Deliver Us from Evil: The Effects of Mortality Salience and Reminders of 9/11 on Support for President George W. Bush

Abstract

According to terror management theory, heightened concerns about mortality should intensify the appeal of charismatic leaders. To assess this idea, we investigated how thoughts about death and the 9/11 terrorist attacks influence Americans' attitudes toward current U.S. President George W. Bush. Study 1 found that reminding people of their own mortality (mortality salience) increased support for Bush and his counterterrorism policies. Study 2 demonstrated that subliminal exposure to 9/11-related stimuli brought death-related thoughts closer to consciousness. Study 3 showed that reminders of both mortality and 9/11 increased support for Bush. In Study 4, mortality salience led participants to become more favorable toward Bush and voting for him in the upcoming election but less favorable toward Presidential candidate John Kerry and voting for him. Discussion focused on the role of terror management processes in allegiance to charismatic leaders and political decision making.

Full text

http://psp.sagepub.com
Personality and Social Psychology Bulletin
DOI: 10.1177/0146167204267988
2004; 30; 1136 Pers Soc Psychol Bull
Daniel M. Ogilvie and Alison Cook
Mark J. Landau, Sheldon Solomon, Jeff Greenberg, Florette Cohen, Tom Pyszczynski, Jamie Arndt, Claude H. Miller,
George W. Bush
Deliver us from Evil: The Effects of Mortality Salience and Reminders of 9/11 on Support for President
http://psp.sagepub.com/cgi/content/abstract/30/9/1136
The online version of this article can be found at:
Published by:
http://www.sagepublications.com
On behalf of:
Society for Personality and Social Psychology, Inc.
can be found at:Personality and Social Psychology Bulletin Additional services and information for
http://psp.sagepub.com/cgi/alerts Email Alerts:
http://psp.sagepub.com/subscriptions Subscriptions:
http://www.sagepub.com/journalsReprints.navReprints:
http://www.sagepub.com/journalsPermissions.navPermissions:
http://psp.sagepub.com/cgi/content/refs/30/9/1136 Citations
at University of Missouri-Columbia on March 31, 2009 http://psp.sagepub.comDownloaded from

10.1177/0146167204267988PERSONALITY AND SOCIAL PSYCHOLOGY BULLETINLandau et al. / TERROR MANAGEMENT AND SUPPORT FOR G. W. BUSH
Deliver Us From Evil:
The Effects of Mortality Salience and Reminders
of 9/11 on Support for President George W. Bush
Mark J. Landau
University of Arizona
Sheldon Solomon
Skidmore College
Jeff Greenberg
University of Arizona
Florette Cohen
Rutgers University
Tom Pyszczynski
University of Colorado, Colorado Springs
Jamie Arndt
University of Missouri
Claude H. Miller
University of Oklahoma
Daniel M. Ogilvie
Rutgers University
Alison Cook
University of Missouri
According to terror management theory, heightened concerns
about mortality should intensify the appeal of charismatic lead-
ers. To assess this idea, we investigated how thoughts about
death and the 9/11 terrorist attacks influence Americans’ atti-
tudes toward current U.S. President George W. Bush. Study 1
found that reminding people of their own mortality (mortality
salience) increased support for Bush and his counterterrorism
policies. Study 2 demonstrated that subliminal exposure to 9/
11-related stimuli brought death-related thoughts closer to con-
sciousness. Study 3 showed that reminders of both mortality and
9/11 increased support for Bush. In Study 4, mortality salience
led participants to become more favorable toward Bush and vot-
ing for him in the upcoming election but less favorable toward
Presidential candidate John Kerry and voting for him. Discus-
sion focused on the role of terror management processes in
allegiance to charismatic leaders and political decision making.
Keywords: terror management theory; terrorism; 9/11; George W.
Bush; election politics; charismatic leaders
It is [fear] that makes people so willing to follow brash,
strong-looking demagogues with tight jaws and loud
voices: those who focus their measured words and their
sharpened eyes in the intensity of hate, and so seem most
capable of cleansing the world of the vague, the weak,
1136
Authors’ Note: We thank the reviewers for their helpful comments.
Correspondence concerning this article should be addressed to Mark
J. Landau, University of Arizona, Department of Psychology, P.O. Box
210068, Tucson, AZ 85721-0068;e-mail: mjlandau@email.arizona.edu.
PSPB, Vol. 30 No. 9, September 2004 1136-1150
DOI: 10.1177/0146167204267988
© 2004 by the Society for Personality and Social Psychology, Inc.
at University of Missouri-Columbia on March 31, 2009 http://psp.sagepub.comDownloaded from

the uncertain, the evil. Ah, to give oneself over to their
direction—what calm, what relief.
—Ernest Becker,
The Birth and Death of Meaning (1971, p. 161)
Starting with Sigmund Freud’s (1921/1965) sugges-
tion that leaders serve as substitute parent figures, psy-
chologists of diverse theoretical persuasions have
argued that the popularity of leaders depends, at least in
part, on the extent to which they meet the pressing psy-
chological needs of their followers (e.g., Becker, 1973;
Bord, 1975; Ehrhart & Klein, 2001; Fromm, 1941;
Kirkpatrick & Locke, 1996; Pillai, 1996; Redl, 1942). Ter-
ror management theory (TMT) (Greenberg, Pyszczynski,
& Solomon, 1986; Solomon, Greenberg, & Pyszczynski,
1991) posits that one of the most basic functions that
leaders serve is that of helping people manage a deeply
rooted fear of death that is inherent in the human condi-
tion. Initial empirical support for this point was provided
by Cohen, Solomon, Maxfield, Pyszczynski, and
Greenberg’s (in press) finding that reminders of mortal-
ity increase support for charismatic leaders in a hypo-
thetical election scenario. This analysis implies that
when reminders of one’s vulnerability and mortality are
highly salient, support for such leaders is likely to
increase.
The terrorist attacks on the World Trade Center and
Pentagon on September 11, 2001, seem highly likely to
have dramatically increased the salience of such death-
related concerns for most of the American people
(Pyszczynski, Solomon, & Greenberg, 2003). As one
might predict from the terror management perspective,
the popularity of the then and current American presi-
dent, George W. Bush, increased dramatically in the days
after the 9/11 terrorist attacks and has remained rela-
tively high well into 2004. In the present article, we use
the popularity of President Bush as a context for an
“experimental case study” of the role of existential fear
in promoting support for government leaders. Specifi-
cally, we examine (a) the effect of reminders of mortality
on support for President Bush, (b) the link between
9/11-related stimuli and the accessibility of death-
related thoughts (which have been shown to mediate the
effect of mortality salience on the pursuit of faith in one’s
cultural worldview; for a review, see Pyszczynski,
Greenberg, & Solomon, 1999), and (c) the impact of
reminders of 9/11 on the popularity of President Bush.
TERROR MANAGEMENT THEORY
Building on the writings of Ernest Becker (1971,
1973, 1975), TMT begins with the assumption that
humans share with all animals a fundamental orienta
-
tion toward continued survival, along with many of the
same biological and psychological systems designed for
self-preservation. Humans are unique, however, in their
capacity for self-consciousness, symbolic thought, and
imagining possible future events. Despite their adaptive
value, these cognitive abilities render humans aware that
their own death is inevitable and possible at any
moment, a recognition that conflicts with the biological
propensity for continued existence and therefore gives
rise to the potential for debilitating anxiety. To manage
the potential for anxiety that this awareness produces,
people deny that physical death implies absolute annihi-
lation by maintaining faith in a personalized version of a
cultural worldview: a set of humanly constructed, cultur-
ally derived, and socially validated beliefs about the
nature of reality that provides meaning and the promise
of literal or symbolic immortality to those who uphold
culturally prescribed standards of value. Faith in cultural
meaning and the perception of oneself as an object of
value within that scheme provide a protective shield
against the potential for anxiety that results from one’s
awareness of the inevitability of death.
The most prominent line of empirical support for
TMT comes from tests of the mortality salience (MS)
hypothesis, which states that if the cultural worldview
functions to provide protection against death-related
concerns, then reminders of death should intensify
efforts to bolster and defend faith in the worldview. This
broad hypothesis has been supported by a wide range of
studies demonstrating the many ways in which MS
increases defense of one’s worldview (for reviews of
empirical support for TMT, see Greenberg, Solomon, &
Pyszczynski, 1997; Solomon, Greenberg, & Pyszczynski,
in press). This body of research has operationalized MS
in a variety of ways and has included control inductions
that prime aversive topics other than death (e.g., physi-
cal pain, social rejection, uncertainty) and that consis-
tently fail to produce effects parallel to MS on the pri-
mary measures of worldview defense. Research also has
shown that effects parallel to MS are not produced by
heightened self-awareness, the salience of cultural val-
ues, meaninglessness, or high cognitive load (e.g.,
Greenberg et al., 1995). In addition, internal analyses
consistently reveal that terror management defenses are
not mediated by the participant’s current emotional
state. This large body of evidence thus strongly suggests
that MS effects result specifically from activating death-
related cognitions.
TMT posits that terror management defenses are ulti-
mately concerned with the implicit knowledge of death
rather than with consciously experienced terror per se
(see Greenberg et al., 2003). Based on a large body of evi-
dence, Pyszczynski et al. (1999) proposed a dual process
model of the cognitive processes through which
Landau et al. / TERROR MANAGEMENT AND SUPPORT FOR G. W. BUSH 1137
at University of Missouri-Columbia on March 31, 2009 http://psp.sagepub.comDownloaded from

thoughts of death affect behavior. This model posits that
conscious contemplation of mortality first arouses direct
threat-focused proximal defenses involving suppression
of death-related thoughts or pushing the problem of
death into the distant future by denying one’s vulnerabil-
ity to various risk factors (e.g., promising to get more
exercise). Once death-related thought is no longer in
consciousness, distal symbolic terror management
defenses, which serve to bolster faith in the meaningful
worldview and one’s sense of self-worth, are activated to
manage the potential for anxiety engendered by height-
ened accessibility of implicit death-related thought.
Once these defenses have been employed, death-
thought accessibility dissipates back to baseline level. In
sum, whereas more straightforward proximal defenses
against death serve to push death out of awareness, it is
the sustained perception of oneself as a person of value
in a world of meaning that allows people to avert the
potential for anxiety that results from the increased
accessibility of death-related thought (for recent reviews
of research supporting this model, see Arndt, Cook, &
Routledge, in press; Solomon et al., in press).
THE POPULARITY OF PRESIDENT GEORGE W. BUSH
Prior to the September 11, 2001, terrorist attacks on
the World Trade Center and Pentagon, President Bush’s
popularity among the American people was tenuous. He
had lost the popular vote in the 2000 election and won
the presidency after a narrow victory in the Electoral Col-
lege that was ultimately decided by the Supreme Court
after a highly controversial near-draw in the critical elec-
toral state of Florida. A collection of national public
opinion polls by PollingReport.com (2004; including
Fox News, CNN/USA Today/Gallup Polls, and ABC
News/Washington Post Polls) indicate that President
Bush’s approval ratings hovered around 50% in the
weeks preceding the terrorist attacks. A front page arti-
cle in The New York Times (Berke & Sanger, 2001) on Sep-
tember 9, 2001, reported intense efforts by the White
House staff to increase the president’s popularity in the
face of a sagging economy and critical evaluations of his
style and strategy by leading Republican legislators. All
of this changed virtually overnight, with polls indicating
an unprecedented 88% to 90% approval rating as early
as September 13, 2001 (Morin & Deane, 2001;
PollingReport.com, 2004). Opinion polls also showed
overwhelming support for Bush’s handling of the terror-
ist crisis, the restriction of civil liberties, the military
action against the Taliban in Afghanistan, and at least ini-
tially, the preemptive war on Iraq (Dworkin, 2002;
Etzioni, 2002; Huddy, Khatid, & Capelos, 2002; Langer,
2001; Moore, 2001; Toner & Elder, 2001).
This is not to say that Bush’s actions and policies have
not been subjected to intense scrutiny, criticism, and
skepticism in some quarters (e.g., Alterman & Green,
2004; Anonymous, 2004; Clarke, 2004; Dean, 2004;
Frank, 2004; Humberman, 2004; Phillips, 2004; Risen,
2004; Suskind, 2004; Unger, 2004; Waldman, 2004;
Woodward, 2004), but Bush’s popularity seemed to with-
stand such attacks well into early 2004, the final year of
his current term as President, despite a variety of poten-
tially damaging allegations that had come to light since
his decision to proceed with a preemptive war on Iraq,
including the reliance on faulty intelligence informa-
tion regarding links of Iraq to al Qaeda and the 9/11
bombings, the presence of weapons of mass destruction,
claims that initial discussion of plans for war on Iraq
appeared to have been held soon after 9/11, and pro-
tracted fighting that lasted long after the war was
declared won.
1
TMT AND SUPPORT FOR PRESIDENT BUSH
How might we understand the psychological forces
bearing on the confidence and trust people are willing
to place in their national leaders, especially in times of
crisis? TMT posits that popular support for leaders is
partly the result of the need to allay a deeply rooted fear
of death. According to TMT, people are motivated to
conceive of themselves as valued participants in a cosmi-
cally significant cultural scheme rather than physical
creatures subject to decay and death. Investing faith in
the leader of that scheme to assure symbolic prosperity
facilitates the maintenance of this buffer. The appeal of
the leader lies in his or her perceived ability to both liter-
ally and symbolically deliver the people from illness,
calamity, chaos, and death as well as to demonstrate the
supremacy of the worldview. Building on the ideas of
Freud, Ferenzci, Rank, Redl, and Fromm, Becker (1971,
1973, 1975) proposed that over the course of socializa-
tion, the primary sources of psychological security, those
entities that sustain the sense of life as meaningful and
one’s self as significant, shift from one’s parents to the
culture and its figures of power, authority, and righteous-
ness. As a result of this transference process, Becker
(1975) viewed secular leaders as representing not only
rational, objective policy makers but, similar to their reli-
gious counterparts, assuming the role of “death-dealer
and death-defier” (p. 42)—the embodiments of
invulnerability and symbolic supremacy (see also Lifton,
1968).
From this perspective, President Bush’s appeal may
lie in his image as a protective shield against death,
armed with high-tech weaponry, patriotic rhetoric, and
the resolute invocation of doing God’s will to “rid the
world of evil” (Purdum, 2001, p. 2; see also Carney &
Dickerson, 2001). Indeed, Woodward (2004) describes
how when asked if he seeks his father’s advice on Iraq,
the president replied, “You know, he is the wrong father
1138 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN
at University of Missouri-Columbia on March 31, 2009 http://psp.sagepub.comDownloaded from

to appeal to in terms of strength. There is a higher father
that I appeal to” (p. 421). President Bush may provide a
sense of protection from the overwhelming terrorist
threat, both literally by ensuring physical safety and sym
-
bolically by acting on America’s behalf in a heroic tri-
umph over evil (symbolized by Bush landing aboard an
aircraft carrier in a Navy jet to declare the war in Iraq to
be functionally over and by the dramatic toppling of
Saddam Hussein’s statue in Baghdad). In sum, TMT sug-
gests that people avoid a potentially debilitating preoc-
cupation with personal frailty and finitude, in part, by
transferring power to and investing faith in a powerful
authority. Such a central locus of control over life and
death appears to assure indefinite perpetuation by
assuming mastery over nature and tragedy and uphold-
ing the cultural meaning system that imbues individual
lives with transcendent meaning, order, and
permanence.
If loyalty to leaders stems in part from terror manage-
ment concerns, then reminders of mortality should
increase Americans’ support for President Bush. A num-
ber of previous findings are generally consistent with this
hypothesis. A substantial body of research indicates that
MS engenders inflated regard for a wide array of people,
concepts, and objects that represent the culture to which
the participants subscribe (reviewed by Greenberg et al.,
1997; Solomon et al., in press). For example, Greenberg
et al. (1990) found that, compared to those primed with
another aversive outcome, death-primed individuals
expressed especially favorable evaluations of essays and
their authors who praised the United States and espe-
cially negative evaluations of anti-U.S. essays and their
authors. MS also has been shown to incite aggressive
behavior against those who impinge on one’s worldview:
McGregor et al. (1998) found that death-primed partici-
pants administered excessive amounts of an aversively
spicy hot sauce to a target who verbally attacked their
political orientation. Research by Greenberg, Simon,
Porteus, Pyszczynski, and Solomon (1995) showed that
following MS, American participants became more
reluctant and uncomfortable when treating an Ameri-
can flag inappropriately. Finally, Arndt, Greenberg, and
Cook (2002) found that subliminal death primes (and
explicit MS treatments after a delay) can increase the
accessibility of nationalistic cognitions. Taken together,
these findings provide convergent support for the role
of intimations of mortality in people’s allegiance to and
defense of the nationalistic aspects of their cultural
worldviews.
Of most direct relevance for present purposes, Cohen
et al. (in press) recently demonstrated that MS
enhanced the appeal of a charismatic leader who pro
-
motes a grand vision and promises citizens a significant
role in a noble mission in a hypothetical election sce
-
nario. More specifically, MS increased preference for a
hypothetical political candidate portrayed as having
charismatic qualities but not for ones portrayed as task
oriented or relationship oriented. The studies reported
here seek to extend these findings by providing an
“experimental case study” of the role of death-related
concerns in general, and 9/11-related concerns in par-
ticular, in promoting support for President George W.
Bush.
STUDY 1
Study 1 was designed to test the hypothesis that to the
extent that support for President Bush derives in part
from terror management needs, then MS should
increase support for Bush and his policies. To test this
hypothesis, we primed American participants with
thoughts of either death or a control topic and then mea-
sured their support for President Bush and his
antiterrorism policies.
Method
PARTICIPANTS
Ninety-seven undergraduates at Rutgers University
(65 women, 32 men) volunteered to participate in
return for extra credit in their psychology class in Octo-
ber 2003.
MATERIALS AND PROCEDURE
The study was run in a single session in a psychology
class. The experiment was described as a short study of
the relationship between personality attributes and
opinions about social issues. Each participant was given a
questionnaire packet and was asked to complete each
question in the booklet in the order in which it
appeared. The MS manipulation followed two filler
questionnaires included to sustain the cover story and
obscure the true purpose of the study. In the MS condi-
tion, participants responded to two open-ended ques-
tions (used in previous TMT studies, e.g., Greenberg
et al., 1990; Rosenblatt, Greenberg, Solomon,
Pyszczynski, & Lyon, 1989): “Please briefly describe the
emotions that the thought of your own death arouses in
you” and “Jot down, as specifically as you can, what you
think will happen to you as you physically die and once
you are physically dead.” Control participants
responded to parallel questions about watching televi-
sion. All participants then completed a self-report mood
scale (PANAS-X; Watson & Clark, 1991) to assess possible
affective consequences of the MS induction and read a
short literary passage to serve as a delay because previous
research (Greenberg, Pyszczynski, Solomon, Simon, &
Breus, 1994) has shown that MS increases worldview
defense most consistently when there is a delay between
the MS induction and dependent variable assessment.
Landau et al. / TERROR MANAGEMENT AND SUPPORT FOR G. W. BUSH 1139
at University of Missouri-Columbia on March 31, 2009 http://psp.sagepub.comDownloaded from

Participants then read the following essay expressing
a highly favorable opinion of the measures taken by Pres-
ident Bush with regard to 9/11 and the Iraqi conflict:
It is essential that our citizens band together and support
the President of the United States in his efforts to secure
our great Nation against the dangers of terrorism. Per-
sonally, I endorse the actions of President Bush and the
members of his administration who have taken bold
action in Iraq. I appreciate our President’s wisdom
regarding the need to remove Saddam Hussein from
power and his Homeland Security Policy is a source of
great comfort to me. It annoys me when I hear other peo-
ple complain that President Bush is using his war against
terrorism as a cover for instituting policies that, in the
long run, will be detrimental to this country. We need to
stand behind our President and not be distracted by citi-
zens who are less than patriotic. Ever since the attack on
our country on September 11, 2001, Mr. Bush has been a
source of strength and inspiration to us all. God bless
him and God bless America.
They were then asked to respond to three questions: “To
what extent do you endorse this statement?” “I share
many of the attitudes expressed in the above statement,”
and “Personally, I feel secure knowing that the President
is doing everything possible to guard against any further
attacks against the United States.” All responses were
made on 5-point scales (1 = strongly agree, 5 = strongly dis-
agree). These responses were then reversed-scored so
that higher numbers were indicative of greater support
for the president.
Results and Discussion
Support for the president. The responses on the three
questions demonstrated good internal reliability (α =
.94) so they were combined to yield a composite agree-
ment score, which was submitted to a 2 (MS vs. control) ×
2 (gender) ANOVA. There was no main effect or interac-
tion for gender, ps > .2. However, the predicted main
effect for MS was obtained, F(1, 93) = 112.48, p < .001;
those in the MS condition reported higher support for
President Bush (M = 4.16, SE = .145) than those in the
control prime condition (M = 2.09, SE = .131). Analyses
conducted on the individual items making up our com-
posite measure revealed significant effects on all three
items, all ps < .01, with MS increasing approval of Bush
and his policies on each item. It is noteworthy that the
mean in the control group is on the disagreement side of
the scale’s midpoint, whereas the mean in the MS condi-
tion is on the agreement side. Also, the effect size is large,
η
2
= .55.
Affect. To assess whether MS affected mood, we per-
formed a MANOVA and ANOVAs on the subscales of the
PANAS-X (Watson & Clark, 1991) and ANOVAs on the
aggregate positive and negative affect scales as well.
2
Consistent with previous TMT research demonstrating
that MS does not engender affect, there were no signifi-
cant differences found for these analyses. To ensure that
the MS effect reported above was not mediated by affect,
we conducted an analysis of covariance (ANCOVA) with
the affect subscale scores (including positive and nega-
tive affect) as covariates and the effect of MS remained
statistically intact. Thus, we are quite confident that, as in
past research, this finding is not caused by affective dif-
ferences between MS and TV salience conditions.
9/11 and Support for Bush
In his 2004 reelection effort, President Bush sought to
establish his effectiveness in combating terrorism and
his ability to protect people from the alleged threats
associated with terrorism. Political advertisements in
support of the Bush campaign have featured images of
9/11 and its aftermath and emphasized the President’s
ability to preserve national security (Rutenberg &
Sander, 2004). Because the prosecution of the war on
terror has become a touchstone of Bush’s reelection
campaign, one might wonder whether references to 9/
11 are effective in increasing support for President Bush
and, if so, why.
Understanding these reactions requires a theoretical
framework from which to assess the psychological signifi-
cance of terrorism-related events. In this vein,
Pyszczynski et al. (2003) recently portrayed the events of
9/11 as a “natural” MS induction and discussed how
many of the symbolic defenses prompted by 9/11 (e.g.,
veneration for culturally sacred objects, such as flags)
paralleled those found in response to MS inductions
used in terror management research. Based on these
affinities, Pyszczynski et al. proposed that symbolic
responses to 9/11 are the product of a potent double-
barreled threat posed by the events: vivid and unceasing
depictions of death and destruction compounded by a
massive threat to America’s sense of military and finan-
cial power and moral righteousness. In concert, these
threats should encourage bolstering of Americans’ faith
in the symbolic structures that constitute the dominant
cultural worldview. From this perspective, many sym-
bolic defenses in response to 9/11 represent efforts to
reaffirm the integrity and absolute validity of the Ameri-
can worldview in response to increased accessibility of
implicit death-related thoughts (see also Miller & Lan-
dau, in press). However, we know of no direct experi-
mental evidence that terrorism-related events parallel
the effect of MS primes in increasing death-thought
accessibility or that 9/11-related thoughts play a role in
increased attraction to a leader. In Study 2, we therefore
1140 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN
at University of Missouri-Columbia on March 31, 2009 http://psp.sagepub.comDownloaded from

sought to discover whether 9/11-related thought func
-
tions in the same way as death-related thought in produc-
ing heightened implicit death accessibility. Then, in
Study 3, we determined if reminders of 9/11 function
like MS primes in increasing support for President Bush.
STUDY 2
Support for the dual process model has been
obtained from studies showing that MS inductions that
completely by-pass the conscious consideration of death
can influence symbolic defenses (Arndt, Greenberg,
Pyszczynski, & Solomon, 1997). More specifically, Arndt
et al. demonstrated that presenting death-related words
beneath conscious awareness led to an immediate
increase in death-thought accessibility relative to neutral
or negative control words. Furthermore, subliminal
death stimuli led to an immediate increase in worldview
defense (increased preference for a pro-American essay
and its author as opposed to an anti-American author),
whereas previous research has shown that supraliminal
death stimuli produce increased death accessibility and
worldview defense only after a delay and distraction. In
short, the increased accessibility of death-related
thought outside of focal attention—the condition
shown to engender symbolic defenses—occurs at an
unconscious level.
Other research has found that threats to terror man-
agement defenses increase death-thought accessibility,
whereas strengthening those defenses tends to reduce it
(Arndt, Greenberg, Solomon, Pyszczynski, & Simon,
1997). For example, increases in death-thought accessi-
bility have been shown to follow explicit reminders of
one’s physical nature (Goldenberg, Cox, Pyszczynski,
Greenberg, & Solomon, 2002), salient threats to a close
relationship (Mikulincer, Florian, & Hirschberger,
2003), and confrontations with events that violate the
belief in a just world (Landau et al., in press). Based on
these lines of research, we predicted that to the extent
that 9/11 functions like an MS prime in activating
unconscious concerns about mortality, subliminal expo-
sure to 9/11-related stimuli should result in an increase
in the accessibility of death-related thoughts. This would
establish the cognitive link between 9/11 and death-
thought accessibility assumed by our analysis of the
effects of 9/11.
Method
PARTICIPANTS AND DESIGN
Participants were 52 introductory psychology stu-
dents at the University of Missouri–Columbia who were
randomly assigned to receive one of three subliminal
primes: 911, WTC, or 573. This study was conducted
approximately 1 month after the 9/11 attacks. The
dependent variable was the accessibility of death-related
words. Six participants were excluded (4 who did not
complete the questionnaires correctly; 2 due to errors
with their materials). Therefore, a total of 46 partici-
pants (26 women, 20 men) were included in the
analyses.
PROCEDURE
Upon participants’ arrival, the experimenter intro-
duced himself and described the experiment as explor-
ing the ability to perceive word relationships and person-
ality traits. Participants were told that they would
complete some personality questionnaires, a computer
task, and then some additional word perception mea-
sures. Participants were assured of their anonymity and
then read and signed a consent form before the
procedures commenced.
After the participants completed some filler personal-
ity questionnaires (which served to maintain the cover
story), they were escorted into smaller cubicles that con-
tained the computers. The experimenter then gave the
instructions for the word-relation task, which required
participants to decide as quickly as possible whether two
words that flashed sequentially on the computer were
related or unrelated by pressing the right shift key or the
left shift key, respectively. For example, if the words flower
and rose were presented, they should press the right shift
key to indicate that they are related, but if the words
sneaker and fajita were presented, they should press the
left shifted key to indicate that they are not. The experi-
menter then turned off the room light to reduce the
glare on the screen. The appropriate keys and the light
switch were illuminated via glow-in-the-dark stickers.
When participants were finished with the computer
task, they were told to turn on the light and complete the
packet of word perception measures lying on a clipboard
next to the computer. The questionnaires included a
death-theme-accessibility measure followed by some
additional word puzzles and then some manipulation
check questions to assess participants’ awareness of the
stimuli display. Participants were told to put the com-
pleted questionnaire in a blank envelope provided and
then put the envelope in the box located under the desk.
When participants were finished with both of these tasks,
they were instructed to crack the door of the cubicle.
Then, participants were probed for suspicion, fully
debriefed, and thanked for their time.
MATERIALS
Apparatus and program for stimulus presentation. Stimuli
were presented on a 15-in. Gateway color monitor con-
trolled by an IBM-compatible computer. The task was
Landau et al. / TERROR MANAGEMENT AND SUPPORT FOR G. W. BUSH 1141
at University of Missouri-Columbia on March 31, 2009 http://psp.sagepub.comDownloaded from

presented using DMASTR display software developed at
Monash University and at the University of Arizona by
K. I. Forster and J. C. Forster. The program synchronizes
the timing of the display and uses normal, bit-mapped
fonts. The first few frames presented instructions and
three practice stimuli centered on the screen. There
were then 10 trials sequentially presenting three words
centered on the screen. The first and third words were
the target words for which participants were supposed to
determine the presence or absence of a relationship.
Actually, these two words served as a forward mask (and
fixation point) and backward mask, respectively, and
were displayed for 356 ms. The critical subliminal prime,
either 911, WTC, or 573 depending on the condition,
was presented between the two masked words for 28.5
ms. Both 911 (referring to the date) and WTC (World
Trade Center) were chosen as the terrorism primes due
to their widespread association with the terrorist attacks
on the World Trade Center on September 11, 2001.
Because 911 is also associated with emergencies, we also
included WTC as a prime, which presumably had little
meaning until its increased media usage following the
terrorist attacks. The control prime, 573, was chosen
because of its familiarity (as the area code for Columbia,
Missouri) and its number of characters matched that of
the other two primes. Which version of the program was
presented was set by another research assistant to keep
the experimenter blind to conditions.
Accessibility. A word fragment completion task, similar
to tasks used by other researchers (e.g., Gilbert & Hixon,
1991; Greenberg et al., 1994), was introduced as a mea-
sure “being pre-tested for future studies” but was actually
used to assess the accessibility of death-related themes.
The measure consisted of 34 word fragments that partici-
pants were instructed to complete with the first word that
came to mind. Six of the 34 fragments could be com-
pleted with either a neutral or death-related word. For
example, the fragment COFF_ _ could be completed as
COFFEE (a neutral word) or COFFIN (a death-related
word). The possible death-related words were buried,
skull, murder, stiff, coffin, and grave.
Results and Discussion
Checks on awareness of subliminal stimuli. To assess par-
ticipants’ awareness of the subliminal stimuli, we exam-
ined their responses to the five questions presented at
the conclusion of the session. In response to the first
question—How many words did you see in each display
(each trial for which you were to make relational judg-
ments)?—Forty of the 46 participants indicated that they
saw two words in each trial display. With the next ques
-
tion—Did you ever see more than two words flash at a
time—only 1 participant indicated that he may have
seen more than two words. Pearson chi-square tests
indicated that there were no differences between prime
conditions, χ
2
(2, N = 46) < 2.10, p > .34. For the questions
that asked if the word was the same or different and
asked participants to list all possibilities, all of the partici
-
pants left them blank. A Pearson chi-square test con-
ducted on the multiple-choice question—Assuming that
there was something flashed between the two target
words, which of the following do you think it may have
been?—revealed no differences between conditions,
χ
2
(10, N = 46) = 11.42, p > .32. Thus, as in previous
research using this manipulation (e.g., Arndt et al.,
1997a, 2002), there was no conscious retrospective
awareness of the prime.
Accessibility of death-related thoughts. A one-way ANOVA
was conducted on the accessibility of death thoughts, as
indicated by the number of death-related word frag-
ments, and showed a marginal difference between
prime groups, F(2, 43) = 3.03, p < .06, η
2
= .12. Of course,
our prediction corresponded to a 2 versus 1 contrast
rather than a linear trend (e.g., Rosenthal, Rosnow, &
Rubin, 2000). Thus, we conducted two orthogonal con-
trasts. The first answers the following question: Does
being subliminally primed with symbols related to the
recent terrorist attacks increase death-thought accessi-
bility compared to a control condition? This planned
contrast confirmed our hypothesis, revealing that partic-
ipants in the terrorism prime conditions showed greater
death-thought accessibility (911 = 2.07, SD = .80; WTC =
1.80, SD = 1.15) than those in the control condition (573 =
1.25, SD = .86), t(43) = 2.34, p < .03. The second contrast
found that although the mean was a bit higher in the 911
condition than in the WTC condition, this difference
was not statistically significant, p > .44.
Study 2 demonstrated that stimuli commonly associ-
ated with the 9/11 attacks (911 and WTC) produce an
increase in death-thought accessibility, much like previ-
ous research has shown subliminal death-related stimuli
do (Arndt et al., 1997a). This establishes the cognitive
linkage between the 9/11 terrorist attacks and death-
thought accessibility that is central to Pyszczynski et al.’s
(2003) analysis of how Americans reacted to these
attacks and that we posit to underlie the effect of remind-
ers of the 9/11 attacks on Americans’ approval of
President Bush and his policies.
STUDY 3
Having established that cognitions related to 9/11
function much like MS primes in increasing death-
thought accessibility, we next examined whether
reminders of 9/11 were functionally equivalent to MS
primes in increasing support for President Bush. To test
this hypothesis, we primed participants with thoughts of
1142 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN
at University of Missouri-Columbia on March 31, 2009 http://psp.sagepub.comDownloaded from

death, 9/11, or an aversive control topic and then mea
-
sured their attitudes toward Bush.
Because a nonaversive control prime (television
salience) was used in Study 1, the possibility remains that
the findings of that study were the result of thinking
about negative events in general and not necessarily
death, per se. To address this alternative, Study 3 used a
control prime likely to be aversive to students: thoughts
of an upcoming exam. Indeed, Greenberg et al. (1995)
found that this prime produced more negative affect
than MS. We also wanted to assess possible effects of
reminders of mortality and 9/11 on general political ori-
entation. From our perspective, the increase in
favorability toward Bush reflects the effects of death
reminders on the appeal of a leader who promotes secu-
rity and the vanquishing of evil, but an alternative possi-
bility is that reminders of death or 9/11 simply make
people more politically conservative, which in turn
makes Bush more appealing. Thus, we included a simple
measure assessing participants’ self-reported position
along a continuum from very conservative to very liberal.
Method
PARTICIPANTS
Participants were 74 (46 women, 28 men) Rutgers
undergraduates who volunteered to be in the study for
extra credit in February 2004.
MATERIALS AND PROCEDURE
The experiment was run in a classroom setting. The
procedure was virtually identical to Study 1: Participants
were told that we were interested in the relationship
between personality attributes and opinions on social
issues. After completing filler questionnaires to sustain
the cover story, participants were randomly assigned to
an MS, exam salience, or terrorism prime condition. MS
participants completed the typical two open-ended ques-
tions about death; exam salience participants completed
parallel questions about an upcoming exam; terrorism
salience participants were asked to “Please describe the
emotions that the thought of the terrorist attacks on Sep-
tember 11, 2001, arouses in you” and “Write down as spe-
cifically as you can what happened during the terrorist
attacks on September 11, 2001.” After completing the
PANAS-X to assess the affective consequences of the MS
and terrorism salience inductions and reading the delay
paragraph, participants were asked to read the same
paragraph praising George Bush used in Study 1 and to
answer the same three questions about it. These items
demonstrated good internal reliability (α = .89), so we
formed a composite index indicative of support for the
president. Finally, after completing some filler demo-
graphic questions, participants were asked to indicate
their political orientation on a scale from 1 (very conserva-
tive) to 9 (very liberal).
Results and Discussion
Support for the president. There was no effect of, or inter-
action with, gender, Fs < 1. A one-way ANOVA (priming
condition) on the liking for Bush composite revealed a
significant effect, F(2, 71) = 43.12, p < .001 (see relevant
means in Figure 1). A 2 versus 1 contrast confirmed our
hypothesis, revealing that participants in MS and 9/11
salience conditions showed greater support for Bush
and his policies compared to those in the control condi-
tion, t(72) = 9.31, p < .001. A second contrast found that
although the mean was a bit higher in the MS condition,
this difference was not statistically significant, p > .8. As in
Study 1, exam salient control participants were on the
disagreement side of the scale, whereas MS and 9/11-
salient participants were on the agreement side of the
scale; there was also a large effect size, η
2
= .55. These
results thus replicate those of Study 1 and show that
reminders of 9/11 have the same effect as MS in increas-
ing the appeal of President Bush. As in Study 1, analyses
of the individual items revealed significant effects on all
three items, all ps < .01, with identical patterns of
statistically significant differences among means.
Affect. To assess whether MS affected mood, we per-
formed a MANOVA and ANOVAs on the subscales of the
PANAS-X (Watson & Clark, 1991) and ANOVAs on posi-
tive affect and negative affect. Consistent with previous
TMT research demonstrating that MS does not engen
-
der affect, there were no significant differences found
for these analyses.
Landau et al. / TERROR MANAGEMENT AND SUPPORT FOR G. W. BUSH 1143
2
2.5
3
3.5
4
Exam Mortality Terrorism
Support for President Bush
Figure 1 Study 3: Support for George W. Bush as a function of prim-
ing condition.
at University of Missouri-Columbia on March 31, 2009 http://psp.sagepub.comDownloaded from

Political orientation. A one-way ANOVA revealed no
effect of the MS or 9/11 prime on political orientation,
F < 1, suggesting that these inductions enhanced affec
-
tion for President Bush without altering political orien
-
tation per se. Overall, the sample was slightly liberal, with
a mean of 5.8. To assess whether the priming effects were
not exclusive to participants with a conservative political
orientation, we conducted a regression in which the
Bush approval composite was the dependent variable
and prime condition, political orientation, and their
interaction served as predictors. We observed the
expected main effects for condition and political orien-
tation, indicating higher approval in the MS and 9/11
conditions and the more conservative the participants
were.
We also found an unanticipated significant interac-
tion, β = .74, SE = .07, t = 1.96, p = .05. We plotted the inter-
action in Figure 2 using 1 standard deviation above (lib-
eral) and below (conservative) the mean of political
orientation (Aiken & West, 1991). The simple slopes for
political orientation within each level of the priming con-
dition are as follows: exam salience control, β = –.48, t =
–4.11, p < .001; mortality salience, β = –.28, t = –1.53, p =
.13; and 9/11 salience, β = –.06, t = –.32, p = .75. To test
whether these slopes were statistically different in pair-
wise fashion, we conducted three separate regressions to
examine how political orientation interacts with exam
versus mortality, exam versus 9/11, and mortality versus
9/11. The only significant interaction to emerge was the
Political Orientation × Exam vs. 9/11 Salience, t = 2.53, p
< .02 (for MS vs. 9/11, t = 1.28, p > .2; exam vs. MS, t =
1.01, p > .29). These results in conjunction with Figure 2
suggest that MS increased approval of Bush similarly for
liberals and conservatives but that the terrorism prime
had a stronger effect on liberals, such that in that condi-
tion, political orientation was a negligible predictor, and
significantly less predictive of approval of Bush than it
was in the exam control condition.
These results support the hypotheses that MS and a
reminder of 9/11 would both increase the appeal of
President Bush regardless of political orientation. The
interaction suggests that 9/11 does so a bit less for the
more conservative participants. We cannot offer a defini-
tive interpretation for this unexpected pattern, but if it is
not spurious, it may reflect the nature of the sample. The
conservative group actually included 27 participants
who circled 5, the midpoint of the scale. It may be that
for politically middle-of-the-road people, certain aspects
of Bush’s handling of 9/11 were not appealing. More
research with a broader sampling of political orienta-
tions is needed to explore this further, but it is important
to keep in mind that both MS and 9/11 significantly
increased the appeal of President Bush for our
conservative and liberal participants.
STUDY 4
We have proposed that through his charismatic lead-
ership style, President Bush represents a protective
authority capable of assuaging existential concerns.
Although the results of Studies 1 and 3 are consistent
with this hypothesis, it is possible that the observed
results are due to increased affection for anyone in a
leadership position or with potential for U.S. leadership
rather than Bush himself. Although Bush’s current posi-
tion as president and his other attributes make it impos-
sible to find a leader for comparison who only differs on
one dimension, because of the upcoming election as of
May 2004 when we conducted this last study, we felt that
Democratic presidential nominee Senator John Kerry
would be the best comparison leader. Thus, participants
in the following study were asked to rate either President
Bush or the current Democratic presidential nominee
John Kerry.
Also, because in Studies 1 and 3 we operationalized
support for Bush as agreement with a paragraph praising
Bush and his policies, it is possible that MS influenced
reactions to particular statements in the paragraph (e.g.,
those that affirmed cultural or moral values), which then
led to the increased favorability toward Bush. This
operationalization also prevents us from ruling out the
possibility that MS increased agreeability to any elabo-
rate, persuasive message rather than a pro-Bush message
per se.
3
We address these alternatives in Study 4 by elimi-
nating the essay and using straightforward questions
designed to more directly assess support for Bush. More-
over, we included a question that allows us to directly
compare the effect of MS on the likelihood of voting for
either President Bush or presidential candidate Kerry in
the upcoming presidential election. To further assess the
1144 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN
1
1.5
2
2.5
3
3.5
4
4.5
Exam Mortality Terrorism
Priming Condition
Support for President Bush
Conservative
Liberal
Figure 2 Study 3: Support for George W. Bush as a function of prim-
ing condition and political orientation.
at University of Missouri-Columbia on March 31, 2009 http://psp.sagepub.comDownloaded from

specificity to concerns about mortality, in this study, we
used a different aversive control topic, intense pain.
Method
PARTICIPANTS AND DESIGN
One hundred and fifty-seven students at Brooklyn
College (95 women, 62 men) were randomly assigned to
conditions in a 2 (mortality salient vs. intense pain
salient control) × 2 (evaluate George W. Bush vs. evalu-
ate John Kerry) design. Participants completed the
experimental materials individually on May 13, 2004.
MATERIALS AND PROCEDURE
The experimenter approached individuals in the col-
lege cafeteria and asked them to participate in a short
study of personality attributes and social judgments.
After giving verbal consent, each participant was given a
questionnaire packet and asked to complete each ques-
tion in the booklet in the order in which it appeared.
The packet began with two filler questionnaires to sus-
tain the cover story and obscure the true purpose of the
study, followed by the manipulation of MS. In the MS
condition, participants responded to the two open-
ended questions used in Studies 1 and 3; in the intense
pain condition, participants responded to two parallel
questions: “Please describe the emotions that the
thought of being in intense pain arouses in you” and
“Write down as specifically as you can what you think will
happen to you physically as you are in intense pain.” All
participants then completed a self-report mood scale
(PANAS-X; Watson & Clark, 1991) to assess possible
affective consequences of the MS induction and read a
short literary passage to serve as a delay.
The next page of the questionnaire booklet was titled
“Opinion Survey.” In the evaluate George W. Bush con-
dition, participants were instructed to “Think for a
moment about President George W. Bush and then
answer the following questions by circling the number
that best approximates your feelings.” Four questions
followed: “How favorably do you view George W. Bush?”
“To what extent do you admire George W. Bush?” “To
what extent do you have confidence in George W. Bush
as a leader?” and “If you vote in the upcoming presiden-
tial election, how likely is it you will vote for George W.
Bush?” In the evaluate John Kerry condition, partici-
pants read identical instructions and responded to iden-
tical questions about presidential candidate (rather than
President) John Kerry. The questions were followed by 9-
point scales with endpoints marked not at all favorably
and extremely favorably for the first question and not at all
and very much for the remaining three questions. Finally,
after completing some filler demographic questions,
participants were asked to indicate their political orien
-
tation on a scale from 1 (very conservative) to 9 (very
liberal).
Results
Opinion survey. The four questions on the opinion sur-
vey demonstrated good internal reliability (α = .94) so we
formed a composite index indicative of support for
either Bush or Kerry. We then subjected the composite
index scores to a 2 (MS vs. intense pain) × 2 (evaluate
Bush vs. evaluate Kerry) × 2 (gender) ANOVA, which
revealed main effects for mortality salience versus in-
tense pain, F
(1, 149)
= 8.52, p = .004, η
2
= .054, and evaluating
Bush versus Kerry, F
(1, 149)
= 8.48, p = .004, η
2
= .054, quali-
fied by an interaction between these factors, F
(1, 149)
=
64.00, p < .001, η
2
= .30 (see relevant means in Figure 3).
The main effects were due to participants in the MS con-
dition giving higher ratings (M = 4.68) to either candi-
date than those in the intense pain control condition
(M = 3.83) and higher ratings of John Kerry (M = 4.67)
than George Bush (M = 3.83). More important, however,
an examination of the interaction revealed that
although John Kerry was significantly more highly
regarded than George Bush in the intense pain control
condition, t(75) = 7.50, p < .001, George Bush’s evalua-
tions increased in response to MS (across the midline of
the scale), t(76) = 7.87, p < .001, whereas John Kerry’s
evaluations declined, t(77) = 3.34, p = .001, such that
Bush was evaluated significantly more positively than
Kerry when mortality was salient, t(78) = 3.69, p < .001.
Analyses conducted on the individual items making up
our composite measure revealed significant effects on all
items, including the voting decision item, all ps < .001,
with identical patterns of statistically significant differ-
ences among means.
Landau et al. / TERROR MANAGEMENT AND SUPPORT FOR G. W. BUSH 1145
0
1
2
3
4
5
6
Pain Mortality
Support for Leader
John Kerry
George W. Bush
Figure 3 Study 4: Support for President George W. Bush and
presidential candidate John Kerry as a function of priming
condition.
at University of Missouri-Columbia on March 31, 2009 http://psp.sagepub.comDownloaded from

Affect. To assess whether MS affected mood, we per
-
formed a MANOVA and ANOVAs on the subscales of the
PANAS-X (Watson & Clark, 1991) and ANOVAs on the
aggregate positive and negative affect scales as well.
These analyses revealed only two univariate main effects.
Compared to the pain salience prime, MS led to less
attentiveness (p = .05), less positive mood (p < .03), and
less sadness (p < .05). To ensure that the MS effect
reported above was not mediated by affect, we con-
ducted separate analyses of covariance (ANCOVAs) with
attentiveness, sadness, and positive affect entered sepa-
rately as covariates. The results of the primary analyses
remained statistically unaffected. Thus, we are quite con-
fident that this finding is not caused by affective differ-
ences between MS and pain salience conditions.
Political orientation. Overall, the sample was moder-
ately liberal, with a mean of 5.8. A 2 (MS vs. pain) × 2
(Bush vs. Kerry) × 2 (gender) ANOVA revealed no
effects on political orientation, suggesting that the MS
induction influenced evaluations of Bush and Kerry
without altering political orientation per se. As in Study
3, we assessed whether the priming effects were exclusive
to participants with a conservative political orientation
by regressing the approval composite on priming condi-
tion, leader condition, political orientation, and their
interactions. We observed main effects for all variables,
the predicted Prime × Leader interaction, and an unsur-
prising Political Orientation × Leader interaction (ps<
.05), but no hint of a three-way interaction (p > .46).
These results indicate that MS intensified support for
President Bush and reduced support for Kerry
regardless of political orientation.
Discussion
The results of Study 4 demonstrate that compared to
an intense pain prime, MS increased support for Presi-
dent Bush. In contrast, MS significantly reduced support
for presidential candidate John Kerry. These results sug-
gest that MS does not heighten affection for anyone asso-
ciated with U.S. leadership, but Bush in particular. These
results are consistent with Cohen et al.’s (in press) find-
ing that MS heightens the appeal of a leader with a char-
ismatic style and go further in demonstrating that MS
can significantly reduce the appeal of some (potential)
leaders. The Cohen et al. study found that MS reduced
attraction to a hypothetical candidate with an egalitar-
ian, relationship-oriented style; perhaps the reduced
appeal of Kerry in this study reflects a perception of him
as having such a style. Beyond comparing Bush to Kerry,
the use of straightforward approval questions rather
than agreement with an elaborate message provides
direct evidence of the effect of MS on favorability of Pres
-
ident Bush. Of importance, these results go beyond
mere approval and reveal that in a complete reversal
from the control condition, mortality-salient partici-
pants actually favored Bush in the upcoming presiden-
tial election.
GENERAL DISCUSSION
Study 1 explored the role of mortality concerns in
American popular support for President Bush and
found that MS increased agreement with a favorable
assessment of Bush and his policies. Because issues of
national security are a substantial component of the
Bush administration’s campaign strategy, we then
explored the psychological impact of reminders of 9/11.
In Study 2, we found that paralleling the effects of sub-
liminal death primes reported in previous research, sub-
liminal presentations of 9/11-related stimuli (9/11 and
WTC) increased death-thought accessibility. Study 3
extended these findings by demonstrating that making
9/11 salient was functionally equivalent to MS in increas-
ing support for President Bush. Study 4 provided more
direct evidence that MS intensifies support for President
Bush but found that it decreased liking for presidential
candidate John Kerry. Study 4 also showed that MS influ-
ences actual voting intentions. In all three studies, MS
increased evaluations of Bush from below the scale mid-
point to above it. In addition, Study 4 revealed that
whereas Kerry was preferred over Bush in the control
condition, mortality salient participants preferred Bush
over Kerry.
The present findings support the views of many theo-
rists (e.g., Becker; Freud, 1921/1965; Fromm, 1941;
Lifton, 1968; Lipman-Bluemen, 1996; Weber) and
researchers (Bord, 1975; Ehrhart & Klein, 2001; Kirk-
patrick & Locke, 1996; Pillai, 1996) who have noted that
political allegiances are not always based on the bal-
anced, rational forces of self-interest suggested by the
Jeffersonian notion of democracy but also on the opera-
tion of nonrational forces of which we are not always
aware. From the perspective of TMT, it is the need to
manage concerns about personal mortality that lead
people to cling to the protection provided by their lead-
ers. By showing that reminders of both death and the
9/11 terrorist attacks increase support for President
Bush and that even very subtle reminders of 9/11 in-
crease the accessibility of death-related thoughts, the
present research provides clear support for this TMT
analysis.
The political orientation measure in Studies 3 and 4
suggests that the increased appeal of President Bush in
response to reminders of death or the events of 9/11 was
not at all limited to conservative individuals and was not
the result of an increase in political conservatism. These
null findings regarding political orientation must, of
course, be interpreted with caution and could reflect
insensitivity of the one-item, albeit face valid, measure.
1146 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN
at University of Missouri-Columbia on March 31, 2009 http://psp.sagepub.comDownloaded from

However, it is consistent with earlier findings. Greenberg
et al. (1992) and McGregor et al. (1998) manipulated
MS in conservative and liberal participants and in nei
-
ther case did the dependent measures suggest a shift in
political orientation of the liberals toward conservatism.
In addition, in Cohen et al. (in press), the descriptions of
the hypothetical candidates did not imply anything
about their political orientation, making it unlikely that
the MS-induced increase in preference for the charis-
matic candidate reflected a conservative shift. These
points notwithstanding, further research on the effects
of MS and 9/11 reminders on political orientation as
well as the appeal of leaders is certainly warranted.
The present findings raise a variety of intriguing ques-
tions for future research. The fact that MS decreased the
appeal of Kerry in Study 4 suggests that there is some-
thing specific about President George W. Bush that
makes him especially useful for terror management pur-
poses. Still, President Bush is a multidimensional stimu-
lus so it is difficult to be definitive about precisely what
makes him more appealing to Americans after a
reminder of death. However, related research points to a
few likely factors. One is his status as the president of the
country. This position makes him the person most repre-
sentative of the United States at the current time. If MS
increases pro-U.S. sentiment in Americans, as a consid-
erable body of evidence indicates, positive responses
toward the president also are likely to be increased. Con-
sistent with this idea that the sitting president will be
more appealing when concerns about death are afoot,
there have been numerous instances of increased presi-
dential popularity at times when the nation is facing a
serious threat. According to Gallup (http://www.gallup.
com/), Franklin D. Roosevelt saw his approval ratings
surge 12 points following the Pearl Harbor bombings,
John F. Kennedy’s approval rose 13 points following the
Cuban missile crisis, and George H. W. Bush (the first
Bush administration) saw a dramatic 18-point surge
following the start of the Gulf war.
Second, President Bush has certain elements of a
charismatic style: He appears highly self-confident and
certain of his views (Feldmann, 2004), appeals to patrio-
tism, and emphasizes the positive qualities of America
and being American and the central role of the United
States in triumphing over evil and defending freedom.
Cohen et al. (in press) found that MS increases prefer-
ence for a hypothetical candidate who embodies pre-
cisely this style, and one could argue that the earlier lead-
ers FDR, JFK, and even Bush senior exemplified certain
elements of this style. A third possible factor is Bush’s
advocacy of strong security and aggressive military mea
-
sures. Although it would be useful to determine the rela
-
tive contributions of these different facets of Bush, this
would be a difficult task because Bush and his positions
are unique. Although in Study 4 we compared Bush to
John Kerry, Kerry’s qualities are less clear, and he seems
to lack all three of those elements. In contrast, a past
leader such as Reagan had all three qualities but differs
in no longer being influential. Indeed, we would suggest
that these three attributes tend to covary in political lead-
ers and act in concert to contribute to who President
Bush is as a public figure.
Although the present findings are consistent with a
TMT analysis, could they also be interpreted as stem-
ming from the more general phenomenon of increased
ingroup cohesion and favoritism in the face of a shared
external threat? More specifically, research both in
(Gaertner & Dovidio, 2000) and outside the laboratory
(Piliavin, Rodin, & Piliavin, 1969; Sherif, 1966; Sherif,
Harvey, White, Hood, & Sherif, 1961) has shown that fac-
ing or even anticipating (Darley & Morris, 1975) a com-
mon enemy or threat can activate superordinate identi-
ties and increase ingroup solidarity. Based on this
research, perhaps reminders of 9/11 and the threat of
terrorism united people in a common cause and a singu-
larly integrated identity, leading them to “rally ‘round
the flag” and support their current leader. Although this
analysis resembles our terror management analysis in
some respects, there are a number of reasons why we
believe that a common enemy explanation cannot ade-
quately account for the current findings. First, applying
this alternative to the MS effects observed in Studies 1, 3,
and 4 requires positing that MS spontaneously activated
cognitions associated with 9/11 and terrorism. Although
possible, this is unlikely because, according to two raters
unaware of the purpose, none of the completions of the
open-ended items about death in any of the three studies
mentioned 9/11, terrorism, war, or violence-related
words or themes. Second, all participants in Studies 1
and 3 were exposed to a paragraph that made explicit
mention of 9/11 and future terrorist threats, so the com-
mon enemy theme was prominent in all conditions of
both studies, yet MS participants showed the highest sup-
port for Bush. Third, in Study 4, we found the same effect
of MS with no mention of 9/11 or terrorism at all. Finally,
in the typical common enemy scenario, the similarity of
circumstance heightens group cohesion even among
ingroup members who previously held conflicting
beliefs. However, multiple studies have demonstrated
that MS has the opposite effect, increasing disdain for or
distancing from those with differing values and beliefs
(e.g., Arndt, Greenberg, Schimel, Pyszczynski, & Solo-
mon, 2002; Florian & Mikulincer, 1997; Greenberg et al.,
1990; Rosenblatt et al., 1989). Although it is possible that
MS and 9/11 salience activated two distinct processes,
heightened accessibility of death-related thought pro
-
vides the most parsimonious account of the full set of
findings. Although the present results are unlikely to
Landau et al. / TERROR MANAGEMENT AND SUPPORT FOR G. W. BUSH 1147
at University of Missouri-Columbia on March 31, 2009 http://psp.sagepub.comDownloaded from

reflect a simple act of “rallying ’round the flag,” one
could view the salience of a common enemy or threat as
one circumstance likely to increase the accessibility of
death-related thought; in this light, TMT may help
explain why external threats intensify ingroup favoritism
and unanimity.
Political Implications
No matter how bad Bush does on the war and 9/11, just
having voters think about it kills us.
—Democratic strategist, quoted in Time,
May 3, 2004, p. 32
The present results clearly show that President Bush’s
popularity is increased when thoughts of death or terror-
ism are especially salient, and this is particularly relevant
to ongoing campaign strategies as the 2004 presidential
election approaches and for future political campaign
strategies as well. Indeed, the results of Study 4 clearly
show that mortality-salient participants were more
inclined to reelect President Bush as opposed to presi-
dential candidate Kerry. The fact that reminders of
death and the events of 9/11 enhanced support for Pres-
ident Bush in the present studies may not bode well for
the philosophical democratic ideal that political prefer-
ences are the result of rational choice based on an
informed understanding of the relevant issues. If the
effect of MS on attraction to leaders is indeed rooted in
the largely irrational symbolic protection that they pro-
vide, the best antidote to this problem may be to take
great pains to encourage people to vote with their
“heads” rather than their “hearts,” because past research
(Simon et al., 1997) has demonstrated that MS effects
are attenuated by instructions to think rationally. Of
course, in these frightening times, when the media is rife
with images of death and the threat of terrorist acts is
increasingly imminent, rationally driven decisions may
be unlikely. But perhaps the knowledge of how concerns
about death influence human behavior can promote
campaign strategies and electoral choices based on the
political issues and qualifications of the candidates
rather than based on rhetoric primarily serving
defensive needs to preserve psychological equanimity in
the face of death.
NOTES
1. Only in mid-May 2004, in light of the global scandal regarding
American treatment of Iraqi detainees, escalating gas prices, and no
end of the fighting in Iraq in sight, has Bush’s approval finally dipped
below pre-9/11 levels.
2. Due to an error, only 32 of the 60 items from the PANAS-X were
included in this questionnaire. Consequently, affect subscales were
constructed from these items. This was rectified for Studies 3 and 4
with the same results as obtained in this study.
3. We thank an anonymous reviewer for pointing this out.
REFERENCES
Aiken, L. S., & West, S. G. (1991). Multiple regressions: Testing and inter
-
preting interactions. Newbury Park, CA: Sage.
Alterman, E., & Green, M. (2004). The book on Bush: How George W.
(mis)leads America. New York: Viking.
Anonymous. (2004). Imperial hubris: Why the West is losing the war on ter
-
rorism. New York: Brasseys, Inc.
Arndt, J., Cook, A., & Routledge, C. (in press). The blueprint of terror
management: Understanding the cognitive architecture of psy
-
chological defense against death-related thought. In J.
Greenberg, S. Koole, & T. Pyszczynski (Eds.), Handbook of empirical
existential psychology. New York: Guilford.
Arndt, J., Greenberg, J., & Cook, A. (2002). Mortality salience and the
spreading activation of worldview-relevant constructs: Exploring
the cognitive architecture of terror management. Journal of Experi-
mental Psychology: General, 131, 307-324.
Arndt, J., Greenberg, J., Pyszczynski, T., & Solomon, S. (1997a). Sub-
liminal exposure to death-related stimuli increases defense of the
cultural worldview. Psychological Science, 8, 379-385.
Arndt, J., Greenberg, J., Schimel, J., Pyszczynski, T., & Solomon, S.
(2002). To belong or not belong, that is the question: Terror man-
agement and identification with ethnicity and gender. Journal of
Personality and Social Psychology, 83, 26-43.
Arndt, J., Greenberg, J., Solomon, S., Pyszczynski, T., & Simon, L.
(1997b). Suppression, accessibility of death-related thoughts, and
cultural worldview defense: Exploring the psychodynamics of ter-
ror management. Journal of Personality and Social Psychology, 73, 5-
18.
Becker, E. (1971). The birth and death of meaning (2nd ed.). New York:
Free Press.
Becker, E. (1973). The denial of death. New York: Free Press.
Becker, E. (1975). Escape from evil. New York: Free Press.
Berke, R. L., & Sanger, D. E. (2001, September 9). Bush’s aides seek to
focus efforts on the economy. The New York Times, Section 1, p. 1.
Bord, R. J. (1975). Toward a social-psychological theory of charis-
matic social influence processes. Social Forces, 53, 485-497.
Carney, J., & Dickerson, J. F. (2001, October 22). A work in progress:
Bush is growing —and graying —before our eyes. An inside report
on the making of a leader. Time Magazine.
Clarke, R. (2004). Against all enemies: Inside America’s war on terror. New
York: Free Press.
Cohen, F., Solomon, S., Maxfield, M., Pyszczynski, T., & Greenberg, J.
(in press). Fatal attraction: The effects of mortality salience on eval-
uations of charismatic, task-oriented, and relationship-oriented
leaders. Psychological Science.
Darley, J., & Morris, W. N. (1975). Effects of stress and commonality of
fate on helping behavior. Journal of Personality and Social Psychology,
31, 145-149.
Dean, J. (2004). Worse than Watergate: The secret presidency of George W.
Bush. Boston: Little, Brown.
Dworkin, R. (2002). The threat to patriotism. In G. Calhoun, P.
Prince, & A. Timmer (Eds.), Understanding September 11 (pp. 273-
284). New York: New Press.
Ehrhart, M. G., & Klein, K. J. (2001). Predicting follower’s prefer-
ences for charismatic leadership: The influence of follower values
and personality. Leadership Quarterly, 12, 153-179.
Etzioni, A. (2002, August 28). How democracy is preserved. Christian
Science Monitor, pp. 1-2.
Feldmann, L. (2004, April 16). Presidents as communicators in war-
time. Retrieved from http://www.csmonitor.com/2004/0416/
p03s01-uspo.htm.
Florian, V., & Mikulincer, M. (1997). Fear of death and the judgment
of social transgressions: A multidimensional of terror manage-
ment theory. Journal of Personality and Social Psychology, 73, 369-380.
Frank, J. (2004). Bush on the couch: Inside the mind of the President. New
York: Regan Books.
Freud, S. (1965). Group psychology and the analysis of the ego. New York:
Bantam Books. (Original work published 1921)
Fromm, E. (1941). Escape from freedom. New York: Avon Books.
1148 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN
at University of Missouri-Columbia on March 31, 2009 http://psp.sagepub.comDownloaded from

Gaertner, S. L., & Dovidio, J. F. (2000). Reducing intergroup bias: The
common ingroup identity model. Philadelphia: U.S. Psychology
Press/Taylor & Francis.
Gilbert, D. T., & Hixon, J. G. (1991). The trouble of thinking: Activa
-
tion and application of stereotypic beliefs. Journal of Personality and
Social Psychology, 50, 269-280.
Goldenberg, J. L., Cox, C. R., Pyszczynski, T., Greenberg, J., & Solo-
mon, S. (2002). Understanding human ambivalence about sex:
The effects of stripping sex of meaning. Journal of Sex Research, 39,
310-320.
Greenberg, Greenberg, J., Simon, L., Solomon, S., Pyszczynski, T., &
Lyon, D. (1995). Assessing alternative explanations of mortality
salience effects: Terror management, value accessibility, or
aversive thoughts? European Journal of Social Psychology, 25, 417-433.
Greenberg, J., Martens, A., Jonas, E., Eisenstadt, D., Pyszczynski, T., &
Solomon, S. (2003). Detoxifying thoughts of death: Eliminating
the potential for anxiety eliminates the effects of mortality
salience on worldview defense. Psychological Science, 14, 516-519.
Greenberg, J., Pyszczynski, T., & Solomon, S. (1986). The causes and
consequences of a need for self-esteem: A terror management the-
ory. In R. F. Baumeister (Ed.), Public self and private self (pp. 189-
212). New York: Springer-Verlag.
Greenberg, J., Pyszczynski, T., Solomon, S., Rosenblatt, A., Veeder,
M., Kirkland, S., et al. (1990). Evidence for terror management,
II: The effects of mortality salience on reactions to those who
threaten or bolster the cultural worldview. Journal of Personality and
Social Psychology, 58, 308-318.
Greenberg, J., Pyszczynski, T., Solomon, S., Simon, L., & Breus, M.
(1994). Role of consciousness and accessibility of death-related
thoughts in mortality salience effects. Journal of Personality and
Social Psychology, 67, 627-637.
Greenberg, J., Simon, L., Harmon-Jones, E., Solomon, S., Pyszczynski,
T., & Chatel, D. (1995). Testing alternative explanations for mor-
tality effects: Terror management, value accessibility, or worri-
some thoughts? European Journal of Social Psychology, 12, 417-433.
Greenberg, J., Simon, L., Porteus, J., Pyszczynski, T., & Solomon, S.
(1995). Evidence of a terror management function of cultural
icons: The effects of mortality salience on the inappropriate use of
cherished cultural symbols. Personality and Social Psychology Bulle-
tin, 21, 1221-1228.]
Greenberg, J., Simon, L., Pyszczynski, T., Solomon, S, & Chatel, D.
(1992). Terror management and tolerance: Does mortality
salience always intensify negative reactions to others who threaten
ones worldview?. Journal of Personality and Social Psychology, 63, 212-
220.
Greenberg, J., Solomon, S., & Pyszczynski, T. (1997). Terror manage-
ment theory of self-esteem and social behavior: Empirical assess-
ments and conceptual refinements. In M. P. Zanna (Ed.),
Advances in experimental social psychology (Vol. 29, pp. 61-139). New
York: Academic Press.
Huddy, L., Khatid, N., & Capelos, T. (2002). Trends: Reactions to the
terrorism attacks of September 11, 2001. Public Opinion Quarterly,
66, 418-450.
Humberman, J. (2004). The Bush-hater’s handbook. New York: Thun-
der’s Mouth Press.
Kirkpatrick, S. A., & Locke, E. A. (1996). Direct and indirect effects of
three core charismatic leadership components on performance
and attitudes. Journal of Applied Psychology, 81, 36-51.
Landau, M. J., Johns, M., Greenberg, J., Pyszczynski, T., Martens, A.,
Goldenberg, J. L., et al. (in press). A function of form: Terror man
-
agement and structuring the social world. Journal of Personality and
Social Psychology.
Langer, G. (2001, May 7). Prayers for victims, support for reprisals. Re-
trieved from http://abcnews.go.com/sections/us/DailyNews/
wtc_abcpoll010911.html.
Lifton, R. J. (1968). Revolutionary immortality: Mao Tse-tung and the Chi
-
nese cultural revolution. New York: Random House.
Lipman-Bluemen, J. (1996). The connective edge: Leading in an interde
-
pendent world. San Francisco: Jossey-Bass.
McGregor, H., Lieberman, J. D., Solomon, S., Greenberg, J., Arndt, J.,
Simon, L., et al. (1998). Terror management and aggression: Evi
-
dence that mortality salience motivates aggression against
worldview threatening others. Journal of Personality and Social Psy
-
chology, 74, 590-605.
Mikulincer, M., Florian, V., & Hirschberger, G. (2003). The existen-
tial function of close relationships: Introducing death into the sci-
ence of love. Personality and Social Psychology Review, 7, 20-40.
Miller, C. H., & Landau, M. J. (in press). Communication and the
causes and costs of terrorism: A terror management perspective.
In D. O’Hair, R. Health, & G. Ledlow (Eds.), Terrorism: Communica
-
tion and rhetorical perspectives. Cresskill, NJ: Hampton Press.
Moore, D. W. (2001, May 7). Americans see terrorist attacks as “acts of war.”
Retrieved from http://www.gallup.com/content/login.aspx?ci=
4879.
Morin, R., & Dean, C. (2001, October 8). Public support is over-
whelming; poll finds 94% favor Bush’s ordering strikes on Afghan-
istan. The Washington Post, p. A5.
Phillips, K. (2004). American dynasty: Aristocracy, fortune, and the politics
of deceit in the house of Bush. New York: Viking.
Piliavin, I. M., Rodin, J. A., & Piliavin, J. (1969). Good Samaritanism:
An underground phenomenon? Journal of Personality and Social
Psychology, 13, 289-299.
Pillai, R. (1996). Crisis and the emergence of charismatic leadership
in groups: An experimental investigation. Journal of Applied Psy-
chology, 26, 543-562.
PollingReport.com. (2004, May 7). President Bush: Job ratings. Re-
trieved from http://www.pollingreport.com/BushJob.htm.
Purdum, T. S. (2001, September 17). After the attacks: The White
House; Bush warns of a wrathful, shadowy and inventive war. The
New York Times, p. A2.
Pyszczynski, T., Greenberg, J., & Solomon, S. (1999). A dual-process
model of defense against conscious and unconscious death-
related thoughts: An extension of terror management theory. Psy-
chological Review, 106, 835-845.
Pyszczynski, T., Solomon, S., & Greenberg, J. (2003). In the wake of
9/11: The psychology of terror. Washington, DC: American Psycholog-
ical Association.
Redl, F. (1942). Group emotion and leadership. Psychiatry, pp. 573-596.
Risen, J. (2004, April 28). How pair’s finding on terror led to clash on
shaping intelligence. The New York Times, p. 1.
Rosenblatt, A., Greenberg, J., Solomon, S., Pyszczynski, T., & Lyon, D.
(1989). Evidence for terror management theory I: The effects of
mortality salience on reactions to those who violate or uphold cul-
tural values. Journal of Personality and Social Psychology, 57, 681-690.
Rosenthal, R., Rosnow, R. L., & Rubin, D. B. (2000). Contrasts and effect
sizes in behavioral research: A correlational approach. Cambridge, UK:
Cambridge University Press.
Rutenberg, L., & Sander, D. E. (2004, March 12). Bush campaign
goes on attack in new anti-Kerry ads. The New York Times, p. A18.
Sherif, M. (1966). In common predicament: Social psychology of intergroup
conflict and cooperation. Boston: Houghton-Mifflin.
Sherif, M., Harvey, O. J., White, B. J., Hood, W. R., & Sherif, C. W.
(1961). Intergroup cooperation and competition: The Robbers Cave exper-
iment. Norman, OK: University Book Exchange.
Simon, L., Greenberg, J., Harmon-Jones, E., Solomon, S., Pyszczynski,
T., Arndt, J., et al. (1997). Cognitive-experiential self-theory and
terror management theory: Evidence that terror management
occurs in the experiential system. Journal of Personality and Social
Psychology, 72, 1132-1146.
Solomon, S., Greenberg, J., & Pyszczynski, T. (1991). A terror man
-
agement theory of social behavior: The psychological functions of
self-esteem and cultural worldviews. In M. P. Zanna (Ed.),
Advances in experimental social psychology (Vol. 24, pp. 93-159). New
York: Academic Press.
Solomon, S., Greenberg, J., & Pyszczynski, T. (in press). The cultural
animal: Twenty years of terror management theory and research.
In J. Greenberg, S. L. Koole, & T. Pyszczynski (Eds.), Handbook of
experimental existential psychology. New York: Guilford.
Landau et al. / TERROR MANAGEMENT AND SUPPORT FOR G. W. BUSH 1149
at University of Missouri-Columbia on March 31, 2009 http://psp.sagepub.comDownloaded from

Suskind, R. (2004). The price of loyalty: George W. Bush, the White House,
and the education of Paul O’Neill. New York: Simon & Schuster.
Toner, R., & Elder, J. (2001, December 12). Public is wary but support
-
ive on rights curbs. The New York Times, p. 1.
Unger, C. (2004). House of Bush, house of Saud. New York: Scribner.
Waldman, P. (2004). Fraud: The strategy behind the Bush lies and why the
media didn’t tell. Naperville, IL: Sourcebooks Trade.
Watson, D., & Clark, L. A. (1991). The PANAS-X: Preliminary manual for
the positive and negative affect schedule-expanded form. Unpublished
manuscript, Southern Methodist University, Dallas, Texas.
Woodward, B. (2004). Plan of attack. New York: Simon & Schuster.
Received April 29, 2004
Revision accepted May 21, 2004
1150 PERSONALITY AND SOCIAL PSYCHOLOGY BULLETIN
at University of Missouri-Columbia on March 31, 2009 http://psp.sagepub.comDownloaded from