ArticlePDF Available

Sadistic cruelty and unempathic evil: Psychobiological and evolutionary considerations



Understanding the origins of evil behaviour is one of our most important intellectual tasks. A distinction can perhaps be drawn between overt sadistic cruelty and the lack of empathy to suffering that is a hallmark of evil. There is increasing data available on the prevalence, proximal psychobiological underpinnings, and distal evolutionary basis for these contrasting phenomena.
Cruelty’s rewards: The gratifications
of perpetrators and spectators
Victor Nell
Institute for Social and Health Sciences, University of South Africa, Pretoria,
South Africa. Correspondence to: West Hill House, 6 Swains Lane, Highgate,
London N6 6QS, United Kingdom.
To the memory of Linda Mealey
Abstract: Cruelty is the deliberate infliction of physical or psychological pain on other living creatures, sometimes indifferently, but
often with delight. Though cruelty is an overwhelming presence in the world, there is no neurobiological or psychological
explanation for its ubiquity and reward value. This target article attempts to provide such explanations by describing three stages in
the development of cruelty. Stage 1 is the development of the predatory adaptation from the Palaeozoic to the ethology of
predation in canids, felids, and primates. Stage 2, through palaeontological and anthropological evidence, traces the emergence of
the hunting adaptation in the Pliocene, its development in early hominids, and its emotional loading in surviving forager societies.
This adaptation provides an explanation for the powerful emotions high arousal and strong affect evoked by the pain-blood-
death complex. Stage 3 is the emergence of cruelty about 1.5 million years ago as a hominid behavioural repertoire that promoted
fitness through the maintenance of personal and social power. The resulting cultural elaborations of cruelty in war, in sacrificial
rites, and as entertainment are examined to show the historical and cross-cultural stability of the uses of cruelty for punishment,
amusement, and social control.
Effective violence prevention must begin with perpetrators, not victims. If the upstream approaches to violence prevention
advocated by the public-health model are to be effective, psychologists must be able to provide violence prevention workers with a
fine-grained understanding of perpetrator gratifications. This is a distasteful task that will compel researchers to interact with
torturers and abusers, and to acknowledge that their gratifications are rooted in a common human past. It is nonetheless an
essential step in developing effective strategies for the primary prevention of violence.
Keywords: compassion; cruelty; entertainment industry; evolutionary psychology; intraspecific killing; pain; predation; punishment;
torture; violence prevention
1. Introduction
Cruelty (from the Latin crudelem, “morally rough”) is the
deliberate infliction of physical or psychological pain on a
living creature; its most repugnant and puzzling feature is
the frequently evident delight of the perpetrators. Cruelty
has an overwhelming presence in the world – in wars and
massacres, in the routine work of police and military inter-
rogators, in children’s play, and in the dealings of men with
women and of adults with children. Although the ease with
which situations can overwhelm values and elicit cruelty in
hitherto irreproachable individuals is empirically (Haney
et al. 1973; Milgram 1969/1974; Zimbardo 2003) and
observationally (Browning 1993; Grossman 1995; Tester
1997) well established, there is no motivational or neuro-
biological explanation for cruelty’s prevalence or the fasci-
nation it holds.
This target article argues that the reinforcement value of
pain and bloodshed derives from the predatory adaptation
from the Middle Cambrian to the Pleistocene. The argu-
ment is therefore as follows:
1. Cruelty is a behavioural by-product of predation.
2. Cruelty is driven by reinforcers that derive from this
3. Because cruelty presupposes the intention to inflict
pain and is therefore exclusively a hominid behaviour, it
dates to no earlier than H. erectus, about 1.5 million
years ago (Ma).
4. Cruelty has fitness benefits in solving problems of
survival and reproduction in forager, pastoral, and urban
5. The enjoyment of cruelty is a culturally elaborated
manifestation of the predatory adaptation.
VICTOR NELL is Emeritus Professor of Psychology at
the University of South Africa, where he directed the
clinical neuropsychology training programme, and led
epidemiological studies of traumatic brain injury and
interpersonal violence in Johannesburg. He estab-
lished the university’s Institute for Social and Health
Sciences and headed a World Health Organisation Col-
laborating Centre for Injury and Violence Prevention,
which is now a South African Medical Research
Council Lead Programme on Crime, Violence, and
Injury. He has published books and book chapters
on neuropsychology, policing and police accountability,
theoretical social psychology, and the psychological
impact of Apartheid.
Printed in the United States of America
#2006 Cambridge University Press 0140-525x/06 $12.50 211
These hypotheses generate a research agenda for affective
neuroscience, for social psychology, and for violence pre-
vention. They also provide a heuristic for understanding
why media violence is attractive, why men find war beau-
tiful, why homicide has been a fixed feature of human
societies from prehistoric times to the present (Buss
2005; Daly & Wilson 1988), and why, despite the human
capacity for compassion, atrocities continue.
1.1. Three stages in the emergence of cruelty
1.1.1. Predation.The predatory adaptation derives from
resource competition between and within species, which,
in the Cambrian, becomes predation, the killing and con-
sumption of one living creature by another. Predation is
hard work: the evidence reviewed in section 3 shows
that it is powerfully reinforced in mammalian carnivores
and in the hunting apes by a set of linked conditioned
stimuli that are carried over to the hunting adaptation in
hominids. The stimuli driving predation and hunting are
the pain-blood-death (PBD) complex: the prey’s terror
and struggles to escape as it is brought down, the shedding
of its blood, and its vocalisations as it is wounded and
eaten, often while it is still alive. A range of anticipatory
and consummatory reinforcers is triggered by the PBD
complex, which is also active in intraspecific killing, and
strikingly so in chimpanzees. The material in section 3
on the neurobiology of predation suggests that predation
is dopaminergic, affectively positive, and distinct from rage.
1.1.2. Hunting.Nutritional killing by hominids is also hard
work: the palaeontological and anthropological evidence
reviewed below suggests that hunting in hominids, as
with predation in canids, felids, and primates, is reinforced
by the PBD complex and that the nonnutritional “other
end” of hunting, for which anthropologists have sought,
derives from these reinforcers.
1.1.3. Power.Cruelty requires a sufficient cognitive basis
for intentionality and a sufficientsocial basis for its disciplin-
ary elaboration (see sect. 5). Once these foundations have
been laid, there are florid social and cultural elaborations
of cruelty as punishment, for amusement, and for social
control. Each of these modalities affirms the power of the
perpetrator – this may be an individual acting alone or as
the agent of a collective over the victim. In hierarchical
states with centralised power, cruelty becomes a vehicle
for public entertainments that buttress the power of the
state and heroise war. The affective loading of these elabor-
ations is described in order to identify parallels between
blood as a principal reinforcer of predators and hunters
on the one hand, and, on the other, of the audiences that
delight in spectacles of pain and bloodshed.
In Stage 3, the use of cruelty is a strongly male-gendered
and contextually sensitive adaptation, which “could remain
dormant for the entire life of an individual, if the relevant
contexts are not encountered” (Buss 1999, p. 284),
promoting inclusive fitness by augmenting the personal
power, survival, and sexual access of cruel individuals.
Historically, the enjoyment of cruelty has been sufficiently
powerful to have led to huge social resources being chan-
nelled into cruel rites and spectacles, and this enjoyment
remains a primary driver of the modern entertainment
industry. The distinction between use and enjoyment has
behavioural and neurobehavioural implications that may
have animal parallels with quiet-biting predation on the
one hand and aggressive rage on the other. However, as
with all behavioural states, the boundaries between instru-
mentality and affectivity are permeable: for example,
hunters may inflict pain on the prey beyond that which
is instrumentally necessary, and the hunt may slip into
surplus killing that continues beyond the satisfaction of
nutritional needs (as with Actaeon in Ovid’s Metamor-
phoses,c. 8 AD/1997, p. 105). Violence is a significant
by-product of cruelty (see sect. 6).
Evidence for the continued salience of the predatory
adaptation for human behaviour is derived from palaeon-
tology and taphonomy (Brain 1981); predator ethology;
primatology, with special reference to chimpanzee preda-
tion and intraspecific killing; cognitive evolution with
special reference to language; the psychology of moti-
vation and learning; the anthropology of provisioning;
societal evolution; cultural history; and the psychology of
individual differences.
The reinforcers of cruelty feed into violence, defined by
the World Health Organisation as the intentional use of
physical force or power – against oneself, another individ-
ual, or a group – that causes injury, death, or psychological
harm (Krug et al. 2002, p. 5): one of this target article’s
purposes is to show cruelty’s relevance to the initiation
and escalation of high-volume everyday violence such as
drunken brawls, child beating, and sexual assault.
The study of cruelty, which is one of the manifestations
of evil, is dangerous on three counts: first, because of the
fear that evil is contagious, and that those who deal with
it become tainted (as, for example, in “The Problem of
Evil” in Coetzee 2003); second, because to probe the psy-
chology of perpetrators fails to condemn, casting a shadow
over the researcher’s rectitude; and third, because rooting
cruelty in the human evolutionary past appears to natura-
lise it, absolving perpetrators and their audiences of moral
responsibility. The study of cruelty neither contaminates
nor condones, and the purpose of this article is compassio-
nate and preventive. Cruelty will not be contained through
obscurantism. Its reinforcers must be understood, and
if these have evolutionary origins, effective prevention
requires that they be revealed.
2. A taxonomy of cruelty
The preconditions for cruelty are a mental state, namely
the intention to inflict pain, which in turn presupposes
a theory of mind (Premack 1988), as well as an action,
which is the deliberate infliction of physical or psychologi-
pain on another living creature, or on the self.
Punishment is cruel if its purpose is not to vanquish or
disable the victim, but to inflict pain; if the victim has no
control over the intensity or duration of the pain; and if
the victim is physically restrained or otherwise rendered
helpless. Punishment may also be used for social control
and discipline: here, the preconditions are that the
reason for the punishment is communicated to the
victim, and that the punishment is derived from a penal
code, is imposed by a higher authority, and is implemented
by agents of that authority.
Affectivity. Cruelty’s affective state is ferocity (from the
Latin ferox, “fierce,” now in the sense of savage violence).
Nell: Cruelty’s rewards: The gratifications of perpetrators and spectators
Cruel acts arouse strong positive or negative emotions in
the perpetrator and the audience, although habituation
and instrumentality may attenuate them. Whether or not
the conditions for punishment are met, an act is cruel if
the perpetrator or the audience experiences physiological
or psychological arousal triggered by the victim’s pain.
Entertainment is cruel if the audience is aroused by the
intentional shedding of blood or infliction of pain; the
infliction of pain for amusement is always cruel.
These definitions hold regardless of the perpetrator’s
position on a continuum ranging from instrumental
cruelty, marked by the perpetrator’s emotional coldness
and distance from the victim, to expressive or affective
cruelty, marked by the perpetrator’s escalating arousal.
Exclusions. These definitions of cruelty exclude pain
that results from fighting, killing, and war,
in which the
goal is not to inflict pain but to cause the adversary’s
flight, submission, or death, and also exclude pain that is
a by-product of treatment intended to cure or heal.
3. Stage 1: The predatory adaptation
3.1. Antecedents of predation
Predation’s precursor is competitive aggression, which
confers fitness by solving an animal’s problems in relation
to self-preservation, protection of the young, and resource
competition (Archer 1988, p. ix); this competition began
three billion years ago with the first primordial cell, a
benthic procaryote, which, “outreproducing its competi-
tors, took the lead in the process of cell division and evol-
ution” and made the world we know (Alberts et al. 1989,
p. 10). Organisms at a primitive level of neural organis-
ation and without specialised effector organs are capable
of aggression. For example, intertidal molluscs such as
limpets and chitons show spatial aggression by crawling
over a rival conspecific and trying to dislodge it from its
rock crevice by backward and forward movements; and
the nematocysts of the solitary anemone Actinia equina
are used for offence against conspecifics, with the loser
detaching from the substrate (Archer 1988, pp. 18 19).
The earliest evidence of predation in the fossil record is
from the terminal Proterozoic, 600 Ma, from which Clou-
dina fossils with tiny rounded holes have been recovered,
suggesting that the attacking organism was a predator,
selecting its prey for size (Brain 2001). With the Middle
Cambrian explosion of animal life, 540 23 Ma, the first
effective predators emerged, with sense organs to locate
prey and the ability to pursue and overpower it. The
largest and most fearsome of these was Anomalocaris,an
active swimmer growing up to 50 cm with two large
eyes; Opabinia, another Burgess Shale organism, “had
five large eyes at the front of the head and a long flexible
proboscis that ended in an array of grasping spines”
(Brain 2001, p. 23).
3.2. The ethology of predation
Predation is widespread in the animal kingdom. Salticids,
the largest family of spiders, have elaborate, vision-
mediated predatory behaviour that is prey-specific, with
behavioural flexibility that includes conditional predatory
strategies, trial-and-error to solve predatory problems,
and detours to reach prey (Jackson & Pollard 1996);
there is similar flexibility in the predatory behaviour of
Pacific white sharks (Klimley 1994) and electric rays
(Lowe et al. 1994).
The ethology of mammalian predation is now reviewed
in relation to the arousal level, sensory feedback, and
biochemical neurobiological drivers of the search-swoop-
kill-feed cycle. Photographic evidence and the field obser-
vations reported below show that this cycle is accompanied
by a range of auditory, visual, olfactory, tactile, gustatory,
and visceral stimuli which together make up the PBD
3.2.1. Hyenas and lions.In his Serengeti notebook,
Kruuk described an adult male wildebeest turning to
confront four spotted hyenas who had pursued it at
speeds of 4050 kph over a 3 km distance:
The hyenas tried to bite him in the hindquarters, sides, and
especially the testicles, while he in turn struggled to horn his
attackers.... All four [hyenas] bit simultaneously at the loins,
testicles, and anal region of the wildebeest, paying little atten-
tion to his horns. The mobility of the victim was much
impaired by the four pursuers hanging onto his hindquarters.
Another two minutes later the wildebeest had a large gash in
the right loin, the testicles had been bitten off, and he stood
as if in a state of shock. Occasionally he made some frantic
movements and was able to struggle free from the hyenas,
but then some member of the pack would renew the
attack.... Eight minutes after the wildebeest had stopped
running he went down and the hyenas stood over him
pulling out his insides. Another two minutes later, the wilde-
beest died. (Kruuk 1972, p. 149)
Like wild dogs (Van Lawick 1977, pp. 24243, 24647),
hyenas “kill the victim by eating it” (Kruuk 1972, p. 153), in
that the animal may be struggling and vocalising as feeding
begins and may die up to a quarter hour later. The belly
and loins are torn open; the fetus is eaten if the prey is
pregnant; the testicles or udder is eaten; the stomach is
pulled out; and the stomach wall is eaten and the contents
spilled on the ground (Kruuk 1972, p. 125). Mills’s
descriptions (1990, p. 103 and Fig. 3.25) and photographs
(Mills & Harvey 2001, pp. 6669) of spotted hyenas
hunting and feeding in the Kalahari, and Van Lawick’s
(1977, pp. 18687) for the Serengeti, show virtually
identical behaviour.
Lions kill by slow strangulation, biting the throat of their
prey: death is rapid for small prey but may take an hour for
an adult wildebeest (Schaller 1973, p. 31) while it struggles
to escape.
Auditory stimuli. Most prey species emit distress calls as
they are wounded during the kill zebras give a high,
intense scream that is quite different from their bark or
snort alarm calls; wildebeest and buffalo bleat or moan,
like an intensified lowing (Mills, personal communication,
November 22, 2001). Schaller (1973) describes the “wild
... frenzied cry of a dying zebra” (p. 97), and Kruuk
(1972) writes of wildebeest “moaning at the incessant ...
bites” inflicted by hyenas (pp. 27, 29). A Thomson’s
gazelle fawn pursued by a hyena “jumped, ran, bleated
until the hyena’s jaws closed around its shoulders”
(Kruuk 1972, p. 25). Lions dig out a warthog burrow, the
animal finally bolts, and, “amid screaming cries from the
pig, the lions ...tear it apart” (Mills & Harvey 2001, p. 46).
Olfactory stimuli are equally rich. Schaller arrived at a
fresh zebra kill to find “the air heavy with the odors of
blood and sour rumen contents” (Schaller 1973, p. 97).
Nell: Cruelty’s rewards: The gratifications of perpetrators and spectators
Visceral reinforcers operate through gastric distention and
satiation. Hyenas, for example, gorge themselves at great
speed: Kruuk describes a pack of 25 hyenas completely
consuming a zebra and her foal within 40 minutes (Kruuk
1972, p. 16). Tactile stimuli include proprioceptive feedback
as the prey is clawed and bitten, and the prey’s bucking,
writhing, kicking, and goring as it attempts to escape.
Arousal level. The predatory cycle is highly energised.
Schaller writes that “at no other time do animals convey
such a high level of mental and physical tension” (1973,
p. 25). Kruuk (1975) describes hyena hunts as “wild and
exciting.... there is the sudden action, the wild run, the
gasps of the victim.... Then the kill, steaming in the
chill air, with a hyena cacophony over and around it”
(pp. 23, 33). Lion hunts are attended by the same high
arousal (Mills & Harvey 2001, pp. 4445).
Arousal during feeding. High arousal is maintained
during the feeding phase as the predators scan for scaven-
gers, chase them off, and jostle one another (Schaller 1973,
p. 83), while hyenas also feed in large competitive groups;
a single hyena may be overwhelmed by vultures (Van
Lawick 1977, pp. 18889). For lions, scavenging from a
hyena kill (and vice versa: Van Lawick 1977, pp. 98 99)
is dangerous work accompanied by loud vocalisations.
3.2.2. Baboons and capuchins.One of the earliest
authenticated cases of baboon predation is an eyewitness
description with photographs in Dart 1957 (Figs. 11 and
12). The Gilgil baboons in Kenya hunt cooperatively and
eat meat once a day more often than any other non-
human primate population (Strum 1981, in Stanford
1999). New World capuchins “hunt as avidly and success-
fully as chimpanzees” (Stanford 1999, p. 30), preying on
squirrels, tamarin monkeys, and immature coatis. Like
chimpanzees, they have a high brain-to-body-mass ratio.
3.2.3. Chimpanzees.At the Gombe, Taı
¨, Mahale, and
Kibale research sites, chimpanzees hunt red colobus
monkeys as well as other primate and ungulate species
(Mitani & Watts 1999). Hunting is coalitionary (Boesch
1994): for example, a group of Gombe chimpanzees
locates a troop of red colobus and posts drivers and block-
ers; the trap closes, and the colobus retreat to the highest
branches: “all the forest is screaming, meat is so rare and
so special, there is huge excitement” (soundtrack, National
Geographic, 1995). The prey is often an immature colobus
“that is grasped by the hands, pinned to the branch, and
bitten through the rear of the skull or the neck” (Stanford
1999, p. 96). Chimpanzees are highly successful hunters
(Stanford 1999; Stanford et al. 1994; Wrangham &
Peterson 1996, p. 216), and arousal during hunts is very
high, with pant-hooting, screaming, whistling, piloerection
to exaggerate body size, charge displays, and the shaking
of tree branches (Michael L. Wilson, personal communi-
cation, April 24, 2001). At all the sites,
the chimpanzees’ visceral reaction to a hunt and kill is intense
excitement. The forest comes alive with the barks and hoots
and cries of the apes, and aroused newcomers race in from
several directions. The monkey may be eaten alive, shrieking
as it is torn apart. Dominant males try to seize the prey,
leading to fights and charges and screams of rage. For one or
two hours or more, the thrilled apes tear apart and devour
the monkey. This is blood lust in its rawest form (Wrangham
& Peterson 1996, p. 216; see also pp. 10–11).
Bonobos, on the other hand, do not prey on monkeys and
are socially more peaceable than their close relations, the
chimpanzees: Wrangham and Peterson (1996, p. 219)
speculate that as predation was suppressed, so was intras-
pecific violence.
3.3. Intraspecific killing
The array of sensory and autonomic reinforcers that
operate during nutritional hunting is also activated when
conspecifics are attacked, wounded, or killed, as with
Norway rats (Blanchard et al. 1995) and wild rats
(Niehoff 1999, p. 61). Hyenas and lions defend their
home ranges vigorously. Kruuk records four sightings of
hyenas dead near the site of a kill “with clear evidence
that they were killed by other hyenas” (1972, p. 256);
Schaller (1973, p. 76) documents territorial killing in
Serengeti lions.
Among chimpanzees, alpha-male unseating can lead
to life-threatening or fatal wounds (De Waal 1989;
Wrangham & Peterson 1996), and territorial defence
may involve lethal violence. As with colobus hunts, these
interband confrontations are marked by intense excite-
ment that appears indistinguishable from that during pre-
dation. Goodall’s early account of such intergroup violence
(Goodall 1990, p. 89) has now been supplemented by
Wrangham and Peterson (1996) and by Wilson et al.
(2001). Though rarer than nutritional hunting, chimpan-
zee intraspecific killing is frequent enough to account
for between 24% and 52% of Gombe male mortality
(Wrangham & Peterson 1996, pp. 271 72).
3.4. The neurobiology of predation
Three distinct aggressive circuits in the mammalian brain
are evoked by electrical stimulation of the brain (ESB) of
three slightly different brain areas, namely predatory
aggression; intermale territorial and sex-related aggres-
sion; and angry aggression (RAGE in the terminology of
Panksepp 1998, pp. 51, 188).
3.4.1. Predatory aggression.Predatory, quiet-biting
aggression is mediated by the SEEKING system, a for-
aging, exploration, curiosity, and expectancy system “that
leads organisms to eagerly pursue the fruits of their
environment.... Energy is delight” (Panksepp 1998, pp.
145, 164), and SEEKING is its vehicle. Predatory aggres-
sion is quiet, with methodical stalking and well-directed
ESB, in the ascending dopamine pathways from mid-
brain nuclei through the extended lateral hypothalamic
corridor from the ventral tegmental area to the nucleus
accumbens, evokes the most highly energized exploratory
and search behaviours of which the animal is capable
(Panksepp 1998, p. 145). The emotional tone of affective
attack is unpleasant (see sect. 3.4.2), but the hunt and
kill are positive emotional experiences for the predator
(Panksepp 1998, p. 188).
The most effective quiet-biting attack electrodes always evoke
self-stimulation.... [Self-stimulation and predatory aggression]
are two behavioural expressions of SEEKING tendencies
that emerge from homologous systems in the brains of differ-
ent species. The species-typical expressions of this system lead
to foraging in some species and predatory stalking in others.
(Panksepp 1998, p. 194)
Nell: Cruelty’s rewards: The gratifications of perpetrators and spectators
Though SEEKING is dopaminergic, the pleasures of
the consummatory processes (feeding, sex) diminish
arousal in the SEEKING system (Panksepp 1998,
p. 147) and are strongly linked to brain opioid systems
which “may participate in every pleasure, serving as a
general neurochemical signal that the body is returning
to homeostasis” (Panksepp 1998, p. 184).
3.4.2. Affective aggression.RAGE circuits run from the
medial areas of the amygdala through the hypothalamus
and down to the dorsal periaqueductal grey (PAG). Affec-
tive attack sites yield escape behaviours (Panksepp 1998,
p. 195), and most animals soon learn to turn off rage-
inducing ESB (Panksepp 1998, p. 194). High testosterone,
high MAO-A, and low serotonin potentiate aggression; in
“tournament species,” testosterone is highest in the breed-
ing season (Panksepp 1998, p. 189).
Affective attack, whether offensive or defensive (this
latter is a mix of RAGE and FEAR), has marked display
features – piloerection with noisy hissing and growling
(the chimpanzee vocalisation during affective attack is the
pant-hoot). Both quiet-biting attack and self-stimulation
are evoked by ESB to the PAG of the midbrain, whereas
the dorsal PAG evokes affective attack and aversive
3.4.3. Predation in relation to aggression.It has been
customary to make a clear distinction between predation
and aggression. Archer (1988) holds that “so-called
predatory aggression is so motivationally and neurally
different from other forms of aggression that it is most
usefully considered as a separate form of behaviour”
(p. 25; also Lorenz 1963/2002 and Niehoff 1999).
Panksepp’s model accords more parsimoniously with
the above behavioural accounts of predation and intra-
specific killing, which suggest that predation and aggres-
sion are closely interwoven (see also Wilson 1975/2000,
p. 243), with quiet stalking (felids) or observation (canids
and hyenids) alternating with noisy defence of the kill. It
also provides a neurobehavioural basis for predation’s
distinctiveness, in that, first, predation and affective
attack have separate circuits in the brain; second, the
RAGE and SEEKING circuits have mutually inhibitory
interactions and cannot therefore co-occur; third,
predatory attack is endogenously generated because the
predatory cycle usually begins before the stimulus is
present – unlike affective attack, which is triggered by
the presence of the target; and fourth, it is accompanied
by positive affect, even though the energising contribution
made by hunger may be aversive, and, “from the animal’s
point of view, there is no apparent anger involved in this
food-seeking response” (Panksepp 1998, p. 198).
3.4.4. Endogenous opioids.The literature on the role of
the neuropeptides in predation, especially endorphins and
enkephalins, is sparse and contradictory: for example, that
microinjection of naloxone at PAG sites at which ESB
evoked quiet-biting attack in cats blocked predatory beha-
viour (Weiner et al. 1991); a later study (Manchanda et al.
1995) showed, on the other hand, that microinjection of an
enkephalin at excitatory PAG sites suppressed both the
somatomotor and affective display components of preda-
tory attack.
There is, however, a copious literature on opioid release
under predatory threat, which entrains a sequence of
defensive responses in prey that include hypoalgesia (in
mice exposed to a cat, Kavaliers & Colwell 1994, and to
insect stings, Kavaliers et al. 1998), and, as a final-stage
response, tonic immobility (Gargaglioni et al. 2001). Pre-
dator odours are highly salient in eliciting innate defensive
analgesia (Williams 1999). In humans, the release of
endogenous opioids in acute traumatic injuries correlates
significantly with physician pain ratings and scores on an
injury severity scale (Bernstein et al. 1995), suggesting
that anecdotal accounts of spontaneous analgesia in sol-
diers wounded in combat have a physiological basis.
The hunt and kill are a dangerous time for predators.
The prey butts, kicks, and gores, and scavengers must be
repulsed. If the predator is the scavenger as often
happens with hyenas and lions – the risk of injury
increases. The known links between consummatory pro-
cesses and brain opioid systems may therefore be augmen-
ted during the killingfeeding cycle by further opioid
release in response to injuries: an aspect of the predatory
adaptation may thus be an opioid “high” that is further
augmented by injury.
3.4.5. Pain and pleasure in predation, hunting, and
sexuality.The predatory cycle makes massive energy
demands of the predator: among them, a sustained high
level of autonomic arousal; the physical exertion of what
may be a prolonged high-speed chase; the act of killing,
during which the predator must overcome the last highly
energised struggle of the prey and the close-in hazards
of the kill; and scavenger threats. The aversive stimuli of
physical exertion to the point of exhaustion are augmented
by this final struggle. Yet the dopaminergic biochemistry
of the predatory cycle and ESB evidence of its reward
value indicate that far from being aversive, predation is a
powerfully rewarding experience even before satiation
One may thus hypothesise that a necessary condition for
the success of the predatory and hunting adaptations is the
conjunction of pain – the stress of exertion and the pain of
injury – with a high level of pleasurable reward intermixed
with sexual arousal, and that this is also true of fighting in
its various forms, including single combat, assaults by indi-
viduals or groups on rivals, and war: though fighting is by
definition not cruel, pain is inseparable from combat.
It is incomprehensible that the infliction of pain on the
self is both pleasurable and also sexually arousing. This
unlikely conjunction has long puzzled moral philosophers
and psychologists. In a famous passage, Freud wrote that
“the existence of a masochistic trend in the instinctual
life of human beings may justly be described as mysterious
from an economic point of view” (Freud 1924/1985,
p. 413). Yet, using functional magnetic resonance
imaging (fMRI), Becerra et al. (2001) report that a pain
stimulus (a probe heated to 468C applied to the skin) acti-
vated the brain’s reward circuitry, following a pathway
similar to that of the pleasure response: protein from the
cfos gene shows “that many neurons in the amygdala
that are aroused by aggressive encounters are also
aroused by sexual activity” (Panksepp 1998, p. 199): the
underlying motivation may be the seeking of safety.
The intertwining of aggression and sexuality is linguisti-
cally and ethologically apparent. The term for the !Kung
Nell: Cruelty’s rewards: The gratifications of perpetrators and spectators
hunting bow, n!au, is “a bawdy metaphor for the penis”
(Lee 1979, p. 207), and !Kung hunters say that “when
one’s heart is sweet with the thought of the kill, intercourse
is particularly good” (Lee 1979, p. 220). In primates and
humans, intermale territorial and dominance-seeking
aggression is driven by reproductive-fitness needs, with
females responding positively to aggressive success so
that the most vigorous males get preferential access to
reproductive opportunities (Panksepp 1998). Men with
absolute power may father several hundred children
(Ridley 1993; Wrangham & Peterson 1996, p. 234).
4. Stage 2: The hunting adaptation
Despite its high costs, the hunting adaptation mediates
powerful social and psychological rewards and is “the
most successful and persistent ... man has ever achieved”
(Lee & DeVore 1968, p. 3). It remains so in 58 surviving
forager societies from the equator to latitudes above 60
degrees, in which the contribution of hunting to annual
food intake converges on 35% (Lee 1968). The following
sections review the emergence of hominid hunting and
then, in two forager societies, the Dobe !Kung of Botswana
(Lee 1979; 1984) and the Yanomamo
¨of southern
Venezuela (Chagnon 1983), consider the reinforcements
that support the expenditure of large time and energy
resources on meat procurement.
4.1. Early hominid hunting
As hominids moved into the dry savannahs of the Pliocene,
the evolutionary shift from gatheringto meat eating required
major changes in sociality, brain size, and weapons (Stanford
1999). The nutritional accommodation of a big brain is
shrinking of the gut, which can be done only if there
has been a switch to easily digested and highly nutritious
foods (Aiello & Wheeler 1995) such as meat and
tubers, with preconsumption processing of chemically or
mechanically protected tubers (O’Connell et al. 1999).
The earliest fossil evidence of hominid meat eating is
the appearance of crude stone tools in east Africa in the
mid-Pliocene about 2.5 Ma, probably representing an
overlay of large mammal scavenging on a tradition of
small mammal hunting (Plummer & Stanford 2000);
killing or meat scavenging without tools may have
occurred much earlier but would not have left fossil
evidence (Stanford 1999). O’Connell et al. (1999) argue
that the earliest hominid meat eating is considerably
later, contemporaneous with the appearance of African
H. erectus about 1.8 Ma. Changes driven by “grand-
mothering” – foraging by postmenopausal women –
would have promoted larger group size, which in turn
brought advantages in defending against predators and
opened the way to aggressive scavenging (O’Connell
et al. 1999). Isotope evidence shows that archaic H.
sapiens were not only scavengers but also top-level carni-
vores, obtaining almost all of their dietary protein from
animal sources (Richards et al. 2000; but see Binford
1987 on H. erectus as primarily an aggressive scavenger).
The controlled use of fire, which Brain (2001) has dated
from 1.42 Ma in East Africa, served both to make meat
more palatable and to keep predators at bay: hominids
were both hunters and hunted (Brain 1981; Frison
1998), with both life-threatening dangers and nutritional
opportunities driving the development of hominid intelli-
gence. Folded within this brain development were the
emotional drivers of the predatory adaptation, responding
powerfully then as now to any opportunity to pursue,
butcher, and consume prey animals, whether as hunters
making the kill or as scavengers drawn to the kills of
other predators.
4.2. The high arousal of the hunt
4.2.1. Learning to hunt.The rough-and-tumble play of
young predators – rats, puppies, kittens – mimics the
techniques for tripping up, gripping, and biting prey
(Van Lawick 1977, pp. 16465; Panksepp 1998,
Fig. 15.2). There are aspects of children’s play that are
also a preparation for predation. Lee (1979, pp. 236 38)
describes the predation games of !Kung children; and
among the Yanomamo
¨, young boys capture a live lizard,
tie it to a stick in the village clearing, and gleefully shoot
featherless arrows at it with their miniature bows: “since
lizards are very quick and little boys are poor shots, the
target practice can last for hours” (Chagnon 1983, p. 118).
4.2.2. The hunter’s arousal.The large antelope species
and giraffe are hunted with bow and poisoned arrows.
The hunt proceeds through a cycle of stalking, wounding,
tracking, killing, and butchering. !Kung hunters, like
felids, are intensely focussed and silent stalkers: when
the prey is sighted, “one man moves forward, crouching
at first, then crawling, then inching forward on his belly”
(Lee 1979, p. 217). If the animal shows any signs of
alarm, the hunter freezes for several minutes at a time;
then, having reached bowshot range (10 m is the
optimum distance), he looses the first arrow. The time
for stealth now over, he breaks cover, running to intercept
the fleeing animal and shooting his remaining arrows at it.
The hunting group now tracks the wounded animal until
the poison takes effect and the animal collapses: “in all
cases a spear is methodically worked in and out of the
throat to ensure that the animal is dead” (Lee 1979,
p. 221). The party immediately sets about butchering the
animal, first skinning it, then removing the heart, liver,
and lungs and emptying the stomach contents. The liver
is cooked and eaten on the spot, and the long leg bones
may be split for the marrow, which is rubbed onto the
body. Blood from abdominal cavity is collected and
carried home in an empty stomach sac; during the night
after a kill, the hunter “is in a ritually heightened state”
(Lee 1979, p. 220). The excitement of the pursuit and
kill is no less for small game (Lee 1979, pp. 21621).
4.3. The status of hunting and the hunter
Lexical and narrative elaboration are markers for the social
salience of a phenomenon; similarly, elaborated memories
of distant events are evidence both of its social significance
and of high arousal at the time of the event: Rolls (1999)
notes that if a powerful reinforcer accompanies a situation,
many details will be stored, including memories of the
emotional state that accompanied that situation. This
storage may be implemented by nonspecific projecting
systems to the cerebral cortex and hippocampus, including
Nell: Cruelty’s rewards: The gratifications of perpetrators and spectators
cholinergic pathways in the basal forebrain and ascending
noradrenergic pathways.
Lee writes that “hunting vocabulary has undergone a
fantastic elaboration in !Kung speech ... there are many
dozens of synonyms, metaphors, and euphemisms”
(1979, p. 207) for stalking, shooting, fleeing, finishing off
the wounded animal, butchering, and so forth. Men tell
the story of the hunt round the fire “until the sky rips
open” [meaning until dawn breaks].... Graphic descrip-
tions of hunts, both recent and distant, constitute an
almost nightly activity.... Men can portray a hunt, step-
by-step, in microscopic and baroque detail” (Lee 1979,
pp. 207, 205) and give lifetime retrospective hunting
histories (pp. 23031).
Hunting success confers direct fitness benefits: Among
the Ache, “better hunters were more often named as
lovers by Ache women and better hunters had more
surviving children.... Better hunters had much higher
fertility than other men” (Hawkes et al. 2001, p. 134;
also Holmberg 1950).
4.4. The high costs of meat eating
The costs of meat eating are high for both predators and
4.4.1. Predators.The balance between the moose and
wolves on Isle Royale in Lake Superior – there are 20
25 wolves and 6001,000 moose – is maintained because
“it is very hard work to trap and kill a moose” (Wilson
1975/2000, p. 86). The wolves travel an average of 25
30 km a day during the winter, and one set of field obser-
vations showed that on 131 separate moose hunts, 77
resulted in a confrontation, and in only 6 of these were
moose killed. The kill success rate is 4.58%, and the
meat yield is 4 kg of meat per wolf per day. The success
rates for Kalahari spotted hyena are 63% for all encounters
with gemsbok calves, 14% for gemsbok adults, 39% for
wildebeest, and 31% for springbok (Mills 1990, pp. 94
110). For the Gombe chimpanzees, hunting is nutritionally
uneconomic: a 1 kg baby monkey is the typical yield
for a hunting party of up to 20, so that the effort expended
“is enormously costly relative to the quantity of meat that is
usually available” (Stanford 1999, p. 97).
4.4.2. Hunters.In human hunter-gatherer societies, the
meat yield is high for the group, but successes for the indi-
vidual hunter are sparse and unpredictable, with the daily
failure rate for individual Hadza hunters at 97%, and a
hunter may go days or weeks without a kill (Hawkes
et al. 2001); the !Kung hunting yield is 1 hr/100 calories.
Nonetheless, “the !Kung ...are willing to devote consider-
able energy to the less reliable and more highly valued
food sources such as medium and large mammals” (Lee
1968, p. 41).
Why do hunters make such large investments in a
nonessential resource? O’Connell et al. (1999) argue that
human paternal provisioning, a key aspect of the man-
the-hunter hypothesis, is absent in primates and is not
the purpose of human big-game hunting. If it were,
hunters would spend more time on small game and
plant foods, which are more reliable food sources: that
they do not “strongly suggests that big game hunting
serves some other end, unrelated to provisioning wives
and children” (p. 464). The material in this section,
taken together with the cultural elaborations of cruelty
(in sect. 5.2) suggest that this other end is the confirmation
of male sexual desirability through shedding the blood and
taking the life of big game, which is both scarce and
4.5. Pain, blood, and death in predation and hunting
Although the predatory cycle is endogenously generated,
usually beginning before the stimulus is present
(Panksepp 1998), the predator is greatly energised by
the prey’s presence and its actual or attempted flight,
which is a powerful trigger for pursuit and attack; by the
prey’s pain (ears, lips, and genitalia are ripped off, and
the prey is disembowelled while alive; hunters snare,
club, and stab living animals); and by the invariable
nexus between the infliction of pain and release of the
prey’s blood, which is a signal for the prey’s imminent
death: the muzzles of two spotted hyenas tearing at a
zebra’s stomach are red with blood (Mills & Harvey
2001, p. 128; Van Lawick 1977, pp. 18687); blood
smears the teeth, mandibles, and snouts of feeding lions
(Mills & Harvey 2001, p. 128; Schaller 1973, p. 21). It is
possible that in forager societies, blood has become a prin-
cipal conditioner of the reward system that drives preda-
tion, deriving its cultural weight (see sect. 5.2.6) from its
centrality in predation, hunting, and intraspecific violence.
Stimuli regularly attached to a rewarding activity
become conditioned reinforcers: During both mammalian
and hominid evolution, the prey’s flight and pain, and then
the sight, smell, and taste of blood, were prominent among
the reinforcers that shaped the predatory and hunting
adaptations. For predators, pain and blood signal satiation;
for humans, they are the harbingers not only of impending
satiation and sexual access, but also of the animal’s death,
which was bound up with the precarious survival of
Pleistocene hunters, who were also the hunted (Brain
4.6. The predatory transition
In historical context, these notions emerged in a largely
forgotten and, in its time, much derided 1953 paper
“The Predatory Transition from Ape to Man,” in which
the South African palaeontologist Raymond Dart charac-
terised Australopithecus africanus as “carnivorous crea-
tures that seized living quarries by violence, battered
them to death, tore apart their broken bodies, dismem-
bered them limb from limb, slaking their ravenous thirst
with the hot blood of victims, and greedily devouring
living writhing flesh” (Dart 1953, p. 209). Towards the
end of the paper, Dart observed that “the taste for
animal meat led inexorably ... to unspeakable cruelties”
(p. 219).
In the 1950s, Dart’s thesis that the australopithecines
were bloodthirsty murderers, and the suggestion that
their bloodlust was the foundation of human cruelty, was
ridiculed by palaeontologists. The orthodox and altogether
more optimistic view was that of Richard Leakey – that
early hominids were food-sharing foragers, and that
violence emerged only “when we became psycho-social
man probably 30 to 40,000 years ago” (in White 1985,
p. 7). Dart writes of himself that because of this onslaught,
Nell: Cruelty’s rewards: The gratifications of perpetrators and spectators
he was presumed to have retired “wounded or wroth....
into some parochial tent to brood upon the unresponsive
attitudes of my overseas colleagues” (1957, p. vii).
R. F. Ewer, Dart’s colleague, once remarked that Dart
was usually right though sometimes for the wrong
reasons. Bob Brain, Dart’s student and collaborator, who
described Dart as “this gentle, yet strangely bloodthirsty
man” (Brain 1993, p. 4), has shown that most of the
reasons Dart gave for his characterisation of the australo-
pithecines were indeed wrong (Brain 1981). But if, as this
paper has argued, there is a wide and accommodating
passage from predation to cruelty, he was right, after all.
5. Stage 3: The social uses of cruelty
Section 5.1 considers the points in hominid evolution at
which punitive and disciplinary cruelty could have
emerged in relation to the preconditions set by the taxo-
nomy of cruelty and examines its design features at these
points. Section 5.2 reviews the fitness benefits of cruelty
through its various social and cultural elaborations.
5.1. Emergence and design features of cruelty in relation
to cognitive and societal evolution
Punishment in the sense of inflicting pain on another crea-
ture has no preconditions: in this sense, the behaviour of
the rats, hyenas, lions, and chimpanzees described in
section 3 is indeed punishment, but the great apes
cannot meet the first of the criteria for disciplinary
cruelty, that the reason for the punishment must be com-
municated to the victim: Donald (1993) argues that
although the great apes are brilliant event perceivers,
with the capacity for social attribution and insight, they
are unable to communicate even their simplest intentions
because they cannot “actively shape and modify their own
actions or ...voluntarily access their own stored represen-
tations” (p. 739; Tomasello et al. 1998 cite field evidence
that apes cannot understand the communicative intentions
of others).
5.1.1. Mimesis as a sufficient basis for intentionality.
Mimetic communication “broke the hold of the environ-
ment on hominid motor behaviour” (Donald 1993,
p. 740). Using the whole body as a communication
device, body actions can be retrieved from memory,
replayed, stopped, and refined, allowing the development
of toolmaking, social expressiveness, and extended com-
petition, all with prosodic intonation of nonlinguistic
vocalisations. The transition to a mimetic culture with
H. erectus would therefore have increased the differences
between individuals and groups in the capacity for social
manipulation, fighting, physical dominance, and rewards
for competitive success; it also would have provided a
sufficient communicative basis for the emergence of the
preconditions for cruelty and disciplinary cruelty.
5.1.2. Stages of societal evolution.Johnson and Earle
(1987) identify three levels of socioeconomic integration,
emerging in sequence as population density increases:
the family-level group, the local group, and the regional
5.1.3. Cruelty in the family-level and local group.Hominid
forager societies, dating to no later than 1 Ma (Brain 2001),
are organised as family-level groups of some 25 50 indi-
viduals, as for example the !Kung San. Leadership in
most surviving forager societies is egalitarian, with consen-
sual decision-making and a strict humility ethic that effec-
tively blocks any aspirations to dominance and leadership
that a high-status and physically powerful individual might
have. In the local group (for example the Yanomamo
¨) with
aggregations of 300500 people, a strong charismatic
leader, the so-called big man, may emerge who maintains
group cohesion and negotiates intergroup alliances – but
in the absence of coercive social mechanisms, his power
lasts only as long as supporters’ loyalty. Disciplinary
cruelty cannot occur at these levels, because the requisite
hierarchical social structure with a penal code and judicial
system is unavailable.
5.1.4. Design features of cruel punishment in the family-
level and local group.However, there is a sufficient
linguistic and organisational basis to punish individuals
who threaten group survival: for example, by deliberately
frightening off game during a hunt or defecting from a
war party. The offended individuals and their kin would
have had the communicative and social resources to
restrain the offender and flog, break bones, or inflict
other exemplary pain. Since group disintegration would
have created major survival threats in relation to food
procurement, predator dangers, and attack by rival
groups, public punishment would have been strongly
adaptative by contributing to effective hunting, defence,
mate guarding, and stable food sharing.
Trinkaus (1992) observes that “the difficult existence of
the Neanderthals is reflected in their high frequency of
traumatic injury.... The remains of all older individuals
show signs of serious wounds, sprains, or breaks”
(p. 838; see also Walker 1989). The palaeopathological
evidence required to date the first emergence of cruel
punishment would be to differentiate bone fracture
caused by combat or accidental injury from the pre-
mortem twisting or other manipulation of broken bones,
which causes intense pain (Edgerton 1985, p. 135) and
would leave identifiable pre-mortem traces in the fossil
record (White 1985). The theory of mind requirement
(sect. 5.1.1) suggests that such evidence would not be
found earlier than H. erectus (c. 2 Ma).
Anthropological investigation of surviving forager and
pastoral societies might determine whether violation of
fundamental social norms elicits cruel punishment in the
family-level group (sect. 5.1.3) and whether disciplinary
cruelty is absent under big-man leadership (sect. 5.1.4),
first appearing in regional polities.
5.1.5. Disciplinary cruelty: The regional polity and early
state.Regional polities, whether as chiefdoms of several
thousand people or the empires that emerged in Egypt,
China, and India in the third millennium BC, brought
“not only dazzling advances in civilisation, but also the
enormously powerful instrument of state power as a new
moving force in history” (Heilbroner 1992, p. 907).
Disciplinary cruelty that meets the conditions in section
2 becomes a political imperative with the establishment of
conscript armies requiring the immediate punishment
of cowardice or desertion, the systematic slaughter of
Nell: Cruelty’s rewards: The gratifications of perpetrators and spectators
and the creation of slave or serf populations: the
costs of maintenance and subjugation are recovered
through coerced labour, necessitating an escalation in
the frequency and severity of punishment to maintain
productivity (on slave penal codes, see Kiefer 1938;
Hornblower & Spawforth 1999). Cruelty thus acquires
an economic driver.
5.1.6. Design features of cruelty in the regional
polity.Kings or emperors affirm their power as social reg-
ulators through carnivalesque public entertainments and
punishments in which the social purpose of cruelty is
manifest. The infliction of prolonged pain is an effective
way to establish and maintain social dominance; the
harsher or more painful the punishment, the greater the
relative status advantage of the perpetrator in relation to
the victim; and the more terrible the punishment, the
more permanent its effects on the social system.
5.1.7. The retained design features of cruelty.The social-
control functions of Roman and mediaeval carnivals of
death continue to the present in public entertainments
that are unwillingly stopped short of frank killing –
boxing and kickboxing, college football, car and motorcycle
racing. Animal baiting continues, in enclave groups in the
west, and more openly in other cultures. The willingness of
military establishments to develop technologies of cruelty
as instruments of war flourishes globally, and the coercive
forces of the state and its opponents use confessional and
disciplinary cruelty for political ends.
Deliberate infliction of pain, as with any other decisive
manifestation of interpersonal power, enhances the
status of the perpetrator. Accordingly, the initiation and
coordination of punishment in the family-level and local
group would have facilitated the emergence of a leadership
figure, whose willingness to injure would have created a
reputation for ferocity with significant resource access
benefits for that individual. Thus with Agathocles the Sici-
lian in the third century BC, whose “barbarous cruelty and
inhumanity, together with his countless atrocities” are
recounted with approval by Machiavelli (1513/1940,
p. 32) and with gang life in Glasgow’s Gorbals district:
describing the latter, Boyle (1977) tells how he slashed a
boy in a fight, the first time a knife had been used in a
gang fight in that area. Within days, Boyle was a force to
be reckoned with and placed on a pedestal by his own
gang. Today as in the past, aggression linked to a readiness
to inflict pain is a route to prestige, leadership, and social
mastery that entrains survival and reproductive benefits.
Empirically, it might be shown that group hierarchy rank-
ings are significantly altered by “cruelty rumours” – for
example, that a low-ranking member had tortured and
killed a rival. Cruelty attributions may elevate status, lea-
dership, and sexual attraction ratings more, for example,
than attributions of physical strength or intelligence.
5.2. Social and cultural elaborations of cruelty
The striking stability of the social uses of cruelty for pun-
ishment, amusement, and social control suggests that the
underlying motivational structures have a species-wide
evolutionary origin. Within the cultural sphere, “quite
new kinds of evolution may occur” with great rapidity,
spreading through a population or becoming extinct
within a single generation (Lea 1999, pp. 1718; cf.
Plotkin 1996). Whether a cultural innovation spreads or
becomes extinct will be determined by its social utility
and its contribution to individual fitness.
This section reviews evidence that throughout recorded
history and in a diversity of cultures, cruel entertain-
ments which as a means of social control also have a
fitness value – have attracted huge audiences. The attrac-
tions of war, the veneration of the warrior hero, and the
symbolic weight of blood are further cultural manifes-
tations of the predatory adaptation.
5.2.1. Cruel punishment.The strong routinely use pain as
punishment (from the Latin poena, penalty) in their
dealings with the weak – masters with slaves, adults with
children, and men with women. When Sarai complained
to Abram of Hagar’s contempt, he replied, “‘Your slave-
girl is in your power, do with her as you please.’ Then
Sarai dealt harshly with her, and she ran away from her”
(Genesis 16.6). Corporal punishment of children and
pupils was part of mediaeval and early modern life. From
the fifteenth century, the birching of school pupils
became increasingly common and brutal “for all offences
and all ages” (Arie
`s 1960/1962, p. 259). If a Yanomamo
woman is tardy in responding to her husband’s needs,
“the husband is within his rights to beat her.... It is not
uncommon for a man to injure his errant wife seriously”
(Chagnon 1983, p. 112). This domestic cruelty has
behavioural parallels among chimpanzees. When a young
male attains the size of an adult female, he is “systematically
brutal towards each female in turn”; a male can almost
always coerce an unwilling female into copulation
(Wrangham & Peterson 1996, pp. 143, 145; Stanford 1999).
5.2.2. Cruelty as amusement.The boundaries between
punishment and amusement are permeable. Caligula
(1241 AD) tortured Roman senators, men he knew
well, not to extract information, but for amusement
(Kiefer 1938). Commodus (177192 AD) was destructive
even in his humorous moments (Scriptores Historiae
Augustae, c. 500 AD/1960), and a chief delight of the
emperor Augustus was to watch boxing matches, “and
not merely professional bouts, in which he used to pit
Italians against Greeks, but slogging matches between
untrained roughs in city alleys” (Suetonius, Augustus,
45); the combatants wore gloves made of leather bands
loaded with balls of lead or iron that often blinded the
fighters (Tertullian 197 AD/1958, p. 97, note 7).
When the state’s official torturers believe that they
will not be held accountable, the oscillation between
instrumental and affective cruelty becomes apparent.
Thus with surreptitiously videotaped scenes of South
African police brutality – setting dogs on prisoners as a
“training exercise” or burning an unconscious hijack
suspect with a cigarette lighter: the perpetrators laugh
uproariously. In one of his notebooks, the artist Francis
Bacon wrote, “The reek of human blood, /it’s laughter to
my heart.”
5.2.3. The escalation of cruelty.A hallmark of cruelty is
its rapid escalation, from a slap to a punch to the smashing
of bones and teeth, from teasing to murder: the closing
scenes of Pasolini’s Salo illustrate the frenzy of the
torturer inflamed by the terror and pain of his victims.
Nell: Cruelty’s rewards: The gratifications of perpetrators and spectators
The underlying mechanisms appear to be, first, that the
affective tone of bullies and mob killers is energised and
exultant. Because RAGE and SEEKING are mutually
inhibitory in animals (Panksepp 1998), it is possible that
in humans, cruelty’s escalation arises from the
SEEKING rather than the affectively aversive RAGE-
aggression circuits. A hypothesis worth investigating is
whether the gratifications of perpetrators are dopamin-
ergic and fuelled by opioid release. Second, though
victims’ distress can inhibit violence (Blair 1997), their
fear and pain may also escalate the perpetrator’s savagery,
paralleling the predator’s escalating ferocity in the prey’s
death struggle as its terror and its vocalisations mount.
5.2.4. Disciplinary cruelty.Judicial punishment to
enforce laws and preserve discipline ranges from verbal
reprimand, shaming, and ostracism (see Note 2), to
death by execution or lethal mutilation. The agent of
these punishments (sometimes formally appointed to this
role: see Applbaum 1995) is emotionally cold. Herodotus
(440 BC, 5:25) tells how the Persian king Cambyses
ordered a corrupt judge to be flayed. Gerard David’s
Justice of Cambyses (1498) portrays the flaying in a
scene as devoid of emotion as a coroner’s autopsy: the
ritualised severity of the executioners and the assembled
court perfectly illustrate the emotional quality of instru-
mental cruelty. The contrast between this high sobriety
and the laughing crowds portrayed in popular woodcuts
of execution scenes (Puppi 1991, passim) is striking.
Mutilative punishments derive from the principle of
talion, retaliation, first codified by Hammurabi (c. 1760
BC) and transmitted through Deuteronomy 19:19 21
(c. 600 BC), and the Roman Law of the Twelve tables
(450 BC): its cruelty led Gibbon to remark that it is
“written in characters of blood” (1776/1903, vol. 4, p. 587).
5.2.5. Social control.The worst cruelties were inflicted on
slaves and the “inferior races” of the New World colonies.
Spartan youths killed helots for sport (Plutarch, c.100 AD/
1988, p. 28), and in Roman law, citizens freely used the
power of life and death they had over their slaves
(Kiefer 1938). Torture to inculcate terror was a favoured
instrument of political control in Europe’s African and
South American colonies: 15 million Africans are reported
to have perished in King Leopold’s Congo (Kimbrough
1972); rubber traders on the Putamayo River, a tributary
of the Amazon, were equally cruel (Mitchell 1997;
Taussig 1986). In the late twentieth century, the Greek
and Argentinian juntas adopted torture as an instrument
of state policy (Haritos-Fatouros 2003; Timerman 1981).
5.2.6. Confessional cruelty.Pain bends the victim’s
will to the torturer’s. Judicial torture in order to obtain
evidentially admissible confessions was recognised by the
Greek, Roman, and mediaeval European legal systems
(Held 1985; Robbins 1960). Criminals torture to uncover
loot: in early-fourteenth-century England, burglars placed
a housewife on a trivet over a fire until she revealed the
goods they sought (Hanawalt 1976).
5.2.7. Cruelty as entertainment. The Roman arena.Cruelty as an instrument of
social control in the form of elaborate, state-sponsored
entertainments (Coleman 1990; Wistrand 1992) reached
its apogee in the late Roman Republic and early Empire.
The elaborate and theatrically sophisticated arena ceremo-
nial (Barton 1993; Lafaye 1896) had a twofold social
purpose. It was educative, teaching the Romans “exactly
what their leaders thought essential to the survival of
Rome” (Wistrand 1992, p. 69): soldiers in training were
obliged to witness the combats in order to harden them
for war (Barton 1993). Second, the extravagant arena spec-
tacles were an extension of the emperor’s power and
benevolence (Coleman 1990). Suetonius (100 AD/1984,
Jul 39.3) records that Caesar, for his triumph in 46 BC,
held five days of animal hunts in the arena and the first
naumachiae (mock naval battles) in a specially excavated
basin near the Tiber: these were mock battles in the
sense that they were theatrical, but the deaths were real:
“thousands of superfluous foreigners were despatched in
a single extravagant display” (Coleman 1993, p. 74).
Gladiatorial shows in the amphitheatre were “the most
prominent and most popular spectacle of all,” writes
Tertullian (197 AD/1958, 12:1). One could not attend
the arena spectacles, he continues, “without his mind
being aroused and his soul being stirred by some unspoken
agitation. No-one ever approaches a pleasure such as this
without passion [and] violent agitation of the soul” (15:2
6). Even sober citizens demanded that “the man who has
been slain be dragged back to feast [their] eyes on him,
taking delight in scrutinising [his death] close at hand”
(21:15). The allusion is to a platform in the middle of
the arena to which wounded victims were dragged, “thus
enabling the spectators to observe more closely their
death struggle” (Tertullian 197 AD/1958, p. 94, note 3).
In his Confessions, St. Augustine tells of his young friend
Alypius, a Christian who had come to Rome to study
law. Augustine’s account captures the delirious contagion
that swept over the arena audience: “some man fell;
there was a great roar from the whole mass of
spectators .... [Alypius] saw the blood and he gulped
savagery .... he was drunk with the lust of blood” (vi, 8).
On occasion, this frenzy tipped spectators into active
killing, as with Pothinus of Lyons (Musurillo 1979,
Martyrs of Lyons, 5.35), and the Oriental monk Telema-
chus, who in 404 AD in Rome leaped from the stands
into the arena demanding that the bloodshed cease: he
was stoned to death (other versions say that he was torn
limb from limb) by the enraged spectators (Durant
1950). Bullfights arouse similar passions: if a matador
has been unsuccessful, writes Hemingway, the spectators
may decide to kill the bull themselves, “swarming on
him ... with knives, daggers, butcher knives and rocks
... cutting up at him until he sways and goes down”
(1939/1994, p. 21).
As the neurobiology of predation predicts, blood and
death have erotic force. Barton (1993) writes that the
raging sexuality of the arena came to a focus in the gladia-
tor’s scarred body, and Rome’s prostitutes gathered at the
arena exits, where they did a brisk trade. Mediaeval carnivals of death.Spectacles of pain
and death were a fixed part of mediaeval life, and there
is a rich popular art of execution scenes (Edgerton 1985;
Puppi 1991). The route followed by the executioner’s
cart was planned so as to draw the whole of the urban
fabric into these public demonstrations of the sovereign’s
Nell: Cruelty’s rewards: The gratifications of perpetrators and spectators
power (Foucault 1975/1986). Great crowds followed the
wagon and gathered at the place of execution, as with
the Catholic conspirator Guido Fawkes, who in 1606 was
drawn backwards through the streets of London at a
horse’s tail with his head near the ground, “being not
entitled to the common air” (Fraser 1996, p. 223). Animal baiting.The conjunction between pleasure
and the pain of animals is especially distressing to western
sensibilities but is ubiquitous across time and cultures.
The Romans scoured their African and Asian provinces
for exotic beasts that were transported to Rome in huge
numbers to be killed by a special class of gladiator called
bestiarii. Indian palace paintings portray elephant fights,
and in 1846 a traveller to Java reported that “one of
the favourite amusements ... is a fight between a tiger
and a native buffalo; the former has often to be urged on
by ... pouring boiling water over it, or pelting it with
lighted straw” (Friedla
¨nder 1871/1964, p. 189). In 1575,
Queen Elizabeth attended a baiting of 13 bears, and in
eighteenth-century England, bullbaiting and cockfighting
drew excited crowds (Malcolmson 1973). In the southern
United States, cockfighting continues (Herzog & Cheek
1979), and South Africans stage dogfights in empty
swimming pools.
5.2.8. War.War may be the most significant social product
of the predatory adaptation. The material that follows
suggests that the emotional state of the warrior in
combat mimics that of predators and hunters, with high
arousal, positive affect, and heightened libido, which in
turn raises the possibility that in the transition from preda-
tion to intraspecific, non-nutritional killing, the reinforcers
of the pain-blood-death complex have become attached to
combat and warfare. The warrior hero.In mythology, ethnography, and
contemporary culture, there are explicit links between
hunting, war, and manhood. Because of the male gender-
ing of hunting (Lee 1979; Lee & DeVore 1968; Hawkes
et al. 2001; Stanford 1999, pp. 4041), it becomes an affir-
mation of manhood: Croesus of Lydia dreamed that his
son Atys would die by the blow of an iron weapon and
accordingly forbade him to hunt a huge boar that troubled
the people of Mysia. “What face meanwhile must I wear as
I walk to the agora or return from it?” lamented Atys.
“What must ... my young bride think of me? What sort
of man will she suppose her husband to be? ... I pray
you, therefore, let me go with them” (Herodotus, 440
BC, 1:3439).
Reciprocally, the great warrior is a great predator, and
combat, like hunting, is a high-risk activity. Warlike bruta-
lity may be invoked through the metaphors of predation, as
in Yanomamo
¨war-party preparations (Chagnon 1983);
Achilles is “a soaring eagle / launching down from the
dark clouds to earth / to snatch some helpless lamb or
trembling hare” (Homer, 800 BC/1990, pp. 22:364 68).
In the hominid past, young “warrior hawks” were highly
prized because of violent interband rivalry that made it
essential for a group to have a contingent of “dawn war-
riors ... healthy, adventurous, and potentially violent
young men.... The most brutal ... have the advantage
over their less ‘sociopathic’ adversaries” (Bailey 1995,
p. 542). As in the arena, killing is erotic: a Vietnam
veteran says, “carrying a gun was like having a permanent
hard-on. It was a pure sexual trip every time you got to pull
the trigger” (Grossman 1995, p. 137). An American tank
commander talks about his first killing of German soldiers:
“The excitement was just fantastic ... the tremendous
feeling of lift, of excitement, of exhilaration, it was like
the first time you go deer hunting” (Grossman 1995,
p. 235). A deer hunt is the central metaphor in Michael
Cimino’s brutal 1978 movie about Pennsylvania steel-
workers serving in Vietnam.
It is possible that in combat and in cruel acts, the inten-
sity of wounding and killing activity is escalated by pain,
just as the dopaminergic biochemistry of predation, in
itself powerfully rewarding, may be augmented by endor-
phin release in response to exertion and pain (4.4 4.5). If
so, this dopaminergic escalation could be experimentally
demonstrated. The beauty of war.In Dispatches (Herr 1978),
Michael Herr describes the nights at Khe Sanh: “Even
the incoming was beautiful at night, beautiful and deeply
dreadful. I remembered the way a Phantom pilot had
talked about how beautiful the surface-to-air missiles
looked as they drifted up towards his plane to kill him.”
A reviewer of Herr’s book wrote that he had returned
from Vietnam “with the worst imaginable news: war
thrives because enough men still love it.” Why? The nove-
list John Coetzee suggests an answer: The gun is “the only
copula we knew of between ourselves and our objects....
The gun saves us from the fear that all life is within us.
It does so by laying at our feet all the evidence we need
of a dying and therefore a living world” (Coetzee 1974,
pp. 17, 79).
5.2.9. The weight of blood.If war is predation’s most
significant social product, its principal cultural product is
the emotional weight of blood in mythology, religion,
literature, and the graphic arts. A fixed feature of early
religions is the gods’ thirst for animal and human blood:
“for the life of the flesh is in the blood: ... for it is the
blood that maketh an atonement for the soul” (Leviticus
17:11). It is the wasting life of the sacrificial victim that
gives the words their power: The Neoplatonist Sallustius
writes, “Prayers divorced from sacrifice are only words,
prayers with sacrifices are animated words, the word
giving power to the life and the life to the word” (Sallustius
361 AD/1926). A Yanomamo
¨creation myth tells that war-
riors were created from the moon’s blood (Chagnon 1983,
p. 95).
Blood feeds frenzy, and frenzy demands blood. In Euri-
pides’ Bacchae (c. 406 BC/1970), the ecstatic women,
bare-handed, attack grazing cattle, “tearing full-grown
cows to pieces” and hurling body parts to and fro in a
scene of bloodlust that parallels the Wrangham and Peter-
son description in section 3 of chimpanzees dismembering
red colobus monkeys (Wrangham & Peterson 1996).
5.3. The walls of shame
Given that the human appetite for cruel spectacles is una-
bated and that arousal by scenes of cruelty remains part of
the human condition, it is remarkable that punishment
and killing, once openly displayed in amphitheatres and
Nell: Cruelty’s rewards: The gratifications of perpetrators and spectators
city streets, have for the past two centuries been banished
from public view and hidden behind prison walls.
What psychosocial mechanisms have operated to
achieve this great shift from the permitted to the taboo?
Part of the answer is given by Norbert Elias (1939/2000),
who writes that the history of western civilisation is of
“an advance in the frontiers of shame, in the threshold
of repugnance”
(p. 172). No shame attached to the
public torment of humans and animals in ancient or
mediaeval times (sect. 5.2.4); the warrior had “extraordi-
nary freedom in living out his feelings and passions, it
allows savage joys ... [and] hatred in destroying and
tormenting anything hostile or belonging to an enemy
[and] a particular pleasure ... in the mutilation of prison-
ers” (Elias 1939/2000, pp. 16263, 371).
But in the seven-
teenth and eighteenth centuries, knights became
courtiers, so that “a warrior nobility [was] replaced by a
tamed nobility with more muted affects” (Elias 1939/
2000, p. 389). Soon after, centralised state power created
pacified social spaces, the restraint of aggressive instincts
was internalised, and “an automatic, blindly functioning
apparatus of self-control [was] established ... [protected]
by a wall of deep-rooted fears” (Elias 1939/2000, p. 368).
Regrettably, these barriers are permeable and crumble
as opportunity and situation allow: the challenge for
violence prevention is to anchor them more deeply in
the life of the instincts.
6. The problem of violence prevention
Though treatment of the victims of cruelty (Basoglu 1992;
De Jong 2002) remains a moral imperative, effective
prevention must begin with perpetrators. How might the
foregoing analysis of cruelty’s reward systems relate to
the prevention of violence, defined by the World Health
Organisation as the intentional use of physical force or
power against oneself or others that threatens or causes
injury, death, or psychological harm (Krug et al. 2002,
p. 5)? Put differently, the question is how many of the
1,659,000 violence-related deaths in the year 2000 (Krug
et al. 2002, p. 270) were driven by delight in pain and
bloodshed, and might therefore have been prevented if
the public health upstream initiatives advocated by Krug
et al. (2002, p. 243) had, however imperfectly, acknow-
ledged and found ways to address the power of cruelty
to inflame violence?
6.1. The voice of the perpetrator
To begin developing answers to these questions would
in the first place require an understanding of the large
individual differences in cruelty’s eliciting triggers and
behavioural expressions on the one hand, and an under-
standing of the needs and gratifications of perpetrators
on the other: if so, the perpetrator’s voice must be
heard. Repugnant though this may be, violence-
prevention workers will need to gather affectively rich
descriptions of the inner experience of police and military
torturers and interrogators. These cannot be affectively
bland public confessions, with amnesty and social
rehabilitation in mind (Gardo 1987; Huggins 2000;
Victor 1981), but clinical data elicited by skilled inter-
viewers under conditions that guarantee confidentiality
(Fanon 1968; Haritos-Fatouros 2003). Some elements of
the required analysis are given below.
6.2. Universal potentials
6.2.1. The potential for cruelty.Current evidence is that
under situational press, readiness to commit cruel acts is
a human universal. In the 1970s, Milgram’s (1969/1974)
“Eichmann experiment” and the Stanford prison experi-
ment (Haney et al. 1973) demonstrated the “enormous
power of situations” (Haney & Zimbardo 1998, p. 709)
to shape and transform the behaviour of perfectly ordinary
people, whose actions are facilitated by a stance of moral
disengagement (Bandura 1990). Obedience makes moral
idiots of otherwise admirable individuals: the men of
Charlie Company who massacred 350 civilians at My Lai
in 1968 are described as “a typical cross section of
American youth assigned to most combat units throughout
the Army.... most would regard [William Calley] as
coming close to the American ideal” (Tester 1997, pp.
8485). It is resistance to situations that makes moral
6.2.2. The potential for compassion.Common wisdom
holds human nature to be fundamentally compassionate:
“Nature hath implanted in our breasts a love of others, a
sense of duty to them, a moral instinct ... which prompts
us irresistibly to feed and succour their distresses”
(Thomas Jefferson 1814, in Fiering 1976, p. 195; Nell
2004). The universal instinct for compassion derives from
genetically based kinshipbonds (Blair 1997; Panksepp 1998).
6.3. Gendering of cruelty
The gendering of hunting and the links between testoste-
rone and aggression suggest that active cruelty would be
strongly male-gendered. Mealey (1995) notes that boys
with high sensation-seeking and high testosterone are
more likely to initiate aggressive behaviour and be success-
ful in dominance interactions, which in turn triggers
further testosterone release.
6.4. Fascination and horror
These universal potentials cause an oscillation between
fascination and horror (see sect. 2.4.1). Fascination may
be dopaminergic, originating in the proximate and distal
rewards of the predatory/hunting adaptations. The
horror is compassionate, “a certain pain at an apparently
destructive or painful evil happening to one who does
not deserve it and which a person might expect himself
or one of his own to suffer” (Aristotle, Rhetoric, c. 330
BC, p. 1385b). This inward-turning, narcissistic quality is
captured by Darwin: “Almost every one would experience
[horror] in the highest degree in witnessing a man being
tortured or going to be tortured” (Darwin 1872/1965,
p. 304).
If there is indeed an oscillation between cruelty and
compassion, experienced by the subject as a switch from
fascinated gratification to horror, the reversal (Apter
1979) might be neurally detectable. One would further
predict that high- and low-readiness individuals would
differ in the location of this reversal on the cruelty conti-
nuum. This location could be determined by construction
Nell: Cruelty’s rewards: The gratifications of perpetrators and spectators
and validation of a Cruelty Readiness Questionnaire
(CRQ), generated through content analysis of experiential
material gathered from perpetrators telling of their
responses to the pain and terror of their victims. The
theoretical prediction is that high scorers would be indi-
viduals at the high-readiness end of the cruelty continuum,
with a low optimal level of arousal, and therefore have a
higher reversal threshold than low scorers. If high CRQ
scores do indeed correlate with high readiness and plea-
surable arousal at cruelty that continues beyond the
point at which low scorers experience a reversal, they
might have utility in the prediction of dangerousness.
6.5. Passive and active cruelty
The actualisation of this universal potential to use and
enjoy cruelty may vary along a continuum from low to
high readiness. At the low-readiness end are those who
passively enjoy media cruelty but refrain from cruel acts;
moving along the continuum are those who respond to
situational cues, inflicting pain if social inhibitions are
removed and role triggers are present, and following a
pathway into affective cruelty through a reversal from
cruelty inhibition to cruelty potentiation, in which the
victim’s cries and pleading activate the PBD complex, aug-
menting the perpetrator’s arousal and escalating cruelty.
At the high-readiness end of the continuum are active
sadists: the crazed monks in Juliette (Sade 1798/1968),
blood-crazed Fritz Haarman and Karl Denke in Weimar
Germany (Tatar 1995), the 1960s Yorkshire Moors mur-
derers Ian Brady and Myra Hindley, and the protagonist
of American Psycho (Ellis 1991): these are the monsters
of history and the most spectacular members of the
criminal class, for whose atrocities there is an endless
public appetite. Mealey (1995, p. 526) notes that crimi-
nality and sociopathy have a substantial and overlapping
heritable component, which suggests that a common
factor may underlie the various expressions of social
deviance, including active sadism.
Behaviourally, individuals with a high readiness for
cruelty are likely to have a predatory, victim-seeking
style, with homologies between predator self-stimulation
in animals, and the behavioural sequence of victim
stalking, capture, and wounding in humans: this novelty-
and harm-seeking sequence may be found to map to the
dopaminergic SEEKING circuit.
6.6. Neurologies of cruelty
“Psychologically, when different states feel different, they
are different” (Klinger 1971, p. 7). If so, there will not be a
single neurology of cruelty, but many, with different points
on the activepassive continuum that may be marked
by the activation of distinctive patterns of neural
drivers: mapping these is a cardinal prevention challenge.
Is the passion of the street-fighter neurally distinct from
the cold interest of the cigarette burner? Can the rapt,
immobile spectator be distinguished from Alypius’
friends, howling as the gladiators fall? Can perpetrators
be distinguished from spectators? Is recall for portrayals
of torture and painful punishment, and rumination on
such scenes, more intense in low- or high-readiness indi-
viduals, and does this recall covary with optimal level of
arousal (Eysenck & Gudjonsson 1989)? As noted above,
this differential affective neuroscience should begin by
gathering affectively rich descriptions of the inner
experience of perpetrators.
7. Is cruelty an adaptation?
Ferocity is a prerequisite for successful competition and
aggression, predation, hunting, and affective cruelty. The
behavioural commonalities between competitive aggres-
sion and early predation, the ferocity of predators, and
the high arousal of hunters, perpetrators, and spectators
suggest that common neural pathways dating to the
Cambrian subserve this cascade of behaviours that
begins with primordial competition and may end with
human cruelty. The symbolic weight of blood and death,
and their retained power to arouse powerful emotions,
may derive from an endlessly repeated scene in early
hominid history: at the kill, the hunting party is flooded
with the fresh blood of the prey, smeared with its bone
marrow, and exposed to its stomach contents. These
conditioned stimuli are preceded by a multitude of
sensations associated with the prey’s pain and death,
followed by proximate physiological rewards and deferred
reproductive advantages. These stimuli make up the pain-
blood-death complex, which continued to have survival
and reproductive benefits at successively more recent
stages of societal evolution.
With appropriate contextual judgment, the use of
cruelty leads to the accretion of social power and its main-
tenance; contextually inappropriate or excessive use will
result in social ostracism and punishment. The former
has survival and fitness advantages, especially if male
violence is under stabilising sexual selection, whereas the
latter will limit or prevent reproductive access.
However, despite its ancient provenance, it is unlikely
that cruelty is an adaptation that emerges through the
activation of a special-purpose evolutionary module
hard-wired in the cortex. A more parsimonious view that
would account for the striking homologies between preda-
tion, hunting, and human fascination with pain, blood, and
death is that all have a common origin in “the archetypal
emotional-motivational processes that all mammals
share” (Panksepp & Panksepp 2000, p. 112): biogenic
amines, present in the nervous systems of many animal
groups, from molluscs through to mammals, provide the
cortical foundation for these processes. Projections from
the neurons that produce these amines “stretch over
large areas of neural tissue and release chemical messages
diffusely, rather then through information-specific synap-
tic transmissions” (Panksepp & Panksepp 2000, p. 120).
These hypothesised continuities between predation and
cruelty would be confirmed if fMRI demonstrated cerebral
pathways, homologous to those that evoke predatory grat-
ification in canid, felid, and primate predators, in human
males exposed to scenes of pursuit, mutilation, and
killing of human victims, and their pain vocalisations:
there is no lack of graphic stimulus material, as, for
example, in films such as Last Exit to Brooklyn (Uli Edel,
Germany, 1989) and Salo (Pier Paolo Pasolini, Italy,
1975). The male gendering of cruelty could be confirmed
by comparison of malefemale responses to these stimuli.
It is therefore plausible that the wide range of beha-
viours linked to the pain-blood-death complex, from the
Nell: Cruelty’s rewards: The gratifications of perpetrators and spectators
passive enjoyment of media violence and blood sports to
the activities of interrogators and abusers, is reinforced
by these diffuse and very old emotional circuits that
humans share with animals, that “are able to imbue
‘cold’ perceptions with a ‘hot’ affective charge” (Panksepp
& Panksepp 2000, p. 115). This would in turn account for
the apparent universality of these emotions, which erupt as
powerfully in the educated and morally exemplary citizens
of the twenty-first century as in the monsters of history.
1. Though not further considered in this paper, psychological
punishments that inflict no physical pain are also cruel, as in
solitary confinement, public shaming, or social ostracism. The
pittura infamanti (defaming portraits) of mediaeval Florence
had “fearsome potency as an instrument of official state punish-
ment” (Edgerton 1985, p. 60; see also Miller 1993).
2. Self-inflicted pain is not the preserve of masochists, but a
pervasive social phenomenon in contests and sports, especially
contact, endurance, and “extreme” sports. Humour and the
mutual vulnerability of lovers also hold cruelty in tension. A life
without reflexive pain would be dull and colourless, but again,
as with psychological pain, and except in passing, I have excluded
this domain from the argument.
3. I have dealt with war and massacres from the perspective of
the individual actors, and not in their political context: the exhi-
laration of the machine gunner is relevant, but, in this target
article, the military command structures that control these
events are not.
4. A wall carving in the north palace at Nineveh shows King
Ashurbanipal and his commanders walking over headless
enemy bodies, with a beheading still in progress (Bersani &
Dutoit 1985, fig. 26). Roman commanders summarily executed
rebels: a stone relief (Andreae 1978, Fig. 536) shows the behea-
ding of rebellious barbarians under Marcus Aurelius in about 170
5. Arie
`s (1981) chronicles a similar process, within a similar
time frame, that has displaced natural deaths from the public
to the private domain.
6. This condition recapitulates the famous passage in Hobbes’
Leviathan: in war, “every man is enemy to every man .... in such
condition, there is no place for industry, because the fruit thereof
is uncertain...; no account of time, no arts, no letters, no society;
and which is worst of all, continual fear, and danger of violent
death; and the life of man, solitary, poor, nasty, brutish, and
short” (Hobbes 1651/1996, p. 84)
Open Peer Commentary
Cruelty may be a self-control device against
George Ainslie
Veterans Affairs Medical Center, Coatesville, PA 19320.
Abstract: Dispassionate cruelty and the euphoria of hunting or battle
should be distinguished from the emotional savoring of victims’
suffering. Such savoring, best called negative empathy, is what puzzles
motivational theory. Hyperbolic discounting theory suggests that
sympathy with people who have unwanted but seductive traits creates a
threat to self-control. Cruelty to those people may often be the least
effortful way of countering this threat.
Victor Nell presents plausible hypotheses about how human
cruelty may have evolutionary roots in carnivores’ emotional pre-
paredness to hunt. However, humans’ greater mental capacity
can be expected to add unique properties to cruelty, as it does
to most other motives. Nell himself suggests that there is a
kind of cruelty that “presupposes a theory of mind” (sect. 2),
henceforth ToM, a condition that would limit it to humans and
a small number of other species with advanced mental develop-
ment. He initially speaks of this condition as necessary for all
cruelty, but much of his subsequent discussion covers species
without ToM. It is not clear whether a cat plays with a mouse
partially in order to savor the distress of the victim, or merely
since it is an optimally challenging game. The common human
projection onto this activity certainly includes the savoring, as
in Tom & Jerry, but since a real Tom has no ToM, he is
presumably not imagining his victim’s suffering, much less
trying to induce it.
I doubt if many human hunters are rewarded by evidence that
their prey is suffering. In the television show Northern Exposure,
the protagonist was introduced to bird hunting, and said after-
wards, “I loved the shooting; it was the dying I couldn’t stand.”
Habitual hunters can obviously stand the dying more, but there
is little evidence that they glory in it. Primitive Amerindian
hunters were not necessarily any more sadistic. Sometimes they
would perform ceremonies before a hunt to apologize to the
spirits of the intended quarry. On the other hand, their enjoy-
ment of torturing captives was clearly on a par with that of the
ancient Roman mobs at the Coliseum (Adair 1736/2005). My
point is that the urge to do injurious things while disregarding
or actively avoiding attention to the suffering of victims is differ-
ent from the urge to seek out and even enhance this suffering –
although the disregarding might sometimes be a reaction against
the latter urge. Killing in war can be intensely pleasurable
(Bourke 1999, pp. 131; Grossman 1995, p. 115) and is more
apt than killing in hunting to intentionally inflict suffering, but
most infantrymen throughout history never even fired their
weapons at the enemy (Grossman, pp. 17 39). Even in the
euphoria of combat, the thrill is not usually that of cruelty but
of winning a mortal contest or of the power of wielding a
“magic sword ... all you do is move the finger so imperceptibly,
just a wish flashing across your mind ... and poof! In a blast of
sound and energy and light a truck or a house or even people
disappear” (William Broyles, quoted in Bourke 1999, p. 2). The
simultaneous perception that the “mutilated and dead [are] sad
and beastly” (Bourke 1999, p. 21) does not enhance the high
for most soldiers, and indeed soon spoils it.
The puzzle for motivational science is Nell’s “affective cruelty,”
as opposed to the kind that is incidental to hunting or war, or the
workmanlike “instrumental” kind practiced dispassionately for
extrinsic reasons, which probably includes that of the obedient
subjects in Milgram-type experiments (sect. 6.2.1). The point of
affective cruelty is to let yourself experience the suffering of
the victim vicariously, but with the kind of attitude that yields
net pleasure rather than pain, an attitude perhaps best called
negative empathy. Intended physical injury and intended
suffering are entirely dissociable. Medea killed her children not
to be cruel to them, but to be cruel to their father, Jason.
crucial question is how this attitude works, that is, how negative
empathy rewards. To discuss this, I need to include the psycho-
logical cruelty that Nell does not cover, which is the only kind
seen in everyday life.
I have argued elsewhere that empathy, the exercise of your
ToM, is itself rewarding (Ainslie 2001, pp. 161 86; 2005;
2006). My basic argument is that emotion is a goal-directed
(rather than conditioned) process that largely serves as its own
reward, but that entertaining emotions at will attenuates them
into daydreams, because the urge to anticipate the high points
undermines any longing or suspense that might make them
even moderately intense. You therefore learn to make adequately
rare and surprising external events the occasions for emotions.
Commentary/Nell: Cruelty’s rewards: The gratifications of perpetrators and spectators
Events interpreted through the models of other people built by
your ToM usually turn out to be the most satisfactory ones for
occasioning emotions.
Emotions cannot be divided strictly into positive and negative,
because all emotions must have a fast-paying reward component
in order to have their characteristic vividness. Some emotions are
usually aversive because initial attention to them leads to longer-
term inhibition of reward, but even fear and grief can be culti-
vated in ways that make them pleasurable, for instance, in
horror movies and tear-jerkers. Anger is often called negative,
but it shares many psychometric and neurophysiological proper-
ties with the more obviously positive emotions (Lerner et al., in
press). I agree with Nell that cruelty need not involve anger
(sect. 3.4), but I have argued that, like anger, it often becomes
preferred despite its spoiling effect on other rewards because it
repairs a felt vulnerability (Ainslie 2001, pp. 183 86). As with
anger, there are people who cultivate cruelty habitually, presum-
ably in default of richer sources of reward, but occasional cruelty
seems to be common to everyone. It is the commonplace
examples that best differentiate negative empathy from Nell’s
examples of predation: the pleasures of seeing the boor get his
comeuppance, the driver who cut us off stopped by the police,
and the pretensions of the poseur punctured, as well as less
respectable examples like schadenfreude and our minor persecu-
tion of people whom we hope we do not resemble.
What sometimes impels us toward cruelty? Because sympathy
is a mental response quickly rewarded by emotion, it is hard
to bring under voluntary control. But there are people with
traits that we fear in ourselves or who might exploit such traits,
sympathy with whom might let them weaken us or even
enchant us. In the absence of more direct controls, cruelty
toward these people might be the handiest way to reduce our
sense of potential seduction. That is, sympathy with the thief or
heretic, with someone who has a sexual taste we are afraid we
might develop, with a painfully naı
¨ve younger sibling who has
traits we have barely overcome, with the rejecting lover we
can’t get over or the needy lover who threatens to become depen-
dent, with any object of envy, even with someone whom we are
conscious of having wronged – sympathy with any of these
people might threaten to weaken us. A solution that hedonically
pays for itself in the short run is to attack positive empathy with
negative empathy, “set affection against affection and master one
by another: even as we used to hunt beast with beast” (Francis
Bacon, quoted by Hirschman 1977, p. 22). The capacity to do
this undoubtedly comes from a more elementary process,
perhaps the sheer arousal occasioned vicariously by anyone
else’s strong feeling as in the fascination of a fight or car
wreck, perhaps by the inherited preparedness for predation
that Nell suggests. However, because of its tendency to spoil
other sources of reward, it is apt to be cultivated only by
people with a need to suppress their sympathy.
The commentator is employed by a government agency and, as such,
this commentary is considered a work of the U.S. government and not
subject to copyright within the United States.
1. This was not just Euripides’ imagination. I professionally encoun-
tered the case of a man who, when his wife served him with divorce
papers, killed their children and himself, “to give her something to
think about.”
A murky portrait of human cruelty
Albert Bandura
Department of Psychology, Stanford University, Stanford, CA 94305-2130.
Abstract: In this commentary, I review diverse lines of research
conducted at both the macrosocial and microbehavioral level that
dispute the view that cruelty is inherently gratifying. Expressions
of pain and suffering typically inhibit rather than reinforce
cruel conduct in humans. With regard to functional value, cruelty has
diverse personal and social effects, not just the alluring benefits
attributed to it.
In the target article, Nell brings an unusually broad perspective
to bear on the possible origins of human cruelty. He reports
that, despite the cultural evolution over the many millennia,
human cruelty is still overwhelmingly present in the contempor-
ary world. The cited examples of contemporary cruelty highlight
the need for further specification of the defining criteria for what
belongs in this category. Boxing may be construed as an attenu-
ated form of cruelty, but why does motorcycle racing qualify as a
vestige of the pain-blood-death complex? If psychic pain is a
modern proxy of physical slaughter, does cruelty essentially
become a boundless category?
There is a difference between behavior motivated and
reinforced by conditioned pain-based gratifications and by its
functional value. For example, motorcycle racing can bring mon-
etary prizes, social status, and a sense of self-pride for a race well
run. But what do these rewarding benefits have to do with cruelty
and pain gratification? Empirical evidence indicates that cruel
behavior can be more readily modified by varying its functional
value than by relying on inherent affective gratifications of pain
cues (Bandura 1973).
The cited support in the target article for the upper stages of
the theory of human cruelty, which are amenable to empirical
test, is largely in terms of biblical quotations, anecdotes, descrip-
tions of ancient Greek practices, medieval carnivals, and arena
spectacles in the ancient Roman era. Except for passing com-
ments, surprisingly little attention is devoted to the third stage
of cruelty. This stage requires the most detailed theoretical spe-
cification because the link from gorging excitedly on prey in the
pain-blood-death complex to the exercise of social power is the
most enigmatic.
The support for the sexualization of cruelty at the hunter stage
is essentially metaphoric and anecdotal. As evidence for the
fusion of sex and aggression, Nell reports that !Kung hunters
use the penis as the metaphor for their hunting bow. He refers
to a hunter who claims that thoughts about the kill produce the
best sex, and a Vietnam veteran who found killing to be erotic.
No evidence is presented, however, on whether these expe-
riences are anomalous or normative ones.
In commenting on the “beauty of war,” Nell cites the example
of a military pilot mesmerized by the beauty of surface-to-air mis-
siles. One can find support for almost any view by careful selec-
tion of cases. The vast numbers of soldiers who experience the
hell of war and suffer posttraumatic stress disorders receive
no mention. The infliction of death and destruction remotely
by satellite and laser-guided missiles actually creates problems
for Nell’s theory. People behave more injuriously when they do
not see and hear the pain and suffering their acts cause. Faceless
hardware wars heighten destructive conduct by eliminating the
restraining effect of human suffering.
Findings of sexual arousal at depictions of rape, as measured
by a penis transducer, further dispute that cruelty is inherently
erotic. Rapists are sexually aroused by depictions of pain and
suffering of a rape victim, whereas non-rapists are aroused by
consensual sex but are turned off by sexual cruelty (Abel et al.
1977). Verification by selective examples of cruelty elevates
atypical reactivity to universal proclivity.
At the macrosocial level, Nell greatly exaggerates the preva-
lence of human cruelty. There exist wide intercultural differences
representing both warring and pacific societies with large intra-
cultural variations and even rapid transformation of warring
societies into peaceful ones (Alland 1972; Bandura 1973;
Gardner & Heider 1969; Levy 1969; Sanday 1981). The Swiss
used to be the main suppliers of mercenary fighters in Europe.
As they transformed into a pacific society, their militaristic
vestige is evident only in the plumage of the Vatican guards.
Commentary/Nell: Cruelty’s rewards: The gratifications of perpetrators and spectators
For ages, the Vikings plundered other nations. After a prolonged
war with Russia that exhausted Sweden’s resources, the populous
rose up and collectively forced a constitutional change that pro-
hibited kings from starting wars (Moerk 1995). This political
act promptly transformed a fighting society into a peaceable one.
According to Nell, cruelty is “strongly male-gendered.” This,
too, is an exaggeration. Most males do not go around mugging
people, and a good number of females are child and spouse
abusers. Meta-analyses reveal that the gender difference is
much smaller than commonly believed and further diminishes
with age, under conditions of provocation, and in the presence
of aggressive cues. (Bettencourt & Kernahan 1997; Bettencourt
& Miller 1996; Hyde 1984). The theoretical challenge is to
explain the substantial diversity within gender groups, which
far exceeds the difference between them.
The cultural evolution of social roles, norms, and sanctions has
long stripped barbaric cruelty of social and reproductive benefits.
This evolutionary social transformation in the coevolution process
requires theoretical specification. Attributing the shift toward
more humane conduct to the development of “walls of shame”
does not provide much theoretical guidance for deeper inquiry.
At the microanalytic level, the findings of experiments in which
the intensity of victims’ suffering and pain are systematically
varied show that expressions of pain typically inhibit rather
than reinforce aggressive conduct (Baron 1971a; 1971c; Geen
1970; Milgram 1969/1974).
Nell refers to the large audiences for violent entertainment as
further evidence that people seek gratification from watching
cruelty inflicted on others. In Nell’s view, the rewards of
cruelty explain the attraction to media violence. However, the
widespread belief that violence is a draw is disputed by empirical
evidence. Television programs that trade on violence rarely
appear in the upper ranks of popularity. As one television execu-
tive explained, “There is one maxim that is always true. The
network with the most comedy shows is the dominant
network.” Diener and DeFour (1978) tested the relation
between level of program violence and its popularity as measured
by the Nielson index of viewership. Program violence was unre-
lated to popularity, r¼0.05. Uncut versions of violent programs
are not liked any more than the same programs with most of the
gratuitous violence deleted.
Nell mentions the infliction of severe pain on oneself as a case
of major puzzlement to philosophers and psychologists alike. He
reasons that as both aggression and sexual activity activate the
brain’s reward system, the underlining motivation for self-
infliction of pain is the fusion of sex and aggression. Empirical
research has identified some of the conditions governing this
perplexing behavior. For example, self-inflicted pain serves a
self-protective function if it averts more painful treatment by
others (Bandura 1986; Stone & Hokanson 1969). Under these
conditions, self-punitive behavior is adopted and maintained
because it is the lesser of two evils. If preventing the external
painful threat requires ever increasing intensities of self-
punishment, it can escalate to the level of the avoided threat
and even be performed persistently, through lack of reality
testing, after the external threat no longer exists (Sandler &
Quagliano 1964). To observers who witness the seemingly sense-
less self-inflicted pain, without knowing its functional origin and
supporting expectations, the behavior appears deranged or
driven by some obscure masochistic pleasure.
A substantial body of research demonstrates that large-scale
inhumanities are heavily rooted in ideology (Bandura 1999;
Haritos-Fatouros 2003; Reich 1990; Zimbardo 2004). Extensive
training and a multitude of social structural influences are
needed to produce cruel perpetrators. A good part of this socia-
lization for cruelty is designed to disengage moral self-sanctions
from inhumane conduct (Bandura 1999). Through selective
moral disengagement, people who behave compassionately in
other areas of their lives can perpetrate ruthless inhumanities
on disfavored groups.
Cruelty as by-product of ritualisation of
intraspecific aggression in cultural evolution
Ralf-Peter Behrendt
MRC Psychiatry, The Retreat Hospital, York, YO10 5BN, United Kingdom.
Abstract: There are few commonalities between intraspecific aggression
and predation and few convincing arguments for the conceptualisation of
blood and pain as rewards for predation. Not cruelty, but ritualised
intraspecific aggression is the predominant mechanism of accretion of
social power and this, not cruelty, is what bestows reproductive
advantages. Enjoyment of media cruelty is not reinforced by
“emotional circuits” adapted to predation, but represents transient
relief from culturally determined inhibition of aggression.
Although in predation unspeakable suffering is inflicted, this does
not mean that the prey’s suffering is enjoyed by the predator.
McDougall (1924), for whom instincts were central to behaviour,
argued comprehensively against hedonistic theories of beha-
viour. The “prey’s terror and struggles to escape” (target
article, sect. 1.1.1), acting as an incentive stimulus, may certainly
lead to greater arousal and increase the predators’ determination
to kill, but “pain” and “blood” are unlikely to represent rewards
for the sake of which the animal kills, not least because they do
not arrest the striving of the animal. “[H]unters may inflict pain
on the prey beyond that which is instrumentally necessary”
(sect. 1.1.3, para. 2) because predation is an intrinsic instinctual
drive that – once set into motion and energised – has to run its
course (according to drive theory), as opposed to it being some-
thing done for the sake of enjoying a reward (hedonism) or
instrumentally to achieve a certain end (teleology).
Dopaminergic transmission from the ventral tegmental area to
the nucleus accumbens is activated not just by stimuli related to
predation but a range of motivationally meaningful and salient
stimuli that require behavioural reorientation and sustained
effort (Horvitz 2000; Parkinson et al. 2000). Dopaminergic acti-
vation invigorates incentive motivation and drives appetitive
approach but is unrelated to the experience of reward itself
(Berridge & Robinson 1998; Horvitz 2000; Ikemoto & Panksepp
1999). Its occurrence during predation does therefore not
indicate that “predation is a powerfully rewarding experience
even before satiation occurs” (sect. 3.4.5, para. 1).
The hunter is in a “ritually heightened state” during the night
after the kill and elaborates in detail on his experience in the pre-
sence of others – not necessarily because of his “high arousal at
the time of the event” (sect. 4.3) but in accordance with his enjoy-
ing the admiration of others (pride) and his potential for rise in
the group’s ranking order. If there is enjoyment in hunting that
is not socially mediated, then one should consider the possibility
that it represents the sense of mastery arising from the successful
solution of a challenging problem rather than “blood lust.” Thus,
the motivation for hunting may not at all include enjoyment of
cruelty or a desire to inflict suffering.
Little justification is given for the notion that predation
derives from intraspecific resource competition, or competitive
aggression. These behaviours are elicited by different sets of
stimuli, are accompanied by different affective states (“quiet-
biting predation” vs. “aggressive rage” [sect. 1.1.3]) and find
their resolution in the enactment of different states of affairs.
Intraspecific aggression can indeed be “marked by intense exci-
tement that appears indistinguishable from that during preda-
tion” (sect. 3.3) but – in the animals’ natural habitat – it does
not lead to bloodshed or death in most cases (Lorenz 1963/
2002). Its evolutionary purpose is to effect a distribution of
territory and access to resources among individuals of the
same species that is advantageous for survival of the species as
a whole (Lorenz 1963/2002). In higher vertebrates, intraspecific
aggression regulates rank order in groups and expresses itself in
ritualised ways in a wide range of social behaviours (Lorenz
Commentary/Nell: Cruelty’s rewards: The gratifications of perpetrators and spectators
Is it true that “apes cannot understand the communicative
intentions of others” (sect. 5.1), and that, therefore, their acts
of intraspecific aggression cannot be seen as punishment?
Primates develop socially in part by learning how to avoid pun-
ishment from other members of the group. Macaque infants
quickly develop an appreciation of the meaning of social
signals, such as direct eye contact; already by the middle of
the second postnatal week, infants show gaze aversion to
another individual’s direct stare (reviewed in Machado &
Bachevalier 2003). During the second year of their lives,
macaques acquire their dominance rank in the troop equipped
with an understanding that each individual has a “unique set
of intensions determined by the combination of kinship,
dominance, gender, environmental conditions, and the
current social context” (Machado & Bachevalier 2003). Such
observations challenge Nell’s assumption that the intention to
inflict pain presupposes a theory of mind.
“Disciplinary cruelty” and “public punishment” inflict pain
deliberately but are not motivated by delight in another
person’s suffering. The motivating affect is anger caused by an
individual’s breach of social conventions or challenge to the
rank order. If spectators or perpetrators do delight in public
punishment then this may be attributable to projection of and
temporary relief from their own punishment anxieties. What
spectators experience in “spectacles of pain and bloodshed” is a
marked, though transient, relief in psychic tension that derives
from constant unconscious death fears in a society that readily
resorts to corporal punishment and wars. Soldiers in war have
to endure continuous fear of death so that success in combat is
bound to produce a surge of relief and feelings of superiority
and invincibility. The mechanism is one of projection, as hinted
at in Nell’s quote (in sect. of the target article) from
Coetzee (1974): “The gun saves us from the fear that all life is
within us. It does so by laying at our feet all the evidence we
need of a dying and therefore a living world.”
It can be argued that it is the suppression of aggression in the
process of cultural evolution – not enjoyment of cruelty per se –
that became “a primary driver of the modern entertainment
industry” (sect. 1.1.3). People are likely to enjoy media cruelty
for the same reason that they show an incessant interest in scan-
dals involving the downfall of people in society, where there is no
role to play for “blood, pain, and death.” Impulses of intraspecific
aggression that are culturally suppressed can find transient relief
also in humour (laughter as a sudden relief of inhibited aggres-
sion, according to Lorenz [1963/2002]), but once the cultural
inhibitory framework is removed (including through “moral dis-
engagement”), intraspecific aggression becomes disinhibited and
can manifest in actual acts of cruelty. Social barriers restrain
aggressive impulses, as Nell acknowledges, and these barriers
“crumble as opportunity and situation allow” (sect. 5.3). Then,
indeed, aggression may be accompanied by “exultant” affective
tone – mostly, though, because of its effect of instilling a sense
of dominance and power in the perpetrator.
Hunting success may not confer “direct,” as Nell argues, but
indirect “fitness benefits” in terms of sexual desirability and
access to females for reproduction – mediated by enhancement
of one’s ranking position within the group; and the same
applies to cruelty. Both hunting success and public acts of
cruelty may signify greater potential for social control over
others or greater likelihood of success in hostile encounters
with the outside world. Indeed, as Nell reviews, females
respond positively to “aggressive success,” that is, the acquisition
of dominance, not “aggression” per se. It is primarily the
“dominance-seeking” aspect of what Nell calls “dominance-
seeking aggression” that “is driven by reproductive-fitness
needs” (sect. 3.4.5, para. 4). Clinically, this is evident in the
association of mania (representing excessive dominance-seeking
and control of others) with hypersexuality, and depression
(marked by low self-esteem and social withdrawal) with loss of
interest in sex.
Sadism does seem to involve the deliberate infliction of pain for
the sake of enjoyment. Unlike competitive aggression, there is no
anger involved, and unlike predation, the perpetrator is not just
aroused by but enjoys the other’s suffering and, perhaps more
importantly, the other’s denigration, allowing him to project
into the victim his sense of inferiority, experience a sense of super-
iority, and, in conjunction with this, become sexually aroused. The
question why the “infliction of pain on the self” can be “pleasur-
able and also sexually arousing” is challenging but not “incompre-
hensible.” Freud (1917) elucidated the mechanism of enjoyable
self-tormenting in melancholia. Masochism plays a role also in
the psychodynamics of narcissism, envy (Joseph 1986; Spillius
1993), and child abuse (Milton 1994).
Make love, not war: Both serve to defuse
stress-induced arousal through the
dopaminergic “pleasure” network
Mary F. Dallman
Department of Physiology, University of California at San Francisco, CA 94143-
Abstract: Nell restricts cruelty to hominids, although good evidence
suggests that secondary aggression in rodents and particularly primates
may be considered cruel. A considerable literature shows that
glucocorticoid secretion stimulated by stress facilitates learning,
memory, arousal, and aggressive behavior. Either secondary aggression
(to a conspecific) or increased affiliative behavior reduces stressor-
induced activity, suggesting the reward system can be satisfied by other
behaviors than cruelty.
Nell writes on important issues that have plagued human
societies from time immemorial. However, he has boxed
himself in with his third point, specifying that cruelty is only a
human endeavor, because of its intention to inflict pain. Thus,
he ignores a considerable literature that documents “cruelty” in
subhuman primates and other mammals. I generally and strongly
agree with the other points he has developed. However, I must
take issue with respect to the notion that cruelty is specifically
a characteristic of hominids, because much applicable behavioral
and neuroscience research is lost through this definition.
Some dictionary definitions of cruel, or cruelty include the
phrase “inhuman” or “inhuman treatment” (cf. The American
Heritage Dictionary of the English Language [American Heritage
Dictionary 1992]; Webster’s Seventh New Collegiate Dictionary
[Webster 1963]); thus, implicit in the definition is the notion
that humans should be a step more morally advanced than sub-
human primates or other animals. I suspect that Nell is probably
right in his assertion that the origin of cruelty is some kind of a
behavioral by-product of predation. However, because of his
exclusion of the use of cruelty to behavior solely in humans, he
ignores the fact that the situations which he documents so well
of cruelty in mankind have strong parallels in other animals.
Unprovoked aggression to group members in animals is, in my
opinion, analogous to cruelty in man. Predatory aggression in
animals (and probably in man) comprises physiological and
psychological stimuli that result in adrenocortical activation.
Rapid actions of cortisol (or glucocorticoids) secreted by the
adrenals increase learning and memory, arousal, and salience
of ongoing activity, and also facilitate ongoing aggression (Bass
& McKibben 2003; Makara & Haller 2001; Pardon et al. 2002;
Roozendaal et al. 2001). Aggression is reinforcing, and many
species will perform instrumental responses that are reinforced
by the opportunity for aggression (cited in Fish et al. 2005). In
mice and rats, aggression is glucocorticoid dependent (Fish
et al. 2005; Haller et al. 2004). Thus, the glucocorticoid response
to stress sensitizes both aggressive behavior and the memory of
aggression, as well as taking arousal to a higher pitch. Such a
Commentary/Nell: Cruelty’s rewards: The gratifications of perpetrators and spectators
state is both dangerous for appropriate future behaviors and
uncomfortable, and it needs to be defused.
In socially vocal midshipmen fish, cortisol increases the
frequency and duration of vocalization in a hierarchical
network in the brain pons and medulla when territory is threa-
tened by conspecifics (Remage-Healey & Bass 2004). In socially
threatened rats, corticosterone also increases aggressive behavior
toward conspecifics (Haller et al. 2000). Under conditions of
acute or chronic stress, there is increased glucocorticoid-
dependent secretion of norepinephrine throughout the cortical
and limbic brain in response to a novel stress (Finlay et al.
1995; Nissenbaum et al. 1991; Valentino et al. 1983). Thus, the
actions of glucocorticoids on the stress response network in the
brain hone vertebrates to a greater pitch of arousal.
There is a need, then, after predation, or during and after
stress, to cool off, reduce arousal, and return to a sustainable
state that is ready to deal with the future. I suggest that this
may occur in both animals and humans by employing aggression,
accompanied by either real or potential cruelty, to available
conspecifics. Reduction of stress-induced glucocorticoids (and
presumably a high state of arousal) is effected in both rats and
baboons through aggressive behavior toward others in the group.
Given the opportunity to aggress with another rat when
electrically shocked, male rats fight. Provided with that opportu-
nity, ACTH and corticosterone concentrations are lower than if
the rats are not given that outlet (Conner et al. 2000; Weinberg
et al. 1980). Other “displacement behaviors,” such as schedule-
induced polydipsia, and biting wooden sticks can also achieve
reduced adrenal activity during stress in rats (Brett & Levine
1979; Hori et al. 2004). More to the point raised by Nell,
baboons that have been defeated by a dominant male have lower
cortisol concentrations when they aggress either other more subor-
dinate males or females, compared to similarly defeated males that
do not aggress (Virgin & Sapolsky 1997). These results suggest
strongly that secondary aggression reduces stress and arousal.
However, when the composition of the troop under study shifted
to one that was more female-dominated, affiliative rather than
aggressive interactions came to dominate behaviors, and again, glu-
cocorticoid concentrations were lower (Sapolsky & Share 2004).
Clearly, in at least two species, intense and uncontrollable
stress stimulates glucocorticoid secretion, which, in turn, aug-
ments and cements aggressive behavior. However, equally
clearly, provided that other outlets to the stress-induced
arousal exist, aggressive behavior is either diminished, or need
not occur. Frans de Waal has beautifully distinguished behavioral
differences between our close relatives, chimpanzees (de Waal
2000), and bonobos (de Waal & Lanting 1997). Although the
former are quite highly aggressive, the latter are generally non-
aggressive but highly sexually affiliative. Tapping into the same
“pleasure” dopaminergic pathways that are invoked by Nell to
explain the pleasures of cruelty, both increased conspecific
aggression and heightened affiliative behavior appears to
reduce arousal and glucocorticoid reactivity. These findings
suggest strongly that a positive response to the issue of human
cruelty could be to push the motto of “make love, not war” into
formal programs that foster affiliative behavior. It seems that
the deliberate promotion of increased affiliative behaviors
could achieve the same tension reduction as secondary
aggression, and might, if widely available, result in reduction of
the amount of pleasure seeking directed toward cruel behaviors.
Neurobiological bases of aggression,
violence, and cruelty
´a Ine
´s de Aguirre
Department of Neurology, Instituto de Investigaciones Me
´dicas Alfredo Lanari,
Combatientes de Malvinas 3150, Buenos Aires, 1427, Argentina.
Abstract: Aggression, violence, and cruelty are symptoms of psychiatric
illness. They reflect abnormalities in the regulation of the stress and
emotion circuitries. The functioning of these circuitries depends upon
the interaction between genetics and environment. Abuse and neglect
during infancy, as well as maternal stress and poor quality of maternal
care, are some of the causes that produce these types of abnormal
behavior. Research on the neurobiological bases of emotion regulation
will allow the detection of the population at risk.
Alterations in the hypothalamic-pituitary-adrenal (HPA) axis and
the sympathetic nervous system (SNS) constitute the stress
system and its connection with a complex circuit that regulates
emotions. These alterations produce psychiatric illness, where
aggression, violence, and cruelty are the most remarkable symp-
toms. The stress system and its interconnected particular brain
structures exist in humans and nonhumans to ensure adaptation
and survival (Chrousos 1995; Kelley 2004). While these systems
generally serve a highly functional and adaptive role in behavior,
they can be affected in maladaptive ways, producing psycho-
pathology (Brady & Sinha 2005; Duman 2002; Goeders 2003;
Sinha et al. 2003; Weiss 2005; Wu¨st et al. 2004).
Difficulties in managing stressful life events associated with
negative emotions and failure in coping to regain control,
without attaining the desired goal, influence the adaptative
processes and produce psychopathology (Sinha et al. 2004).
Evidence from studies performed on animals and humans have
substantiated the belief that maternal stress or anxiety in
pregnancy is associated with general, rather than specific,
susceptibility to psychopathology in offspring, as a result of an
overactivity and impaired negative feedback regulation of the
HPA axis. Reduced activity of the opioid GABA/benzodiazepine,
serotonin, and dopamine and increased activity of the sympathico-
adrenal systems have also been found.
The serotonergic system has been associated with mood dis-
orders, anxiety, aggression, and impulsivity. The noradrenergic
system is involved in attentional processes, memory, and stress
responses. The dopaminergic system is involved in cognition,
affects, and control of locomotion. The amygdala mediates fear,
anxiety, and mood regulation. In addition, other brain structures
complete the emotion regulatory circuitry, such as the orbital
frontal and the prefrontal cortex, amygdala, hippocampus, hypo-
thalamus, anterior cingulated cortex, insular cortex, as well as the
ventral striatum and periaqueductal gray (Huizink et al. 2004).
The orbital frontal cortex is also involved in the modulation of
antisocial behavior.
We have to contrast reactive from instrumental aggression.
Reactive aggression (impulsive aggression) is triggered by a
frustrating or threatening event, often culminating in physical
violence. It is associated with a low threshold for activating nega-
tive affects (a mixture of emotions and mood that include anger,
distress, and agitation). It is initiated regardless of any potential
goal. Patients with borderline personality disorders characterized
by impulsive aggressive behavior, affective instability, inappropri-
ate intense anger, and unstable interpersonal relationship
present this type of aggression. Conversely, instrumental aggres-
sion (proactive aggression) is purposeful and goal directed. It is
premeditated and is used instrumentally to achieve a specific
desired goal, which is not always the pain of the victim
(cruelty), but rather the victim’s possession, or the increased
status within a group hierarchy. Psychopathic individuals
present a breakdown in moral socialization and impairment in
the affective system, thus showing especially this type of aggres-
sion. Dysfunction in the orbital frontal cortex has been described
in reactive aggression as the amygdala in instrumental aggression.
The activity of this circuitry will depend on the interaction estab-
lished between genetics and environment (Blair 2004; Davidson
et al. 2000b; Moya-Albiol 2004).
It is known that the quality of maternal care received during
infancy determines the adult social competence and ability to
cope with stress. The development of a neurochemical system
within the brain that regulates mothering, aggression,
Commentary/Nell: Cruelty’s rewards: The gratifications of perpetrators and spectators
and other types of social behavior, such as the oxytocin and
vasopressin systems, are affected by parental nurturing received
during infancy. It is hypothesized that the neural bonding system
may be important for the development of loyalty in individuals
towards a social group and its culture. Neglect and abuse
during early life may cause the bonding system to develop abnor-
mally and compromise the capacity for rewarding interpersonal
relationships and commitment to societal and cultural values
later in life. Other means of stimulating reward pathways, such
as drugs, sex, aggression, and intimidating others, as well as the
involvement in sects or gang activities, could become relatively
more attractive as a way of life (Pedersen 2004). We have to
recognize that aggression, violence, and cruelty reflect abnorma-
lities in the emotion regulatory circuitry of the brain.
I would like to extend my appreciation to Dr. S. Finkielman, Director of
Instituto de Investigaciones Me
´dicas Alfredo Lanari, for his continuous
Compassion as an antidote to cruelty
Michael Allen Fox
School of Social Science, University of New England, Armidale, NSW 2351,
Abstract: The impulse toward violence and cruelty is endemic to the
human species. But so, likewise, is the impulse toward compassionate
behavior. Victor Nell acknowledges this, but he does not explore the
matter any further. I supplement his account by discussing how
compassion, specifically in the moral education of children, can help
remedy the problem of violence and cruelty in society.
Aggressive impulses are innate in the human species. And
perhaps all humans are prone to exhibiting and enjoying
cruelty under certain conditions (even if these conditions are
never met in their particular life circumstances). The commonly
told story of human evolution centers upon positing such
impulses as the drivers of our destiny (humans as aggressive,
self-seeking, greedy, “man the hunter,” and so on). Other
accounts, however, recognize the obvious (but often neglected)
truths that: (1) humans are also cooperative, caring, nurturing
beings; and (2) if it were not for these latter traits, our
species’ history – although undeniably bloody – would arguably
have been far less characterized by peaceful periods and every-
day cooperation, far less filled with constructive, creative accom-
plishment, and much shorter (Kropotkin 1908; Mead, G. H. 1934,
Pt. 4; Mead, M. 1937; Skyrms 1996, Ch. 3). However gloomy
today’s human scenario may appear, then, it is important to
keep a sense of balance and hope.
I have no doubt whatsoever that Victor Nell would agree. His
research project is in no way intended to contest the foregoing
perspective, but rather, to take on the challenge of understanding
and, to the extent possible, help counteract and neutralize the
violent and cruel tendencies that are so evident and widespread,
cross-culturally, in our contemporary world. These are laudable
objectives, for two reasons. First, Nell’s approach forces us to
look at violent and cruel acts as having positive reinforcement
for the perpetrators. Only if we unravel how this process works
can we improve our chances of control and rehabilitation in
this arena. Second, his project has the underlying significance
that ordinary people should not have to live in fear for their
safety, and victims of violence should not have to bear the onus
of readjusting to a hostile social environment taken as the
Yet I wonder whether, in addition to Nell’s approach,
we might still need to give careful consideration to the compas-
sionate side of our nature
as an antidote to “the rewards of
Nell presents strong evidence for regarding the perpetration
and enjoyment of cruelty as having deep evolutionary, neuro-
logical, and biochemical underpinnings, and we probably must
accept this. However, within the context of his thoughts on miti-
gating the problem of violence and cruelty in human life, very
little is said about the role of compassion in defining who and
what we are, and in describing the human potential. In
section 6.2, Nell acknowledges that the human motivational
repertoire comprises both cruelty and compassion. But the
impulse toward compassion is not addressed further. Granted,
Nell has another research agenda. I respect that and wish to
make it clear here that my purpose is not to try to pick holes
in his argument, but rather to supplement it, in the positive
spirit of collaborative scholarly exploration.
Compassion is defined as “deep awareness of the suffering of
another coupled with the wish to relieve it” (American Heritage
Dictionary 2000). This formulation illustrates that compassion
is a deeper feeling than either sympathy or empathy (with
which it is often compared), inasmuch as compassion entails
not just resonant fellow-feeling, but also the desire to ameliorate
another’s negative situation. While the above definition is an
etymologically faithful rendering (com/passion as “suffering
with”), I believe compassion has come to have an even richer
meaning than is suggested here, namely, one that embraces
what the philosopher Arthur Schopenhauer calls “loving-
kindness” (Schopenhauer 1841/1965, sect. 17), or everyday
caring concern for the other and his or her interests, and the
desire to see the other flourish. I have given careful thought to
the idea that compassion might be, as Schopenhauer claims,
the sole motivational foundation of ethics (Schopenhauer
1841/1965, sect. 19:4), the primary moral emotion. Two
principles of right conduct, above all others, might generally be
considered as absolutely basic: “Do no harm” (nonmaleficence)
and “Do good whenever possible” (beneficence or benevolence).
Both can be construed as following naturally from the compassio-
nate side of our nature (Schopenhauer 1841/1965, sect. 16). This
is not the place to debate whether compassion is the whole of
morality. But it may be of interest to reflect upon just how
central it is to a better world.
What is special about compassion is that it appears to be an
innate tendency. Children show caring concern toward their
peers, those younger than themselves, and animals. They are
naturally inclined toward acceptance of others and toward
being nonjudgmental (what later may blossom as tolerance, cele-
bration of diversity, etc.). Lack of compassion and intolerance
are, in the broadest sense of the term, learned responses that
stem from certain kinds of conditioning, life experiences, and cir-
cumstantial events, such as cultural images of violence, examples
set by role models, neglect, emotional impoverishment, and
abuse and bullying. These influences lead to desensitization
and may allow other natural impulses – toward violence and
aggression – to predominate in thought and action. But if lack
of compassion, intolerance, and cruelty, for example, can be
learned, they can also be unlearned – or better still, prevented
by different learning.
To teach and instill a cultural ethos of compassion and respect
for others, as well as animals and the environment, is the goal of
the humane education movement (Selby 1995; World Animal
Net, n.d.). The simple precept here is that if we want desirable
personality qualities and dispositions to flourish, we need to
nurture and reinforce these. Imaginary exercises, thought experi-
ments, and role-playing can help create and strengthen the
capacity to put oneself in another’s shoes, and to prevent the
development of a closed mind characterized by compartmentali-
zation, distancing, and the objectification, marginalizing, and
inferiorizing of the other.
The aim of making children more compassionate by
enhancing their natural tendencies in this direction can posi-
tively contribute to creating a society in which people are less
prone to violence and cruelty. It would, however, be naı
¨ve to
Commentary/Nell: Cruelty’s rewards: The gratifications of perpetrators and spectators
assume that such an endeavor could succeed without other
large-scale measures being put in place to remove or at least
restrict the conditions that motivate and encourage violent and
cruel behavior. In this respect, Nell is completely correct to
stress that we must do what we can to make such behavior
less rewarding – a gargantuan task, but one on which the
human future depends. Yet I would maintain that peaceful
social reconstruction hinges equally on the careful, systemic
cultivation of compassion.
1. I owe this victims-of-violence point to Fiona Utley.
2. As I trust will become evident, compassion is to be distinguished
from, and is much more profound than, the altruism that some theorists
explain in terms of evolutionary reproductive strategies, simple recipro-
city, prisoner’s dilemma gamesmanship, and so forth.
Cruelty: A dispositional or a situational
behavior in man?
Mika Haritos-Fatouros
Department of Psychology, Aristotle University of Thessaloniki, Salonika,
Greece 54640.
Abstract: Presentation of evidence from multiple disciplines is the most
impressive feature of Nell’s article. I have observations and objections,
however, about the following issues: (1) violence as a by-product of
cruelty; (2) the equation of animal and human cruelty; (3) social
psychological evidence contrary to the biological model; (4) whether
prevention of cruelty best arises from predispositional or situational factors.
By presenting admirable, extensive evidence from paleontology,
predator ethology, primatology, anthropology, and cognitive and
experimental psychology related to motivation and learning, as
well as social psychology and cultural evolution, Nell traces the
evolutionary origins of cruelty and violence to present-day
human beings. Hypothesizing continuity between the behavior
of predation in animals and contemporary cruelty in humans,
he links a wide range of behaviors into a “pain-blood-death
complex,” a very important and useful thesis. However, several
observations and objections should be stated.
Nell notes that violence is the by-product of cruelty and main-
tains that if effective prevention is to be applied, such origins
must be revealed. But cruelty may also be the by-product of vio-
lence; in war, a general climate of violence may lead to cruelty
and torture by military personnel on their victims without any
previous preparation for it (see, e.g., the Abu Ghraib torturers;
Haritos-Fatouros & Zimbardo 2005). Archetypal emotional-
motivational processes common to all mammals may well influ-
ence human behavior, as Jung has proposed many years ago.
But human behavior is also greatly influenced by cognitive pro-
cesses, and by the resulting situations produced. The Freudian
biological model which proposes a destructive, biologically deter-
mined, death-seeking force, a human “instinct” that produces
aggressive behavior and violence has long been with us and has
been repeatedly challenged and largely refuted by experiments
as well as field studies.
In particular, torturers do not have to have a certain kind of
personality, only exposure to certain kinds of psychological,
social, and political conditions, (Haritos-Fatouros 2003). Similarly,
gender differences, with greater male violence, and sex-related
aggression, and abuse, cannot be attributed mainly to high testos-
terone and low serotonin in males. Albert Bandura (1973; 1990)
and followers of social learning models have shown evidence that
aggression is a behavior pattern largely learned through positive
or negative reinforcement. Disengagement mechanisms are also
used in situations of cruelty, and their importance is indeed
acknowledged by Nell in the target article.
Finally, prevention of human cruelty and violence clearly
requires more than detecting high and low scorers on any type
of questionnaire – Nell proposes a Cruelty Readiness Question-
naire (target article, sect. 6.4) to predict high readiness and
pleasurable arousal in situations of potential cruelty. Neither
would MRIs’ demonstrating individual differences in cerebral
pathway involvement to differentiating stimuli predict cruelty,
or go far to prevent cruelty from occurring. I certainly agree
with the author that cruelty will not be contained through obscur-
antism and that effective prevention requires that its reinforcers
are revealed. However, it is also important not to lose sight of
classic works emphasizing cultural and situational factors, for
example, Foucault’s Discipline and Punish (Foucault 1975/
1979/1986), Milgram’s work on obedience to authority
(Milgram 1969/1974), and Zimbardo’s Stanford Prison Exper-
iment (Zimbardo 1970).
On the other hand, the target article offers an abundance of
possible hypotheses for research. Why certain kinds of behavior
confer direct fitness benefits is of interest; Nell informs us that
among the Ache, better hunters are more often chosen by the
Ache women and have much higher fertility. The basic question
remains, however: How far are aggression, violence, and cruelty
in humans today the result of predisposition factors, or biological
or archetypal processes, and how far are they the result of
cognitive/emotional processes evoked by situational factors? To
paraphrase Voltaire: I do not agree with you, but I shall do every-
thing within my power to help you express your point of view.
Humananimal connections: Recent findings
on the anthrozoology of cruelty
Harold Herzog
and Arnold Arluke
Department of Psychology, Western Carolina University, Cullowhee, NC
Department of Sociology and Anthropology, Northeastern University,
Boston, MA 02115.
Abstract: Recent findings in anthrozoology – the study of human –
animal interactions – shed light on psychological and social aspects of
cruelty. Here we briefly discuss four areas that connect animal
cruelty and cruelty directed toward humans: (1) voices of perpetrators
and their audiences, (2) gender differences in cruelty, (3) cruelty as
play, and (4) the putative relationship between animal abuse and
interpersonal violence.
To support his contention that the roots of cruelty lie in
predation, Nell invokes findings from psychology, ethology,
neurobiology, history, and paleoanthropology. Curiously, given
the central importance of inter-specific interactions to his
theory, Nell neglects anthrozoology – the study of human
animal relationships. Of special relevance are current findings
on animal abuse. Here we briefly raise several findings from
this literature that are relevant to understanding cruelty
1. Voices of perpetrators and their audiences. Nell cor-
rectly calls for greater understanding of the perspectives of
those involved with cruelty, although his idea for doing so
seems narrowly psychological. Anthrozoological studies of
animal cruelty have examined the mistreatment of animals as it
is defined in the course of social interaction in groups. People
arrive at shared agreements about what things mean in given situ-
ations, and cruelty is no exception, whether this includes conven-
tional groups, such as adolescent males, or unconventional
groups of purported abusers, such as “kill-shelter” workers who
are considered to be cruel by their “no-kill” peers (Arluke
2006). Second, when studying their voices, the gratifications of
perpetrators and their audiences must not be limited to psycho-
logical ones such as “escalating arousal.” For example, members of
Commentary/Nell: Cruelty’s rewards: The gratifications of perpetrators and spectators
humane societies constitute an audience for cruelty and can
experience increased solidarity after egregious instances of
harm come to their attention (Arluke 2006), and humane law
enforcement officers, who are another collective audience, can
find ways to elevate their professional status as a result of their
contact with cruelty (Arluke 2004). Finally, by focusing on
intent as the basis for defining cruelty, serious forms of animal
harm such as hoarding are minimized because the perpetrator
lacks clear intent to harm, and it is also important to capture
the voice of those who commit such forms of passive cruelty
(e.g., Vaca-Guzman & Arluke 2006).
2. Gender and animal cruelty. According to Nell, sex differ-
ences in testosterone levels and in hunting mean that cruelty
should be a predominantly male enterprise. When it comes to
animal abuse, this is indeed the case. Gerbasi (2004) reviewed
gender differences in media reports of animal abuse convictions.
The male/female ratios were as follows: beating, 38 to 1; shoot-
ing, 16 to 1; mutilation/torture, 20 to 1; and burning, 17 to
1. Arluke and Luke (1997) reported that virtually all cruelty
cases prosecuted in Massachusetts courts over a 10-year period
involved males. While these represent extreme forms of
cruelty, males are also more likely to be involved in lesser
forms of cruelty. For example, both parental reports and child
self-reports indicate that boys are more likely to abuse animals
than girls. Of course, this preponderance of males is likely to
change in the future if the demographics of animal abusers
follow the trend toward increasing violent crimes in general by
3. Animal cruelty as play. Consistent with Nell’s hypothesis,
cruelty can be a recreational extension of hunting. An apparent
example of this relationship is found in Jared Diamond’s depic-
tion of animal abuse among traditional hunters in the highlands
of New Guinea. Diamond (1993) observed captive animals
systematically tortured by hunters, much to the amusement of
onlookers. Arluke (2002) found that childhood animal cruelty
frequently takes the form of “dirty play,” akin to other forms
of play such as the use of sexual or racial epithets that are objec-