present purposes, it is convenient to consider the behavior of individuals
as caused in part by internal propensities, while
acknowledging both that such propensities are not to be seen as
unitary entities and that they may be influenced by external factors.
Certain behavioral propensities, including the capacity for aggression,
are common to virtually all humans. This does not mean that they
are genetically determined. "Human nature" is a consequence of common
genetic factors and of ubiquitous factors in the environment. The
combination of genetic endowment and environmental factors almost
inevitably encountered in development lead to the presence of a
capacity for aggression in virtually every individual. But that
does not mean that aggressive behavior stems from an innate "drive"
that must be discharged in some way; there is neither psychological
1963) nor cross-cultural evidence for such a view. Humans have
the capacity to be both aggressive and altruistic, cooperative and
cantankerous; the behavior shown depends on a host of developmental,
experiential, social, and circumstantial factors.
Within that framework, aggressive acts are seldom due solely
to aggressive motivation; other motivations are usually present.
For instance, the behavior may involve an attempt to acquire an
object or situation, which for present purposes we may call acquisitiveness.
There may also be a tendency to show off — assertiveness. Furthermore,
aggression usually involves risk of injury for the attacker, so
that it is combined with self-protective or withdrawal responses.
Thus, whether or not aggression actually occurs will depend not
only on the individual's aggressiveness, but also on motivations
of other types.
Individual aggression is often categorised into a number of types.
For instance, one system distinguishes "instrumental aggression,"
deliberate and concerned primarily with obtaining an object or position
or access to a desirable activity; "emotional aggression," hot-headed
and angry; "felonious aggression," occurring in the course of a
crime: and "dyssocial aggression," regarded as appropriate by the
reference group or gang, but not so regarded by outsiders (e.g.,
& Ochberg, 1981). Such categories, though useful for some
purposes, usually turn out to be less clear-cut than they might
appear for an obvious reason: a variety of motivations may contribute
to a single act, and they may be present in various strengths and
combinations. The very fact that such categorization systems can
be only partially satisfactory is in itself an indication of the
motivational complexity of even apparently simple aggressive acts.
It is convenient to divide the factors contributing to an aggressive
act into three categories, with dialectical relations between the
levels of social complexity operating in each case. These three
factors. The tendency of an individual to behave aggressively
depends in part on genetic factors and in part on experience. Physical
aggressiveness tends to be greater in boys than in girls, to increase
with age up to adolescence or early adulthood, and then to decline.
In our own culture, attention has focused on the roles of classical
conditioning, operant conditioning, and observational learning,
and on relationships within the family. These affect both motivational
propensities and the acquisition of cognitive capacities, the latter
including abilities for conflict resolution. Relationships with
individuals outside the family may also be important, including
especially those who serve as role models and the peer group with
its norms. The behavior of the socialising agent, whether intra-
or extra-familial, will be influenced by the norms and values of
the group and the society to which the agent belongs, and these
norms and values may differ with the nature of the targeted individual.
Thus parents may apply different norms for boys and girls, or for
firstborns and later looms. Furthermore, the norms and values operating
will be influenced by, and will influence, the mass media and other
channels of social influence. Thus, the aggressive propensities
of individuals can be understood only through the dialectical relations
between individuals, their relationships and group membership, and
the socio-cultural structure or structures operating.
factors. Across societies, violence is more frequent in those
that tolerate or extol violent acts by individual or state, do not
distribute income or wealth equitably, and lack social and political
institutions linking their members in networks of communal obligation
in press). However, the issues here are complex. While political
violence may provide a context for increase in criminal violence
Kemp & Moema, 1993; Straker et
al., 1996), homicide rates tend to decrease in countries actually
at war, probably because of the increased integration (Lester, 1992).
After the war, however, homicide rates tend to increase. In addition,
the propensity of an individual at any particular time may be influenced
also by a variety of contextual factors, including the current social
situation and its attendant norms, and the presence and density
of other individuals.
factors. Whether an aggressive act is actually elicited depends
on further factors, including the individual's current motivational
state; frustration of current goals; pain, fear, and other aversive
factors; and arousal, the nature of the opponent or victim, and
the availability of weapons. It depends also on a variety of inhibitory
factors, such as fear of punishment and the possibility of alternative
courses of action (Goldstein,
The preceding paragraphs do no more than hint at the complexity
of the factors involved in individual aggression, but they may serve
to indicate that full understanding even of interactions between
individuals requires analysis of individual characteristics and
their bases, a variety of situational factors, and coming to terms
with the dialectical relations between the levels of complexity
and the socio-cultural structure.
between groups requires cooperation between the individuals within
But beyond that, it involves principles additional to those pertaining
to individual aggression, principles that arise from the very nature
of groups and from the relations between the group and its individual
members. The literature on the nature of psychological groups and
inter-group relations is now vast (see e.g. Brewer & Brown,
in press), but some issues important in the present context must
be reviewed briefly.
Individuals see themselves both as autonomous individuals and as
members of groups. In addition to seeing himself as John Smith,
born in such-and-such a place, cleverer than most, and not so good
with his hands, an individual may see himself as a member of a variety
of groups — the middle classes, Jewish, a citizen of this or that
country. Thus Tajfel and Turner
(1986) distinguished between an individual's personal identity
(involving comparisons with other individuals) and social identity
(derived from membership in emotionally significant social groups
or categories). The greater the salience of the latter, the less
that of the former; perception of the self as an interchangeable
unit in a social group involves diminution in perception of the
self as special or unique. It has been suggested that this partial
depersonalisation is basic to many group phenomena (Turner et
al., 1987; Turner et
al., 1994); such a view is certainly in harmony with the methods
used to instill discipline and group loyalty into military recruits.
Members of a psychological group not only label themselves (and
usually are defined by others) as a group, but see themselves as
more similar in group-relevant and distinctive respects to each
other than to outsiders. And they see themselves as in some degree
interdependent, and often as having a common task or goal. There
is some disagreement as to which, if any, of these is primary or
fundamental (see e.g. Rabbie, 1989;
Turner, 1986; Turner, 1981).
Membership in a psychological group has certain consequences on
individual behavior. Members of the in-group tend to be treated
as heterogeneous differentiated individuals, members of the out-group
as undifferentiated units. Individuals who see themselves as members
of a group tend to elaborate, and to subscribe to, group norms and
values and to conform to them (Tajfel &
A frequently important aspect of group membership stems from the
fact that individuals need to find support for their beliefs (Festinger,
1954), and this may be obtained from those who share those beliefs.
Finding that others share one's beliefs may increase one's liking
for and feeling of solidarity with them, especially if the convictions,
such as religious beliefs, are otherwise unverifiable (Byrne, Nelson
& Reeves, 1966); and reciprocally common group membership
authenticates the potential of other group members to provide consensual
validation (Gorenflo &
The self-esteem of group members is influenced by group membership.
Individuals seek a positive social identity, but membership of a
group will contribute to that only if it can be evaluated favorably
relative to other groups. People therefore tend to identify with
groups that they evaluate favorably, and to evaluate favorably groups
with which they identify, even in the absence of objective evidence
for their qualities. The more individuals identify with a group,
the more they are likely to strive to enhance their own self-image
by contributing to the group. Group cohesiveness and in-group cooperation
are thus facilitated, and are likely to be greater in groups that
are seen as successful. In-group membership provides a sense of
security for individuals, and outsiders may be stereotyped and denigrated.
Individuals receive "reflected glory" from the achievements of their
fellow group members, even though not contributing themselves (Cialdini et
al., 1976; Tesser, 1988).
Again, negative acts by the out-group are more likely to be ascribed
to characteristics of that group than similar acts by in-group members,
while achievements and positive acts by the in-group are more likely
to be ascribed to shared in-group characteristics than to external
1990). Reciprocally, negative evaluations of the out-group may
enhance the self-esteem of in-group members and their tendency to
identify with their own (highly esteemed) group (Tajfel &
Turner, 1986; see discussion by Brewer &
Brown, in press).
Because individuals want both to see themselves as individuals and
as related to another or others (Baxter,1990),
groups that are exclusive tend to be especially cohesive. A distinctive
social identity can satisfy both the need to feel part of a group
and the need to feel special and different from others (Brewer, 1991).
There are, it must be noted, exceptions to these generalisations.
For instance, in-group preference may be diluted or reversed in
lower status groups. Minority groups may perceive themselves to
be homogeneous, and yet evaluate the out-group more favorably than
the in-group (Sachdev &
Bourhis, 1991). But in general, individuals tend to show loyalty
and preference to the in-group, to exaggerate differences from out-groups,
and to evaluate the in-group favorably. Recent evidence suggests
that such characteristics of inter-group behavior are most pronounced
when (a) intra-group cooperation, collective achievement, and interdependence
with fellow group members are emphasized; and (b) the existence
or importance of the group depends on the existence of other groups
(Brown et al.,
It will be apparent that many aspects of inter-group relations
are of special importance in times of conflict or war. This
evidence from social psychology is not undermined by the tenuous
evidence from anthropology. While conflict situations by definition
involve two or more groups with conflicting goals, it is in the
interests of leaders to enhance the distinctive identity, and thus
the integrity and cohesiveness, of their group. This is especially
the case within a fighting unit, where the need for in-group cooperation
makes it essential for social identity to be augmented and the perceptions
that individuals have of themselves as unique and autonomous downplayed.
The image of the in-group must be enhanced, the out-group denigrated,
and the difference between them emphasized. Effective action is
more likely if it is seen as likely to be successful, and this depends
on the morale of those involved, which in turn influences and is
influenced by group processes, as discussed above. As long ago as
1800 the regulations for the British Rifle Corps laid down that
every man should have a comrade, and that a family feeling and loyalty
to comrades should be fostered (Richardson,
1978). More recently, unit leaders have been advised to exercise
"professional paternalism" (Rodine, 1977).
Camaraderie, kameradschaft, comradeship are seen as essential elements
in morale — though sometimes less effective on modern battlefields
where men are widely dispersed.
Other differences from individual aggression arise from the fact
that the very presence of other group members affects the behavior
of each individual, with the internal dynamics of the group escalating
or inhibiting tendencies towards violent action. In an unstructured
group, individuals may be more ready to behave aggressively because
of the arousal produced by the group situation or by the relative
anonymity and sharing of responsibility bestowed by group membership.
If the group values violence, individuals may assert themselves
and show off by behaving aggressively, hoping thereby to rise in
the esteem of their associates; but if the group is predisposed
towards restraint, hot-heads may be inhibited. Because of the dialectical
relations between the propensities of individuals and group norms,
the aggressive potential of a group is not simply the sum of that
of its members.
While many aspects of group aggression can be understood by regarding
individuals as units in a collective, that is by no means the whole
story. For one thing, long-standing cultural precedents may set
the scene for violence (Liddell, Kemp
& Moema, 1993). For another, there is likely to be some
differentiation of roles even in quite small groups. The role of
leaders may be critical. Leaders may achieve their position because
they epitomize group values. or they may inculcate their own values
into the group. They may represent the group in negotiations. In
some circumstances, the psychology of the leaders may be crucial.
Beyond the differentiated leaders, the other individuals are not
identical units and are likely to have very different personalities.
In a study of violence in South African townships before the end
of Apartheid, Straker (1992)
identified the following types in the groups of "Youths".
Well-balanced, idealistic, dedicated, skilled, popular. Expressed
Searching for a script as warrior heroes, attempting to achieve
an ego-ideal. Labile. Could become leaders. Potentially reckless.
Lacked sense of self, and used group to define it. Could lead,
but would not intervene or change course of group.
Motivated by social conformity rather than by ideals. Sought
group acceptance, camaraderie, but passions not engaged.
Casualties. Anti-social but might justify criminal acts as political.
Perceived negatively by the group.
diverse personalities are held together in group action in part
because group action satisfies for each his or her particular needs.
But cohesion may be augmented by outside forces, for instance by
the very fact that they are treated as a group, or that they perceive
themselves to be unjustly portrayed as an inchoate mob by the media.
It will be apparent that many of the factors that influence individual
aggression also influence group aggression, but additional social
factors are also involved. While for some purposes groups can be
personified and treated as individual entities, the dynamics of
relationships within the groups may be as important as the relations
between the groups. We shall see that individual aggression is of
even less importance in institutionalised modern war.
L. (1990). "Dialectical Contradictions in Relationship Development."
Journal of Social & Personal Relationships, 7, 69-88.
L. (1963). Aggression. New York: McGraw Hill.
Jones, N. (1972). Ethological Studies of Child behavior. Cambridge:
Cambridge University Press.
M.B. (1991). "The Social Self: on Being the Same and Different at the
Same Time." Personality and Social Psychology Bulletin, 17, 475-82.
M.B. & Brown, R.J. (in press). Intergroup relations. In Handbook
of Social Psychology.
R., Hinkle, S., Ely, P.G., Fox-Cardamone, L., Maras, P. & Taylor.
L.A. (1992). "Recognizing Group Diversity: Individualist-collectivist
and Autonomous-relational Social Orientations and their Implications
for Group Processes." British Journal of Social Psychology, 31,
D., Nelson, D., & Reeves, K. (1966). "Effects of Consensual Validation
and Invalidation on Attraction as a Function of Verifiability." Journal
of Experimental Social Psychology, 2, 98- 107.
R.B., Borden. R.J.. Thorne, A.. Walker. M.R., Freeman. S. & Sloan.
S.R. (1976). "Basking in Reflected Glory: Three (Football) Field Studies."
Journal of Personality & Social Psychology, 39, 406-15.
L. (1954). A Theory of Cognitive Dissonance. Evanston, IL: Row
R. (1996). "Cross-cultural Aspects of Violence." Paper given at Conference
on Violence and War, Valencia.
Gartner, R. (in press). Crime: Variations Across Cultures and Nations.
In C. & M. Ember (eds.), "Cross-cultural Research for Social Science."
Englewood Cliffs, N.J.: Prentice Hall.
J. (1986). Aggression and Crimes of Violence. New York: Oxford
D.W. & Crano, W.D. (1989). "Judgmental Subjectivity/Objectivity
and Locus of Choice in Social Comparison." Journal of Personality
and Social Psychology, 57, 605- 14.
M. (1990). "The 'Ultimate Attribution Error'? A Review of the Literature
on Intergroup Causal Attribution." European Journal of Social Psychology,
D. (1992). "War and Personal Violence." In G. Ausenda (ed.) Effects
of War on Society, pp. 211-22. San Marino: AIEP Editore. Liddell,
C., Kemp, J. & Moema, M. (1991). "The Young Lions: South African
Children and Youth in Political Struggle." In L.A. Leavitt & N.A.Fox
(eds), The Psychological Effects of War and Violence on Children,
pp.199-214. Hillsdale, NJ; Erlbaum.
J.M. (1989). "Group Processes as Stimulants of Aggression." In J. Groebel
R.A.Hinde (eds.), Aggression and War. pp. 141-55. Cambridge:
Cambridge University Press.
F.M. (1978). Fighting Spirit. London: Lee Cooper.
Rodine, L.B. (1977). The Taming of the Troops. London:
I & Bourhis, R.Y. (1991). "Power and Status Differentials in Minority
and Majority Group Relations." European Journal of Social Psychology.
M. (1966). Group Conflict and Cooperation: their Social Psychology.
London: Routledge & Kegan Paul.
G. (1992). Faces in the Revolution. Cape Town: David Philip.
Straker, G., Mendelsohn. M,. Moosa, F. & Tudin, P. (1996). "Violent
Political Contexts and the Emotional Concerns of Township Youth." Child
Development, 67, 46-54.
H. & Turner. J. (1986). "The Social Identity Theory of Intergroup
behavior." In S.Worchel & W.G.Austin. (eds.). Psychology of Intergroup
Relations. pp. 7-24. Chicago: Nelson.
A. (1988). "Toward a Self-evaluation Maintenance Model of Social Behavior."
Advances in Experimental Social Psychology. 21, 181-228.
J.R. & Ochberg, (1981). "Patterns of Adolescent Violence." In D.A.Hamburg
& M.B.Trudeau (eds.),Biobehavioral Aspects of Aggression,
pp. 129-40. New York: Liss.
J.C.. Hogg. M.A., Oakes. P.J. Reicher. S.D. & Wethrell. M.S. (1987).
Rediscovering the Social Group: a Self-categorisation Theory.
Turner, J.C.. Oakes. P.J., Haslam, S.A. & McGarty, C. (1994). "Self
and Collective: Cognition and Social Context." Personality and Social
Psychology Bulletin, 20, 454- 63.