an
established part of warfare throughout history, with for example,
the Norman invasion of 1066, in the first and second world
wars, and in the civil war in the former Yugoslavia. It might not
be the primary motivating factor for men going to war but se-
curing access to women is certainly part of the promise of re-
ward. The point being made, as it has been throughout this
publication, is that males (especially young ones) will do
whatever it takes to access females, regardless of the risks in-
J. ROACH ET AL. 401
Male “sexual jealousy” is the most frequently given explana-
tion for intimate partner violence (Dobash & Dobash, 1979;
Daly & Wilson, 1988; Buss, 2000, 2005; Goetz et al., 2008)
and so will only be briefly mentioned here. For example, Polk
refers to male on female violence as being primarily motivated
by “jealousy/control” on the part of the male (1994: p. 18). Put
simply, men appear to use violence against women as a tactic to
restrict their sexual behaviour. Primarily as means of enforcing
sexual (i.e. reproductive) “exclusivity” (Daly & Wilson, 1988;
Wilson & Daly, 1996; Buss & Malamuth, 1996). For example,
Fiona Brookman found that more than 80 per cent of femicides
occurred where the female was either planning to leave her
partner, or where he perceived her at least to have been “un-
faithful” with another sexual intimate, thereby compromising
sexual exclusivity (Brookman, 2000). Some suggest that vio-
lence against women is best understood as being “a behavioural
output of male sexual jealousy” (Goetz et al., 2008: p. 67).
Occasionally, that violence is lethal resulting in female homi-
cide (Daly & Wilson, 1988; Polk, 1994).
The foremost writer on this topic from an evolutionary per-
spective is Anne Campbell. Campbell makes the observation
that in trying to explain the gender difference in crime the
male-centered approach has dominated evolutionary psychol-
ogy, where there appears to be a broad consensus that the mo-
tivation to achieve status and ‘surplus resources’ is more criti-
cal to male than female reproductive success (Campbell, 2002,
2009). However, where a female-centered approach is instead
taken a different perspective on male on female violence is
“Her most important proximal goal is to stay alive because it
is she who, given 100 per cent maternal certainty, limited re-
productive years and high replacement costs, has most to gain
by ensuring that her offspring survive” (Campbell, 2009: p. 124).
Reproduction (and therefore, by virtue, sexual intercourse),
poses more of a risk to female safety and survival than for
males (Campbell, 2009). Lew and colleagues rather eloquently
describes this as a male-female genetic arms race in which fe-
males must evolve defences against the lethal potential of the
sex drive of males (Lew et al., 2006).Intimate partner violence
being one such defence as when women kill it is more often
than not an intimate (or previously intimate) partner that is the
victim (Daly & Wilson, 1988).
So very briefly how can evolutionary thinking contribute to
preventing domestic violence against women?
The most obvious answer is that women who experience
domestic violence should be encouraged not only to report it as
soon as possible, but they must also be able to access immedi-
ate help, in order that they and their children “stay alive”. Re-
search has consistently shown that a significant number of fe-
male victims of domestic homicide have experienced domestic
violence previously (Wallace, 1986; Campbell, 1992; Smith et
al., 1998). Lethal domestic violence against women being often
the final tragedy in a long-running sequence of violent and
emotional attacks by an intimate (or ex) partners. Mercifully, in
the UK campaigns encouraging women to report domestic vio-
lence have been frequent and with some degree of success,
hopefully exemplified by the year on year increase in recorded
domestic violence incidents. Spousal homicide reflecting an
extreme manifestation of the same basic conflicts that inspire
sub-lethal marital violence on a much larger scale (Daly &
Wilson, 1988; Brookman, 2005).
The important finding domestic violence is generally perpe-
trated by males against female sexual intimates planning to
leave them is also crucial with regards preventing violence
against women. Research showing that a substantial proportion
of femicides are connected to separation (or the threat of) sup-
port strongly the idea of male sexual proprietariness and the
need for males to control female reproduction (Daly & Wilson,
We may not be able to predict on an individual basis whether
domestic violence is likely to occur in specific households, but
evolutionary thinking has highlighted when it is most likely to
occur and who is the likely perpetrator. Such knowledge must
be used to prevent the escalation to lethal violence by, for ex-
ample, making it easier and more practical for women to leave
their violent partners without fear of reprisal, as we know plan-
ning to leave is common flash-point in violence by men against
women. We give Daly and Wilson a deserved last word;
“Killing is just the tip of the iceberg: For every murdered
wife, hundreds are beaten, coerced and intimidated. Although
homicide probably does not often serve the interests of the per-
petrator, it is far from clear that the same can be said of
sub-lethal violence” (1988, p. 205).
From Just So Stories to Evidence-Based Crime
Perhaps the first task is education of the next generation of
criminologists. This is no small order given the accumulated
hostility, sociological and religious and the complexity of mod-
ern evolutionary biology, whose mastery is a necessary condi-
tion of defensible contributions on violence prevention made by
criminologists. Some are tentatively introducing Darwinian
thinking into their books (see for example Wortley, 2011) but
they are rare, and the present writers stress their own inadequa-
cies in treating with evolutionary biology writings. The Catch
22 is that the points of possible connection have to be simple
enough to be comprehensible to a skeptical audience, which
makes them vulnerable to accusations of superficiality and the
Just So criticism. So what is presented below is a set of possi-
bilities ranked from those which many criminologists may per-
haps be prepared to consider to those which are currently re-
garded as difficult and alien.
A further complication is that those measures which have
most often proven effective in violence and harm reduction are
situational in character, and hence are best addressed in the
more sophisticated outcrops of Darwinian thinking. Laland and
Brown (2002) distinguished four major contemporary ap-
proaches; human behavioural ecology, evolutionary psychology,
memetics, and gene-culture co-evolution. The latter two adopt
the meme concept, i.e. the unit for carrying cultural ideas, sym-
bols, or practices, which can be transmitted from one mind to
another, cultural analogues to genes, in that they self-replicate,
mutate, and respond to selective pressures. Changes in crime
tactic memes co-evolving with thresholds of inclination to ag-
gress will certainly be necessary for short and middle-term
variation in manifestations of violence.
Returning to a desensitization process for criminologists,
three stages may be envisaged.
Defense mechanism analogies. Just as the products Smoke
J. ROACH ET AL.
Cloak and Smoke Bandit simulate the defence tactic of the
squid, and the panic alarm can be regarded as equivalent to
alarm calls, are there other analogues? Is defecation an effective
last ditch tactic to avoid rape? Are there diversionary tactics to
avoid attack on one’s family equivalent to birds feigning injury
to divert predators from the nest? In what circumstances are
mimicking less vulnerable species (as the hoverfly the wasp)
relevant in avoiding violence (see Felson, 2006 for some sug-
gestions), Interesting criminologists in such questions may
invite derision from evolutionary biologists, but may be a use-
ful hook for criminological interest, and can be presented
purely as what they are, namely no more than analogies.
Bootstrapping on current interests. Places differ in their
proneness to crime, including violent crime. One key variable
appears to be the permeability of residential areas. Cul-de-sacs
(particularly sinuous cul-de-sacs) are massively less prone to
victimization (Johnson & Bowers, 2010). Intuitively this is
because people enter such roads because they live there or to
visit. The have a reason they can express or defend. In evolu-
tionary psychology, Dunbar’s number. Robin Dunbar (1992)
surveyed village and tribe sizes and settled on 150 as the esti-
mated size of a neolithic farming village; 150 as the splitting
point of Hutterite settlements; 200 as the upper bound on the
number of academics in a discipline’s sub-specialization; 150
as the basic unit size of professional armies in Roman antiquity
and in modern times since the 16th century; and notions of
appropriate company size. In short, he concluded that 150 was
the neocortex-limited community size. Dunbar thinking could
be applied to people recognition patterns by street network and
layout to determine how they linked to patterns of assault and
other victimization. The second writer was unfortunate to be
present at a meeting of geographers and social scientists at a
prestigious university at which the Dunbar number was intro-
duced, to ill-deserved derisive laughter.
A second possible avenue of infiltration is surely due for the
trialing of child protection work on kin selection principles and
Daly and Wilson research.
Getting serious, this would consist of agent-based modeling
of gene-meme coevolution models to assess fit to distributions
of violence by time, place and person (see Cavalli-Sforza &
Feldman, 1981) for background. Such an approach would revo-
lutionise criminal career understanding and preventive sen-
tencing—but that would require persuading judges about evolu-
tionary understanding of crime. That is perhaps a bridge too far
to contemplate now.
Adler, C., & Polk, K. (2001). Child victims of homicide. Cambridge:
Cambridge University Press.
Allis, C. D., Jenuwein, T., & Reinberg, D. (2007). Epigenetics. New
York, NY: Cold Spring Harbour Laboratory Press.
Baron-Cohen, S. (2011). Zero degrees of empathy. London: Allen Lane.
Barr, C. S., Newman, T. K., Becker, M. L., Parker, C. C., Champoux,
M., Lesch, K. P., et al. (2003). The utility of the non-human primate
model for studying gene by environmental interactions in behav-
ioural research. Genes, Brain, and Behaviour, 2, 336-340.
Bernet, W., Vnencak-Jones, C., Farahany, N., & Montgomery, S. A.
(2007). Bad nature, bad nurture and testimony regarding MAOA and
SLC6A4 genotyping at murder trials. Journal of Forensic Psychiatry
and Psychology, 52, 1-9.
Berkowitz, L. (1989). Frustration-aggression hypothesis: Examination
and reformulation. Psychological Bulletin, 106, 59-73.
Brantingham, P. J., & Faust, F. L. (1976). A conceptual model of crime
prevention. Crime and Deli n q u e n c y, 22, 130-146.
Brookman, F. (2000). Dying for control: Men, murder and sub-lethal
violence in England and Wales. British Criminology Conference
Proceedings, 3. URL (last checked 23 June 2010).
Brookman, F. (2005). Understanding homicide. London: Sage Publica-
Buss, D. M., & Shackelford, T. K. (1997). Human aggression in evolu-
tionary psychological perspective. Clinical Psychology Review, 17,
Buss, D. M. (2000). The dangerous passion. New York, NY: Free
Buss, D. M. (2004). Evolutionary psychology: The new science of the
mind (2nd ed.). Boston: Pearson.
Buss, D. M. (2005). The murderer next door. New York, NY: Penguin
Buss, D. M., & Malamuth, N. M. (1996). Sex, power, conflict. New
York, NY: Oxford University Press.
Campbell, A. (2002). A mind of her own: The evolutionary psychology
of women. Oxford: Oxford University Press.
Campbell, A. (2009). Gender and crime. In A. Walsh and K. M. Beaver
(Eds.), Biosocial criminology: New directions in theory and research.
New York, NY: Routledge.
Campbell, J. (1992). If I can’t have you, no one can: Power and control
in homicide of female partners. In J. Radford and D. E. H. Russell
(Eds.), Femicide: The politics of woman killing. Buckingham: Open
Caspi, A., McClay, J., Mofﬁtt, T. E., Mill, J., Martin, J., Craig, I., Tay-
lor, A., & Poulton, R. (2002). Evidence that the cycle of violence in
maltreated children depends on genotype. Scien ce , 297, 851-854.
Cavalli-Sforxa, L. J., & Feldman, M. W. (1981). Cultural transmission
and evolution: A quantitative approach. Princeton, NJ: Princeton
Chagnon, N. A. (1983). Yanamamö: The fierce people (3rd ed.). New
York, NY: Holt, Rinehart and Winston.
Clarke, R. V. (1992). Situational crime prevention: Successful case
studies. New York, NY: Harrow & Heston.
Clarke, R. V. (1997) Situational crime prevention: Successful case
studies. (2nd ed.). New York, NY: Harrow & Heston.
Clarje R. V., & Mayhew, P. (1988). The British gas suicide story and
its criminological implications. Crime and Justic e, 10, 79-116.
Cohen, L. E., & Felson, M. (1979). Social change and crime rate trends.
American Sociological Review, 44, 588-608. doi:10.2307/2094589
Collins, W. A., Maccoby, E. E., Steinburg, L., Hetherington, E. M., &
Bornstein, M. H. (2000). Contemporary research on parenting: The
case for nature and nurture. American Psychologist , 55, 218-232.
Coomaraswamy, K. S., & Shepherd, J. P. (2003). Predictors and sever-
ity of injury in assaults with barglasses and bottles. Injury prevention,
9, 81-84. doi:10.1136/ip.9.1.81
Coyne, J. A. (2009). Why evolution is true. Oxford: Oxford University
Crawford, A. (1998). Crime prevention and community safety: Politics,
policies and practices . Harlow: Longman
Creighton, S.J. (1989). Child Abuse Trends in England and Wales
1983-1987. London: NSPCC.
Daly, M., & Wilson, M. (1988). Homicide. Hawthorne, CA: Aldine de
Daly, M., & Wilson, M. (1998). The truth about Cinderella: A Dar-
winian view of parental lo ve . London: Orion Publishing.
J. ROACH ET AL. 403
Dobash, R. E., & Dobash, R. P. (1979). Violence against wives. New
York, NY: Free Press.
Dollard, J., Doob, L., Miller, N., Mowrer, O., & Sears, R. (1939).
Frustration and aggression. New Haven, CT: Yale University Press.
Dunbar, R. I. M. (1992). Neocortex size as a constraint on group size in
primates. Journal of Human Ev olu tion, 22, 469-493.
Ekblom, P. (1997). Gearing up against crime: A dynamic framework to
help designers keep up with the adaptive criminal in a changing
world. International Journal of Risk, Security and Crime Prevention,
Ekblom, P. (1999). Can we make crime prevention adaptive by learning
from other evolutionary struggles? Studies on Crime and Crime Pre-
vention, 8, 27-51.
Ekblom, P., & Sidebottom, A. (2007). What do you mean, “Is it se-
cure?” Redesigning language to be fit for the task of assessing the
security of domestic and personal electronic goods. European Jour-
nal on Criminal Policy and Research, 14 , 61-87.
Farrington, D. P. (1988). Studying changes within individuals: The
causes of offending. In M. Rutter (Ed.), Studies of psychosocial risk:
The power of longitudinal data. Cambridge: Cambridge University
Farrington, D. P. (2003). Developmental and life-course criminology.
Criminology, 41, 201-235. doi:10.1111/j.1745-9125.2003.tb00987.x
Felson, M. (1994). Crime and everyday life (1st ed.). Thousand Oaks,
CA: Pine Forge Press.
Felson, M. (2002). Crime and everyday life (3rd ed.). London: Pine
Felson, M. (2006). Crime and nature. London: Sage.
Foley, D., Eaves, L., Wormley, B, Silberg, J., Maes, H., Hewitt, J.,
Kuhn, J., & Riley, B. (2004). Childhood adversity, MAOA genotype,
and risk for conduct disorder. Archives of General Psychiatry, 61,
Gao, Y., Raine, A., Venab les, P. H., Dawson, M. E., & Mednick, S. A.
(2010). Association of poor childhood fear conditioning and adult
crime. American Journal of Psychiatry, 167, 156-160.
Goetz, A. T., Shackelford, T. K., Starratt, V. G., & Mckibbin, W. F.
(2008). Intimate partner violence. In J. D.Duntley and T. K.
Shackelford (Eds.), Evolutionary forensic psychology: Darwinian
foundations of crime and law. New York, NY: Oxford University
Gibson, J. J. (1950). The perception of the visual world. Boston, MA:
Hall, P., & Innes, J. (2011). Violent and sexual crime. In J. Flatley, C.
Kershaw, K. Smith, R. Chaplin and D Moon, (Eds.), Crime in Eng-
land and Wales 2009/10: Findings from the British Crime Survey
and police recorded crim e (3rd ed.). London: Home office.
Holmes, R. (2009). The age of wonder. London: H arper.
Gilling, D. (1997). Crime prevention. London: Routledge
Goldblatt, P., & Lewis, C. (Eds) (1998). Reducing offending: An as-
sessment of research evidence on ways of dealing with offending
behaviour. Home Office Research Study, 187. London: HMSO.
Gottfredson, M., & Hirschi, T. (1990). A general theory of crime.
Stanford: Stanford University Press.
Gottschall, J. (2008). The rape of troy. Cambridge: Cambridge Univer-
Grossman, D. (2009). On killing (rev ed.) New York: Little Brown.
Hawton, K., Townsend, E., Deeks, J., Appleby, L., Gunnell, D., Ben-
newith, O., & Cooper, J. (2001). Effects of legislation restricting
pack sizes of paracetomol and salicylate on self-poisoning in the
United Kingdom: Before and after study. British Medical Journal,
322, 1-7. doi:10.1136/bmj.322.7296.1203
Heils, A., Teufel, A., Petri, S., Stober, G., Riederer, P., Bengel, B., &
Lesch, K. P. (1996). Allelic variation of human serotonin transporter
gene expression. Journal of N e urochemistry, 6, 2621-2624.
Hinshaw, S. P. (2002). Intervention research, theoretical mechanisms,
and causal processes related to externalizing behavior patterns. De-
velopment and Psychopathology, 14, 789-818.
Hirschi, T., & Gottfredson, M. (1983). Age and the explanation of
crime. American Journal of Sociology, 89, 552-584.
Innes, M. R. (2003). Investigating murder: Detective work and the
police response to criminal homicide. Oxford: Oxford University
Johnson, S., & Bowers, K. J. (2010). Permeability and burglary risk:
Are cul-de-sacs safer? Quantitative Journal of Criminology, 26,
Johnson, S. D., Summers, L., & Pease, K. (2009). Offender as forager?
A direct test of the boost account of victimisation. Journal of Quan-
titative Criminology, 25, 181-200.
Jones, O. D. (2005). Evolutionary psychology and the law. In D. M.
Buss (Ed.), The handbook of evolutionary psychology. New York,
Junger, M., Feder, L., & Cote, S. M. (2007). Policy implications of
present knowledge on the development and prevention of physical
aggression. European Journal on Criminal Policy and Research, 13,
Kendler, K. S., and Eaves, L. J. (1986). Models for the joint effect of
genotype and environment on liability to psychiatric illness. Ameri-
can Journal of Psychiatry, 143, 273-297.
Keverne, E. B., & Curley, J. P. (2008). Epigenetics, brain evolution and
behaviour. Frontiers in Neuroendo crinology, 29, 398-412.
Killias, M. (1993). International correlations between gun ownership
and rates of homicide and suicide. Canadian Medical Association
Journal, 148, 1721-1725.
Kim-Cohen, L., Caspi, A., Taylor, A., Williams, B., Newcombe, R.,
Craig, I. W., & Moffitt, T..E. (2006). MAOA, maltreatment, and
gene-environment interaction predicting children’s mental health:
New evidence and a meta-analysis. Molecular Psychiatry, 11,
Kramer, D. A. (2005). Nature, nurture and epigenetics. American Asso-
ciation of Child and Adolescent Psychiatry News, 24, 294-297.
Krug, E.G. et al. (2002). World report on violence and health. Geneva:
Lesch, K. P., Bengel, D., Heils, A., Sabol, S. Z., Greebderg, B. D., Petri,
S. et al. (1996). Association of anxiety-related traits with a polymor-
phism in the serotonin transporter gene regulatory region. Science,
274, 1527-1531. doi:10.1126/science.274.5292.1527
Lester, D., & Clarke, R. V. (1989). Effects of the reduced toxicity of
car exhaust on accidental deaths: A comparison of the United States
and Great Britain. Accident Analysis & Prevention, 21, 191-196.
Lew, T. A., Morrow, E. H., & Rice, W. R. (2006). Standing genetic
variance for female resistance to harm from males and its relation-
ship to intralocus se xual conflict. Evolution, 60, 97-105.
Loader, I., & Sparks, R. (2010). Public criminology. London: Routledge.
Lorenz, K. (1966). On aggression. London: Methuen.
Lott, J. R. (2006). More guns, less crime (3rd ed.). Chicago, IL: Uni-
versity of Chicago Press.
Maguire, M., & Brookman, F. (2005). Violent and sexual crime. In N.
Tilley (Ed.) Handbook of crime prevention and community safety.
Mednick, S. A., & Christiansen, K. O. (1977). Biosocial bases of
criminal behaviour. New York, NY: Gardner Press.
Mesquida, C., & Wiener, N. (1996). Human collective aggression: A
behavioural ecology perspective. Ethology and sociobiology, 17,
Mesquida, C., & Wiener, N. (1999). Male age composition and severity
of conflicts. Politics and the life sciences, 18, 181-189.
Milgram, S. (1977, 2010). The individual in a social world: Essays and
experiments (3rd ed.). New York, NY: Pinter and Martin
Mischel, W. (1968). Personality and assessment. New York, NY:
J. ROACH ET AL.
Mofﬁtt, T. E. (1993). Adolescence-limited and life-course-persistent
antisocial-behavior a developmental taxonomy, Psychological Re-
view, 100, 674-701. doi:10.1037/0033-295X.100.4.674
Mofﬁtt, T. E. (1997). Adolescent-limited and life-course persistent
offending: A complementary pair of developmental theories. In T.
Thornberry (Ed.), Developmental theories of crime and delinquency.
Advances in criminological theory. New Brunswick, NJ: Transaction
Mofﬁtt, T. E. (2003). Life-course-persistent and adolescent-limited
antisocial behavior. In B. B. Lahey et al. (Eds.), Causes of conduct
disorder and juvenile del i n qu e nc y. New York, NY: Guilford Press.
Moffitt, T. E., & Caspi, A. (2006). Evidence from behavioural genetics
for environmental contributions to antisocial conduct. In P. O. H.
Wikström and R. J. Sampson (Eds.), The explanation of crime: Con-
text mechanisms and development. Cambridge: Cambridge Univer-
Norman, D. (1998). The design of everyday things. Boston, MA: MIT
Nowak, M. (2011). Super cooperators. Lond on: C anongate.
Olweus, D. (1979). Stability of aggressive reaction patterns in males: A
review. Psychological Bulletin, 86, 852-875.
Pease, K. (2005). Science in the service of crime reduction. In N. Tilley
(Ed.), Handbook of crime prevention and community safety. Cul-
Pease, K. (2010). Crime Science. In S. Shoham (Ed.) International
handbook of penology and criminal justice. London: Taylor and
Polk, K. (1994). When men kill: Scenarios of masculine violence. Cam-
bridge: Cambridge University Press.
Rich, H. J. (1998). The nurture ass u mption. London: Bloomsbury.
Roach, J., & Pease, K. (in press). Evolution and crime. Cullompton:
Roach, J., & Shepherd, A. (2010). Thirty years of homicide in West
Yorkshire 1979-2009. A Report Commissioned by the Homicide and
Major Enquiry Team, West Yorkshire Police. Huddersfield: Univer-
sity of Huddersfield.
Rutter, M. (2001). How can we know environment really matters? In F.
Lamb-Parker, J. Hagen and R. Robinson (Eds.), Developmental and
contextual transition of children and families: Implications for re-
search, policy, and practice. New York, NY: Columbia University
Rutter, M., Pickles, A., Murray, R., & Eaves, L. (2001). Testing hy-
potheses on specific environmental causal effects on behavior. Psy-
chological Bulletin, 1 2 7 , 291-324 doi:10.1037/0033-2909.127.3.291
Rutter, M., (2003a). Commentary: Causal processes leading to antiso-
cial behavior. Develop me nt al Psychology, 39, 372-378.
Rutter, M. (2003b). Crucial paths from risk indicator to causal mecha-
nism. In B. Lahey, T. E. Moffitt and A. Caspi (Eds.), The causes of
conduct disorder and serious juvenile delinquency. New York, NY:
Serbin, L. A., & Karp, J. (2003). Intergenerational studies of parenting
and the transfer of risk from parent to child. Current Directions in
Psychological Science, 1 2 , 138-142. doi:10.1111/1467-8721.01249
Sherman, L. W, Gottfredson, D., Mackenzie, D., Eck, J., Reuter, P., &
Bushway, S. (1998). Preventing crime: What works, what doesn’t,
what’s promising. Was hington, DC: National Institute of Justice.
Smith, B., & Stevens, R. (2002). Evolutionary psychology. In D. Miell,
A. Phoenix and K. Thomas, (Eds.), Mapping Psychology, 1, Milton
Keynes: Open University Press.
Smith, P. H., Moracco, K., & Butts, J. (1998). Partner homicide in
context: A population-based perspective. Homicide Studies, 2/4,
Suomi, S. J. (1991). Up-tight and laid-back monkeys: Individual dif-
ferences in the response to social challenges. In S. Brauth, W. Hall
and R. Dooling (Eds.), Plasticity of development. Cambridge, MA:
Suomi, S. J. (1995). Influence of Bowlby’s attachment theory on re-
search on nonhuman primate biobehaviourial development. In S.
Goldberg, R. Muir and J. Kerr (Eds.), Attachment theory: Social, de-
velopmental, and clini cal perspectives. Hillsdale, NJ: Analytic Press
Szyf, M., Weaver, I., Proven cal, N., McGowan, P. O., Tremblay, R. E.
and Meaney, M. J. (2009). Epigenetics and behaviour. In R. E.
Tremblay, M. A. G.van Aken and W. Koops (Eds.), Development
and prevention of behaviour problems. New York, NY: Psychology
Suomi, S. J. (2009). How gene-environment interactions shape biobe-
havioural development. In R. E. Tremblay, van M. A. G.Aken and W.
Koops (Eds.), Development and prevention of behaviour problems.
New York, NY: Psychology Press
Thaler, R., & Ornstein, C. R. (2008) . Nudge. London: Penguin.
Thomson, H. (2010). Empathetic mirror neurons found in humans at
last. New Scientist, 2756, 12. doi:10.1016/S0262-4079(10)60911-6
Tilley, N. (Ed.) (2005). Handbook of crime prevention and community
safety. Cullompton: Willan.
Tilley, N. (2010). Crime prevention. Cullompton: Willan.
Tooby, J., & Cosmides, L. (1990). The past explains the present: Emo-
tional adaptations and the structure of ancestral environments.
Ethology and Sociobiology, 11, 375-424
Van Dijk, J. J. M.,& De Waard, J. (1991). A two-dimensional typology
of crime prevention projects: With a bibliography. Criminal Justice
Abstracts, 23, 483-503.
Wallace, A. (1986). Homicide: The social reality. Bureau of Crime
Statistics and Research. R esearch Study No. 5. Sydney: BCSR.
Warburton, A. L., & Sherherd, J. P. (2000). Effects of toughened
glassware in terms of reducing injury in bars: A randomized con-
trolled trial. Injury Prevention, 6, 36-40. doi:10.1136/ip.6.1.36
Weldon, S. L. (2002). Protest, policy and the problem of violence
against women: A cross-national comparison. Pittsburgh: University
of Pittsburgh Press.
Welsh, B., & Farrington, D. P. (2006). Preventing crime. Dordrecht:
Widom, C. S., & Brzustowicz, L. M. (2006). MAOA and the “cycle of
violence”: Childhood abuse and neglect, MAOA genotype, and risk
for violent and antisocial behavior. Biological Psychiatry, 60,
Wilson, M., & Daly, M. (1996). Male sexual proprietariness and vio-
lence against women. Current Directions in Psychological Science, 5,
Wortley, R. (2011). Psychological criminol o gy. London: Routledge.
Wright, R. (1994). The moral animal: The new science of evolutionary
psychology. New York, NY: Pantheon.
Zhong, H. (2005). The age-crime relationship across time and offense
types: A comparison of the United States and Taiwan. Doctoral Dis-
sertation. University Park, PA: Pennsylvania State University.
Zimbardo, P. (2007). The lucifer effect: How good people turn evil.
Zimring, F. E., & Hawkins, G. (1987). The citizen’s guide to gun con-
trol. London: Macmillan.