Self-Regulatory Mechanisms and War
1
RUNNING HEAD: Self-Regulatory Mechanisms and War
A Grotesque and Dark Beauty:
How the Self-Importance of Moral Identity and Mechanisms of Moral Disengagement Influence
Cognitive, Emotional, and Behavioral Reactions to War
Karl Aquino
Americus Reed II
University of Delaware
University of Pennsylvania
Stefan Thau
University of Groningen
Author Note:
Address all correspondences to Karl Aquino, Department of Business Administration, Alfred
Lerner College of Business, University of Delaware, Newark, DE 19716. Phone: (302) 8316144. e-mail: [email protected] Co-author contact information: Americus Reed II, Arthur
Anderson Term Assistant Professor of Marketing, The Wharton School, University of
Pennsylvania, 3730 Walnut Street, 700 Jon M. Huntsman Hall, Philadelphia, Pennsylvania
19104-6371. Phone: (215) 898-0651. E-mail: [email protected]. Stefan Thau,
Graduate School ICS, University of Groningen 9712 TG Groningen, The Netherlands. Phone:
+31-50-3636237. E-mail: [email protected]. The authors would like to thank Lisa Bolton,
Dawn Iacobucci, Paul Eder, Carl Turner, and participants of the Zicklin Center for Business
Ethics Research Seminar Series at the Wharton School for either administrative and data analysis
assistance with studies or comments on an earlier version of the article.
Self-Regulatory Mechanisms and War
2
Abstract
Three studies examine the extent to which moral disengagement and moral identity
jointly drive cognitive, emotional and behavioral reactions to war. Moral identity decreased the
negative relationship between moral disengagement and feelings of guilt and shame toward
invading Iraq (Study 1). Two additional studies replicated this effect and showed that it was
mediated by moral judgments about out-groups and in-groups. Moral identity decreased the
positive relationship between moral disengagement and the perceived morality of vengeful
responses toward out-group members responsible for the September 11th attacks, which lead to
stronger feelings of pride, strength, determination and inspiration (Study 2). Moral identity
decreased the positive relationship between moral disengagement and the perceived morality of
supportive responses toward in-group members responsible for the Abu Ghraib prison abuse,
which led to donating relatively more money to a U.S. military charity (Study 3).
Key Words: Moral Identity, Moral disengagement, War, Identity, Self-Regulatory Mechanisms,
Inter-group Conflict, Aggression, In-group favoritism, Out-group denigration
Self-Regulatory Mechanisms and War
3
A Grotesque and Dark Beauty:
How the Self-Importance of Moral Identity and Mechanisms of Moral Disengagement Influence
Cognitive, Emotional, and Behavioral Reactions to War
Let me have a war, say I:
It exceeds peace as far as day
Does night; it’s sprightly, waking, audible, full of vent.
William Shakespeare
Coriolanus, Act IV, Scene V
People are often ambivalent about war. To some, war epitomizes all that is disturbing
about human nature. War magnifies our capacity for cruelty, renders us indifferent to the
suffering of others, and incites an unreflective and virulent nationalism that stigmatizes dissent,
subverts truth, and condones moral hypocrisy. From this perspective, war is a reckless and
despicable act that should evoke feelings of sadness and shame over humankind’s failure to solve
conflicts without violence and force. But for others, a just war –one waged to defend a country,
preserve freedom, or prevent the slaughter of innocents– is an admirable and necessary
expression of a nation’s power and moral goodness. Those who extol the glories of war also
speak of how it ennobles its participants by drawing from them exceptional acts of courage, selfsacrifice, determination, resourcefulness, and loyalty—virtues that often lie dormant during
times of peace.
Yet even its most ardent defenders usually recognize war for what it is: an organized
form of mass killing. The tragedy of modern war is that the majority of those killed are not
soldiers but civilians. It has been estimated that in the wars of the 1990s, 75 to 90 percent of all
war casualties were civilian deaths (Hedges, 2003). It can be disturbing for supporters of war to
contemplate the loss of innocent lives because most people strive to maintain the belief that they
and the groups they identify with are morally good (Taylor & Brown, 1988; Tajfel & Turner,
Self-Regulatory Mechanisms and War
4
1979). For obvious reasons, lending support to an activity that causes death and destruction on a
massive scale can undermine these beliefs. And yet despite whatever doubts or emotional
conflicts they may be experiencing, soldiers, politicians, and citizens sometimes manage to carry
on with the business of war, often with great enthusiasm and even patriotic fervor.
An ongoing topic in social psychology has been the question of how ordinary people
reconcile their support for war and other destructive activities with a self-view of being decent,
caring, and kind. Over the years, psychologists have offered many different answers to this
question. They include highly deterministic theories of aggression based on instinctive (e.g.,
Lorenz, 1967) and socio-biological models (e.g., Ghiglieri, 1999; Tooby & Cosmides, 1990);
situational explanations such as difficult life conditions, submission to authority and role
expectations (Staub, 1989; Milgram 1974; Zimbardo, 1972) as well as more cognitively-oriented
theories emphasizing self-regulatory mechanisms (Bandura, 1990; Bandura, 1999; Bandura,
Barbaranelli, Caprara, & Pastorelli, 1996) and beliefs about the world (Eidelson & Eidelson,
2003).
In this paper, we adopt and build upon the socio-cognitive model (Bandura, 1990;
Bandura, et al., 1996) to examine peoples’ reactions to the U.S. led wars in Afghanistan and Iraq.
We take a socio-cognitive approach not because we believe it is the only theoretical perspective
that can shed light on these reactions, but because the socio-cognitive model explicitly
recognizes the psychic need for people to resolve two seemingly inconsistent cognitions when
they decide whether or not to support war: the desire to maintain a favorable view of the self and
the need to justify actions that violate socialized self-sanctions against harming others.
Self-Regulatory Mechanisms and War
5
Moral Disengagement and Disconnection of the Self from Moral Consequences
From a socio-cognitive perspective, one explanation for why people participate in
activities that cause harm to others is that they execute various mechanisms of moral
disengagement that allow them to justify harmful conduct while maintaining a positive selfimage (Bandura 1991, 1999 Bandura, et al., 1996). The moral disengagement construct captures
a variety of dissonance reducing rationalizations that people use to shield the self from the
consequences of inhumane conduct and the self-condemnations it may impose. For example,
dehumanizing the enemy allows soldiers to disable their empathetic reactions thereby making it
easier for them to overcome natural and socialized aversions to killing (Bandura, et. al, 1996;
Bandura, 1992; Grossman, 1995). Civilians who support war may execute similar disengagement
maneuvers, although to a lesser extent since they are seldom directly involved in the act of
killing.
The mechanisms of moral disengagement fall into several broad categories (Bandura, et
al., 2001). No one study can examine all of them, so we chose to focus on the cognitive
“reconstrual” of the conduct itself through moral justification and advantageous comparison.
Moral justification occurs when harmful behavior is made personally and socially acceptable by
depicting it as serving a valued or righteous social purpose (e.g., defending freedom, ridding the
world of “evil-doers”) (Kelman & Hamilton, 1989). When making an advantageous comparison,
a “cognitive contrast effect” allows a person to convince him or herself or others that harmful
conduct is relatively minor when compared to something more injurious within a similar context
(e.g., inadvertently killing a few innocent civilians during a war is not as bad as a dictator
executing thousands of his people). The use of either or both of these mechanisms not only
Self-Regulatory Mechanisms and War
6
weakens self-deterrents against harming others, but can also provide a sense of self-approval for
such behavior.
But moral disengagement maneuvers do not always enable people to violate selfsanctions against harming others. Anecdotal evidence for this comes from cases where someone
has an opportunity and perhaps even good reason to act inhumanely but does not do so. For
example, during the infamous My Lai massacre in Vietnam two American helicopter crewmen
named Lawrence Colburn and Hugh Thompson refused to participate with their fellow American
soldiers in the killing of unarmed civilians. Instead, Colburn and Thompson attempted to airlift
victims of the slaughter to safety. Such acts beg the question of what social psychological factors
might neutralize the “effectiveness” of moral disengagement maneuvers that allow people to
overcome their aversion to harming others. We sought to answer this question by examining
another theoretical postulate of socio-cognitive theory; namely, that moral cognitions are
translated into action through various self-regulatory mechanisms rooted in standards and selfsanctions associated with the moral self (Bandura, 1999). Drawing from recent theoretical
conceptions in moral psychology, (e.g., Aquino & Reed, 2002; Blasi, 1984, 1993) we refer to
this construct as a person’s moral identity.
Moral Identity Neutralizes the Effectiveness of Moral Disengagement
Socio-cognitive theory suggests that people are more likely to act inhumanely when
moral disengagement allows them avoid confronting the inconsistency between their personal
moral standards and their willingness to engage in or support behavior that may violate these
standards. War is an obvious example of a situation where disengagement practices are not only
likely to be executed by those who endorse war, but even systematically encouraged by military
leaders to ensure that soldiers are able to kill the enemy without hesitation or remorse
Self-Regulatory Mechanisms and War
7
(Grossman, 1995). In this paper, we explore the possibility that moral identity can neutralize the
effects of moral disengagement on reactions to war. Our main hypothesis is that while the
willingness to execute moral disengagement will be positively (negatively) related to pro- (anti)
war thoughts, emotions, and behaviors, these relationships will be weaker among persons whose
moral identity has sufficiently high self-importance. Below we present the conceptualization of
moral identity adopted in this paper and our rationale for predicting its influence on the
effectiveness of moral disengagement.
The Concept of a Moral Identity. Following other writers (e.g., Aquino & Reed, 2002;
Blasi, 1984; Damon, 1984; Damon & Hart, 1992; Lapsley & Lasky, 2001) we conceptualize
moral identity as one of many possible identities that people use as a basis for self-definition.
According to Lapsley and Lasky (2001), a person who has a moral identity is one for whom
moral schemas are chronically available, readily primed, and easily activated for processing
social information. Building on this argument, Aquino and Reed (2002) proposed that people
possess a cognitive schema of the moral self that is organized around a set of moral trait
associations. Both Aquino and Reed’s (2002) and Lapsley and Lasky’s (2001) conceptions of the
moral self are socio-cognitive models because they suggest that the activation of mental
representations of the self is critical for processing social information and providing guidelines
for action (Cervone & Shoda, 1999a). 1
However, it is not enough to argue that some people possess a “moral identity” that can
become activated in certain situations. We should also ask when a particular person’s moral
identity will impact thoughts, feelings and actions. Contemporary cognitive theories of identity
view the self as comprised of many different, hierarchically ordered identities (Stryker, 1980).
However, only a subset of them, known as the working self-concept (Markus & Kunda, 1986) is
Self-Regulatory Mechanisms and War
8
accessible in memory at any given time. An implication of this model of the self is that identities
with the highest self-importance are likely to have the strongest and most consistent influence on
thoughts, emotions, and behavior.2 Consistent with socio-cognitive theory this argument assumes
that self-regulatory mechanisms like moral identity do not come into play unless they are made
psychologically salient. Based on this logic, we propose that the mechanisms of moral
disengagement do not operate in a vacuum, but in the presence of different identities that people
use as a basis for self-definition and that also serve a self-regulatory function. Although some of
these identities may be irrelevant for thinking about war (e.g., athlete identity, occupational
identity), others are likely to be highly relevant (e.g., moral identity, national identity) and should
therefore interact with moral disengagement processes to influence war-related cognitions,
emotions, and behavior. War has serious consequences and confronts people with complex moral
dilemmas. Thus, it seems logical to propose that a person’s moral identity is at least one of the
identities that can directly relate to war-relevant judgments, if it assumes high importance within
the working self-concept.
As mentioned earlier, we hypothesize that a highly self-important moral identity can
neutralize the effectiveness of justifying inhumane conduct through moral disengagement. Our
rationale is based on theoretical arguments and empirical evidence suggesting that the selfimportance of this particular identity is associated with stronger feelings of social responsiveness
and moral obligation toward others.
Moral Identity and the Self-Other Relation Some people are experienced as being closer
to the self in terms of mutual affection and obligation and others are farther away (Sahlins,
1965). This feeling of solidarity has been described using various terms such as Allport’s (1954)
“circle of inclusion”, Opotow’s (1995) “scope of justice”, and Reed and Aquino’s (2003) “circle
Self-Regulatory Mechanisms and War
9
of moral regard”. What these conceptions share is the idea that people establish psychological
group boundaries that define the people with whom they feel connected and obligated to show
social responsiveness and moral concern. In all cultures, feelings of moral obligation tend to be
much stronger towards kin, friends, or members of an in-group, than towards strangers, or
members of an out-group (Sahlins, 1965). Yet from time to time people do respond to the needs
of strangers and there are some people who exhibit such concern consistently throughout their
lives.
Reed and Aquino (2003) theorized one reason why people sometimes show greater
concern for strangers and other socially distant out-group is that their moral identities occupy
high self-importance within their working self-concept. As a result, they are motivated to extend
the boundaries of solidarity and moral obligation to include more groups and individuals within
their moral community because it reinforces their mental representation of what it means to be a
moral person. Younis and Yates (1999) made a similar argument when explaining the actions of
European adults known to have protected and sheltered Jews during the Nazi occupation of
World War II (e.g., Oliner & Oliner, 1988; Monroe & Epperson, 1994). According to Younis
and Yates (1999), these studies revealed that rescuers were different from non-rescuers in that
the former tended to perceive Jews as fellow human beings who merited the same privileges and
community rights. In contrast, non-rescuers viewed Jews as outsiders, as “they” rather than nonJewish “we.” Monroe and Epperson (1994) suggested that these actions reflected the self’s
relationship to others and flowed spontaneously from deep-seated dispositions to extend
psychological group boundaries.
Reed and Aquino (2003) provided one of the few empirical tests of this hypothesis. In
four studies using both attitudinal (e.g., expressing positive attitudes towards out-groups, Study
Self-Regulatory Mechanisms and War
10
2) and behavioral reactions (e.g., financially supporting out-groups, Study 3) they showed that
persons whose moral identity had high, as opposed to low, self-importance felt a stronger moral
obligation to help out-group members. These findings provided evidence for Reed and Aquino’s
(2003) hypothesis that when moral identity has high self-importance people feel a stronger
obligation to respond to the needs and interests of socially distant others because the perceiver’s
circle of moral regard becomes more inclusive.
Based on Reed and Aquino’s (2003) argument, we propose that when a person’s country
wages war, having a highly self-important moral identity will be positively associated with antiwar cognitions, emotions, and behaviors, whereas the inverse relationships hold when a person’s
moral identity has low self-importance. This prediction follows from evidence showing that a
person whose moral identity has high self-importance is more responsive to the needs of others.
Hence, high moral identifiers should be more willing to take into account the suffering that war
inflicts on “others” even if these others are deemed by their country as “the enemy” (Reed &
Aquino, 2003). In turn, this may influence their reactions to war such that they may, for example,
experience more ambivalent emotions and thoughts about war. We suggest that this ambivalence
can weaken or neutralize the effect of moral disengagement on pro-war emotions, thoughts and
behaviors3.
Overview of Studies
In three studies conducted as the U.S. waged wars against Afghanistan and Iraq, we
examine of the extent to which moral disengagement and moral identity interact to influence proand anti-war cognitions, emotions, and behavior. In study 1 we used the prospect of a U.S. led
invasion against Iraq as a backdrop for testing whether a negative relationship between moral
disengagement and feelings of guilt and shame regarding such an invasion would be weaker for
Self-Regulatory Mechanisms and War
11
those whose moral identity has high rather than low self-importance. The data for study 1 were
collected two weeks before the actual invasion of Iraq, which began on March 20, 2003. We
conducted studies 2 and 3 to explore mediating moral judgments of out- and in-groups that link
moral disengagement and moral identity to pro- and anti-war reactions. In study 2 we tested
whether moral identity weakened the positive relationship between moral disengagement and the
endorsement of a highly punitive response toward out-group members responsible for the
September 11th attacks, and whether this endorsement would be associated with pro-war feelings
of pride, strength, determination, and inspiration. This data were collected in January 2002,
shortly after the American invasion of Afghanistan.
Finally, Study 3 examines how Americans would react to abuses to Iraqi prisoners
inflicted by U.S. troops. The focus of study 3 is how multiple and potentially conflicting
identities might impact the relationship between moral disengagement and reactions to abuses
perpetrated by the in-group (i.e., the U.S. military). Study 3 examined this question by testing
whether moral identity would weaken the positive relationship between moral disengagement
and the perceived morality of rewarding American soldiers responsible for the Abu Ghraib
prison abuses. In turn, we expected this judgment to influence peoples’ willingness to donate
money to a charity that benefits the U.S. military (the USO). But we also examined whether
having a strong American identity – an identity that presumably restricts the circle of moral
regard towards out-groups during conflict – would influence cognition and behavior in a manner
opposite to that of moral identity. This data were collected in July of 2004 amidst continued
fighting between the occupying U.S. forces and Iraqi citizens.
Self-Regulatory Mechanisms and War
12
Study 1: Feelings of Shame and Guilt as a Function of Moral Disengagement
Practices and Moral Identity
I, too, belong to this species. I am ashamed not only of my own deeds,
not only of my nation’s deeds, but of human deeds as well. I am ashamed to be a man.
J. Glenn Gray
The Warriors: Reflections of Men in Battle
In March 2003, the U.S. and its coalition allies invaded the nation of Iraq. The
justifications given for this war were both complex and controversial. Among others, they
included preventing Iraq from developing weapons of mass destruction, ending the rule of a
murderous dictator, and bringing freedom and democracy to the Iraqi people and that part of the
world. Although a majority of Americans supported the invasion (Jones, 2003), a considerable
number of Americans opposed it as did millions of world citizens. Those who opposed the war
did so for many reasons, one of which was that an invasion against a country that had a large and
presumably well-equipped army would result in an unacceptable level of military and civilian
casualties. Moreover, since Iraq had not attacked the United States, some viewed a pre-emptive
war as morally unjustified. Despite these objections, the invasion of Iraq seemed to have a
generally positive effect on the way many Americans felt about their country, their President,
and the men and women of the armed forces. For example, national opinion polls revealed that
58 % of Americans were supportive of the anticipated invasion of Iraq (Benedetto, 2003). In
April 2003, shortly after the invasion, 73 % of Americans believed that it “was worth going to a
war in Iraq” (CNN, 2003).
It is not uncommon for some people to experience feelings of shame or even guilt over
the death and destruction war causes. But even those who support war, including soldiers who
have been trained to be remorseless killers, are not always immune from such emotions. Author
Self-Regulatory Mechanisms and War
13
and World War II marine veteran William Manchester recounted his experiences the first time he
killed an enemy at close range:
Then I began to tremble, and next to shake all over. I sobbed, in a voice still
grainy with fear: “I’m sorry.” Then I threw up all over myself. I recognized the
half-digested C-ration beans dribbling down my front, smelled the vomit above
the cordite. At the same time I noticed another odor; I had urinated in my
skivvies…I knew I had become a thing of tears and twitchings and dirtied pants. I
remembering wondering dumbly: Is this what they mean by “conspicuous
gallantry”? (Manchester, 1980: 7).
If a trained soldier like William Manchester can feel shame and guilt during war, it seems
possible for civilians to experience such feelings as well, although probably to a much lesser
degree given their distance from the actual scene of killing and death. Shame has been defined as
a helpless emotion aroused by self-related aversive events (Ferguson & Stegge, 1998). The
ashamed person focuses on devaluing or condemning the entire self, experiences the self as
fundamentally flawed, and thus avoids or hides from others (Ferguson and Stegge 1998). Guilt is
an emotion that one feels after having accepted responsibility for a behavior that violates
internalized standards or causes another person to suffer (Ferguson & Stegge, 1998; Tangney,
1991). Both emotions can be responses to the same situation and both can arise from concerns
about the effects of one’s behavior on others (Eisenberg 2000). They are similar in that they
involve a sense of responsibility and the feeling that one has violated a moral standard
(Eisenberg, 2000). For this reason, shame and guilt are considered members of the same
“family” of self-conscious moral emotions that prompt people to avoid harming others or to
repair the effects of harm-doing once it has occurred (Eisenberg, 2000; Haidt, 2001). As William
Manchester’s account illustrates, these emotions can arise when a person confronts an
inconsistency between the moral standards internalized and espoused since childhood and the act
of killing another human being. If sufficiently powerful, such feelings may motivate people to
Self-Regulatory Mechanisms and War
14
resist participating in or endorsing acts of war. For this reason, we treat them as anti-war
emotions.
In Study 1, we tested whether moral disengagement and moral identity interact to predict
guilt and shame when people are asked to report on how they would be likely to feel if the U.S.
decided to wage war against another nation. Based on the theoretical argument that moral
disengagement allows people to disable or ignore internalized standards that condemn acts of
violence against others, our first hypothesis is that by expressing and maintaining cognitive
justifications of war through moral disengagement, feelings of guilt and shame associated with a
U.S. decision to invade another country will be reduced. However, building on the circle of
moral regard hypothesis, our second prediction is that the effectiveness of moral disengagement
as a mechanism for reducing feelings of shame and guilt will be weaker for people whose moral
identity has high rather than low self-importance.
Sample, Method and Procedure
Study one was conducted with two independent samples from two northeastern
universities who completed two surveys for course credit. The surveys were collected at two
time periods. In the first time period participants answered a segmentation exercise
questionnaire. This questionnaire included the measure of moral identity. This measure was part
of a general opinions questionnaire that contained several items not related to the current study.
Between six and eight weeks later, participants completed a “perspective-taking survey” asking
for their views “on going to war”. These data were collected approximately two weeks prior to
the invasion of Iraq, during a time when experts considered an invasion of Iraq as imminent.
After reading some brief introductory information, participants were asked to generate reasons
for and against war in a foreign country. Since the issue of invading Iraq was mentioned daily by
Self-Regulatory Mechanisms and War
15
the media at the time of this study, we were concerned that participants would be unable to think
even-handedly about the issue. Therefore, we used a perspective taking task (Bolton & Reed
2004) to encourage participants to be as “balanced” as possible in the generation of thoughts
used to make their judgments. To help establish a more even playing field, we asked each
participant to generate reasons for both going to war and not going to war as follows:
•
Please take a few moments to think about reasons in favor of going to war in a foreign
country. Assume that you are an American who supports war in a foreign country. As an
American who supports war in a foreign country, why is it a good idea? Write down as many
reasons as you can in the space below.
•
Please take a few moments to think about reasons against war in a foreign country.
Assume that you are a religious person who does not support going to war in a foreign country.
As a religious person who is against war in a foreign country, why is it a bad idea? Write down
as many reasons as you can in the space below.
Participants were randomly assigned to one of two order conditions: either generating pro-war
thoughts or con-war thoughts first. Additionally, because we wished to avoid thought
polarization (Bolton & Reed, 2004) that may result from asking questions about a topic that most
people were exposed to on a daily basis, each of the two thought tasks described above lasted 3
minutes and was timed by the experimental proctor. Participants then responded to the general
issue of the U.S. going to war in a foreign country by answering several questions designed to
asses the extent to which they had a propensity to morally disengage from a U.S. led war in Iraq.
Participants were then asked to indicate the extent to which they would experience aversive
emotional reactions (guilt and shame) in response to the U.S. going to war with a foreign
country. The time two survey concluded with various demographic questions. Of the students
who completed the initial identity measures, 137 provided usable data for all study variables. The
sample consisted of 56% males, with an average age of 21.1 years (sd = 2.12).
Self-Regulatory Mechanisms and War
16
Independent Variables
Moral identity. We used the five-item Internalization subscale of Aquino and Reed’s
(2002) moral identity instrument to measure this construct. According to Aquino and Reed
(2002) the Internalization subscale captures the degree to which a person’s moral identity is
rooted at the core of one’s being. We used this subscale of their two-dimensional measure
because it appears to be the most robust predictor of morally-relevant behavior (cf. Aquino &
Reed, 2002; Reed & Aquino, 2003) and is consistent with the theoretical definition of moral
identity we adopt in this paper. Respondents answered each item on a 7-point Likert scale
(1=Strongly Disagree; 7=Strongly Agree). Items were averaged to form a scale α=(.79).
Moral disengagement We wrote six items that justify going to war as serving a valued or
righteous purpose – “I believe that America should bring freedom and democracy to foreign
countries, I believe a conflict with a foreign country will uphold International law, I believe that
some countries threaten the well-being of the US, I believe that the US should do whatever it
must to protect the Free World, I believe that a more humane regime in a foreign country is
worth going to war against a tyrannical leader, and I believe that history will show that the
decision to go to war against foreign countries is the right one.” To answer these questions,
participants were asked to think about the general issue for the U.S. going to war against a
foreign country and to indicate their agreement with each of the justifications on a 7-point
(1=strongly disagree, 7=strongly agree) Likert format. The items were averaged to form a scale
where a high score indicated higher moral justification (α =.86).
Control Variables. We assessed possible primacy and recency effects associated with our
perspective taking task (cf. Bolton & Reed 2004) by creating an “order” variable to control for
whether or not participants first generated pro thoughts or con thoughts toward war with the
Self-Regulatory Mechanisms and War
17
latter as the reference category (0=pro thoughts first, 1 = con thoughts first). We also controlled
for sex because there is evidence that women are less supportive of the use of military
instruments in foreign policy than men (Fite, Genest & Wilcox, 1990; Bendyna, Finucane, Kirby,
O’Donnell, & Wilcox, 1996) .
Dependent Variable
Anti-war emotions. After respondents answered the items justifying a U.S. led war
against a foreign country, we asked them to indicate the extent to which they believed they
would experience feelings of “shame” and “guilt” if the U.S. decided to go to war. These
emotions were assessed on a 7-point Likert scale (1=Not at all, 7 = Very much) and were
combined to form a scale (α=.79 )4 with higher scores indicating stronger anti-war emotions.
Results
Descriptive statistics and correlations among the study variables are shown in Table 1.
----------------------------Insert Table 1 here
------------------------------We tested the study hypotheses using multiple regression. Our first hypothesis was that
moral disengagement would be negatively related to self-reported feelings of guilt and shame;
our second hypothesis was that this effect would be weaker among persons whose moral identity
had high rather than low self-importance (measured two months earlier). We tested this second
hypothesis by including a moral disengagement x moral identity interaction in our model along
with the independent variables and the order control variable. The variables forming the
interaction were centered to minimize multicollinearity (Aiken & West, 1991). The results of the
regression analysis are shown in Table 2.
Self-Regulatory Mechanisms and War
18
-----------------------------Insert Table 2 here
------------------------------Table 2 shows that moral disengagement was negatively related to self-reported feelings
of guilt and shame (b=-.63, p <.001), which provides support for our first hypothesis. As
predicted, this effect is qualified, by a marginally significant moral disengagement x moral
identity interaction (b=.33, p = .06). We analyzed the form of the interaction following
procedures recommended by Aiken and West (1991). Specifically, we tested the relationship
between moral disengagement and self-reported feelings of guilt and shame for participants who
scored one standard deviation above and below the mean on the internalization measure of moral
identity. The simple slopes for each of these groups reveals that while moral disengagement was
negatively related to self-reported feelings of guilt and shame regarding the U.S. going to war in
a foreign country for low moral identifiers (b = -.74, p < .001) , this relationship was weaker for
high moral identifiers (b = -.36, p < .01). These results support Hypothesis 2.
Discussion of Study 1
Study 1 showed that people who endorsed various justifications for going to war expected
to experience less guilt and shame when asked to imagine a foreign invasion conducted by their
country. However, we also found that a highly self-important moral identity weakened this
relationship. We explained this neutralizing effect by drawing from Reed and Aquino’s (2003)
argument that having a self-important moral identity expands a person’s circle of moral regard.
Thus, an act of war waged against an out-group elicits self-conscious emotions because the
person experiences even potential enemies as deserving of some moral concern. Whether these
anti-war emotions are sufficiently powerful to discourage people from supporting war cannot be
answered with our data because we only examined peoples’ emotional reactions to a hypothetical
Self-Regulatory Mechanisms and War
19
decision. We did not ask people whether or not they would choose to go to war, or to support
war, if asked to do so. Nevertheless, our findings do suggest that moral disengagement in the
form of justifying a war in the service of higher righteous virtues and principals may not
necessarily lead to pro-war behaviors, since the higher level of self-conscious emotions of guilt
and shame experienced by high moral identifiers may suppress support of actions that can cause
harm to others.
Although the findings of Study 1 provide preliminary empirical support for our
theoretical arguments, there are several limitations of the study that we hoped to address in
Studies 2 and 3. First, study 1 examined only one class of emotions. Second, even though we
tried to enhance the vividness of the decision by encouraging participants to generate concrete
reasons for and against going to war in a foreign country, it was likely that many respondents had
difficulty imagining the real moral consequences of war using an abstract and purely
hypothetical scenario. This may explain why the relationships between moral disengagement and
the anti-war emotions were highly significant for both high and low moral identifiers. Third, the
interaction between moral disengagement and moral identity did not quite reach the conventional
.05 significance level.
We tried to address these limitations in Studies 2 and 3 by presenting participants with
more powerful stimuli toward which they could react: images of inter-group aggression
associated with the attacks of 9/11 and the prison abuses at Abu Ghraib. We also extend study 1
by assessing mediating cognitive processes that might link the self-regulatory constructs of moral
disengagement and moral identity to anti-war outcomes. Specifically, we examined whether
moral judgments people make about how to respond to harm-doers mediates the relationship
between the independent variables and war-related emotions, cognitions, and behavior. These
Self-Regulatory Mechanisms and War
20
studies also update the previous boundary conditions of Reed and Aquino’s (2003) circle of
moral regard hypothesis by directly exploring the out-group and in-group implications of a
highly self-important moral identity.
In response to an act of inter-group aggression directed towards one’s in-group (U.S.
citizens), we propose that people whose moral identity has low self-importance should advocate
a more punitive response towards the harm-doers than people whose moral identity has high selfimportance. The reason is that low moral identifiers should feel little or no obligation to show
concern for suffering of out-group aggressors (terrorists). Indeed, they may even be highly
motivated to inflict the maximum amount of suffering on these aggressors as a punishment for
attacking the in-group. In either case, these moral judgments should evoke stronger pro-war
sentiments. On the other hand, if an act of inter-group aggression is committed against an outgroup by members of their in-group, low moral identifiers should be more willing to reward
such behavior than high moral identifiers. Again, this prediction follows directly from the circle
of moral regard hypothesis. Since low moral identifiers have a restrictive circle of moral regard,
they will be less concerned than high moral identifiers about the injuries their in-group members
(U.S. soldiers) inflict upon members of an out-group (Iraqi citizens). Consequently, low moral
identifiers should be relatively more willing support such acts by rewarding in-group members
who commit them. However, because an out-group member is more likely to be included within
their circle of moral regard, high moral identifiers may be more troubled by harmful acts directed
against out-groups (Reed & Aquino, 2003). As a result, they will be less motivated to glorify,
reward, and support those who perpetrate such acts even if they occur during war.
Self-Regulatory Mechanisms and War
21
Study 2: Reactions to the 9/11 attacks by Out-Group Members
as a Function of Moral Disengagement Practices and Moral Identity
There is no denying the aesthetic satisfaction, the sense of poetic justice that pleasures
us when evil-doers get the comeuppance they deserve...
Arthur Lelyveld Punishment: For and Against
If asked, many people might not admit that they would enjoy taking revenge against
those who harm them. This is because revenge is often portrayed in the media, in religious
writings, and in the academic literature as an emotional and irrational act that should not be
tolerated in civilized society (Jacoby, 1983; Solomon, 1990). Yet the impulse to retaliate against
transgressors is among the most ancient and universal of human motives. The famous “law of the
talion“ cited in Exodus 21:23-25, called for “life for life, eye for eye, tooth for tooth, hand for
hand, foot for foot, burning for burning, wound for wound, stripe for stripe.“ The occurrence of
revenge in many primate species (de Waal, 1991; Silk, 1992) suggests a biological basis for
human vengeance and may explain why, despite what they may publicly proclaim, many people
do in fact find great delight and even pleasure in a well-executed act of revenge (Tripp, Bies, &
Aquino, 2002). The enduring popularity of Hollywood “action“ movies in which the hero
punishes a wrong-doer (usually in an imaginative and exceedingly violent way) supports the
notion that there is an “aesthetic“ to revenge grounded in moral intuitions about the nature of
justice (Tripp, et al., 2002).
Revenge can be deeply satisfying and empowering because, as Nietzsche (1887) once
observed, it elevates one’s ego and demonstrates one’s ability to assert their right to a just
outcome. But revenge also satisfies other motives like the desire to maintain equity in
relationships (Berscheid, Boye, & Walster, 1968) and to defend against future harm-doing by
Self-Regulatory Mechanisms and War
22
demonstrating that aggressive acts will not go unanswered (Axelrod, 1984). For these reasons,
people sometimes judge revenge as being justified and even morally “good” as long as it follows
several principles (Tripp & Bies, 1997; Tripp, et al., 2002). Among these are that the act of
revenge: 1) allows the victim to feel empowered for doing something rather than nothing; 2)
protects the victim or others from further harm; 3) is not disproportionate in severity relative to
the original transgression, 4) does not hurt innocent bystanders; and 5) serves a purpose larger
than the avenger’s own interests (Tripp & Bies, 1997; Tripp, et al, 2002). Indeed, it appears that
in some circumstances people believe that too little revenge is as unacceptable as too much
(Tripp, et al., 2002).
The September 11th terrorist attacks would seem to be a situation where taking revenge
against those responsible would be viewed by many Americans not only as a pragmatic
necessity, but also a moral imperative. In this case, revenge took the form of the U.S. invasion of
Afghanistan in October 2001 and the effort by U.S. armed forces to capture or kill Osama Bin
Laden, his Al Qaeda followers, and the Taliban leaders who supported them. Supporting the
argument that people derive great satisfaction from revenge, national polls indicated war was
strongly supported by the American people (Newport, 2001). Further evidence is provided by the
proliferation of patriotic bumper stickers, billboards, and t-shirts exhorting people to stand
behind President Bush and to find, capture, or even kill those responsible for the 9/11 attacks.
There are many ways the country and its leaders could have dealt with the perpetrators of
the 9/11 attacks. For example, they might strictly follow the “law of the talion” by hunting down
and killing all of the perpetrators without pity or remorse. This was a strategy followed by the
Israeli government’s creation of the Mivtzan Elohim (“Wrath of God“) in response to the murder
of its athletes during the 1972 Munich Olympics (Livingstone, 1986). The members of this
Self-Regulatory Mechanisms and War
23
Israeli organization devoted themselves solely to retaliating in an “eye for an eye” fashion
against the Palestinian group, Black September, which was responsible for murdering the
Olympic participants.
The question we sought to answer in study 2 was: How would people who believed a
highly punitive response (killing them) to the perpetrators of the 9/11 attacks was also the most
moral would feel after admitting this preference? Would they experience pro-war emotions after
expressing their desire to see the perpetrators killed? Or would they feel badly about wanting to
kill others even if these others were guilty of inhumane acts? Also, how might the willingness to
kill the perpetrators of 9/11 be influenced by moral disengagement and the self-importance of
their moral identities? To answer these questions, we tested three hypotheses. First, we
hypothesize that people who morally disengage will be more likely to judge killing the
perpetrators of the 9/11 attacks as more moral than non-lethal alternatives (e.g., life
imprisonment, forced financial victim compensation, forgiving them). Second, we hypothesize
that this relationship will be weaker among people whose moral identity has high rather than low
self-importance. Third, we hypothesize that people who judge killing the perpetrators as more
moral than other responses will experience more pro-war emotions after admitting to this
preference than those who endorse non-lethal responses. However, this relationship will be
stronger if their moral identity has low rather than high self-importance. This last hypothesis
implies that judgments about the morality of killing the perpetrators mediate the relationship
between the interaction of moral disengagement and moral identity on pro-war emotions.
Method
The data for this study were part of a multi-survey data collection effort that took place
over a period of several months following the 9/11 attacks on the World Trade Center. This
Self-Regulatory Mechanisms and War
24
procedure involved the collection of multiple measures using different surveys from the same
sample of University of Pennsylvania students enrolled in a marketing research class. All of the
data from this collection effort can be linked because the same students completed the different
surveys over time. However, not all of the measures collected were meant to address the same
theoretical questions. Rather, a host of variables were collected to allow an examination of
evolving questions concerning peoples’ reactions to 9/11. Some of the data from this data
collection effort were analyzed and reported in a previous study by Reed and Aquino (2003).
However, with the exception of sex, none of the data reported in the current study have ever been
previously analyzed or reported.
Sample Method and Procedure
One-hundred and four participants from a Northeastern university participated in this
study and provided usable data on all study variables. Sixty-two were female. Of those who
reported race, 58 were Caucasian, 8 were African-American, 31 were Asian/Pacific Islanders, 3
were Hispanic, and 4 self-identified as “other.“
The data were collected using on-line surveys at two time points. At time one,
participants answered a battery of “personality tests” administered at the beginning of the
semester including a measure of the self-importance of moral identity. Three months later,
participants completed a computer exercise intended to measure reactions toward “historical
events.” They were told that the purpose of the time two study was to gain a comprehensive
understanding of how people respond to and remember major events. Twenty pictures were
taken from various online media regarding the World Trade Center attacks. The pictures were
shown on a computer screen in a random order. Participants were then asked the affective, moral
judgment, and moral disengagement questions described below.
Self-Regulatory Mechanisms and War
25
Independent Variables
Moral identity. As in study 1, Aquino and Reed’s (2002) Internalization subscale was
used to measure this construct. Respondents answered the items on 5-point (1=Strongly Disagree
5=Strongly Agree) Likert type scales. Items were averaged to form a scale score (α =.85).
Moral disengagement. We used four items from the moral justification subscale of
Bandura et al’s., (1999) measure. We used this subscale because we wanted to present
participants with more general rationalizations for violent retaliation (e.g., “It’s alright to fight to
protect your friends,” “It’s alright to beat someone up who badmouths your family,” “It is okay
to treat badly someone who behaved like a worm,” “It’s alright to fight when your group’s honor
is threatened“). Respondents answered on a 5-point (1=Strongly disagree, 5=Strongly agree)
Likert scale. We also wished to make a direct link to established psychometric work on moral
disengagement. Items were averaged to form a scale (α=.75).
Control. As in study 1, we controlled for sex using dummy coding (0=Male, 1=Female).
Dependent Variables
Moral judgment. We assessed moral judgments about several possible responses that
varied in punitiveness toward the perpetrators of the 9/11 attacks. We presented participants with
five hypothetical choices: “Use any means necessary to kill those responsible for these acts,”
“Try to capture, but not kill those responsible so that they can be tried, and if found guilty,
imprisoned for the rest of their lives,” “Force those responsible to pay economic damages to the
victims of their acts but do not kill them, “Forgive those responsible for these acts, meaning
negative emotions like hatred and anger should be replaced with positive emotions like
compassion and love,” “Extend acts of goodwill towards those held responsible in an effort to
Self-Regulatory Mechanisms and War
26
promote reconciliation and mutual understanding between the parties in conflict.” The first
choice – the one of primary interest— follows the “eye for an eye” principle and represents the
most punitive response; the second and third choices represent efforts to seek retributive justice
through non-lethal means; and the fourth and fifth choice represent the least vengeful and
punitive alternatives. Indeed, they represent the antithesis of vengeance. Participants were asked
to make a forced choice response by indicating which of these alternatives they thought was the
“most moral”. Consistent with our theorizing, we wanted to examine the effects of endorsing the
most punitive option, so we collapsed all responses other than killing the perpetrators into a
single category to represent non-lethal response options. For our analysis, we coded killing the
perpetrators as “1” and other responses as “0.”
Pro-war emotions. We used four adjective descriptors – inspired, strong, determined, and
proud – to measure the extent to which participants’ experienced positive emotions that are more
likely to motivate them to support rather than oppose war. We measured these emotions after
participants had both judged the morality of killing the perpetrators relative to other possible
responses and after they answered the moral justification items. Participants were asked to
indicate how they felt “right at this moment” on a 5–point Likert scale (1=not at all,
5=extremely). Items were averaged to form a scale α=(.88).
Results
Descriptive statistics and correlations among the study variables are shown in Table
three.
----------------------------Insert Table 3 here
-------------------------------
Self-Regulatory Mechanisms and War
27
Our first hypothesis was that moral disengagement would be positively related to the
choice of killing the 9/11 perpetrators as being the most moral response; our second was that this
effect would be weaker among persons whose moral identity had high rather than low selfimportance. We tested these hypotheses by regressing moral judgment on moral disengagement,
moral identity, the moral disengagement x moral identity interaction, and sex. The variables
forming the interaction were centered to minimize multicollinearity between the interaction term
and its components. The results of this analysis are shown in Table 4.
------------------------------------Insert Table 4 here
----------------------------------Table 4 shows that moral disengagement was positively related to the judgment that
killing the perpetrators was the most moral response (b=.17, p <.001), which supports our first
hypothesis. This effect is qualified, however, by a significant moral disengagement x moral
identity interaction (b=-.31, p < .001). We tested the simple slopes of the relationship between
moral disengagement and the retaliatory response of killing the perpetrators for persons who
scored one standard deviation above and below the mean on the moral identity measure to
analyze the form of the interaction. This analysis reveals while moral disengagement was
positively related to the morality of killing for low moral identifiers (b=.32, p < .001), there was
no relationship between these variables for high moral identifiers (b=.02,n.s). This supports our
second hypothesis.
Our third hypothesis was that the judgment of how to respond to the 9/11 perpetrators
mediates the relationship between the moral disengagement x moral identity interaction and prowar emotions. We tested this hypothesis by performing Baron and Kenny’s (1986) test for
mediation. According to Baron and Kenny (1986) support for mediation is indicated if: 1) the
Self-Regulatory Mechanisms and War
28
independent variable significantly predicts the dependent variable in the first equation; 2) the
independent variable significantly predicts the mediator variable in the second equation, and 3)
the mediator significantly predicts the dependent variable in the third equation and the effect of
the independent variable on the dependent variable is weaker when the mediator is added. If the
relationship between the independent and dependent variable becomes non-significant when the
mediator is added, this indicates full mediation.
We already performed step 2 of the test for mediation. We performed steps 1 and 3 by
performing a hierarchical regression predicting pro-war emotions. We entered the independent
variables (moral disengagement and moral identity) and their interaction in the first step, along
with sex. In step 2, we added the judgment of the morality of killing the perpetrators. The results
of this analysis are also shown in Table 5.
-------------------------------Insert Table 5 here
--------------------------------Our third hypothesis is supported as indicated by the significant moral identity x moral
disengagement interaction in step 1 (b=-.52, p < .01), which then becomes non-significant when
the mediator is added to the equation (b=-.31, n.s.). The mediator itself is significant (b=.67, p <
.05) and in the expected direction, thus satisfying Baron and Kenny’s (1986) third condition for
establishing mediation.
Discussion of Study 2
In Study 2, we tested an observation made by social theorists, biologists, and
philosophers: taking revenge against a harm-doer can be highly satisfying, empowering, and
even aesthetically pleasing. We found that while morally endorsing a highly punitive act of
vengeance was indeed related to more positive emotions, not everyone felt equally good after
Self-Regulatory Mechanisms and War
29
making such an endorsment. Those who felt best were people who had a general tendency to
justify aggressive behavior through mechanisms of moral disengagement. However, as in Study
1, the strength of this relationship was moderated by the self-importance of moral identity such
that these justifications had a weaker effect on pro-war emotions among people whose moral
identity had high as compared to low self-importance.
Our third hypothesis was that moral judgment would mediate the relationship between the
moral disengagement x moral identity interaction and pro-war emotions. We found support for
this hypothesis indicating that moral cognitions concerning appropriate responses to an outgroup transgressor may partly explain the link between the self-regulatory mechanisms of moral
disengagement and moral identity and war-related emotions found in both studies 1 and 2.
A key limitation of the first two studies is that they focused on only one kind of identity
as interacting with moral disengagement mechanisms. While moral identity is probably among
the identities that can strongly influence peoples’ responses to war, it is likely that other selfimportant identities may lead to different types of responses (Greenwald & Pratkanis, 1984).
Following the logic of the circle of moral regard hypothesis, we theorize that any identity that
restricts the psychological boundaries of one’s moral community should interact with moral
disengagement in a manner opposite to that of moral identity. Among the many identities that are
likely to have this effect, we chose to investigate American identity as a potential source of selfregulation. Using the context of the Abu Ghraib prison scandal, study 3 examines the possibility
that both moral and American identity can influence the effect of moral disengagement on how
people judge the morality of in-humane acts committed by their in-group, but in opposite
directions. A second objective of study 3 was to test whether these variables affect actual proand anti-war behavior.
Self-Regulatory Mechanisms and War
30
Study 3: Reactions to Iraqi Prison Abuse by In-Group Members
as a Function of Moral Disengagement Practices and Contrasting Identities
“Two souls, alas, are dwelling in my breast, and one is striving to forsake his brother.”
Goethe Faust
The treatment of the enemy in war has always been a practical and moral challenge for
military personnel because they operate within a context of extreme violence and inter-group
hostility. Recognizing that the prisoner of war is a helpless target for acts of cruelty and torture,
and a convenient outlet for the anger and hatred one may feel towards one’s enemy, the nations
of the world established the rules of the Geneva Convention to protect the rights of soldiers
captured during wartime. For example, Article 3 of the Convention prohibits aggressions, torture,
cruel treatment, and outrages upon a person’s personal dignity (Geneva Convention, 1949).
In January 2004, a U.S. Army MP came upon digital images of apparent Iraqi prisoner
abuse. He reported the pictures to his superiors, prompting coalition commander Lt. General
Ricardo Sanchez to order one of three independent investigations by United States Army Major
General Antonio Taguba. The 53-page report resulting from this investigation concluded that
U.S. soldiers had committed "egregious acts and grave breaches of international law" at the Abu
Ghraib prison (http://www.globalsecurity.org/intell/library/reports/2004/800-mp-bde.htm) The
report outlined details that between October and December 2003 there were numerous instances
of "sadistic, blatant, and wanton criminal abuses" of Iraqi prisoners. In violation of Army
regulations, intelligence officers asked military police to "loosen up" inmates before questioning.
The report estimates that 60% of the prisoners at the site were "not a threat to society" and yet
some were subjected to acts of torture and humiliation many observers believed violated the
Articles of the Geneva Convention.
Self-Regulatory Mechanisms and War
31
The abuse of prisoners, like the death of innocents, is an unavoidable consequence of
war. Past research in social psychology suggests that acts like those committed at Abu Ghraib
should not have been surprising given the power of role expectations to shape the behavior of
people in prison settings (Zimbardo, 1972). Once such acts occurred and were publicly disclosed,
an interesting question that builds on the findings of Studies 1 and 2 is: How would American
citizens respond to seeing the abuses directed against Iraqi prisoners by American troops? The
essayist Lewis Lapham posed the nature of this question more pointedly:
“If it is our intention to rule the world from the throne of military empire, how
willing are the American people to tolerate or ignore, perhaps even admire and
applaud, the cruelties necessary to the maintenance of so great a glory?” (Lapham,
2004: 10)
The conceptual arguments put forward thus far suggest that emotional reactions will
partly depend on the ability to morally disengage, but that the effectiveness of such rationalized
cognitions should be influenced by the self-importance of moral identity. However, because the
self is multifaceted, other identities may also exert self-regulatory influence over the individual
dealing with a complex situation like war (Greenwald & Pratkanis, 1984).
During inter-group conflict, social identities can heighten protective actions toward the
in-group and increase hostility toward out-groups (Tajfel & Turner, 1979). Perhaps the most
powerful social identity in war is national identity, which led us to consider the possibility that
American identity might have a self-regulatory influence that works in a direction opposite to
that of moral identity. National identity can be defined as an affiliation with the nation that gives
people a sense of who they are in relation to others. This identity may operate as part of a
person’s self-concept in much the same way as moral identity. Unlike moral identity, though, we
propose that a strong American identity should amplify distinctions between American in-groups
and Iraqi out-groups thereby restricting rather than expanding the circle of moral regard. Based
Self-Regulatory Mechanisms and War
32
on this argument, we predict that rather than neutralizing the effects of moral disengagement and
pro-war reactions, a highly self-important American identity should strengthen them. The reason
is that the directional effects of moral disengagement and American identity are mutually
reinforcing because they may both motivate people to hold cognitions, express emotions, and
behave in ways that protect the in-group.
We examined two kinds of reactions to the Abu Ghraib abuses to test this argument:
moral judgment and pro-war behavior. Following the theoretical model proposed in Study 2, we
tested whether moral disengagement and the two contrasting identities of moral and American
identity would influence moral judgment. In contrast to Study 2, which examined a moral
judgment about punishing a transgressor, we asked people in Study 3 to make a moral judgment
about rewarding the U.S. soldiers who had committed the abuses at Abu Ghraib. We hypothesize
that people who morally disengage should more strongly endorse the morality of rewarding the
soldiers. However, our second hypothesis is that this relationship is weaker for people whose
moral identity has high rather than low self-importance but stronger for those whose American
identity has high rather than low self-importance. In turn, we expected moral judgment to
influence the willingness to provide financial support to an organization whose primary purpose
is to support U.S. military objectives (USO). Our third hypothesis is that (as in Study 2) moral
judgment mediates the relationship between the interaction of moral disengagement and the two
identities on pro-war behavior.
Sample Method and Procedure.
The sample consisted of 154 community residents; students and staff from a Northeastern
university. They were recruited as walk-ins by flyers posted in the neighborhood and were paid
$10 to participate in 45 minutes to one hour of various research tasks in a summer behavioral lab
Self-Regulatory Mechanisms and War
33
conducted by the university. Upon entering the lab, they were given a brief introduction as to the
nature of the lab, types of information collected therein, etc. Each participant completed five
separate and ostensibly unrelated research tasks. In the first task, respondents completed a
number of different scales. As part of this task, they completed measures of moral identity, and
measures of American identity, buried within other measures unrelated to the current study. The
experimental proctor then handed out the next exercise which was a filler task involving the
assessment of different creative car ads. After completing the second task, participants were
placed in front of a private computer screen where the third task took place. The third task was a
study of reactions to current NEWS events. Participants were told to please view a forthcoming
picture show and that when the show ended, to answer the questionnaire provided by the proctor.
The slide show consisted of eleven pictures taken from the Abu Ghraib prison scandal, presented
on screen to subjects at four seconds per picture. Following the slide show, experimental proctors
handed participants a questionnaire that assessed their moral judgments and cognitions about the
pictures they just viewed as well as several other current events opinion items not of direct
interest to the present study. The next task was a filler task in which participants made price
fairness judgments in a number of hypothetical consumer scenarios. At the end of the session,
participants completed a Resource allocation task in which we attempted to capture a measure of
actual behavior. Participants were then debriefed, remunerated and thanked for their
participation. The three tasks of this study were matched up with random participant numbers
provided to respondents upon entry into the lab. 136 study participants completed all measures
and 18 were eliminated due to either administration or participant errors in correctly filling out
their identification numbers.
Self-Regulatory Mechanisms and War
34
Independent Variables
Moral identity. Aquino and Reed’s (2002) 5-item Internalization subscale was used to
measure this construct. Respondents answered on 7-point (1=Strongly Disagree 7=Strongly
Agree) Likert type scale. Items were averaged to form a scale (α =.85).
American identity. The strength of American identity was measured with four items
adapted from previous research (Reed & Aquino 2003). These items appeared in the first part of
the survey, before the measures of moral disengagement and the resource allocation task. The
items were “It’s great to be an American,” “I am extremely proud of my affiliation with the
United States of America,” “Being a member of this country makes me feel like I share a
common goal with others,” and “Being an American is an important part of who I am”. These
items were answered on a 7-point Likert (1=Strongly disagree, 7=Strongly agree) scale and were
averaged to form a scale (α=.88).
Moral disengagement. We developed four items that measured the use of advantageous
comparisons to justify the abuse of Iraqi prisoners. These items were “Compared to the atrocious
things Saddam Hussein would have done to our troops, the treatment of Iraqi prisoners was very
mild,” Taking embarrassing photos of Iraqi prisoners is no big deal when you consider the harm
Iraqis have brought to so many people,” “Humiliating Iraqi prisoners is not too serious
considering that they would have killed our soldiers on the battlefield,” and “Compared to the
attacks of Iraqis on American troops, the treatment of the prisoners was not that extreme.” The
items were answered on a 7-point Likert format (1=Completely disagree, 7=Completely agree)
and were averaged to form a scale (α=.85).
Self-Regulatory Mechanisms and War
35
Dependent variables
Pro-U.S. Military Moral Judgments. Group membership and the centrality of in-group
identity may be a common way by which people maintain or increase their self-esteem (Turner,
Hogg, Oakes, Reicher, & Wetherell 1987). Hence, it is logical for people to ascribe favorable
characteristics to the groups to which they belong (Hewstone, et al., 2002). In the extreme, they
may even reward members of their in-group for behavior that may be in-humane. Based on this
reasoning, we posed the following moral judgment to participants after they viewed the Abu
Ghraib prison images in the second task:
Described below is a hypothetical response that a person might take against those
believed to be responsible for abuse against Iraqi prisoners at Abu Ghraib. Assuming you
could successfully execute this response, to what extent do you believe doing so would be
a moral or immoral response? Use the scale below to indicate your opinion.
The judgment we asked participants to make was the morality of rewarding “the American
soldiers with medals for doing an excellent job of protecting our country and keeping us safe
from terrorists.” We asked participants to rate the morality of this response on a 5-point scale
(1=Extremely immoral, 5=Extremely moral).
Pro-US Military Behavior. We examined the extent to which people were likely to
provide financial support to a US oriented charity that specifically aided the US military. For the
final task of the session, participants were handed a manila envelope that contained four 1$ bills,
three envelopes and the following instructions:
In this packet you will find four $1 bills, descriptions of three groups that provide
assistance to others, and three envelopes with the names of these groups printed on them.
What we would like you to do in this task is to help us decide how we should allocate a
monetary donation among these three groups. The donation is real. At the end of this
experiment, we will actually donate money to each of these groups based on how you and
others in this experiment decided to distribute the $4 in this packet among the three
groups.
Self-Regulatory Mechanisms and War
36
You will make your allocation decision by first reading the descriptions of the three
groups and then deciding how much of the $4 you would like us to donate to each group.
You can divide the $4 any way you want among the three groups. This means that you
can decide to give all the money to one group if you wish or you can decide to divide it
more evenly among the three groups. However, you can only make a donation in
increments of $1. This means you cannot donate $1.50 or $1.10 and you will not be able
to donate an equal amount to all three groups.
To indicate how you want to distribute the $4 among the three groups, all you have to do
is take the money and put the amount you want to donate to each group in the envelope
on which that group’s name is printed. You will then place all the materials back into this
packet and return it to the experiment. Again, we will be donating the actual amount of
money that you and others decide to allocate to each of the three groups so it is important
that you take this decision seriously.
When participants flipped the page, they received a description of the three charitable
organizations5 along with a picture of each organization’s logo taken from their respective
websites:
The United Services Organization (USO)
The USO (United Service Organizations) is chartered by Congress as a nonprofit,
charitable corporation. It is not a part of the U.S. Government. The USO is endorsed by
the President of the United States and the Department of Defense.
The USO mission is to provide morale, welfare and recreation-type services to uniformed
military personnel. The original intent of Congress—and enduring style of USO
delivery—is to represent the American people by extending a touch of home to the
military.
The USO relies solely on the generosity of the general public. More than 12,000 members
in the USO international corps of volunteers provide an estimated 450,000 hours of
service annually, a worldwide personal contribution of over $3 million.
Amnesty International
Amnesty International (AI) is a worldwide movement of people who campaign for
internationally recognized human rights.
AI’s vision is of world in which every person enjoys all of the human rights enshrined in
the Universal Declaration of Human Rights and other international human rights
standards.
Self-Regulatory Mechanisms and War
37
In pursuit of this vision, AI’s mission is to undertake research and action focused on
preventing and ending grave abuses of the rights to physical and mental integrity,
freedom of conscience and expression, and freedom from discrimination, within the
context of its work to promote all human rights.
The Global Fund
The Global Fund to Fight AIDS, Tuberculosis and Malaria was created to dramatically
increase resources to fight three of the world's most devastating diseases, and to
direct those resources to areas of greatest need.
As a partnership between governments, civil society, the private sector and affected
communities, the Global Fund represents an innovative approach to international health
financing.
The Global Fund's purpose is to attract, manage and disburse resources to fight AIDS, TB
and malaria. They do not implement programs directly, relying instead on the knowledge
of local experts.
At their own pace, participants were then allowed to privately allocate the four dollar bills by
sealing the money in any of three white envelopes which had the names of each of the three
charities printed on the outside. They then sealed the manila envelope and returned it to the
experimental proctor. Later on the money was taken out and coded to determine how each
participant allocated the money across the three charities and the total monies allocated to each
charity were calculated and donated.
We constructed a behavioral measure that reflected support for the U.S. Armed forces by
comparing how much more money participants allocated to the USO than to either Amnesty
International or the Global fund. We did this by subtracting the amount allocated to each of the
latter two organizations from the amount allocated to the USO. This procedure produced two
dependent measures: USO vs. Amnesty and USO vs. Global Fund. Positive values on each
measure indicated that people preferred to support the USO over the other focal organization.
We calculated a third measure – Amnesty vs. Global Fund – for comparison purposes to see if
Self-Regulatory Mechanisms and War
38
the independent variables predicted the willingness of participants to donate money to an
organization devoted to protecting prisoner rights after witnessing acts of prison abuse.
Results
Descriptive statistics and correlations among the study variables are shown in Table 6.
----------------------------Insert Table 6 here
------------------------------We tested the study hypotheses using regression. Our first hypothesis was that moral
disengagement, operationalized as the use of advantageous comparisons, would be positively
related to the judgment that rewarding the soldiers who abused their Iraqi prisoners was moral.
Our second hypothesis was that this relationship would be weaker among persons whose moral
identity had high rather than low self-importance, but stronger among people whose American
identity had high rather than low self-importance. We tested this second hypothesis by including
a moral disengagement x moral identity interaction and a moral disengagement x American
identity interaction in our model. The variables forming the interaction were centered to
minimize multicollinearity. The results of the regression analysis are shown in Table 7.
-----------------------------Insert Table 7 here
------------------------------Table 7 shows that moral disengagement was positively related to the judgment that
rewarding the abusive soldiers was a moral response (b=.30, p <.001). This supports our first
hypothesis. This effect is qualified, however, by a significant moral disengagement x moral
identity interaction (b=-.14, p < .05). We analyzed the form of the interaction by examining the
relationship between moral disengagement and judgments about rewarding the soldiers for
prisoner abuse for participants who were one standard deviation above and below the mean on
Self-Regulatory Mechanisms and War
39
moral identity. The simple slopes for each of these groups reveals that moral disengagement was
positively related to the morality of rewarding the abusive soldiers for both low (b = .42, p <
.001) and high moral identifiers (b = .18, p < .05). However, the relationship was weaker among
high moral identifiers. This result supports our second hypothesis. The self-importance of
American identity was positively related to the judgment that rewarding the abusive soldiers was
a highly moral response (b= .14, p < .05). But like moral identity, this effect was qualified by a
significant moral disengagement x American identity interaction (b=.09, p < .05). Analysis of
the simple slopes for participants one standard deviation above and below the mean on American
identity showed that moral disengagement was positively related to the morality of rewarding the
abusive soldiers for both high (b = .40, p < .001) and low American identifiers (b = .20, p < .01).
However, as was the case for moral identity, the relationship was stronger among high American
identifiers. This result supports hypothesis 2.
Our third hypothesis was that the judgments about the moral response to rewarding the
abusive soldiers would mediate the relationship between the interactions and behavior indicating
support for the U.S. Armed Forces. Having previously established the relationship between the
independent variables, the moral disengagement and identity interactions, and moral judgment,
we performed the second and third steps of the test for mediation by first regressing the three
comparative allocation variables – USO vs. Amnesty, USO vs. Global Fund, and Amnesty vs.
Global Fund – on the independent variables and the two-interactions. We then added the moral
judgment variable in the second step of the regression. The results of these regressions are
presented in Table 8.
-----------------------Insert Table 8 here
-------------------------
Self-Regulatory Mechanisms and War
40
Table 8 shows that the moral disengagement x moral identity interaction was significant
in the model predicting USO vs. Amnesty (b = -.30, p < .05), and USO vs. Global Fund (b = .34, p < .05), but not in the model predicting Amnesty vs. Global. The moral disengagement x
American identity interaction was not significant in any of the regression models, however.
When moral judgment is added to the equations, the moral disengagement x moral identity
interaction is unaffected in the regression predicting USO vs. Amnesty (b = -.30, p < .05).
Moreover, the direct effect of moral judgment on this measure is not significant (b = -.01, n.s.).
These results fail to support our hypothesis that moral judgment mediates the effect of the moral
disengagement x moral identity interaction on the preference for supporting the USO over
Amnesty. In contrast, the moral disengagement x moral identity interaction becomes smaller (b =
-.26, p < .05) in the regression predicting USO vs. Global Fund and the moral judgment variable
is significant and in the expected direction (b = .52, p < .01). These results indicate that moral
judgment partially mediates the preference for supporting the USO over the Global Fund, which
is consistent with our third hypothesis. Although we found no significant mediated effect of the
moral disengagement x American identity interaction on our behavioral measures, we did find a
significant direct effect of the self-importance of American identity on USO vs. Amnesty
comparative allocation variables (b = .25, p = .05). The direction of the effect indicated that
people whose American identity was highly self-important allocated more money to the USO
than Amnesty.
Discussion of Study 3
This study replicated the interactive effects of moral identity and moral disengagement in
a third inter-group aggression context. Moreover, Study 3 showed that these self-regulating
mechanisms not only predict emotions (Study 1 and 2) and moral judgments (Study 2, and 3) but
Self-Regulatory Mechanisms and War
41
also behaviors towards an organization that benefits members of the national in-group. In line
with the idea that a person’s moral identity reconnects the self to moral consequences on an
emotional, judgmental, and behavioral level, this study also demonstrated the neutralizing effect
of moral identity on the relationship between making advantageous comparisons that justify the
abuse of Iraqi prisoners and beliefs about the morality of rewarding in-group members
responsible for the abuse. Furthermore, we tested the hypothesis that within a person’s working
self-concept, contrasting identities can co-exist and lead people to experience conflicting
judgmental and behavioral reactions to war. The different effects and opposing effects of moral
and American identity on the relationship between advantageous comparisons and moral
judgment supported this hypothesis.
General Discussion
Moral disengagement allows people to disconnect themselves from the negative selfimplications of hurting others thereby making it easier for them to support war. The primary goal
of this research was to examine if two forms of moral disengagement maneuvers – moral
justification and advantageous comparison – become less effective at evoking pro-war emotions,
cognitions, and behaviors when peoples’ moral identity has high rather than low self-importance.
This effect was hypothesized to occur because having a highly self-important moral identity
influences how people define the self-other relation. Cumulatively, our studies show that three
self-regulatory mechanisms —the ability to deactivate one’s inhibitions against harming others
through moral disengagement, the self-importance of moral identity, and the self-importance of
national identity—jointly predict pro- and anti-war reactions. This premise was supported across
all three studies. These findings contribute to an emerging literature on moral disengagement and
support the basic arguments of social cognitive theory. However, the interactions between
Self-Regulatory Mechanisms and War
42
different identities and moral disengagement mechanisms represent a heretofore untested
empirical extension of these literatures.
Theoretical Implications at the Individual Level
Our findings raise a number of questions for researchers interested in how the self and the
identities of which it is comprised regulates emotions, judgment, and behavior, particularly when
people are confronted with the reality of inter-group aggression and violence. Past research has
shown that moral disengagement can motivate and justify aggression (Bandura et al., 1999).
When a person can successfully justify aggression, there are theoretical reasons to believe that
his or her self-conception is less likely to be threatened even if they express support for activities
that can harm or even kill other human beings. However, what we showed is that this effect
might be bounded to the extent that part of this self-conception is organized around moral
identity. What this suggests is that studying disengagement processes without considering the
role of self-definition provides an incomplete picture of what might allow people to engage in inhumane acts.
In this regard, our findings update and qualify Reed and Aquino’s (2003) conclusions.
They found a direct main effect of moral identity such that moral identity was negatively
(positively) correlated with perceptions of morality of killing (forgiving) hostile perpetrators in
war. They also found a direct main effect between moral identity and behavioral preferences
toward monetary donation to out-group charities. The current research qualifies these findings by
suggesting complex self-regulatory mechanisms linking different identities to moral
disengagement. Self regulatory effects of moral identity and national identity appear in these data
to be at least partially predictive of moral judgments about in-group and out-group behavior
Self-Regulatory Mechanisms and War
43
during war, which translate into both emotional and behavioral reactions that reinforce
competing identities.
Although our results advance moral disengagement research within a broader framework
of self-regulation, a question worth pursuing that we cannot address with our data is how shifts
from one self-regulatory mechanism to another take place and what social psychological factors
might motivate this shift. For example, one way a person might execute a regulatory shift from
one mechanism to another would be to alter the hierarchical ordering of the identities in their
working self concept so that one of the mechanisms– presumably the one that brings the least
self-condemnation when one knowingly supports the harming of another person (i.e., moral
identity)—comes to dominate the other and thereby serves as the primary basis for selfregulation (Fiske & Taylor, 1991). Future studies might explore this possibility experimentally
by using primes to either expand or restrict the self-other relation (e.g., Stapel & Koomen, 2001)
and then further manipulate the salience of these primes with situational cues that differentially
increase their cognitive accessibility.
In proposing this line of future investigation, we acknowledge the question of whether
moral identity as measured by Aquino and Reed’s (2000) instrument is distinct from other
measures of identity centrality. Identity theorists have speculated that research should investigate
how identities might functionally substitute for one another (Reid &Deaux, 1996; Deaux et. al
1995). Identities within a common cluster of evaluative meanings (i.e., that map onto similar trait
associations) could be interchangeable, and the satisfactory expression of one identity within the
cluster might reduce the need to express other similar identities (e.g., woman, man, boyfriend,
homemaker, etc.) (Deaux, et. al 1995). Given the nature of our empirical approach to assessing
moral identity, we cannot rule out the possibility that moral identity co-varying with some other
Self-Regulatory Mechanisms and War
44
unmeasured variable or interchangeable identity. For example, Aquino and Reed (2002) argued
that while moral identity may overlap with other identities (e.g., religious identity, ethnic
identity) it is a unique identity and should therefore predict outcomes these other identities do
not. Specifically, the self-importance of moral identity should be more strongly related to
outcomes that have high moral relevance compared to other kinds of identities that have little or
no a priori association with moral behavior. As it stands, the question of whether and how moral
identity relates to the infinite number of other possible identities a person may hold is
unanswered, but theoretically important and deserves further investigation.
The dynamic nature of the self begs an additional question of how a person can manage
the conflict arising from competing self-identities with different behavioral prescriptions. One
theory for explaining such intra-psychic phenomena can be drawn from contemporary models of
dual attitudes. According to these frameworks, people regularly hold more than one attitude
toward important objects and they do not necessarily integrate these into a unified predisposition
(e.g. Cohen & Reed, 2005; Wilson, Hodges & LaFleur 1995; Wilson, Lindsey & Schooler 2000;
Wilson, Kraft & Dunn 1989). The conceptual and empirical justification for the existence of dual
attitudes is linked to the program of research on implicit versus explicit attitudes (see Fazio et al
1986, Greenwald & Banaji 1995 and Greenwald, McGhee & Schwartz 1998). According to these
models, people often have long-standing (implicit) attitudes that are well-practiced and are
typically easy to retrieve. In addition, recently-constructed evaluations (explicit attitudes) often
develop and are essentially isolated rather than integrated. Wilson and colleagues explain the
coexistence of these two attitudes in terms of the greater capacity and motivation needed to
retrieve explicit attitudes, thus making it less likely that people will be forced to confront
attitudinal inconsistency.
Self-Regulatory Mechanisms and War
45
Following the dual-attitude model, it may be that a self-important moral identity may
operate like an implicit attitude. That is, it acts as a general, habitually activated cognition that
automatically influences a persons’ evaluation of situations in which the well-being of other
human beings is at stake. In contrast, the mechanisms of moral disengagement are more flexible
and are activated only under certain conditions, just like explicit attitudes in the dual attitude
model. Thus, when confronted with a situation in which a person finds it necessary or
appropriate to behave in ways that harm others, he or she may be able to consciously execute
moral disengagement maneuvers, but if that person’s moral identity is highly self-important then
its automatic activation may neutralize the effectiveness of moral disengagement. This may
explain why people who executed the same moral disengagement strategies in our study did not
experience similar pro-war reactions depending on the self-importance of their moral identities.
Theoretical Implications at the Group Level
Our research also has implications for theories of inter-group aggression. That moral
identity and moral disengagement act in concert to influence perceptions of group permeability
(Ellemers, Spears & Dosje, 1997) implies that a highly self-important moral identity may
override automatic social categorization processes such that members of socially distant outgroups might still be considered worthy of moral concern. The circle of moral regard hypothesis
and the results from Study 3 extend Opotow’s (1995) speculations about how categorization
processes influence the “scope of justice.” It is not that group membership per se influences a
person’s scope of justice via a simple categorization process. Rather, Study 3 suggests that
persons who morally disengaged were willing to withhold money from in-group members in
favor of an out-group when their moral identity had high self-importance. In contrast, persons
who disengaged and whose moral identity had low self-importance showed an in-group
Self-Regulatory Mechanisms and War
46
favoritism preference by allocating more money to an organization that supports U.S. soldiers
than one that benefits AIDS victims on a global scale.
Our research might also offer theoretical insights into the reasons for and consequences
of subjective beliefs about the natural dominance of in-groups over members of out-groups
(Pratto, Sidanius, Stallworth, & Malle, 1994). If a highly self important moral identity increases
the permeability of psychological group boundaries, then it might also attenuate the effects of
seemingly universal social motivations like the drive for dominance and in-group favoritism.
This implies that when beliefs consistent with dominance orientations are pervasively
transmitted through the public media, a person’s moral identity may dilute conformity or
adherence to these beliefs. The effect of moral identity on the processing of public opinion
campaigns and advertisements is a stream of research worth examining in future research.
The results of our studies also suggest that invoking moral identity through symbolic
representations that make the moral self more salient may help reduce out-group hostility. It
seems natural for many people to sharpen psychological group boundaries by expressing ingroup favoritism (Hewstone, 2000). Past research suggests that a pervasive problem with
cognitively oriented approaches for suppressing stereotypes and prejudice is that their
effectiveness is limited by the availability of cognitive resources for overcoming these
stereotypes (e.g., Fiske & Neuberg, 1990). For example, when persons experience cognitive
overload, they are more likely to apply stereotypic beliefs about out-group members (Wyer,
Sherman & Stroessner, 2000). This line of research has consistently shown how difficult it is to
reduce stereotypic responses towards minorities, members of out-groups, or otherwise
stereotyped groups (Hewstone, 2000). It may be that an alternative way to discourage negative
stereotypes is to make a person’s moral identity more active within the working self-concept.
Self-Regulatory Mechanisms and War
47
Conclusions
Plato reminds us that only the dead have seen the end of war. Thus, the psychological
consequences of war will be with us for a very long time. This research tried to shed new light on
how self-regulatory factors can explain why some people become fervent and uncritical
supporters of war, while others are plagued by feelings of guilt and take a more expansive
accounting of the toll of war on human lives. The study of how people respond to war warrants
the attention of psychologists because while war is humankind’s most destructive activity, it is
also among its most compelling. As the war correspondent Chris Hedges describes it:
The rush of battle is a potent and often lethal addiction, for war is a drug…It is
peddled by mythmakers—historians, war correspondents, filmmakers, novelists,
and the state—all of whom endow it with qualities it often does possess:
excitement, exoticism, power, chances to rise above our small stations in life, and
a bizarre and fantastic universe that has a grotesque and dark beauty (Hedges,
2002: 3).
For this reason, understanding what leads people to become intoxicated by the “drug of war”
may help us discover how the inner voices that impel groups toward war might be momentarily
silenced so that other voices – like those that remind them of the tragic costs of war – might also
have a chance to speak.
Self-Regulatory Mechanisms and War
48
References
Allport, G.L. (1954). The nature of prejudice. New York: Basil Blackwell.
Aiken, L. S. & West, S. G. (1991). Multiple regression: Testing and interpreting interactions,
Sage, Thousand Oaks, CA
Aquino, K. F. & Reed, II, A. (2002) The self-importance of moral identity. Journal of
Personality and Social Psychology, 83, 1423–1440.
Axelrod, R. (1984). The evolution of cooperation. New York: Basic Books.
Bandura, A. (1990). Subjective activation and disengagement of moral control. Journal of Social
Issues, 46, 27-46.
Bandura, A. (1991). Social cognitive theory of moral thought and action. In: W.M. Kurtines &
J.L. Gewirtz (Eds.), Handbook of moral behavior and development: Theory, research and
applications (pp. 71–129). Hillsdale, NJ: Erlbaum.
Bandura, A. (1999). Moral disengagement in the perpetration of inhumanities. Personality and
Social Psychology Review, 3, 193–209.
Bandura, A., Barbaranelli, C., Caprara, G. V. & Pastorelli, C. (1996). Mechanisms of moral
disengagement in the exercise of moral agency. Journal of Personality and Social
Psychology, 71, 364–374.
Bandura, A., Caprara, G. V., Barbaranelli, C., Pastorelli, C. & Regalia, C. (2001). Sociocognitive
self-regulatory mechanisms governing trasngressive behavior. Journal of Personality and
Social Psychology, 80, 125-135.
Baron, R. M. & Kenny, D. A. (1986). The moderator-mediator variable distinction in social
psychological research: Conceptual, strategic, and statistical considerations. Journal of
Personality and Social Psychology, 51, 1173-1182.
Self-Regulatory Mechanisms and War
49
Baumeister, R.F., Stillwell, A.M., & Heatherton, T.F. (1994). Guilt: An interpersonal approach.
Psychological Bulletin, 115, 243-267.
Benedetto, R. (2003). Most back war, but want U.N. support. Retrieved November 30, 2004
from http://www.usatoday.com/news/world/iraq/2003-03-16-poll-iraq_x.htm.
Bendyna, M. E., Finucane, T. & Kirby, L. (1996). Gender differences in public attitudes towards
the Gulf war: A test of competing hypotheses. The Social Science Journal, 33, 1-22.
Berscheid, E., Boye, D. & Walster, E. (1968). Retaliation as a means of restoring equity. Journal
of Personality and Social Psychology, 10, 370-376.
Blasi, A. (1984). Moral identity: Its role in moral functioning. In: W. Kurtines and J. Gewirtz
(Eds.), Morality, moral behavior and moral development (pp. 128–139). Wiley, New
York.
Bolton, L. & Reed, A. II (2004). Sticky priors: The perseverance of identity effects on judgment.
Journal of Marketing Research.
Cervone, D. & Shoda, Y. (1999a). Beyond traits in the study of personality coherence. Current
Directions in Psychological Science, 8, 27-32.
CNN (2003). Poll: Americans less positive on Iraq. Retrieved November 30, 2004 from
http://www.cnn.com/2003/US/06/30/sprj.irq.iraq.poll.
Cohen, J. & Reed, (2005) Accounting for attitude instability: A model of attitude generation and
recruitment. Manuscript under review.
Damon, W. (1984). Self-understanding and moral development from childhood to adolescence.
In W. Kurtines and J. Gewirtz (Eds.), Morality, moral behavior and moral development
(pp. 109-127). New York: Wiley.
Self-Regulatory Mechanisms and War
50
Damon, W., & Hart, D. (1992). Self-understanding and its role in social and moral development.
In M. Bornstein & M. E. Lamb (Eds.), Developmental psychology: An advanced
textbook, (3rd ed., pp. 421-464). Hillsdale, NJ: Lawrence Erlbaum.
Deaux, K., Reid, A., Mizragi K., & Ethier A. K. (1995). Parameters of social identity. Journal of
Personality & Social Psychology, 68(2), 280-291
De Waal, F.B.M. (1991). The chimpanzee’s sense of social regularity and its relation to the
human sense of justice. American Behavioral Scientist, 34, 335-439.
Eidelson, R. J. & Eidelson, J. I. (2003). Dangerous ideas: Five beliefs that propel groups toward
conflict. American Psychologist, 58, 182-192.
Eisenberg, N. (2000). Emotion, regulation, and moral development. Annual Review of
Psychology, 51, 665–697.
Ellemers, N., Spears, R. & Dosje, B. (1997). Sticking together or falling apart: In-group
identification as a psychological group determinant of group commitment versus
individual mobility. Journal of Personality and Social Psychology, 72, 617-626.
Erikson, E.H. (1964). Insight and Responsibility. New York: Norton.
Fazio, R.H., Sanbonmatsu, D. W., Powell, M. C. & Kardes, F. R. (1986). On the automatic
activation of attitudes. Journal of Personality and Social Psychology, 50, 229-238.
Ferguson, T.J. & Stegge, H.(1998). Measuring guilt in children: A rose by any other name still
has thorns. In J. Bybee (Ed.) Guilt and children, pp. 19-74. San Diego: Academic.
Fiske, S. T. & Neuberg, S. L. (1990). A continuum of impression formation from category-based
to individuating processes: Influences of information and motivation on attention and
interpretation. In M. P. Zanna (Ed.), Advances in Experimental Social Psychology (Vol.
23, pp. 1-74). New York: Academic Press.
Self-Regulatory Mechanisms and War
51
Fiske, S. T. & Taylor, S. E. (1991). Social cognition. New York: McGraw-Hill.
Geneva Convention. (1949).
Fite, D., Genest, M., Wilcox, C. (1990). Gender differences in foreign policy attitudes: A
longitudinal analysis. American Politics Quarterly, 18, 492-512.
Ghiglieri, M. P. (1999). The dark side of man: Tracing the origins of male violence. Cambridge,
MA: Perseus Publishing.
Global Security Site Retrieved March 2004 from
http://www.globalsecurity.org/intell/library/reports/2004/800-mp-bde.htm
Gray, J.G. (1998). The warriors: Reflections of men in battle. Lincoln: University of Nebraska
Press.
Greenwald, A. G., & Pratkanis, A. R. (1984). The self. In R. S. Wyer & T. K. Srull (Eds.),
Handbook of social cognition (pp. 129-178). Hillsdale, NJ: Erlbaum.
Greenwald, A. G., & Banaji, M. R. (1995). Implicit social cognition: Attitudes, self-esteem, and
stereotypes. Psychological Review, 102, 4-27.
Greenwald, A.G., McGhee, D.E. & Schwartz, J.L.K. (1998). Measuring individual differences in
implicit cognition: The Implicit Association Test. Journal of Personality and Social
Psychology, 1464–1480.
Grossman, D. (1995). On killing. New York: Little, Brown, and Co.
Haidt, J. (2001). The emotional dog and its rational tail: A social intuitionist approach to moral
judgment. Psychological Review, 108, 814–834.
Hedges, C. (2002). War is a force that gives us meaning. New York: Anchor Books.
Hedges, C. (2003). What every person should know about war. New York: Free Press.
Self-Regulatory Mechanisms and War
52
Hewstone, M. (2000). Contact and categorization: Social psychological interventions to change
intergroup relations. In C. Stangor (Ed.), Stereotypes and prejudice: Essential readings
(pp. 394-418). Philadelphia, PA: Psychology Press.
Hewstone, M., Rubin, M. & Willis, H. (2002). Intergroup bias. Annual Review of Psychology
53, 575–604.
Hinkle, S. & Brown, R. (1990). Intergroup comparisons and social identity: Some links and
lacunae. In D. Abrams & M. A. Hogg (Eds), Social identity theory: Constructive and
critical advances (pp. 48-70).New York: Harvester/Wheatsheaf.
Jacoby, S. (1983). Wild justice: The evolution of revenge. New York: Harper & Row.
Jones, J. M. (2003). Latest update shows no change in support for invasion of Iraq: Vast majority
still say Iraq not complying with United nations requirements. Retrieved October 16,
2004 from http://www.gallup.com/poll/content/login.aspx?ci=7942.
Kelman, H. C. & Hamilton, V. L. (1989). Crimes of obedience: Toward a social psychology of
authority and responsibility. New Haven, CT: Yale University Press.
Lapham, L.H. (2004). R & D. Harper’s, 309, 9-11.
Lapsley, K. & Lasky, B. (2001). Prototypic moral character. Identity, 1, 345–363.
Lelyveld, A. (1971). Punishment: For and against. In A.S. Neill (Ed.), Punishment: For and
against. New York: Hart.
Livingstone, N.C. (1986). Proactive responses to terrorism: Reprisals, preemption, and
retribution. In N.C. Livingstone & T. E. Arnold (Eds.) Fighting back: Winning the war
against terrorism. Lexington, MA: Lexington Books.
Lorenz, K. (1967). On aggression. New York: Bantam Books.
Manchester, W. (1980). Goodbye darkness: A memoir of the Pacific War. New York: Dell.
Self-Regulatory Mechanisms and War
53
Markus, H. & Kunda, Z. (1986). Stability and malleability of the self-concept. Journal of
Personality and Social Psychology, 51 (4), 858-866.
Milgram, S. (1974). Obedience to authority: An experimental view. New York: Harper & Row.
Monroe, K. R. & Epperson, C. (1994). “But what else could I do?” Choice, identity, and
cognitive perceptual theory. Political Psychology, 15, 201–226.
Newport, F. (2001). Public opinion of the war in Afghanistan: 88 % of Americans approve of the
military action. Retrieved October 16, 2004 from
http://www.gallup.com/poll/content/login.aspx?ci=9994.
Nietzsche, F.W. (1887). The genealogy of morals. (translated by P. Watson). London: S.P.C.K.
Oliner, S.P. & Oliner, P. M. (1988). The altruistic personality: Rescuers of the Jews in Nazioccupied Poland. New York: Free Press.
Opotow, S. (1995). Drawing the line: Social categorization, moral exclusion, and the scope of
justice. In J. Z. Rubin & B. B. Bunker (Eds.), Conflict, cooperation, and justice: Essays
inspired by the work of Morton Deutsch (pp. 374-369). San Francisco, CA: Jossey-Bass.
Pratto, F., Sidanius, J., Stallworth, L.M., & Malle, B.F. (1994). Social dominance orientation: A
personality variable predicting social and political attitudes. Journal of Personality and
Social Psychology, 67, 741-763.
Reed, II, A. & Aquino, K. (2003). Moral identity and the expanding circle of moral regard
towards out-groups. Journal of Personality and Social Psychology, 84, 1270-1286.
Reid, A., & Deaux, K. (1996). Relationship Between Social and Personal Identities: Segregation
or Integration? Journal of Personality and Social Psychology, 71, 1084-1091.
Sahlins, M. (1965). On the sociology of primitive exchange. In B. Banton (Ed.) The relevance of
models for social anthropology. Association of Social Anthropologists, Monograph 1.
Self-Regulatory Mechanisms and War
54
London: Tavistock. (Reprinted in M. Sahlins (1972). Stone Age Economics. New York:
Aldine).
Selman, R. L. (1980). The growth of interpersonal understanding. New York: Academic Press.
Sivard, R.L. (1991). World military and social expenditures 1991. Washington, D.C: World
Priorities.
Solomon, R. (1990). A passion for justice: Emotions and the origins of the social contract.
Reading, MA: Addison-Wesley.
Silk, J.B. (1992). The patterning of intervention among male bonnett macaques: reciprocity,
revenge, and loyalty. Current Anthropology, 33, 318-324.
Stapel, D. A. & Koomen, W. (2001). I, we, and the effects of others on me: How self-construal
moderates social comparison effects. Journal of Personality and Social Psychology, 80,
766-781.
Staub, E. (1989) The roots of evil: The origins of genocide and other group violence. Cambridge:
Cambridge University Press.
Stryker, S. (1980). Symbolic interactionism: A social structural version. Menlo Park, CA:
Benjamin/Cummings.
Tajfel, H. & Turner, J. C. (1979). An integrative theory of intergroup conflict. In W.G. Austin &
S. Worchel (Eds.), The social psychology of intergroup relations (pp. 33–47). Monterey,
CA: Brooks-Cole.
Tangney, J. P. (1991). Moral affect: the good, the bad, and the ugly. Journal of Personality and
Social Psychology, 61, 598-607.
Taylor, S. E. & Brown, J. (1988). Illusion and well-being: A social psychological perspective on
mental health. Psychological Bulletin, 103, 193–210.
Self-Regulatory Mechanisms and War
55
Tripp, T. M. & Bies, R. J. (1997). What’s good about revenge? The avenger’s perspective. In R.
J. Bies & R. J. Lewicki (Eds.), Research on negotiation in organizations (Vol. 6, pp. 145160).
Tripp, T. M., Bies, R. J. & Aquino, K. (2002). Poetic justice or petty jealousy? The aesthetics of
revenge. Organizational Behavior and Human Decision Processes, 89, 966-984.
Tooby, J. & Cosmides, L. (1990). The past explains the present: Adaptations and the structure of
ancestral environments. Ethology and sociobiology, 11, 375-424.
Turner, C., Hogg, M. A., Oakes, P. J., Reicher, S.D. & Wetherell, M.S. (1987). Rediscovering
the social group: A self-categorization theory. Oxford, England: Blackwell.
Wilson, T. D., Hodges, S. D. & LaFleur, S. J. (1995). Effects of introspecting about reasonsInferring attitudes from accessible thoughts. Journal of Personality and Social
Psychology, 69, 16-28.
Wilson, T.D. & Lindsey, S. & Schooler, T.T. (2000). A model of dual attitudes. Psychological
Review, 107, 101–126.
Wilson, T. D., Kraft, D. & Dunn, D. S. (1989). The disruptive effects of explaining attitudes-The
moderating effect of knowledge about the attitude object. Journal of Experimental Social
Psychology, 25, 379-400.
Wyer, N. A., Sherman, J. W. & Stroessner, S. J. (2000). The roles of motivation and ability in
controlling the consequences of stereotype suppression. Personality and Social
Psychology Bulletin, 26, 13-25
Younis, J. & Yates, M. (1999). Youth service and moral-civic identity: A case of everyday
morality. Educational Psychology Review, 11, 361–376.
Zimbardo, P. (1972). Pathology of imprisonment. Society, 9, 4-8.
Self-Regulatory Mechanisms and War
56
Table 1: Descriptive Statistics and Correlations for Study 1 Variables
SD
1.57
1
.79
2
3
4
1
Anti-War Emotions
Mean
2.67
2
Male
1.59
.49
-.04
3
Order
.49
.50
-.01
-.07
4
Moral Identity
6.46
.55
-.05
.19
-.16
.79
5
Moral Disengagement
4.22
1.34
-.53
-.06
.07
-.02
5
.86
Note: (N = 106). All correlations greater than .16 are significant at p < .05. Alpha reliabilities are
shown along the diagonal.
Self-Regulatory Mechanisms and War
57
Table 2: Regression of Anti-War Emotions on Moral Identity, Moral Disengagement, and their
Interaction
Anti-War Emotions
Anti-War Emotions
Step 1
Step 2
Variables
1. Male
B
-.17
t
-.63
B
-.25
t
-.92
2. Order
.06
.22
.06
.25
3. Moral Identity
-.15
-.59
-.10
-.42
4. Moral Disengagement
-.63
-6.40***
-.56
-5.33***
.33
1.89†
5. Moral Identity X
Moral Disengagement
2
.29
.02†
.29
.31
∆R
2
R
10.38***
F
Note: Entries are unstandardized regression weights.
†p < .10, * p<.05, **p<.01, ***p <.001.
9.23***
Self-Regulatory Mechanisms and War
58
Table 3. Descriptive Statistics and Correlations Among Study 2 Variables
Mean
2.12
SD
1.02
1
.88
2
Moral Judgment
.13
.34
.51
--
3
Female
.60
.49
.06
-.08
--
4
Moral Identity
4.50
.49
-.32
-.41
.14
.85
5
Moral Disengagement
2.61
.79
.60
.41
-.15
-.29
1
Pro-War Emotions
2
3
4
5
.75
Note: (N=104). All correlations greater than .29 are significant at p < .01. Alpha reliabilities are
shown along the diagonal.
Self-Regulatory Mechanisms and War 59
Table 4: Regression of Moral Judgment on Moral Identity, Moral Disengagement, and Their
Interaction
Moral Judgment
Variables
B
t
1. Female
.02
.43
2. Moral Identity
-.03
-.45
3. Moral Disengagement
.17
4.64***
4. Moral Identity X
Moral Disengagement
-.31
-4.59***
R2
.39
F
15.95***
Note: Entries are unstandardized regression weights. * p<.05, **p<.01, ***p <.001
Self-Regulatory Mechanisms and War 60
Table 5: Regression of Pro-War Emotions on the Independent Variables and Moral Judgment
Pro-War Emotions
Pro-War Emotions
Step 1
Step 2
Variables
1. Female
B
.38
t
2.43*
B
.37
t
2.38*
2. Moral Identity
-.07
-.34
-.05
-.24
3. Moral Disengagement
.80
7.67***
.68
6.10***
4. Moral Identity X
Moral Disengagement
-.52
-2.68**
-.31
-1.49
5. Moral Judgment
.67
2.40*
∆ R2
.03*
R2
.45
.48
F
20.55***
18.38***
Note: Entries are unstandardized regression weights. * p<.05, **p<.01, ***p <.001
Self-Regulatory Mechanisms and War 61
Table 6. Descriptive Statistics and Correlations Among Study 3 Variables
SD
1.65
1
--
2
3
4
1
Favor USO over Amnesty
2
Favor USO over Global Fund
-1.13
1.74
.54
--
3
Favor Amnesty over Global Fund -.57
1.62
-.44
.52
--
4
Moral Identity
6.22
.85
-.17
-.14
.02
.85
5
American Identity
3.66
1.05
.18
.11
-.06
.21
.88
6
Moral Disengagement
2.45
1.44
.38
.30
-.06
-.16
.17
.85
7
Reward (Moral Judgment)
1.41
.93
.27
.40
.15
-.23
.19
.56
Note: N=138. All correlations greater than .16 are significant at (p < .05) .Reliabilities for
each scale are reported on the diagonal.
5
6
Mean
-.57
7
--
Self-Regulatory Mechanisms and War 62
Table 7: Regression of Rewarding Soldiers on Moral Identity, Moral Disengagement, and
Their Interaction
Moral Judgment
Variables
1.Moral Identity
B
-.17
t
-2.18*
2. American Identity
.14
2.14*
3. Moral Disengagement
.30
6.40***
4. Moral Identity X
Moral Disengagement
-.14
-2.39*
5. American Identity X
Moral Disengagement
.09
2.11*
R2
.38
F
16.36***
Note: Entries are unstandardized regression weights. * p<.05, **p<.01, ***p <.001
Self-Regulatory Mechanisms and War 63
Table 8: Regression of Allocation Decisions on Moral Judgment, Moral Identity, and Moral Disengagement
USO vs. Amnesty
Step 1
USO vs. Global
Step 2
Step 1
Amnesty vs. Global
Step 2
Step 1
Step 2
Variables
1. Moral Identity
B
-.23
t
-1.42
B
-.23
t
-1.40
B
-.17
t
-.95
B
-.08
t
-.44
B
.06
t
.33
B
.15
t
.86
2. American Identity
.25
1.95
.25
1.92
.15
1.10
.08
.60
-.09
-.67
-.17
-1.19
3. Moral
Disengagement
.37
3.95*** .38
3.46**
.31
3.00**
.15
1.33
-.06
-.59
-.22
-1.89
4. Moral Identity X
Moral Disengagement
-.30
-2.48*
-.30
-2.43*
-.34
-2.55*
-.26
-1.99*
-.04
-.29
.04
.28
5. American Identity X
Moral Disengagement
.02
.17
.02
.18
.05
.53
.003
.03
.04
.37
-.01
-.14
-.01
-.04
.52
2.78**
.53
2.80**
6. Moral Judgment
∆R
R2
2
.22
.00
.15
.05
.01
.06
.22
.22
.15
.20
.01
.07
7.30***
6.03***
4.61**
5.32***
.24
1.51
F
Note: Entries are unstandardized regression weights. * p<.05, **p<.01, ***p <.001
Self-Regulatory Mechanisms and War 64
Endnotes
1
By emphasizing the self as a regulator of ethical behavior, the social-cognitive approach
does not argue that moral reasoning and cognitive capabilities are unimportant. The ability to
conceptualize a consistent set of traits that determine a stable sense of self probably does
require a certain level of cognitive sophistication and an understanding of personality (cf.
Selman, 1980). However, it seems unlikely that the sample to which we want to generalize
our particular set of studies lacks this capacity, unless it is a sample representative of very
young children or perhaps the mentally disabled. Consequently, we are not arguing here that
cognitive ability is completely unimportant in shaping an identity, but rather that it is
relatively less important in the moral identity framework than in frameworks that have
traditionally assumed more rationalistic explanations of unethical behavior.
2
This fundamental postulate of identity theory (Erikson, 1964; Hinkle and Brown, 1990;
Stryker, 1980) assumes that people are motivated to think, feel, and act in ways that reinforce
and are consistent with their self-definitions.
3
This does not mean that high moral identifiers will necessarily object to war, or even refuse
to fight in a war. Rather, it simply means that their responses to war will be relatively less
positively valenced than those of low moral identifiers. For example, we can imagine
someone who supports war as a practical and perhaps even a moral necessity, but who does
so with a heavy heart and with feelings of sympathy and compassion towards those who
suffer as a result. It is also possible to imagine a supporter of war who expresses contempt for
the enemy and feels little or no emotional disturbance upon witnessing the suffering of those
outside his or her national in-group.
Self-Regulatory Mechanisms and War 65
4
Some theorists have argued that guilt can be theoretically distinguished from shame in the
sense that guilt is more strongly linked to a specific action, while shame may pertain more
generally to evaluations of the self (e.g., Baumeister, Stillwell & Heatherton, 1994; Tangney,
1991). In our data, they were highly correlated so we collapsed them into a single measure of
anti-war emotions. When examined separately, they produce identical results.
5
One of three orders (USO, Amnesty, Global; Amnesty, Global, USO or Global, USO,
Amnesty) was randomly given to participants as the second page of the final task.