Emerging Trends in The Social and Behavioral Sciences

Public Opinion and International Conflict

Item

Title

Public Opinion and International Conflict

Author

Berinsky, Adam J.

Research Area

Class, Status and Power

Topic

Violence, Conflict and Conflict Resolution

Abstract

Should the opinions of citizens in a democracy matter in decisions of war and peace? The answer to this critical question depends on the stock we place in the ability of the mass public to come to meaningful decisions regarding the conduct of foreign affairs. In this essay, I examine public opinion about war over the past 75 years and make the case that our assessment of the mass public depends in large part on the nature of the information it receives from political leaders. Contrary to the conventional wisdom regarding public opinion and foreign policy that emerges from scholarly and journalistic accounts, events do not directly influence the public. Instead, citizens learn about wars largely from political leaders. Public opinion during times of war is therefore shaped by many of the same attachments and enmities that matter in domestic politics. As in other areas of politics, public opinion is primarily structured by the ebb and flow of partisan and group‐based political conflict.

Related Essays

Mediation in International Conflicts (Political Science), Kyle Beardsley and Nathan Danneman

Interdependence, Development, and Interstate Conflict (Political Science), Erik Gartzke

Genetic Foundations of Attitude Formation (Political Science), Christian Kandler et al.

Reconciliation and Peace‐Making: Insights from Studies on Nonhuman Animals (Anthropology), Sonja E. Koski

Why Do States Sign Alliances? (Political Science), Brett Ashley Leeds

Attitude: Construction versus Disposition (Psychology), Charles G. Lord

Political Psychology and International Conflict (Political Science), Rose McDermott

Implicit Attitude Measures (Psychology), Gregory Mitchell and Philip E. Tetlock

War and Social Movements (Political Science), Sidney Tarrow

Trends in the Analysis of Interstate Rivalries (Political Science), William R. Thompson

Identifier

etrds0267

extracted text

Public Opinion and International
Conflict
ADAM J. BERINSKY

Abstract
Should the opinions of citizens in a democracy matter in decisions of war and peace?
The answer to this critical question depends on the stock we place in the ability of the
mass public to come to meaningful decisions regarding the conduct of foreign affairs.
In this essay, I examine public opinion about war over the past 75 years and make
the case that our assessment of the mass public depends in large part on the nature
of the information it receives from political leaders. Contrary to the conventional
wisdom regarding public opinion and foreign policy that emerges from scholarly
and journalistic accounts, events do not directly influence the public. Instead, citizens
learn about wars largely from political leaders. Public opinion during times of war
is therefore shaped by many of the same attachments and enmities that matter in
domestic politics. As in other areas of politics, public opinion is primarily structured
by the ebb and flow of partisan and group-based political conflict.

Should the opinions of citizens in a democracy matter in decisions of war and
peace? The answer to this critical question depends on the stock we place in
the ability of the mass public to come to meaningful decisions regarding the
conduct of foreign affairs. In this essay, I examine public opinion about war
over the past 75 years and make the case that our assessment of the mass
public depends in large part on the nature of the information it receives from
political leaders. Contrary to the conventional wisdom regarding public
opinion and foreign policy that emerges from scholarly and journalistic
accounts, events do not directly influence the public. Instead, citizens learn
about wars largely from political leaders. Public opinion during times of
war is therefore shaped by many of the same attachments and enmities
that matter in domestic politics. As in other areas of politics, public opinion
is primarily structured by the ebb and flow of partisan and group-based
political conflict.
In this essay, I first review the current state of knowledge about public opinion and war. I then briefly describe the patterns of public opinion during
Emerging Trends in the Social and Behavioral Sciences. Edited by Robert Scott and Stephen Kosslyn.
© 2015 John Wiley & Sons, Inc. ISBN 978-1-118-90077-2.

1

2

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

wartime, contrasting World War II and Iraq, to demonstrate how the public’s reaction to these wars illustrates several important lessons about public
reaction to foreign affairs.
PUBLIC OPINION AND FOREIGN POLICY
Scholarly and popular valuations of the competence of the American public have waxed and waned considerably over the past 50 years. In the early
days of survey research, scholars argued that the public opinion concerning
foreign policy was volatile and irrational—a fickle and changing “mood,” in
Almond’s (1960) words. Indeed, much of the research from the early days of
polling in the 1940s and 1950s painted a fairly dim view of democracy. Survey
researchers found that Americans were remarkably uninformed about basic
facts and developments in the international world. This view, which Holsti
(2004) has termed the “Almond–Lippmann consensus,” ruled the discussion
of public opinion concerning foreign affairs.1
Starting in the 1970s, however, this traditional view began to come under
attack by a new generation of scholars. The catalyst for a more charitable
view of the mass public might have arisen as a reaction to the Vietnam War,
as Holsti (2004) argues, or it may have been part of a more general backlash
against the dismal findings of the founders of the survey research enterprise.
Regardless of its source, the conventional wisdom that has emerged over the
past 30 years in the public opinion and foreign policy literature has taken a
dramatically different turn. Most scholars currently claim that the course of
events in a conflict directly determines public support for war.
Consider the “casualties hypothesis”—the view that the American people
will shy away from international involvement in the face of war deaths. This
hypothesis reflects Mueller’s (1973) contention that public support for war
is inversely related to the number of casualties. As the human costs of war
rise, Mueller argues, public support for conflict falls.2 Policymakers share the
belief that casualties shape public opinion concerning the wisdom of war. In
a series of interviews with political elites, Kull and Destler (1999) found that
three-fifths of congressional staff members in 1995 believed that the loss of
US lives during a UN peace keeping operation would lead to the immediate withdrawal of US troops. Leaders of other countries also subscribe to
1. For an excellent review of these early studies, see Holsti (2004).
2. Mueller actually argued that public support is related to the logged cumulative casualties, allowing early deaths to have greater effects than later deaths. As Burk (1999) notes, precise statements of
the casualties hypothesis are hard to find. But references to the general mechanism at the heart of this
hypothesis—that war deaths directly drive down public support for war—permeate popular discourse
concerning war and even the writings of military strategists and policymakers (see also Kull & Destler,
1999). Some modifications have been made to this basic theory over time. Gartner and Segura (1997, 2000),
for instance, argue that casualty rates of military personnel in local communities are an important determinant of support for the war.

Public Opinion and International Conflict

3

Mueller’s logic. In the lead-up to the first Gulf War, for instance, Saddam
Hussein claimed, “In the event of war, there will be great losses … when
5000 Americans have fallen, Bush will have to end such a war … ” (St. Louis
Dispatch, December 22, 1990). Thus, the story advanced by Mueller remains
a dominant view among both academics and policymakers (see Feaver &
Gelpi, 2004; Kull & Destler, 1999).
Moreover, in recent years, scholars have considered not only the human
costs of war but also other costs and benefits associated with military interventions. For instance, some authors contend that the possible benefits arising from the ultimate success of a mission—whether the United States will
prevail in a given conflict—determines public support for conflict (Feaver &
Gelpi, 2004; Gelpi, Feaver, & Reifler, 2005/2006; Kull & Ramsay, 2001). Other
authors argue that the motivation behind a given mission may at least indirectly affect levels of support for that conflict. For example, Jentleson and
Britton (1998) make the case that the policy objective of an intervention plays
a large role in determining whether the costs of intervention weigh greatly
in the public mind. In particular, they argue that military interventions that
are designed either to stop foreign aggression against America and its allies
or to intervene in humanitarian crises will engender greater support than
missions designed to effect change in the governments of other countries.3
Larson (1996) takes the most general stance, arguing that the collective public
decides whether to support a conflict based on a rational cost/benefit calculation. According to Larson, the greater the perceived stakes the clearer the
objectives, and the higher the probability of success the greater the level of
public support for war. These theories differ in their particulars, yet all share
the belief that “events” directly determine public support for war. Thus, even
for scholars who consider factors beyond casualties, the basic logic underlying Mueller’s argument remains the dominant position. These “event response” theories argue that the mass public is rational and will support war if,
and only if, the events of war ensure that the costs of military action are outweighed by the perceived benefits of a successful outcome.
Although event–response theories of public support for war have made
important contributions, they have several potentially serious problems.
First, such theories presume that members of the mass public at least
implicitly incorporate knowledge of political developments into their political judgments. However, this assumption does not square with political
3. Jentleson’s theory is different than other theories relating to the ongoing costs and benefits of war
because it considers the overall justification that leads to war, not the continuing development of the war.
Put another way, Jentleson’s theory speaks to the baseline levels of support for interventions sparked by
different initiating events, but cannot explain the dynamics of support for a given intervention, short of a
change in the policy objective of that intervention. Thus, while the objective of a mission may play a role
in determining absolute levels of public support, that consideration is distinct from other theories of war
discussed here.

4

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

scientists’ findings regarding the knowledge levels of the mass public.
Harkening back to the early studies of public opinion discussed earlier,
while researchers have long known that Americans know little about
politics, knowledge levels are even lower when the focus turns to specific
factual information. For instance, in 1998, only 12% of Americans knew that
the crime rate in the United States had declined over the previous decade
(Gilens, 2001; see also Ansolabehere, Snowberg, & Snyder, 2005; Wong, 2007).
Second, much research on the relationship between the particulars of
war—such as mission objective and casualty levels—and support for war
has examined differences in aggregate public support for intervention either
across different wars or at different moments in time during the same war
(Jentleson, 1992; Jentleson & Britton, 1998; Klarevas, 2002; Larson, 1996;
Mueller, 1973). With some important exceptions (Gartner & Segura, 2000;
Gartner, Segura, & Wilkerning, 1997; Gelpi et al., 2005/2006), analysis has
ignored differences among groups within the mass public during the course
of particular conflicts.4 Put another way, almost all the work described
earlier ignores the American political process. Treating the mass public as
an undifferentiated whole—innocent of attachments to political parties and
relevant social groups—leaves no room for the effect of domestic politics.
Many researchers who study public opinion and war—even those scholars
who conduct individual-level analysis—often talk about “the public” as if
it were a monolithic entity. But foreign policy is often as contentious and
partisan as domestic politics. Our understanding of war and politics must
account for the effects of the domestic political process (Baum & Groeling,
2004; Schultz, 2001).
MEDIATED REALITY: THE PRIMACY OF POLITICAL COMPETITION
Acknowledging the relative shortcoming of event–response theories of
war support does not mean that we must retreat to the dismal conclusions
regarding public opinion and foreign policy prevalent in the 1950s and
1960s. Event–response theories, after all, are not the only explanation for the
dynamics of public support for war. Another possibility is to examine the
influence of competition among political leaders on public opinion. After
all, political scientists have long known that partisan attachments shape
political judgments across a host of issues.
Zaller (1992) has used the relative positions of politicians to explain public
opinion across a number of domestic political controversies. This approach
4. In a related vein, Hurwitz and Peffley (1987) look more generally at foreign policy attitudes and
advance an individual-level hierarchical model of attitude structure. This model is incompatible with the
event-driven theories of war support, but is complementary to the elite-driven view advanced in this
essay.

Public Opinion and International Conflict

5

has become the dominant approach in the study of domestic public opinion.
Zaller argues that the balance of persuasive messages carried in the political
media determines the shape of opinion on a given policy controversy. Individuals who are most politically knowledgeable are most likely to receive
political messages and accept those messages that fit personal political predispositions. The greater the volume of elite discourse favoring a particular
policy position from leaders of a particular political stripe, the more likely it
is that the members of the mass public who share the political predispositions
of those elites will adopt that position.
Perhaps elite discourse is the key to explaining war support as well. Zaller
makes such a case in the context of the Vietnam War, arguing that the decline
in support for that war was driven by a change in the balance of elite discourse across the 1960s. In the early phase of the war, when political leaders
were almost uniform in their support for the US policy in Vietnam, Zaller
found a positive relationship between political awareness and support for
the war; those most attentive to elite discourse were most supportive of the
current policy, regardless of their other political views. Zaller terms this phenomenon the mainstream pattern of political support. On the other hand, in the
later phases of the Vietnam War, when the mainstream consensus dissolved
into elite disagreement, a “polarization pattern” emerged. Here, the effect of
political awareness on support for the war was conditional on an individual’s
political values. Citizens attentive to politics followed the path of those leaders who shared their political views. For the Vietnam War, greater awareness
led to higher levels of support among hawks and higher levels of opposition among doves. Zaller’s story is not particular to Vietnam. Zaller (1994)
also found that the changing positions of Congressional Democrats from the
summer of 1990 until the eve of the first Gulf War could explain shifts in
support for military action among Democratic identifiers in the mass public.
Theories of elite competition bring to bear lessons from the domestic arena
to the realm of foreign policy and explicitly take into account how patterns
of political competition shape public opinion. Individuals with different
political predilections react in response to different forms of elite discourse.
However, Zaller’s explanation is an incomplete account of elite influence.
Certainly, there are cases when political actors on both sides of a controversy
take clear stands, leading to polarized opinions among the mass public. But
even in the absence of arguments from both sides, individuals might have
the information they need to come to a judgment regarding the fit between
the policy options on the table and their political predispositions. As long
as one side takes a stand, citizens can use the positions of leading political
figures to come to reasonable political decisions. Specifically, even in the
absence of explicitly contradictory messages, citizens can use the position
of prominent politicians as reference points and decide whether to support

6

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

or oppose a policy. For instance, if I am a Democrat, I need only know that
George Bush supports a policy initiative to recognize that I should oppose
such a course of action; I do not need Democratic leaders to explicitly make
such a case. In effect, citizens delegate the difficult process of arriving at
an opinion on a complicated policy matter to trusted political experts.
Presidents serve as such cuegivers, especially in the realm of foreign policy.
But to use this cue requires that citizens have knowledge of the positions of relevant political actors.5 Thus, Zaller’s basic insight holds. As an
individual’s level of political information increases, her awareness of the
positions of particular leaders—and the distinctiveness of those positions
relative to other political actors—increases. Thus, a pattern of opinion
polarization could occur even in the absence of vocal opposition, provided a
strong cuegiver takes a clear position on that policy. As I illustrate later, this
alternative mechanism of elite influence—what I the elite cue theory—can
explain the pattern of opinion in World War II, when both Franklin D.
Roosevelt (FDR) and his Republican opponents took distinct positions
(Berinsky, 2007, 2009). Of course Zaller’s theory can explain this course of
events as well. However, unlike Zaller’s original formulation, this theory
can also explain the polarized pattern of opinion concerning the war in
Iraq in 2004, a situation in which Bush and Republican Party leaders took
a strong pro-war position, but Democratic Party leaders failed to express
either strong support or opposition.
WORLD WAR II
World War II provides a useful forum for demonstrating the power of the elite
cue theory. Contrary to the expectations of the casualties hypothesis, over the
almost 4 years of US involvement in that War, support for the effort did not
wane, even as war deaths mounted, in particular after the spring of 1944.6
Campbell and Cain (1965) use a number of questions to measure support for
the government’s stated military aims and demonstrate that at no point did
public support fall below 75%.
Other explanations that find the roots of continued support of the American public in wartime events are also problematic. There is a broad sense in
popular accounts and some academic treatments of World War II, that this
5. This position was actually advanced in a somewhat different form by Mueller (1973). Mueller discusses the importance of partisan cues in structuring wartime opinion and argues that well-informed
partisans are most likely to adopt the positions of the political leaders of their respective parties (see
pp. 120–121).
6. The Department of the Army recorded monthly casualties (although not war deaths) from December 7, 1941 to December 31, 1946. From December 1941 until December 1943, casualties only exceed 10,000
in 1 month. Over the course of 1944, however, monthly casualty rates increased greatly, reaching 55,000
in June and ranged from 50,000 to 80,000 until April, 1945 (Army Battle Casualties and Nonbattle Death
in World War II, Final Report).

Public Opinion and International Conflict

7

conflict was the “good war” where the benefits of intervention were clear.
According to this account, the United States, shaken by the Japanese attack
at Pearl Harbor, quickly rallied to the cause of protecting democracy. Larson
(1996), for instance, writes:
In the Second World War—‘the good war’—the public had an excellent cause.
Of course Japan’s attack on Pearl Harbor and Germany’s declaration of war on
the United States contributed greatly to support for U.S. entry into the war. But
support also derived from the shared perception of important stakes and vast
benefits of eliminating a grave threat to U.S. security and from optimism that
the outcome would be a decisive victory and punishment of the Axis powers
… Further contributing to support for the war was a desire for punishment
as a consequence of the Japanese sneak attack on Pearl Harbor, such atrocities
as the Bataan Death March, reports of the Japanese torture of U.S. prisoners of
war, and Germany’s holocaust (pp. 14–15).

These explanations may seem plausible in retrospect, but public opinion
data from the 1940s does not provide support for such accounts. Knowledge of the atrocities such as the Holocaust, discussed by Larson, was thin
during the war. In January 1943, only 47% of the population thought that
Germany was engaged in the mass destruction of Europe’s Jewish population.7 Even when a belief in the existence of concentration camps became
widespread in late 1944—when 76% of the public believed that “the Germans
have murdered many people in concentration camps”—only about a third
of respondents thought that the toll at the camps would rise above 100,000.8
Furthermore, at several points in time, Gallup and Hadley Cantril’s Office of
Public Opinion Research (OPOR) asked the public if they had “a clear idea
what the war is about.” In March 1942, almost 4 months after the attack on
Pearl Harbor, only 43% of Americans felt they had such an idea. By July, that
figure rose to 62% but, for the rest of the war, the percentage of Americans
who agreed with the statement largely fluctuated in the 65–70% range, rising to 75% in June 1944, but falling below 60% in March 1944. Thus, while
a majority of Americans could identify a war aim, a sizeable minority could
not. Certainly, the specific context of the Second World War helped engender high levels of support for the war. However, support for the US effort
at the time was not as self-evident as it was in retrospect. As Mueller (1973)
aptly notes, “the major reasons for supporting [World War II] were largely
unappreciated while it was going on” (p. 65). Thus, the existing accounts
7. The question read, “It is said that two million Jews have been killed in Europe since the war began.
Do you think this is true, or just a rumor?”
8. These figures (and the other World War II—era public opinion figures presented in this chapter)
are from the author’s analysis of opinion polling data.

8

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

that attribute continued public support to the benefits made clear by ongoing
wartime events are not supported by the data.
What, then, can explain continued public support for the war? As argued
earlier, my contention is that it was not the direct influence of events that
determined support for World War II, but rather the patterns of elite conflict
during the 1930s and 1940s. In Berinsky (2009), I describe my evidence on
this point more fully, but Legro’s (2000) study of political rhetoric in the 1930s
and 1940s is illustrative of this point. Legro finds that from 1938 through the
end of 1941, support among elites for some form of US involvement in World
War II increased generally over time. However, the gap between FDR and his
critics on the necessity and wisdom of US involvement in the Second World
War remained large. For instance, Legro finds that FDR’s critics—represented
by the editorial page of the Chicago Tribune, a paper that can be seen as
the mouthpiece of the isolationist wing of the Republican Party—moved in
an internationalist direction through 1941. However, FDR’s position consistently outpaced that of his critics. Beginning in 1939, FDR moved in a strongly
internationalist direction, but it was not until 1942 that the Tribune expressed
any support for military commitments abroad. Conversely, from 1942 on,
“the collective orthodoxy embraced the necessity of international cooperation and multilateralism” (Legro, 2000, p. 261).
Thus, elite discourse split along the lines of support for FDR before Pearl
Harbor, but presented a largely united front after the United States entered
the war. This line of argument is not intended to minimize the importance of
Pearl Harbor in shaping opinion on the war. However, given the ephemeral
nature of rally effects (Brody, 1991) it is clear that greater attention needs to
be paid to how support for the war was sustained through the nearly 4 years
of US involvement. The elite cue theory suggests that the key is the nature
of elite discourse from this time; we should see the polarization pattern of
support before US entry into the war, and—following the pattern of elite
discourse—the mainstream pattern after that point.
The full results of my analysis are presented in Berinsky (2009). But it is
rather easy to summarize the results. As predicted by the elite cue theory,
the polarization pattern largely characterizes opinion through the middle of
1941. Consistent with expectation, the pattern of public support for military
action changed greatly after the United States entered the war. In line with
the expectations of the elite cue theory, as discourse moved from a two-sided
to a one-sided flow in 1941, the public followed suit. Measured in a variety of
ways—whether the United States should send its army abroad, whether the
United States should take an active role in world affairs after the War, and,
most critically, whether the United States should make peace with Germany
if Hitler was overthrown—individuals more attuned to elite discourse were

Public Opinion and International Conflict

9

more supportive of an active United States role, regardless of their predispositions regarding FDR.9
THE WAR IN IRAQ
The Iraq War provides another interesting window into the study of public opinion concerning war. Just as with World War II, the elite cue theory
leads to clear expectations regarding the relative role of events and elites
in structuring opinion concerning the war. To repeat the argument, consistent with recent work on US public opinion, but contrary to the expectations
of many scholars and policymakers, events have little effect on the public’s
day-to-day judgments regarding the wisdom of war. Conversely, patterns of
elite discourse—the stated positions of leading Democratic and Republican
politicians—should play a large role in determining public support for war.
Individuals will use positions of prominent politicians as reference points
that guide their opinions concerning war. Moreover, contrary to Zaller, we
should expect to find divergence between partisans on the wisdom of the
Iraq War even without outspoken opposition on the part of Democratic politicians. In the rest of this section, I demonstrate this phenomenon using data
collected during 2004.
THE INABILITY OF CASUALTIES TO EXPLAIN WAR SUPPORT
The Iraq War Casualty Survey, conducted from July 23 to August 2, 2004 by
Knowledge Networks, asked a nationally representative sample of respondents:
Please give your best guess to this next question, even if you are not sure of the
correct answer. As you know, the United States is currently involved in a war
in Iraq. Do you happen to know how many soldiers of the U.S. military have
been killed in Iraq since the fighting began in March 2003?

At first glance, it appears that the public was informed about the level of
troop deaths in Iraq. The mean estimate of deaths in the sample was 952
deaths, while the median response was 900 deaths.10 Both of these figures
are extraordinarily close to the true casualty count, which rose from 901 to
915 over the span of the survey. The accuracy of the median respondent,
however, obscures large variation in the casualty estimates. Respondents
9. The 1940 election casts additional light on the importance of elite discourse in determining support
for war. See Berinsky (2009) for further discussion.
10. The mean and median estimates were generated using the poststratification weights provided by
Knowledge Networks. Fewer than 2% declined to answer the initial questions and, following a probe
asking respondents to provide their best guess even if they were not sure of the correct answer, every
respondent answered the question.

10

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

gave answers ranging from 0 deaths to 130,000 deaths. Even setting aside the
extreme responses (casualty guess under 10 and over 10,000) the standard
deviation of the casualty estimate was 802.11
A simple tabulation of the estimates illuminates the pattern of responses to
the casualty question. I scored those respondents who estimated the number
of war deaths to be between 801 and 1015 (the true estimate ±100 deaths) as
“correct.” Those who gave an estimate of 800 or lower were scored as “underestimators,” while those who guessed higher than 1015 are considered
“overestimators.”12 The modal response (47%) is a correct answer. However,
nearly as many respondents (42%) underestimated the number of war deaths
(11% overestimated the number of deaths).
Importantly, for present purposes, this variation is not random; elite
cues play a significant role in shaping casualty estimates. I predicted that
Democrats should overestimate the number of war deaths because of their
ambivalence about war, while Republicans should systematically downplay
the costs because of their support for war. Indeed, as demonstrated in
Berinsky (2007), compared to strong Republicans, strong Democrats are less
likely to underestimate and are slightly more likely to overestimate casualty
levels (see also Gaines, Kuklinski, Quirk, Peyton, & Verkuilen, 2007).
Having demonstrated that the respondents’ perceptions of events in the
Iraq War are influenced by partisanship, I next moved to the more important
question of whether casualty estimates have any influence on opinions concerning war. I measured attitudes toward the Iraq war with two commonly
used measures of war support. The first question asked, “Do you think the
U.S. made the right decision or the wrong decision in using military force
against Iraq?” The second question asked, “All in all, considering the costs
to the United States versus the benefits to the United States, do you think the
current war with Iraq has been worth fighting, or not?”
Embedded in the Iraq War Survey was an experiment in which one half
of those people who were asked to estimate how many soldiers died in
Iraq were then told, “Many people don’t know the answer to this question,
but according to the latest estimates, 901 soldiers have been killed in Iraq
since the fighting began in March 2003.”13 In other words, one-half of the
people who were asked to estimate the number of American deaths were
11. With the extreme responses included, the standard deviation was 3012.
12. I tried other methods of scoring a “correct” response—increasing and decreasing the band of
acceptable answers incrementally from ±50 deaths to ±200 deaths—and found essentially the same
results. The robustness of these findings extends beyond this particular dataset. Cobb (2007) examines a
number of surveys that measure knowledge of cumulative casualty rates from 2003 to 2006, using errors
greater than 20% as his cutoff for a “correct” answer and finds similar patterns of misperception.
13. This number was updated once on July 30th, moving the casualty figure to 908. In the analyses
presented in the paper, I set the range of the one “correct answer” (independent of the range of acceptable
answers that were scored as “correct” estimates) from the low point (901) to the high point (915) of war
deaths from this period.

Public Opinion and International Conflict

11

given the correct answer before answering the questions concerning support
for the Iraq War. This experimental design allows me to contrast levels
of support for the war between two otherwise comparable groups: (i) the
respondents in the “uncorrected” condition who underestimate casualties
(e.g., those who said that there were fewer than 800 casualties) but were
not told the correct number of war deaths; and (ii) the respondents in the
“corrected” condition who underestimate war deaths but were then told
the number of US soldiers who died.14 I can similarly compare respondents
who overestimate casualties. This is a powerful comparison, because the
only difference between the “estimate” group and the “corrected” group is
that respondents in the “corrected” condition were subsequently told the
true casualty rates. By comparing these two groups, I can assess the effect of
introducing the correct information on support for war for individuals who
are similarly misinformed about casualty rates.
The results of these analyses are presented in Berinsky (2007), but are easily
summarized here. There are essentially no differences between the respondents in the two conditions. Events—such as casualties—do not determine
whether the public supports war.
THE PRIMACY OF PARTISAN CUES
What, then, shapes public support for war? The central argument of this
essay has been that in all wars, the public reaction to a given conflict has
been deeply colored by partisan politics, as the elite cue theory predicts. The
Iraq war is no exception. From the early days of the war, public opinion was
best characterized by the polarization pattern.
In Berinsky (2009), I performed analysis similar to Zaller’s—and similar to
the World War II analysis discussed—using my Iraq War survey. Specifically,
I examined the relationship between political information levels and support
for the war, broken down by party affiliation.15 As I show in Berinsky (2009),
as a person’s attention to political discourse increases, he adopts diametrically opposed positions on the war, depending on whether he is a Democrat
or a Republican. Although there is a gap between Democrat and Republican
support for war at the lowest information levels, this gap grows as information levels increase, indicating that differences in elite positions are reflected
14. I employed a between-subjects design rather than a within-subjects design (in which support for
the war would be measured both before and after the treatment) in part because I was worried that respondents would try to maintain consistency in their answers to the war question, given the short time-span
of the interview.
15. Following Zaller (1992, 1994), I modeled the measures of support for war as a function of partisanship, information, the interaction between information and partisanship, and several control variables.
The graphs present the predicted positions for a white male with some college education. I ran the analysis separately for each of the four experimental conditions and found that the results were nearly identical
across each of the conditions (see Berinsky, (Forthcoming), for details).

12

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

in individuals’ positions on war. In Iraq, as in World War II, partisan cues
were the primary determinant of opinion concerning war.
CONCLUSION: THE PAST AND THE PRESENT
In June 2006, White House Press Secretary Tony Snow appeared on CNN
Late Edition with Wolf Blitzer. Speaking about public opinion concerning the
Iraq War, he said, “If someone had taken a poll in the Battle of the Bulge, I
dare say people would have said, ‘Wow, my goodness, what are we doing
here?’”16 Surely Snow was surprised to learn in the following days that, in
fact, someone did take a poll during the Ardennes offensive that represented
Germany’s last push of the war. From December 31, 1944 until January 4,
1945, George Gallup’s American Institute of Public Opinion asked, “If Hitler
offered to make peace now and would give up all land he has conquered,
should we try to work out a peace or should we go on fighting until the German army is completely defeated?” Contrary to Snow’s speculation, 73% of
the public expressed support for the stated US policy of unconditional surrender; the American people wanted to continue fighting until victory was
complete.
A look below the surface reveals important lessons for modern times. Support for the war crossed party lines in 1945. Of those respondents who had
voted to re-elect FDR in the 1944 election, 78% wished to continue fighting.
Among those who voted for the Republican candidate, Thomas Dewey, 73%
wanted to fight until the Germany army met complete defeat. Thus, although
war support was slightly higher among President Roosevelt’s supporters
than his opponents, this gap pales in comparison to partisans’ opinions on
the war in Iraq.
The roots of these differences can be found, in part, in the different
approaches FDR and Bush took toward the wars begun under their watch.
From 1938 through the end of 1941, support among politicians of both
parties for some form of US involvement in World War II increased generally
over time. However, the gap between FDR and his critics on the necessity
and wisdom of US involvement in the Second World War remained large.
But after United States entry into the war, FDR secured the support of
his Republican opponents, and both parties expressed a strong pro-war
message. Conversely, even before it began, the war in Iraq was strongly
associated with President Bush and his Republican allies in Congress.
Although Democratic politicians did not express open opposition to the war
effort until relatively late in the conflict, they had never joined en masse
with their Republican counterparts in openly supporting the war.
16. Available at http://transcripts.cnn.com/TRANSCRIPTS/0606/18/le.01.html. Accessed August
20, 2007.

Public Opinion and International Conflict

13

Currently, as in other times in American history, patterns of agreement and
disagreement among partisan political actors play a critical role in shaping
popular responses to war. Without the support of politicians from across the
aisle, the American people as a whole will never support any war. Among
both politicians and the mass public, the Iraq War is an exclusively Republican war.
This essay leads also to an important general lesson. If, as I have argued
here, it is the dynamics of elite conflict, rather than the events of war, that
determine public support for war, then to properly understand the decision
to go to and wage war, we need to understand how domestic politics and
partisan divisions influence the political thinking of ordinary citizens. Foreign policy and domestic policy are two sides of the same political coin. By
explicitly accounting for the elite mediation of foreign events, we can better
understand how citizens in democracies can guide and constrain the government’s ability to wage war. Even if one does not agree with Page and
Shapiro’s (1992) contention that government affairs can “conceal or misrepresent reality without being challenged,” surely political leaders have the
agency and flexibility to interpret ambiguous wartime events. Perhaps early
survey researchers took too dim a view of the mass public, but the revisionist work in the field of public opinion and foreign policy has since painted
an unrealistically optimistic picture of the ability of ordinary citizens to constrain the actions of their leaders in times of war.
REFERENCES
Almond, G. A. (1960). The American people and foreign policy. New York, NY: Praegar.
Ansolabehere, S., Snowberg, E. C., & Snyder, J. M., Jr. (2005). Unrepresentative information: The case of newspaper reporting on campaign finance. Public Opinion
Quarterly, 69, 213–231.
Baum, M.A., & Groeling, T. (2004). What gets covered?: How media coverage of elite
debate drives the rally-’Round-the-Flag Phenomenon, 1979–1998. Paper resented at the
Annual Meeting of the American Political Science Association.
Berinsky, A. J. (2007). Assuming the costs of war: Events, elites, and American public
support for military conflict. Journal of Politics, 69, 975–997.
Berinsky, A. J. (2009). In time of war: Understanding american public opinion from world
war II to Iraq. Chicago, IL: University of Chicago Press.
Berinsky, A. J. (Forthcoming). Beyond the water’s edge? American public opinion during
wartime, from World War II to Iraq. Chicago, IL: University of Chicago Press.
Brody, R. A. (1991). Assessing the President: The media, elite opinion and public support.
Stanford, CA: Stanford University Press.
Burk, J. (1999). Public support for peacekeeping in Lebanon and Somalia: Assessing
the casualties hypothesis. Political Science Quarterly, 114, 53–78.
Campbell, J. T., & Cain, L. S. (1965). Public opinion and the outbreak of war. The
Journal of Conflict Resolution, 9(3), 318–333.

14

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Cobb, M. D. (2007). Dispelling casualty myths: Citizens’ knowledge and news media’s coverage about casualties. Presented at the Annual Meeting of the Midwest Political
Science Association, Chicago, Illinois.
Feaver, P. D., & Gelpi, C. (2004). Choosing your battles: American civil-military relations
and the use of force. Princeton, NJ: Princeton University Press.
Gaines, B. J., Kuklinski, J. H., Quirk, P. J., Peyton, B., & Verkuilen, J. (2007). Same facts,
different interpretations: Partisan motivation and opinion on Iraq. The Journal of
Politics, 69(4), 957–974.
Gartner, S. S., & Segura, G. M. (1997). Appearances can be deceiving: Self-selection,
social group identification, and political mobilization. Rationality and Society, 9(2),
133–161.
Gartner, S. S., & Segura, G. M. (2000). Race, casualties, and opinion in the Vietnam
War. The Journal of Politics, 62, 115–146.
Gartner, S. S., Segura, G. M., & Wilkerning, M. (1997). All politics are local: Local
losses and individual attitudes toward the Vietnam War. The Journal of Conflict
Resolution, 41(5), 669–694.
Gelpi, C., Feaver, P. D., & Reifler, J. (2005/2006). Casualty sensitivity and the War in
Iraq. International Security, 30(3), 7–46.
Gilens, M. (2001). Political ignorance and collective policy preferences. The American
Political Science Review, 95(2), 379–396.
Holsti, O. R. (2004). Public opinion and American foreign policy. Ann Arbor, MI: University of Michigan Press.
Hurwitz, J., & Peffley, M. (1987). How are foreign policy attitudes structured? A hierarchical model. American Political Science Review, 81, 1099–1120.
Jentleson, B. W. (1992). The pretty prudent public: Post post-Vietnam American opinion on the use of military force. International Studies Quarterly, 36, 49–73.
Jentleson, B. W., & Britton, R. L. (1998). Still pretty prudent: Post-Cold War American public opinion on the use of military force. Journal of Conflict Resolution, 42,
395–417.
Klarevas, L. (2002). The ‘essential domino’ of military operations: American public
opinion and the use of force. International Studies Perspectives, 3(4), 417–37.
Kull, S., & Destler, I. M. (1999). Misreading the public: They myth of a new isolationism.
Washington, DC: Brookings.
Kull, S., & Ramsay, C. (2001). The myth of the reactive public: American public attitudes on military fatalities in the post-Cold War period. In P. Everts & P. Isneria
(Eds.), Public opinion and the international use of force. London, England: Routledge.
Larson, E. V. (1996). Casualties and consensus: The historical role of casualties in domestic
support for US Wars and military operations. Santa Monica, CA: The RAND Corporation, MR-726-RC.
Legro, J. W. (2000). Whence American internationalism? International Organization,
54(2), 253–289.
Mueller, J. E. (1973). War, Presidents, and public opinion. New York, NY: John Wiley &
Sons, Inc.
Page, B. I., & Shapiro, R. Y. (1992). The rational public: Fifty years of trends in Americans’
policy preferences. Chicago, IL: University of Chicago Press.

Public Opinion and International Conflict

15

Schultz, K. (2001). Democracy and coercive diplomacy. Cambridge, England: Cambridge
University Press.
Wong, C. (2007). ‘Little’ and ‘big’ pictures in our heads: Race, local context and innumeracy about racial groups in the U.S. Public Opinion Quarterly, 71, 392–412.
Zaller, J. R. (1992). The nature and origins of mass opinion. New York, NY: Cambridge
University Press.
Zaller, J. R. (1994). Elite leadership of mass opinion: New evidence from the Gulf
War. In L. Bennett & D. Paletz (Eds.), Taken by storm: The media, public opinion, and
U.S. foreign policy in the Gulf War. Chicago, IL: University of Chicago Press.

ADAM J. BERINSKY SHORT BIOGRAPHY
Adam J. Berinsky is a Professor of Political Science at MIT. Berinsky received
his PhD from the University of Michigan in 2000. He is a specialist in the fields
of political behavior and public opinion. He is the author of In Time of War:
Understanding American Public Opinion from World War II to Iraq (University
of Chicago Press, 2009) and Silent Voices: Public Opinion and Political Participation in America (Princeton University Press, 2004) and has published articles
in many scholarly journals. He has won several scholarly awards, is the recipient of grants from the National Science Foundation, and was a fellow at the
Center for Advanced Study in the Behavioral Sciences.
Webpage: http://web.mit.edu/berinsky/www/
RELATED ESSAYS
Mediation in International Conflicts (Political Science), Kyle Beardsley and
Nathan Danneman
Interdependence, Development, and Interstate Conflict (Political Science),
Erik Gartzke
Genetic Foundations of Attitude Formation (Political Science), Christian
Kandler et al.
Reconciliation and Peace-Making: Insights from Studies on Nonhuman
Animals (Anthropology), Sonja E. Koski
Why Do States Sign Alliances? (Political Science), Brett Ashley Leeds
Attitude: Construction versus Disposition (Psychology), Charles G. Lord
Political Psychology and International Conflict (Political Science), Rose
McDermott
Implicit Attitude Measures (Psychology), Gregory Mitchell and Philip E.
Tetlock
War and Social Movements (Political Science), Sidney Tarrow
Trends in the Analysis of Interstate Rivalries (Political Science), William R.
Thompson



Public Opinion and International
Conflict
ADAM J. BERINSKY

Abstract
Should the opinions of citizens in a democracy matter in decisions of war and peace?
The answer to this critical question depends on the stock we place in the ability of the
mass public to come to meaningful decisions regarding the conduct of foreign affairs.
In this essay, I examine public opinion about war over the past 75 years and make
the case that our assessment of the mass public depends in large part on the nature
of the information it receives from political leaders. Contrary to the conventional
wisdom regarding public opinion and foreign policy that emerges from scholarly
and journalistic accounts, events do not directly influence the public. Instead, citizens
learn about wars largely from political leaders. Public opinion during times of war
is therefore shaped by many of the same attachments and enmities that matter in
domestic politics. As in other areas of politics, public opinion is primarily structured
by the ebb and flow of partisan and group-based political conflict.

Should the opinions of citizens in a democracy matter in decisions of war and
peace? The answer to this critical question depends on the stock we place in
the ability of the mass public to come to meaningful decisions regarding the
conduct of foreign affairs. In this essay, I examine public opinion about war
over the past 75 years and make the case that our assessment of the mass
public depends in large part on the nature of the information it receives from
political leaders. Contrary to the conventional wisdom regarding public
opinion and foreign policy that emerges from scholarly and journalistic
accounts, events do not directly influence the public. Instead, citizens learn
about wars largely from political leaders. Public opinion during times of
war is therefore shaped by many of the same attachments and enmities
that matter in domestic politics. As in other areas of politics, public opinion
is primarily structured by the ebb and flow of partisan and group-based
political conflict.
In this essay, I first review the current state of knowledge about public opinion and war. I then briefly describe the patterns of public opinion during
Emerging Trends in the Social and Behavioral Sciences. Edited by Robert Scott and Stephen Kosslyn.
© 2015 John Wiley & Sons, Inc. ISBN 978-1-118-90077-2.

1

2

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

wartime, contrasting World War II and Iraq, to demonstrate how the public’s reaction to these wars illustrates several important lessons about public
reaction to foreign affairs.
PUBLIC OPINION AND FOREIGN POLICY
Scholarly and popular valuations of the competence of the American public have waxed and waned considerably over the past 50 years. In the early
days of survey research, scholars argued that the public opinion concerning
foreign policy was volatile and irrational—a fickle and changing “mood,” in
Almond’s (1960) words. Indeed, much of the research from the early days of
polling in the 1940s and 1950s painted a fairly dim view of democracy. Survey
researchers found that Americans were remarkably uninformed about basic
facts and developments in the international world. This view, which Holsti
(2004) has termed the “Almond–Lippmann consensus,” ruled the discussion
of public opinion concerning foreign affairs.1
Starting in the 1970s, however, this traditional view began to come under
attack by a new generation of scholars. The catalyst for a more charitable
view of the mass public might have arisen as a reaction to the Vietnam War,
as Holsti (2004) argues, or it may have been part of a more general backlash
against the dismal findings of the founders of the survey research enterprise.
Regardless of its source, the conventional wisdom that has emerged over the
past 30 years in the public opinion and foreign policy literature has taken a
dramatically different turn. Most scholars currently claim that the course of
events in a conflict directly determines public support for war.
Consider the “casualties hypothesis”—the view that the American people
will shy away from international involvement in the face of war deaths. This
hypothesis reflects Mueller’s (1973) contention that public support for war
is inversely related to the number of casualties. As the human costs of war
rise, Mueller argues, public support for conflict falls.2 Policymakers share the
belief that casualties shape public opinion concerning the wisdom of war. In
a series of interviews with political elites, Kull and Destler (1999) found that
three-fifths of congressional staff members in 1995 believed that the loss of
US lives during a UN peace keeping operation would lead to the immediate withdrawal of US troops. Leaders of other countries also subscribe to
1. For an excellent review of these early studies, see Holsti (2004).
2. Mueller actually argued that public support is related to the logged cumulative casualties, allowing early deaths to have greater effects than later deaths. As Burk (1999) notes, precise statements of
the casualties hypothesis are hard to find. But references to the general mechanism at the heart of this
hypothesis—that war deaths directly drive down public support for war—permeate popular discourse
concerning war and even the writings of military strategists and policymakers (see also Kull & Destler,
1999). Some modifications have been made to this basic theory over time. Gartner and Segura (1997, 2000),
for instance, argue that casualty rates of military personnel in local communities are an important determinant of support for the war.

Public Opinion and International Conflict

3

Mueller’s logic. In the lead-up to the first Gulf War, for instance, Saddam
Hussein claimed, “In the event of war, there will be great losses … when
5000 Americans have fallen, Bush will have to end such a war … ” (St. Louis
Dispatch, December 22, 1990). Thus, the story advanced by Mueller remains
a dominant view among both academics and policymakers (see Feaver &
Gelpi, 2004; Kull & Destler, 1999).
Moreover, in recent years, scholars have considered not only the human
costs of war but also other costs and benefits associated with military interventions. For instance, some authors contend that the possible benefits arising from the ultimate success of a mission—whether the United States will
prevail in a given conflict—determines public support for conflict (Feaver &
Gelpi, 2004; Gelpi, Feaver, & Reifler, 2005/2006; Kull & Ramsay, 2001). Other
authors argue that the motivation behind a given mission may at least indirectly affect levels of support for that conflict. For example, Jentleson and
Britton (1998) make the case that the policy objective of an intervention plays
a large role in determining whether the costs of intervention weigh greatly
in the public mind. In particular, they argue that military interventions that
are designed either to stop foreign aggression against America and its allies
or to intervene in humanitarian crises will engender greater support than
missions designed to effect change in the governments of other countries.3
Larson (1996) takes the most general stance, arguing that the collective public
decides whether to support a conflict based on a rational cost/benefit calculation. According to Larson, the greater the perceived stakes the clearer the
objectives, and the higher the probability of success the greater the level of
public support for war. These theories differ in their particulars, yet all share
the belief that “events” directly determine public support for war. Thus, even
for scholars who consider factors beyond casualties, the basic logic underlying Mueller’s argument remains the dominant position. These “event response” theories argue that the mass public is rational and will support war if,
and only if, the events of war ensure that the costs of military action are outweighed by the perceived benefits of a successful outcome.
Although event–response theories of public support for war have made
important contributions, they have several potentially serious problems.
First, such theories presume that members of the mass public at least
implicitly incorporate knowledge of political developments into their political judgments. However, this assumption does not square with political
3. Jentleson’s theory is different than other theories relating to the ongoing costs and benefits of war
because it considers the overall justification that leads to war, not the continuing development of the war.
Put another way, Jentleson’s theory speaks to the baseline levels of support for interventions sparked by
different initiating events, but cannot explain the dynamics of support for a given intervention, short of a
change in the policy objective of that intervention. Thus, while the objective of a mission may play a role
in determining absolute levels of public support, that consideration is distinct from other theories of war
discussed here.

4

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

scientists’ findings regarding the knowledge levels of the mass public.
Harkening back to the early studies of public opinion discussed earlier,
while researchers have long known that Americans know little about
politics, knowledge levels are even lower when the focus turns to specific
factual information. For instance, in 1998, only 12% of Americans knew that
the crime rate in the United States had declined over the previous decade
(Gilens, 2001; see also Ansolabehere, Snowberg, & Snyder, 2005; Wong, 2007).
Second, much research on the relationship between the particulars of
war—such as mission objective and casualty levels—and support for war
has examined differences in aggregate public support for intervention either
across different wars or at different moments in time during the same war
(Jentleson, 1992; Jentleson & Britton, 1998; Klarevas, 2002; Larson, 1996;
Mueller, 1973). With some important exceptions (Gartner & Segura, 2000;
Gartner, Segura, & Wilkerning, 1997; Gelpi et al., 2005/2006), analysis has
ignored differences among groups within the mass public during the course
of particular conflicts.4 Put another way, almost all the work described
earlier ignores the American political process. Treating the mass public as
an undifferentiated whole—innocent of attachments to political parties and
relevant social groups—leaves no room for the effect of domestic politics.
Many researchers who study public opinion and war—even those scholars
who conduct individual-level analysis—often talk about “the public” as if
it were a monolithic entity. But foreign policy is often as contentious and
partisan as domestic politics. Our understanding of war and politics must
account for the effects of the domestic political process (Baum & Groeling,
2004; Schultz, 2001).
MEDIATED REALITY: THE PRIMACY OF POLITICAL COMPETITION
Acknowledging the relative shortcoming of event–response theories of
war support does not mean that we must retreat to the dismal conclusions
regarding public opinion and foreign policy prevalent in the 1950s and
1960s. Event–response theories, after all, are not the only explanation for the
dynamics of public support for war. Another possibility is to examine the
influence of competition among political leaders on public opinion. After
all, political scientists have long known that partisan attachments shape
political judgments across a host of issues.
Zaller (1992) has used the relative positions of politicians to explain public
opinion across a number of domestic political controversies. This approach
4. In a related vein, Hurwitz and Peffley (1987) look more generally at foreign policy attitudes and
advance an individual-level hierarchical model of attitude structure. This model is incompatible with the
event-driven theories of war support, but is complementary to the elite-driven view advanced in this
essay.

Public Opinion and International Conflict

5

has become the dominant approach in the study of domestic public opinion.
Zaller argues that the balance of persuasive messages carried in the political
media determines the shape of opinion on a given policy controversy. Individuals who are most politically knowledgeable are most likely to receive
political messages and accept those messages that fit personal political predispositions. The greater the volume of elite discourse favoring a particular
policy position from leaders of a particular political stripe, the more likely it
is that the members of the mass public who share the political predispositions
of those elites will adopt that position.
Perhaps elite discourse is the key to explaining war support as well. Zaller
makes such a case in the context of the Vietnam War, arguing that the decline
in support for that war was driven by a change in the balance of elite discourse across the 1960s. In the early phase of the war, when political leaders
were almost uniform in their support for the US policy in Vietnam, Zaller
found a positive relationship between political awareness and support for
the war; those most attentive to elite discourse were most supportive of the
current policy, regardless of their other political views. Zaller terms this phenomenon the mainstream pattern of political support. On the other hand, in the
later phases of the Vietnam War, when the mainstream consensus dissolved
into elite disagreement, a “polarization pattern” emerged. Here, the effect of
political awareness on support for the war was conditional on an individual’s
political values. Citizens attentive to politics followed the path of those leaders who shared their political views. For the Vietnam War, greater awareness
led to higher levels of support among hawks and higher levels of opposition among doves. Zaller’s story is not particular to Vietnam. Zaller (1994)
also found that the changing positions of Congressional Democrats from the
summer of 1990 until the eve of the first Gulf War could explain shifts in
support for military action among Democratic identifiers in the mass public.
Theories of elite competition bring to bear lessons from the domestic arena
to the realm of foreign policy and explicitly take into account how patterns
of political competition shape public opinion. Individuals with different
political predilections react in response to different forms of elite discourse.
However, Zaller’s explanation is an incomplete account of elite influence.
Certainly, there are cases when political actors on both sides of a controversy
take clear stands, leading to polarized opinions among the mass public. But
even in the absence of arguments from both sides, individuals might have
the information they need to come to a judgment regarding the fit between
the policy options on the table and their political predispositions. As long
as one side takes a stand, citizens can use the positions of leading political
figures to come to reasonable political decisions. Specifically, even in the
absence of explicitly contradictory messages, citizens can use the position
of prominent politicians as reference points and decide whether to support

6

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

or oppose a policy. For instance, if I am a Democrat, I need only know that
George Bush supports a policy initiative to recognize that I should oppose
such a course of action; I do not need Democratic leaders to explicitly make
such a case. In effect, citizens delegate the difficult process of arriving at
an opinion on a complicated policy matter to trusted political experts.
Presidents serve as such cuegivers, especially in the realm of foreign policy.
But to use this cue requires that citizens have knowledge of the positions of relevant political actors.5 Thus, Zaller’s basic insight holds. As an
individual’s level of political information increases, her awareness of the
positions of particular leaders—and the distinctiveness of those positions
relative to other political actors—increases. Thus, a pattern of opinion
polarization could occur even in the absence of vocal opposition, provided a
strong cuegiver takes a clear position on that policy. As I illustrate later, this
alternative mechanism of elite influence—what I the elite cue theory—can
explain the pattern of opinion in World War II, when both Franklin D.
Roosevelt (FDR) and his Republican opponents took distinct positions
(Berinsky, 2007, 2009). Of course Zaller’s theory can explain this course of
events as well. However, unlike Zaller’s original formulation, this theory
can also explain the polarized pattern of opinion concerning the war in
Iraq in 2004, a situation in which Bush and Republican Party leaders took
a strong pro-war position, but Democratic Party leaders failed to express
either strong support or opposition.
WORLD WAR II
World War II provides a useful forum for demonstrating the power of the elite
cue theory. Contrary to the expectations of the casualties hypothesis, over the
almost 4 years of US involvement in that War, support for the effort did not
wane, even as war deaths mounted, in particular after the spring of 1944.6
Campbell and Cain (1965) use a number of questions to measure support for
the government’s stated military aims and demonstrate that at no point did
public support fall below 75%.
Other explanations that find the roots of continued support of the American public in wartime events are also problematic. There is a broad sense in
popular accounts and some academic treatments of World War II, that this
5. This position was actually advanced in a somewhat different form by Mueller (1973). Mueller discusses the importance of partisan cues in structuring wartime opinion and argues that well-informed
partisans are most likely to adopt the positions of the political leaders of their respective parties (see
pp. 120–121).
6. The Department of the Army recorded monthly casualties (although not war deaths) from December 7, 1941 to December 31, 1946. From December 1941 until December 1943, casualties only exceed 10,000
in 1 month. Over the course of 1944, however, monthly casualty rates increased greatly, reaching 55,000
in June and ranged from 50,000 to 80,000 until April, 1945 (Army Battle Casualties and Nonbattle Death
in World War II, Final Report).

Public Opinion and International Conflict

7

conflict was the “good war” where the benefits of intervention were clear.
According to this account, the United States, shaken by the Japanese attack
at Pearl Harbor, quickly rallied to the cause of protecting democracy. Larson
(1996), for instance, writes:
In the Second World War—‘the good war’—the public had an excellent cause.
Of course Japan’s attack on Pearl Harbor and Germany’s declaration of war on
the United States contributed greatly to support for U.S. entry into the war. But
support also derived from the shared perception of important stakes and vast
benefits of eliminating a grave threat to U.S. security and from optimism that
the outcome would be a decisive victory and punishment of the Axis powers
… Further contributing to support for the war was a desire for punishment
as a consequence of the Japanese sneak attack on Pearl Harbor, such atrocities
as the Bataan Death March, reports of the Japanese torture of U.S. prisoners of
war, and Germany’s holocaust (pp. 14–15).

These explanations may seem plausible in retrospect, but public opinion
data from the 1940s does not provide support for such accounts. Knowledge of the atrocities such as the Holocaust, discussed by Larson, was thin
during the war. In January 1943, only 47% of the population thought that
Germany was engaged in the mass destruction of Europe’s Jewish population.7 Even when a belief in the existence of concentration camps became
widespread in late 1944—when 76% of the public believed that “the Germans
have murdered many people in concentration camps”—only about a third
of respondents thought that the toll at the camps would rise above 100,000.8
Furthermore, at several points in time, Gallup and Hadley Cantril’s Office of
Public Opinion Research (OPOR) asked the public if they had “a clear idea
what the war is about.” In March 1942, almost 4 months after the attack on
Pearl Harbor, only 43% of Americans felt they had such an idea. By July, that
figure rose to 62% but, for the rest of the war, the percentage of Americans
who agreed with the statement largely fluctuated in the 65–70% range, rising to 75% in June 1944, but falling below 60% in March 1944. Thus, while
a majority of Americans could identify a war aim, a sizeable minority could
not. Certainly, the specific context of the Second World War helped engender high levels of support for the war. However, support for the US effort
at the time was not as self-evident as it was in retrospect. As Mueller (1973)
aptly notes, “the major reasons for supporting [World War II] were largely
unappreciated while it was going on” (p. 65). Thus, the existing accounts
7. The question read, “It is said that two million Jews have been killed in Europe since the war began.
Do you think this is true, or just a rumor?”
8. These figures (and the other World War II—era public opinion figures presented in this chapter)
are from the author’s analysis of opinion polling data.

8

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

that attribute continued public support to the benefits made clear by ongoing
wartime events are not supported by the data.
What, then, can explain continued public support for the war? As argued
earlier, my contention is that it was not the direct influence of events that
determined support for World War II, but rather the patterns of elite conflict
during the 1930s and 1940s. In Berinsky (2009), I describe my evidence on
this point more fully, but Legro’s (2000) study of political rhetoric in the 1930s
and 1940s is illustrative of this point. Legro finds that from 1938 through the
end of 1941, support among elites for some form of US involvement in World
War II increased generally over time. However, the gap between FDR and his
critics on the necessity and wisdom of US involvement in the Second World
War remained large. For instance, Legro finds that FDR’s critics—represented
by the editorial page of the Chicago Tribune, a paper that can be seen as
the mouthpiece of the isolationist wing of the Republican Party—moved in
an internationalist direction through 1941. However, FDR’s position consistently outpaced that of his critics. Beginning in 1939, FDR moved in a strongly
internationalist direction, but it was not until 1942 that the Tribune expressed
any support for military commitments abroad. Conversely, from 1942 on,
“the collective orthodoxy embraced the necessity of international cooperation and multilateralism” (Legro, 2000, p. 261).
Thus, elite discourse split along the lines of support for FDR before Pearl
Harbor, but presented a largely united front after the United States entered
the war. This line of argument is not intended to minimize the importance of
Pearl Harbor in shaping opinion on the war. However, given the ephemeral
nature of rally effects (Brody, 1991) it is clear that greater attention needs to
be paid to how support for the war was sustained through the nearly 4 years
of US involvement. The elite cue theory suggests that the key is the nature
of elite discourse from this time; we should see the polarization pattern of
support before US entry into the war, and—following the pattern of elite
discourse—the mainstream pattern after that point.
The full results of my analysis are presented in Berinsky (2009). But it is
rather easy to summarize the results. As predicted by the elite cue theory,
the polarization pattern largely characterizes opinion through the middle of
1941. Consistent with expectation, the pattern of public support for military
action changed greatly after the United States entered the war. In line with
the expectations of the elite cue theory, as discourse moved from a two-sided
to a one-sided flow in 1941, the public followed suit. Measured in a variety of
ways—whether the United States should send its army abroad, whether the
United States should take an active role in world affairs after the War, and,
most critically, whether the United States should make peace with Germany
if Hitler was overthrown—individuals more attuned to elite discourse were

Public Opinion and International Conflict

9

more supportive of an active United States role, regardless of their predispositions regarding FDR.9
THE WAR IN IRAQ
The Iraq War provides another interesting window into the study of public opinion concerning war. Just as with World War II, the elite cue theory
leads to clear expectations regarding the relative role of events and elites
in structuring opinion concerning the war. To repeat the argument, consistent with recent work on US public opinion, but contrary to the expectations
of many scholars and policymakers, events have little effect on the public’s
day-to-day judgments regarding the wisdom of war. Conversely, patterns of
elite discourse—the stated positions of leading Democratic and Republican
politicians—should play a large role in determining public support for war.
Individuals will use positions of prominent politicians as reference points
that guide their opinions concerning war. Moreover, contrary to Zaller, we
should expect to find divergence between partisans on the wisdom of the
Iraq War even without outspoken opposition on the part of Democratic politicians. In the rest of this section, I demonstrate this phenomenon using data
collected during 2004.
THE INABILITY OF CASUALTIES TO EXPLAIN WAR SUPPORT
The Iraq War Casualty Survey, conducted from July 23 to August 2, 2004 by
Knowledge Networks, asked a nationally representative sample of respondents:
Please give your best guess to this next question, even if you are not sure of the
correct answer. As you know, the United States is currently involved in a war
in Iraq. Do you happen to know how many soldiers of the U.S. military have
been killed in Iraq since the fighting began in March 2003?

At first glance, it appears that the public was informed about the level of
troop deaths in Iraq. The mean estimate of deaths in the sample was 952
deaths, while the median response was 900 deaths.10 Both of these figures
are extraordinarily close to the true casualty count, which rose from 901 to
915 over the span of the survey. The accuracy of the median respondent,
however, obscures large variation in the casualty estimates. Respondents
9. The 1940 election casts additional light on the importance of elite discourse in determining support
for war. See Berinsky (2009) for further discussion.
10. The mean and median estimates were generated using the poststratification weights provided by
Knowledge Networks. Fewer than 2% declined to answer the initial questions and, following a probe
asking respondents to provide their best guess even if they were not sure of the correct answer, every
respondent answered the question.

10

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

gave answers ranging from 0 deaths to 130,000 deaths. Even setting aside the
extreme responses (casualty guess under 10 and over 10,000) the standard
deviation of the casualty estimate was 802.11
A simple tabulation of the estimates illuminates the pattern of responses to
the casualty question. I scored those respondents who estimated the number
of war deaths to be between 801 and 1015 (the true estimate ±100 deaths) as
“correct.” Those who gave an estimate of 800 or lower were scored as “underestimators,” while those who guessed higher than 1015 are considered
“overestimators.”12 The modal response (47%) is a correct answer. However,
nearly as many respondents (42%) underestimated the number of war deaths
(11% overestimated the number of deaths).
Importantly, for present purposes, this variation is not random; elite
cues play a significant role in shaping casualty estimates. I predicted that
Democrats should overestimate the number of war deaths because of their
ambivalence about war, while Republicans should systematically downplay
the costs because of their support for war. Indeed, as demonstrated in
Berinsky (2007), compared to strong Republicans, strong Democrats are less
likely to underestimate and are slightly more likely to overestimate casualty
levels (see also Gaines, Kuklinski, Quirk, Peyton, & Verkuilen, 2007).
Having demonstrated that the respondents’ perceptions of events in the
Iraq War are influenced by partisanship, I next moved to the more important
question of whether casualty estimates have any influence on opinions concerning war. I measured attitudes toward the Iraq war with two commonly
used measures of war support. The first question asked, “Do you think the
U.S. made the right decision or the wrong decision in using military force
against Iraq?” The second question asked, “All in all, considering the costs
to the United States versus the benefits to the United States, do you think the
current war with Iraq has been worth fighting, or not?”
Embedded in the Iraq War Survey was an experiment in which one half
of those people who were asked to estimate how many soldiers died in
Iraq were then told, “Many people don’t know the answer to this question,
but according to the latest estimates, 901 soldiers have been killed in Iraq
since the fighting began in March 2003.”13 In other words, one-half of the
people who were asked to estimate the number of American deaths were
11. With the extreme responses included, the standard deviation was 3012.
12. I tried other methods of scoring a “correct” response—increasing and decreasing the band of
acceptable answers incrementally from ±50 deaths to ±200 deaths—and found essentially the same
results. The robustness of these findings extends beyond this particular dataset. Cobb (2007) examines a
number of surveys that measure knowledge of cumulative casualty rates from 2003 to 2006, using errors
greater than 20% as his cutoff for a “correct” answer and finds similar patterns of misperception.
13. This number was updated once on July 30th, moving the casualty figure to 908. In the analyses
presented in the paper, I set the range of the one “correct answer” (independent of the range of acceptable
answers that were scored as “correct” estimates) from the low point (901) to the high point (915) of war
deaths from this period.

Public Opinion and International Conflict

11

given the correct answer before answering the questions concerning support
for the Iraq War. This experimental design allows me to contrast levels
of support for the war between two otherwise comparable groups: (i) the
respondents in the “uncorrected” condition who underestimate casualties
(e.g., those who said that there were fewer than 800 casualties) but were
not told the correct number of war deaths; and (ii) the respondents in the
“corrected” condition who underestimate war deaths but were then told
the number of US soldiers who died.14 I can similarly compare respondents
who overestimate casualties. This is a powerful comparison, because the
only difference between the “estimate” group and the “corrected” group is
that respondents in the “corrected” condition were subsequently told the
true casualty rates. By comparing these two groups, I can assess the effect of
introducing the correct information on support for war for individuals who
are similarly misinformed about casualty rates.
The results of these analyses are presented in Berinsky (2007), but are easily
summarized here. There are essentially no differences between the respondents in the two conditions. Events—such as casualties—do not determine
whether the public supports war.
THE PRIMACY OF PARTISAN CUES
What, then, shapes public support for war? The central argument of this
essay has been that in all wars, the public reaction to a given conflict has
been deeply colored by partisan politics, as the elite cue theory predicts. The
Iraq war is no exception. From the early days of the war, public opinion was
best characterized by the polarization pattern.
In Berinsky (2009), I performed analysis similar to Zaller’s—and similar to
the World War II analysis discussed—using my Iraq War survey. Specifically,
I examined the relationship between political information levels and support
for the war, broken down by party affiliation.15 As I show in Berinsky (2009),
as a person’s attention to political discourse increases, he adopts diametrically opposed positions on the war, depending on whether he is a Democrat
or a Republican. Although there is a gap between Democrat and Republican
support for war at the lowest information levels, this gap grows as information levels increase, indicating that differences in elite positions are reflected
14. I employed a between-subjects design rather than a within-subjects design (in which support for
the war would be measured both before and after the treatment) in part because I was worried that respondents would try to maintain consistency in their answers to the war question, given the short time-span
of the interview.
15. Following Zaller (1992, 1994), I modeled the measures of support for war as a function of partisanship, information, the interaction between information and partisanship, and several control variables.
The graphs present the predicted positions for a white male with some college education. I ran the analysis separately for each of the four experimental conditions and found that the results were nearly identical
across each of the conditions (see Berinsky, (Forthcoming), for details).

12

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

in individuals’ positions on war. In Iraq, as in World War II, partisan cues
were the primary determinant of opinion concerning war.
CONCLUSION: THE PAST AND THE PRESENT
In June 2006, White House Press Secretary Tony Snow appeared on CNN
Late Edition with Wolf Blitzer. Speaking about public opinion concerning the
Iraq War, he said, “If someone had taken a poll in the Battle of the Bulge, I
dare say people would have said, ‘Wow, my goodness, what are we doing
here?’”16 Surely Snow was surprised to learn in the following days that, in
fact, someone did take a poll during the Ardennes offensive that represented
Germany’s last push of the war. From December 31, 1944 until January 4,
1945, George Gallup’s American Institute of Public Opinion asked, “If Hitler
offered to make peace now and would give up all land he has conquered,
should we try to work out a peace or should we go on fighting until the German army is completely defeated?” Contrary to Snow’s speculation, 73% of
the public expressed support for the stated US policy of unconditional surrender; the American people wanted to continue fighting until victory was
complete.
A look below the surface reveals important lessons for modern times. Support for the war crossed party lines in 1945. Of those respondents who had
voted to re-elect FDR in the 1944 election, 78% wished to continue fighting.
Among those who voted for the Republican candidate, Thomas Dewey, 73%
wanted to fight until the Germany army met complete defeat. Thus, although
war support was slightly higher among President Roosevelt’s supporters
than his opponents, this gap pales in comparison to partisans’ opinions on
the war in Iraq.
The roots of these differences can be found, in part, in the different
approaches FDR and Bush took toward the wars begun under their watch.
From 1938 through the end of 1941, support among politicians of both
parties for some form of US involvement in World War II increased generally
over time. However, the gap between FDR and his critics on the necessity
and wisdom of US involvement in the Second World War remained large.
But after United States entry into the war, FDR secured the support of
his Republican opponents, and both parties expressed a strong pro-war
message. Conversely, even before it began, the war in Iraq was strongly
associated with President Bush and his Republican allies in Congress.
Although Democratic politicians did not express open opposition to the war
effort until relatively late in the conflict, they had never joined en masse
with their Republican counterparts in openly supporting the war.
16. Available at http://transcripts.cnn.com/TRANSCRIPTS/0606/18/le.01.html. Accessed August
20, 2007.

Public Opinion and International Conflict

13

Currently, as in other times in American history, patterns of agreement and
disagreement among partisan political actors play a critical role in shaping
popular responses to war. Without the support of politicians from across the
aisle, the American people as a whole will never support any war. Among
both politicians and the mass public, the Iraq War is an exclusively Republican war.
This essay leads also to an important general lesson. If, as I have argued
here, it is the dynamics of elite conflict, rather than the events of war, that
determine public support for war, then to properly understand the decision
to go to and wage war, we need to understand how domestic politics and
partisan divisions influence the political thinking of ordinary citizens. Foreign policy and domestic policy are two sides of the same political coin. By
explicitly accounting for the elite mediation of foreign events, we can better
understand how citizens in democracies can guide and constrain the government’s ability to wage war. Even if one does not agree with Page and
Shapiro’s (1992) contention that government affairs can “conceal or misrepresent reality without being challenged,” surely political leaders have the
agency and flexibility to interpret ambiguous wartime events. Perhaps early
survey researchers took too dim a view of the mass public, but the revisionist work in the field of public opinion and foreign policy has since painted
an unrealistically optimistic picture of the ability of ordinary citizens to constrain the actions of their leaders in times of war.
REFERENCES
Almond, G. A. (1960). The American people and foreign policy. New York, NY: Praegar.
Ansolabehere, S., Snowberg, E. C., & Snyder, J. M., Jr. (2005). Unrepresentative information: The case of newspaper reporting on campaign finance. Public Opinion
Quarterly, 69, 213–231.
Baum, M.A., & Groeling, T. (2004). What gets covered?: How media coverage of elite
debate drives the rally-’Round-the-Flag Phenomenon, 1979–1998. Paper resented at the
Annual Meeting of the American Political Science Association.
Berinsky, A. J. (2007). Assuming the costs of war: Events, elites, and American public
support for military conflict. Journal of Politics, 69, 975–997.
Berinsky, A. J. (2009). In time of war: Understanding american public opinion from world
war II to Iraq. Chicago, IL: University of Chicago Press.
Berinsky, A. J. (Forthcoming). Beyond the water’s edge? American public opinion during
wartime, from World War II to Iraq. Chicago, IL: University of Chicago Press.
Brody, R. A. (1991). Assessing the President: The media, elite opinion and public support.
Stanford, CA: Stanford University Press.
Burk, J. (1999). Public support for peacekeeping in Lebanon and Somalia: Assessing
the casualties hypothesis. Political Science Quarterly, 114, 53–78.
Campbell, J. T., & Cain, L. S. (1965). Public opinion and the outbreak of war. The
Journal of Conflict Resolution, 9(3), 318–333.

14

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Cobb, M. D. (2007). Dispelling casualty myths: Citizens’ knowledge and news media’s coverage about casualties. Presented at the Annual Meeting of the Midwest Political
Science Association, Chicago, Illinois.
Feaver, P. D., & Gelpi, C. (2004). Choosing your battles: American civil-military relations
and the use of force. Princeton, NJ: Princeton University Press.
Gaines, B. J., Kuklinski, J. H., Quirk, P. J., Peyton, B., & Verkuilen, J. (2007). Same facts,
different interpretations: Partisan motivation and opinion on Iraq. The Journal of
Politics, 69(4), 957–974.
Gartner, S. S., & Segura, G. M. (1997). Appearances can be deceiving: Self-selection,
social group identification, and political mobilization. Rationality and Society, 9(2),
133–161.
Gartner, S. S., & Segura, G. M. (2000). Race, casualties, and opinion in the Vietnam
War. The Journal of Politics, 62, 115–146.
Gartner, S. S., Segura, G. M., & Wilkerning, M. (1997). All politics are local: Local
losses and individual attitudes toward the Vietnam War. The Journal of Conflict
Resolution, 41(5), 669–694.
Gelpi, C., Feaver, P. D., & Reifler, J. (2005/2006). Casualty sensitivity and the War in
Iraq. International Security, 30(3), 7–46.
Gilens, M. (2001). Political ignorance and collective policy preferences. The American
Political Science Review, 95(2), 379–396.
Holsti, O. R. (2004). Public opinion and American foreign policy. Ann Arbor, MI: University of Michigan Press.
Hurwitz, J., & Peffley, M. (1987). How are foreign policy attitudes structured? A hierarchical model. American Political Science Review, 81, 1099–1120.
Jentleson, B. W. (1992). The pretty prudent public: Post post-Vietnam American opinion on the use of military force. International Studies Quarterly, 36, 49–73.
Jentleson, B. W., & Britton, R. L. (1998). Still pretty prudent: Post-Cold War American public opinion on the use of military force. Journal of Conflict Resolution, 42,
395–417.
Klarevas, L. (2002). The ‘essential domino’ of military operations: American public
opinion and the use of force. International Studies Perspectives, 3(4), 417–37.
Kull, S., & Destler, I. M. (1999). Misreading the public: They myth of a new isolationism.
Washington, DC: Brookings.
Kull, S., & Ramsay, C. (2001). The myth of the reactive public: American public attitudes on military fatalities in the post-Cold War period. In P. Everts & P. Isneria
(Eds.), Public opinion and the international use of force. London, England: Routledge.
Larson, E. V. (1996). Casualties and consensus: The historical role of casualties in domestic
support for US Wars and military operations. Santa Monica, CA: The RAND Corporation, MR-726-RC.
Legro, J. W. (2000). Whence American internationalism? International Organization,
54(2), 253–289.
Mueller, J. E. (1973). War, Presidents, and public opinion. New York, NY: John Wiley &
Sons, Inc.
Page, B. I., & Shapiro, R. Y. (1992). The rational public: Fifty years of trends in Americans’
policy preferences. Chicago, IL: University of Chicago Press.

Public Opinion and International Conflict

15

Schultz, K. (2001). Democracy and coercive diplomacy. Cambridge, England: Cambridge
University Press.
Wong, C. (2007). ‘Little’ and ‘big’ pictures in our heads: Race, local context and innumeracy about racial groups in the U.S. Public Opinion Quarterly, 71, 392–412.
Zaller, J. R. (1992). The nature and origins of mass opinion. New York, NY: Cambridge
University Press.
Zaller, J. R. (1994). Elite leadership of mass opinion: New evidence from the Gulf
War. In L. Bennett & D. Paletz (Eds.), Taken by storm: The media, public opinion, and
U.S. foreign policy in the Gulf War. Chicago, IL: University of Chicago Press.

ADAM J. BERINSKY SHORT BIOGRAPHY
Adam J. Berinsky is a Professor of Political Science at MIT. Berinsky received
his PhD from the University of Michigan in 2000. He is a specialist in the fields
of political behavior and public opinion. He is the author of In Time of War:
Understanding American Public Opinion from World War II to Iraq (University
of Chicago Press, 2009) and Silent Voices: Public Opinion and Political Participation in America (Princeton University Press, 2004) and has published articles
in many scholarly journals. He has won several scholarly awards, is the recipient of grants from the National Science Foundation, and was a fellow at the
Center for Advanced Study in the Behavioral Sciences.
Webpage: http://web.mit.edu/berinsky/www/
RELATED ESSAYS
Mediation in International Conflicts (Political Science), Kyle Beardsley and
Nathan Danneman
Interdependence, Development, and Interstate Conflict (Political Science),
Erik Gartzke
Genetic Foundations of Attitude Formation (Political Science), Christian
Kandler et al.
Reconciliation and Peace-Making: Insights from Studies on Nonhuman
Animals (Anthropology), Sonja E. Koski
Why Do States Sign Alliances? (Political Science), Brett Ashley Leeds
Attitude: Construction versus Disposition (Psychology), Charles G. Lord
Political Psychology and International Conflict (Political Science), Rose
McDermott
Implicit Attitude Measures (Psychology), Gregory Mitchell and Philip E.
Tetlock
War and Social Movements (Political Science), Sidney Tarrow
Trends in the Analysis of Interstate Rivalries (Political Science), William R.
Thompson


Public Opinion and International
Conflict
ADAM J. BERINSKY

Abstract
Should the opinions of citizens in a democracy matter in decisions of war and peace?
The answer to this critical question depends on the stock we place in the ability of the
mass public to come to meaningful decisions regarding the conduct of foreign affairs.
In this essay, I examine public opinion about war over the past 75 years and make
the case that our assessment of the mass public depends in large part on the nature
of the information it receives from political leaders. Contrary to the conventional
wisdom regarding public opinion and foreign policy that emerges from scholarly
and journalistic accounts, events do not directly influence the public. Instead, citizens
learn about wars largely from political leaders. Public opinion during times of war
is therefore shaped by many of the same attachments and enmities that matter in
domestic politics. As in other areas of politics, public opinion is primarily structured
by the ebb and flow of partisan and group-based political conflict.

Should the opinions of citizens in a democracy matter in decisions of war and
peace? The answer to this critical question depends on the stock we place in
the ability of the mass public to come to meaningful decisions regarding the
conduct of foreign affairs. In this essay, I examine public opinion about war
over the past 75 years and make the case that our assessment of the mass
public depends in large part on the nature of the information it receives from
political leaders. Contrary to the conventional wisdom regarding public
opinion and foreign policy that emerges from scholarly and journalistic
accounts, events do not directly influence the public. Instead, citizens learn
about wars largely from political leaders. Public opinion during times of
war is therefore shaped by many of the same attachments and enmities
that matter in domestic politics. As in other areas of politics, public opinion
is primarily structured by the ebb and flow of partisan and group-based
political conflict.
In this essay, I first review the current state of knowledge about public opinion and war. I then briefly describe the patterns of public opinion during
Emerging Trends in the Social and Behavioral Sciences. Edited by Robert Scott and Stephen Kosslyn.
© 2015 John Wiley & Sons, Inc. ISBN 978-1-118-90077-2.

1

2

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

wartime, contrasting World War II and Iraq, to demonstrate how the public’s reaction to these wars illustrates several important lessons about public
reaction to foreign affairs.
PUBLIC OPINION AND FOREIGN POLICY
Scholarly and popular valuations of the competence of the American public have waxed and waned considerably over the past 50 years. In the early
days of survey research, scholars argued that the public opinion concerning
foreign policy was volatile and irrational—a fickle and changing “mood,” in
Almond’s (1960) words. Indeed, much of the research from the early days of
polling in the 1940s and 1950s painted a fairly dim view of democracy. Survey
researchers found that Americans were remarkably uninformed about basic
facts and developments in the international world. This view, which Holsti
(2004) has termed the “Almond–Lippmann consensus,” ruled the discussion
of public opinion concerning foreign affairs.1
Starting in the 1970s, however, this traditional view began to come under
attack by a new generation of scholars. The catalyst for a more charitable
view of the mass public might have arisen as a reaction to the Vietnam War,
as Holsti (2004) argues, or it may have been part of a more general backlash
against the dismal findings of the founders of the survey research enterprise.
Regardless of its source, the conventional wisdom that has emerged over the
past 30 years in the public opinion and foreign policy literature has taken a
dramatically different turn. Most scholars currently claim that the course of
events in a conflict directly determines public support for war.
Consider the “casualties hypothesis”—the view that the American people
will shy away from international involvement in the face of war deaths. This
hypothesis reflects Mueller’s (1973) contention that public support for war
is inversely related to the number of casualties. As the human costs of war
rise, Mueller argues, public support for conflict falls.2 Policymakers share the
belief that casualties shape public opinion concerning the wisdom of war. In
a series of interviews with political elites, Kull and Destler (1999) found that
three-fifths of congressional staff members in 1995 believed that the loss of
US lives during a UN peace keeping operation would lead to the immediate withdrawal of US troops. Leaders of other countries also subscribe to
1. For an excellent review of these early studies, see Holsti (2004).
2. Mueller actually argued that public support is related to the logged cumulative casualties, allowing early deaths to have greater effects than later deaths. As Burk (1999) notes, precise statements of
the casualties hypothesis are hard to find. But references to the general mechanism at the heart of this
hypothesis—that war deaths directly drive down public support for war—permeate popular discourse
concerning war and even the writings of military strategists and policymakers (see also Kull & Destler,
1999). Some modifications have been made to this basic theory over time. Gartner and Segura (1997, 2000),
for instance, argue that casualty rates of military personnel in local communities are an important determinant of support for the war.

Public Opinion and International Conflict

3

Mueller’s logic. In the lead-up to the first Gulf War, for instance, Saddam
Hussein claimed, “In the event of war, there will be great losses … when
5000 Americans have fallen, Bush will have to end such a war … ” (St. Louis
Dispatch, December 22, 1990). Thus, the story advanced by Mueller remains
a dominant view among both academics and policymakers (see Feaver &
Gelpi, 2004; Kull & Destler, 1999).
Moreover, in recent years, scholars have considered not only the human
costs of war but also other costs and benefits associated with military interventions. For instance, some authors contend that the possible benefits arising from the ultimate success of a mission—whether the United States will
prevail in a given conflict—determines public support for conflict (Feaver &
Gelpi, 2004; Gelpi, Feaver, & Reifler, 2005/2006; Kull & Ramsay, 2001). Other
authors argue that the motivation behind a given mission may at least indirectly affect levels of support for that conflict. For example, Jentleson and
Britton (1998) make the case that the policy objective of an intervention plays
a large role in determining whether the costs of intervention weigh greatly
in the public mind. In particular, they argue that military interventions that
are designed either to stop foreign aggression against America and its allies
or to intervene in humanitarian crises will engender greater support than
missions designed to effect change in the governments of other countries.3
Larson (1996) takes the most general stance, arguing that the collective public
decides whether to support a conflict based on a rational cost/benefit calculation. According to Larson, the greater the perceived stakes the clearer the
objectives, and the higher the probability of success the greater the level of
public support for war. These theories differ in their particulars, yet all share
the belief that “events” directly determine public support for war. Thus, even
for scholars who consider factors beyond casualties, the basic logic underlying Mueller’s argument remains the dominant position. These “event response” theories argue that the mass public is rational and will support war if,
and only if, the events of war ensure that the costs of military action are outweighed by the perceived benefits of a successful outcome.
Although event–response theories of public support for war have made
important contributions, they have several potentially serious problems.
First, such theories presume that members of the mass public at least
implicitly incorporate knowledge of political developments into their political judgments. However, this assumption does not square with political
3. Jentleson’s theory is different than other theories relating to the ongoing costs and benefits of war
because it considers the overall justification that leads to war, not the continuing development of the war.
Put another way, Jentleson’s theory speaks to the baseline levels of support for interventions sparked by
different initiating events, but cannot explain the dynamics of support for a given intervention, short of a
change in the policy objective of that intervention. Thus, while the objective of a mission may play a role
in determining absolute levels of public support, that consideration is distinct from other theories of war
discussed here.

4

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

scientists’ findings regarding the knowledge levels of the mass public.
Harkening back to the early studies of public opinion discussed earlier,
while researchers have long known that Americans know little about
politics, knowledge levels are even lower when the focus turns to specific
factual information. For instance, in 1998, only 12% of Americans knew that
the crime rate in the United States had declined over the previous decade
(Gilens, 2001; see also Ansolabehere, Snowberg, & Snyder, 2005; Wong, 2007).
Second, much research on the relationship between the particulars of
war—such as mission objective and casualty levels—and support for war
has examined differences in aggregate public support for intervention either
across different wars or at different moments in time during the same war
(Jentleson, 1992; Jentleson & Britton, 1998; Klarevas, 2002; Larson, 1996;
Mueller, 1973). With some important exceptions (Gartner & Segura, 2000;
Gartner, Segura, & Wilkerning, 1997; Gelpi et al., 2005/2006), analysis has
ignored differences among groups within the mass public during the course
of particular conflicts.4 Put another way, almost all the work described
earlier ignores the American political process. Treating the mass public as
an undifferentiated whole—innocent of attachments to political parties and
relevant social groups—leaves no room for the effect of domestic politics.
Many researchers who study public opinion and war—even those scholars
who conduct individual-level analysis—often talk about “the public” as if
it were a monolithic entity. But foreign policy is often as contentious and
partisan as domestic politics. Our understanding of war and politics must
account for the effects of the domestic political process (Baum & Groeling,
2004; Schultz, 2001).
MEDIATED REALITY: THE PRIMACY OF POLITICAL COMPETITION
Acknowledging the relative shortcoming of event–response theories of
war support does not mean that we must retreat to the dismal conclusions
regarding public opinion and foreign policy prevalent in the 1950s and
1960s. Event–response theories, after all, are not the only explanation for the
dynamics of public support for war. Another possibility is to examine the
influence of competition among political leaders on public opinion. After
all, political scientists have long known that partisan attachments shape
political judgments across a host of issues.
Zaller (1992) has used the relative positions of politicians to explain public
opinion across a number of domestic political controversies. This approach
4. In a related vein, Hurwitz and Peffley (1987) look more generally at foreign policy attitudes and
advance an individual-level hierarchical model of attitude structure. This model is incompatible with the
event-driven theories of war support, but is complementary to the elite-driven view advanced in this
essay.

Public Opinion and International Conflict

5

has become the dominant approach in the study of domestic public opinion.
Zaller argues that the balance of persuasive messages carried in the political
media determines the shape of opinion on a given policy controversy. Individuals who are most politically knowledgeable are most likely to receive
political messages and accept those messages that fit personal political predispositions. The greater the volume of elite discourse favoring a particular
policy position from leaders of a particular political stripe, the more likely it
is that the members of the mass public who share the political predispositions
of those elites will adopt that position.
Perhaps elite discourse is the key to explaining war support as well. Zaller
makes such a case in the context of the Vietnam War, arguing that the decline
in support for that war was driven by a change in the balance of elite discourse across the 1960s. In the early phase of the war, when political leaders
were almost uniform in their support for the US policy in Vietnam, Zaller
found a positive relationship between political awareness and support for
the war; those most attentive to elite discourse were most supportive of the
current policy, regardless of their other political views. Zaller terms this phenomenon the mainstream pattern of political support. On the other hand, in the
later phases of the Vietnam War, when the mainstream consensus dissolved
into elite disagreement, a “polarization pattern” emerged. Here, the effect of
political awareness on support for the war was conditional on an individual’s
political values. Citizens attentive to politics followed the path of those leaders who shared their political views. For the Vietnam War, greater awareness
led to higher levels of support among hawks and higher levels of opposition among doves. Zaller’s story is not particular to Vietnam. Zaller (1994)
also found that the changing positions of Congressional Democrats from the
summer of 1990 until the eve of the first Gulf War could explain shifts in
support for military action among Democratic identifiers in the mass public.
Theories of elite competition bring to bear lessons from the domestic arena
to the realm of foreign policy and explicitly take into account how patterns
of political competition shape public opinion. Individuals with different
political predilections react in response to different forms of elite discourse.
However, Zaller’s explanation is an incomplete account of elite influence.
Certainly, there are cases when political actors on both sides of a controversy
take clear stands, leading to polarized opinions among the mass public. But
even in the absence of arguments from both sides, individuals might have
the information they need to come to a judgment regarding the fit between
the policy options on the table and their political predispositions. As long
as one side takes a stand, citizens can use the positions of leading political
figures to come to reasonable political decisions. Specifically, even in the
absence of explicitly contradictory messages, citizens can use the position
of prominent politicians as reference points and decide whether to support

6

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

or oppose a policy. For instance, if I am a Democrat, I need only know that
George Bush supports a policy initiative to recognize that I should oppose
such a course of action; I do not need Democratic leaders to explicitly make
such a case. In effect, citizens delegate the difficult process of arriving at
an opinion on a complicated policy matter to trusted political experts.
Presidents serve as such cuegivers, especially in the realm of foreign policy.
But to use this cue requires that citizens have knowledge of the positions of relevant political actors.5 Thus, Zaller’s basic insight holds. As an
individual’s level of political information increases, her awareness of the
positions of particular leaders—and the distinctiveness of those positions
relative to other political actors—increases. Thus, a pattern of opinion
polarization could occur even in the absence of vocal opposition, provided a
strong cuegiver takes a clear position on that policy. As I illustrate later, this
alternative mechanism of elite influence—what I the elite cue theory—can
explain the pattern of opinion in World War II, when both Franklin D.
Roosevelt (FDR) and his Republican opponents took distinct positions
(Berinsky, 2007, 2009). Of course Zaller’s theory can explain this course of
events as well. However, unlike Zaller’s original formulation, this theory
can also explain the polarized pattern of opinion concerning the war in
Iraq in 2004, a situation in which Bush and Republican Party leaders took
a strong pro-war position, but Democratic Party leaders failed to express
either strong support or opposition.
WORLD WAR II
World War II provides a useful forum for demonstrating the power of the elite
cue theory. Contrary to the expectations of the casualties hypothesis, over the
almost 4 years of US involvement in that War, support for the effort did not
wane, even as war deaths mounted, in particular after the spring of 1944.6
Campbell and Cain (1965) use a number of questions to measure support for
the government’s stated military aims and demonstrate that at no point did
public support fall below 75%.
Other explanations that find the roots of continued support of the American public in wartime events are also problematic. There is a broad sense in
popular accounts and some academic treatments of World War II, that this
5. This position was actually advanced in a somewhat different form by Mueller (1973). Mueller discusses the importance of partisan cues in structuring wartime opinion and argues that well-informed
partisans are most likely to adopt the positions of the political leaders of their respective parties (see
pp. 120–121).
6. The Department of the Army recorded monthly casualties (although not war deaths) from December 7, 1941 to December 31, 1946. From December 1941 until December 1943, casualties only exceed 10,000
in 1 month. Over the course of 1944, however, monthly casualty rates increased greatly, reaching 55,000
in June and ranged from 50,000 to 80,000 until April, 1945 (Army Battle Casualties and Nonbattle Death
in World War II, Final Report).

Public Opinion and International Conflict

7

conflict was the “good war” where the benefits of intervention were clear.
According to this account, the United States, shaken by the Japanese attack
at Pearl Harbor, quickly rallied to the cause of protecting democracy. Larson
(1996), for instance, writes:
In the Second World War—‘the good war’—the public had an excellent cause.
Of course Japan’s attack on Pearl Harbor and Germany’s declaration of war on
the United States contributed greatly to support for U.S. entry into the war. But
support also derived from the shared perception of important stakes and vast
benefits of eliminating a grave threat to U.S. security and from optimism that
the outcome would be a decisive victory and punishment of the Axis powers
… Further contributing to support for the war was a desire for punishment
as a consequence of the Japanese sneak attack on Pearl Harbor, such atrocities
as the Bataan Death March, reports of the Japanese torture of U.S. prisoners of
war, and Germany’s holocaust (pp. 14–15).

These explanations may seem plausible in retrospect, but public opinion
data from the 1940s does not provide support for such accounts. Knowledge of the atrocities such as the Holocaust, discussed by Larson, was thin
during the war. In January 1943, only 47% of the population thought that
Germany was engaged in the mass destruction of Europe’s Jewish population.7 Even when a belief in the existence of concentration camps became
widespread in late 1944—when 76% of the public believed that “the Germans
have murdered many people in concentration camps”—only about a third
of respondents thought that the toll at the camps would rise above 100,000.8
Furthermore, at several points in time, Gallup and Hadley Cantril’s Office of
Public Opinion Research (OPOR) asked the public if they had “a clear idea
what the war is about.” In March 1942, almost 4 months after the attack on
Pearl Harbor, only 43% of Americans felt they had such an idea. By July, that
figure rose to 62% but, for the rest of the war, the percentage of Americans
who agreed with the statement largely fluctuated in the 65–70% range, rising to 75% in June 1944, but falling below 60% in March 1944. Thus, while
a majority of Americans could identify a war aim, a sizeable minority could
not. Certainly, the specific context of the Second World War helped engender high levels of support for the war. However, support for the US effort
at the time was not as self-evident as it was in retrospect. As Mueller (1973)
aptly notes, “the major reasons for supporting [World War II] were largely
unappreciated while it was going on” (p. 65). Thus, the existing accounts
7. The question read, “It is said that two million Jews have been killed in Europe since the war began.
Do you think this is true, or just a rumor?”
8. These figures (and the other World War II—era public opinion figures presented in this chapter)
are from the author’s analysis of opinion polling data.

8

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

that attribute continued public support to the benefits made clear by ongoing
wartime events are not supported by the data.
What, then, can explain continued public support for the war? As argued
earlier, my contention is that it was not the direct influence of events that
determined support for World War II, but rather the patterns of elite conflict
during the 1930s and 1940s. In Berinsky (2009), I describe my evidence on
this point more fully, but Legro’s (2000) study of political rhetoric in the 1930s
and 1940s is illustrative of this point. Legro finds that from 1938 through the
end of 1941, support among elites for some form of US involvement in World
War II increased generally over time. However, the gap between FDR and his
critics on the necessity and wisdom of US involvement in the Second World
War remained large. For instance, Legro finds that FDR’s critics—represented
by the editorial page of the Chicago Tribune, a paper that can be seen as
the mouthpiece of the isolationist wing of the Republican Party—moved in
an internationalist direction through 1941. However, FDR’s position consistently outpaced that of his critics. Beginning in 1939, FDR moved in a strongly
internationalist direction, but it was not until 1942 that the Tribune expressed
any support for military commitments abroad. Conversely, from 1942 on,
“the collective orthodoxy embraced the necessity of international cooperation and multilateralism” (Legro, 2000, p. 261).
Thus, elite discourse split along the lines of support for FDR before Pearl
Harbor, but presented a largely united front after the United States entered
the war. This line of argument is not intended to minimize the importance of
Pearl Harbor in shaping opinion on the war. However, given the ephemeral
nature of rally effects (Brody, 1991) it is clear that greater attention needs to
be paid to how support for the war was sustained through the nearly 4 years
of US involvement. The elite cue theory suggests that the key is the nature
of elite discourse from this time; we should see the polarization pattern of
support before US entry into the war, and—following the pattern of elite
discourse—the mainstream pattern after that point.
The full results of my analysis are presented in Berinsky (2009). But it is
rather easy to summarize the results. As predicted by the elite cue theory,
the polarization pattern largely characterizes opinion through the middle of
1941. Consistent with expectation, the pattern of public support for military
action changed greatly after the United States entered the war. In line with
the expectations of the elite cue theory, as discourse moved from a two-sided
to a one-sided flow in 1941, the public followed suit. Measured in a variety of
ways—whether the United States should send its army abroad, whether the
United States should take an active role in world affairs after the War, and,
most critically, whether the United States should make peace with Germany
if Hitler was overthrown—individuals more attuned to elite discourse were

Public Opinion and International Conflict

9

more supportive of an active United States role, regardless of their predispositions regarding FDR.9
THE WAR IN IRAQ
The Iraq War provides another interesting window into the study of public opinion concerning war. Just as with World War II, the elite cue theory
leads to clear expectations regarding the relative role of events and elites
in structuring opinion concerning the war. To repeat the argument, consistent with recent work on US public opinion, but contrary to the expectations
of many scholars and policymakers, events have little effect on the public’s
day-to-day judgments regarding the wisdom of war. Conversely, patterns of
elite discourse—the stated positions of leading Democratic and Republican
politicians—should play a large role in determining public support for war.
Individuals will use positions of prominent politicians as reference points
that guide their opinions concerning war. Moreover, contrary to Zaller, we
should expect to find divergence between partisans on the wisdom of the
Iraq War even without outspoken opposition on the part of Democratic politicians. In the rest of this section, I demonstrate this phenomenon using data
collected during 2004.
THE INABILITY OF CASUALTIES TO EXPLAIN WAR SUPPORT
The Iraq War Casualty Survey, conducted from July 23 to August 2, 2004 by
Knowledge Networks, asked a nationally representative sample of respondents:
Please give your best guess to this next question, even if you are not sure of the
correct answer. As you know, the United States is currently involved in a war
in Iraq. Do you happen to know how many soldiers of the U.S. military have
been killed in Iraq since the fighting began in March 2003?

At first glance, it appears that the public was informed about the level of
troop deaths in Iraq. The mean estimate of deaths in the sample was 952
deaths, while the median response was 900 deaths.10 Both of these figures
are extraordinarily close to the true casualty count, which rose from 901 to
915 over the span of the survey. The accuracy of the median respondent,
however, obscures large variation in the casualty estimates. Respondents
9. The 1940 election casts additional light on the importance of elite discourse in determining support
for war. See Berinsky (2009) for further discussion.
10. The mean and median estimates were generated using the poststratification weights provided by
Knowledge Networks. Fewer than 2% declined to answer the initial questions and, following a probe
asking respondents to provide their best guess even if they were not sure of the correct answer, every
respondent answered the question.

10

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

gave answers ranging from 0 deaths to 130,000 deaths. Even setting aside the
extreme responses (casualty guess under 10 and over 10,000) the standard
deviation of the casualty estimate was 802.11
A simple tabulation of the estimates illuminates the pattern of responses to
the casualty question. I scored those respondents who estimated the number
of war deaths to be between 801 and 1015 (the true estimate ±100 deaths) as
“correct.” Those who gave an estimate of 800 or lower were scored as “underestimators,” while those who guessed higher than 1015 are considered
“overestimators.”12 The modal response (47%) is a correct answer. However,
nearly as many respondents (42%) underestimated the number of war deaths
(11% overestimated the number of deaths).
Importantly, for present purposes, this variation is not random; elite
cues play a significant role in shaping casualty estimates. I predicted that
Democrats should overestimate the number of war deaths because of their
ambivalence about war, while Republicans should systematically downplay
the costs because of their support for war. Indeed, as demonstrated in
Berinsky (2007), compared to strong Republicans, strong Democrats are less
likely to underestimate and are slightly more likely to overestimate casualty
levels (see also Gaines, Kuklinski, Quirk, Peyton, & Verkuilen, 2007).
Having demonstrated that the respondents’ perceptions of events in the
Iraq War are influenced by partisanship, I next moved to the more important
question of whether casualty estimates have any influence on opinions concerning war. I measured attitudes toward the Iraq war with two commonly
used measures of war support. The first question asked, “Do you think the
U.S. made the right decision or the wrong decision in using military force
against Iraq?” The second question asked, “All in all, considering the costs
to the United States versus the benefits to the United States, do you think the
current war with Iraq has been worth fighting, or not?”
Embedded in the Iraq War Survey was an experiment in which one half
of those people who were asked to estimate how many soldiers died in
Iraq were then told, “Many people don’t know the answer to this question,
but according to the latest estimates, 901 soldiers have been killed in Iraq
since the fighting began in March 2003.”13 In other words, one-half of the
people who were asked to estimate the number of American deaths were
11. With the extreme responses included, the standard deviation was 3012.
12. I tried other methods of scoring a “correct” response—increasing and decreasing the band of
acceptable answers incrementally from ±50 deaths to ±200 deaths—and found essentially the same
results. The robustness of these findings extends beyond this particular dataset. Cobb (2007) examines a
number of surveys that measure knowledge of cumulative casualty rates from 2003 to 2006, using errors
greater than 20% as his cutoff for a “correct” answer and finds similar patterns of misperception.
13. This number was updated once on July 30th, moving the casualty figure to 908. In the analyses
presented in the paper, I set the range of the one “correct answer” (independent of the range of acceptable
answers that were scored as “correct” estimates) from the low point (901) to the high point (915) of war
deaths from this period.

Public Opinion and International Conflict

11

given the correct answer before answering the questions concerning support
for the Iraq War. This experimental design allows me to contrast levels
of support for the war between two otherwise comparable groups: (i) the
respondents in the “uncorrected” condition who underestimate casualties
(e.g., those who said that there were fewer than 800 casualties) but were
not told the correct number of war deaths; and (ii) the respondents in the
“corrected” condition who underestimate war deaths but were then told
the number of US soldiers who died.14 I can similarly compare respondents
who overestimate casualties. This is a powerful comparison, because the
only difference between the “estimate” group and the “corrected” group is
that respondents in the “corrected” condition were subsequently told the
true casualty rates. By comparing these two groups, I can assess the effect of
introducing the correct information on support for war for individuals who
are similarly misinformed about casualty rates.
The results of these analyses are presented in Berinsky (2007), but are easily
summarized here. There are essentially no differences between the respondents in the two conditions. Events—such as casualties—do not determine
whether the public supports war.
THE PRIMACY OF PARTISAN CUES
What, then, shapes public support for war? The central argument of this
essay has been that in all wars, the public reaction to a given conflict has
been deeply colored by partisan politics, as the elite cue theory predicts. The
Iraq war is no exception. From the early days of the war, public opinion was
best characterized by the polarization pattern.
In Berinsky (2009), I performed analysis similar to Zaller’s—and similar to
the World War II analysis discussed—using my Iraq War survey. Specifically,
I examined the relationship between political information levels and support
for the war, broken down by party affiliation.15 As I show in Berinsky (2009),
as a person’s attention to political discourse increases, he adopts diametrically opposed positions on the war, depending on whether he is a Democrat
or a Republican. Although there is a gap between Democrat and Republican
support for war at the lowest information levels, this gap grows as information levels increase, indicating that differences in elite positions are reflected
14. I employed a between-subjects design rather than a within-subjects design (in which support for
the war would be measured both before and after the treatment) in part because I was worried that respondents would try to maintain consistency in their answers to the war question, given the short time-span
of the interview.
15. Following Zaller (1992, 1994), I modeled the measures of support for war as a function of partisanship, information, the interaction between information and partisanship, and several control variables.
The graphs present the predicted positions for a white male with some college education. I ran the analysis separately for each of the four experimental conditions and found that the results were nearly identical
across each of the conditions (see Berinsky, (Forthcoming), for details).

12

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

in individuals’ positions on war. In Iraq, as in World War II, partisan cues
were the primary determinant of opinion concerning war.
CONCLUSION: THE PAST AND THE PRESENT
In June 2006, White House Press Secretary Tony Snow appeared on CNN
Late Edition with Wolf Blitzer. Speaking about public opinion concerning the
Iraq War, he said, “If someone had taken a poll in the Battle of the Bulge, I
dare say people would have said, ‘Wow, my goodness, what are we doing
here?’”16 Surely Snow was surprised to learn in the following days that, in
fact, someone did take a poll during the Ardennes offensive that represented
Germany’s last push of the war. From December 31, 1944 until January 4,
1945, George Gallup’s American Institute of Public Opinion asked, “If Hitler
offered to make peace now and would give up all land he has conquered,
should we try to work out a peace or should we go on fighting until the German army is completely defeated?” Contrary to Snow’s speculation, 73% of
the public expressed support for the stated US policy of unconditional surrender; the American people wanted to continue fighting until victory was
complete.
A look below the surface reveals important lessons for modern times. Support for the war crossed party lines in 1945. Of those respondents who had
voted to re-elect FDR in the 1944 election, 78% wished to continue fighting.
Among those who voted for the Republican candidate, Thomas Dewey, 73%
wanted to fight until the Germany army met complete defeat. Thus, although
war support was slightly higher among President Roosevelt’s supporters
than his opponents, this gap pales in comparison to partisans’ opinions on
the war in Iraq.
The roots of these differences can be found, in part, in the different
approaches FDR and Bush took toward the wars begun under their watch.
From 1938 through the end of 1941, support among politicians of both
parties for some form of US involvement in World War II increased generally
over time. However, the gap between FDR and his critics on the necessity
and wisdom of US involvement in the Second World War remained large.
But after United States entry into the war, FDR secured the support of
his Republican opponents, and both parties expressed a strong pro-war
message. Conversely, even before it began, the war in Iraq was strongly
associated with President Bush and his Republican allies in Congress.
Although Democratic politicians did not express open opposition to the war
effort until relatively late in the conflict, they had never joined en masse
with their Republican counterparts in openly supporting the war.
16. Available at http://transcripts.cnn.com/TRANSCRIPTS/0606/18/le.01.html. Accessed August
20, 2007.

Public Opinion and International Conflict

13

Currently, as in other times in American history, patterns of agreement and
disagreement among partisan political actors play a critical role in shaping
popular responses to war. Without the support of politicians from across the
aisle, the American people as a whole will never support any war. Among
both politicians and the mass public, the Iraq War is an exclusively Republican war.
This essay leads also to an important general lesson. If, as I have argued
here, it is the dynamics of elite conflict, rather than the events of war, that
determine public support for war, then to properly understand the decision
to go to and wage war, we need to understand how domestic politics and
partisan divisions influence the political thinking of ordinary citizens. Foreign policy and domestic policy are two sides of the same political coin. By
explicitly accounting for the elite mediation of foreign events, we can better
understand how citizens in democracies can guide and constrain the government’s ability to wage war. Even if one does not agree with Page and
Shapiro’s (1992) contention that government affairs can “conceal or misrepresent reality without being challenged,” surely political leaders have the
agency and flexibility to interpret ambiguous wartime events. Perhaps early
survey researchers took too dim a view of the mass public, but the revisionist work in the field of public opinion and foreign policy has since painted
an unrealistically optimistic picture of the ability of ordinary citizens to constrain the actions of their leaders in times of war.
REFERENCES
Almond, G. A. (1960). The American people and foreign policy. New York, NY: Praegar.
Ansolabehere, S., Snowberg, E. C., & Snyder, J. M., Jr. (2005). Unrepresentative information: The case of newspaper reporting on campaign finance. Public Opinion
Quarterly, 69, 213–231.
Baum, M.A., & Groeling, T. (2004). What gets covered?: How media coverage of elite
debate drives the rally-’Round-the-Flag Phenomenon, 1979–1998. Paper resented at the
Annual Meeting of the American Political Science Association.
Berinsky, A. J. (2007). Assuming the costs of war: Events, elites, and American public
support for military conflict. Journal of Politics, 69, 975–997.
Berinsky, A. J. (2009). In time of war: Understanding american public opinion from world
war II to Iraq. Chicago, IL: University of Chicago Press.
Berinsky, A. J. (Forthcoming). Beyond the water’s edge? American public opinion during
wartime, from World War II to Iraq. Chicago, IL: University of Chicago Press.
Brody, R. A. (1991). Assessing the President: The media, elite opinion and public support.
Stanford, CA: Stanford University Press.
Burk, J. (1999). Public support for peacekeeping in Lebanon and Somalia: Assessing
the casualties hypothesis. Political Science Quarterly, 114, 53–78.
Campbell, J. T., & Cain, L. S. (1965). Public opinion and the outbreak of war. The
Journal of Conflict Resolution, 9(3), 318–333.

14

EMERGING TRENDS IN THE SOCIAL AND BEHAVIORAL SCIENCES

Cobb, M. D. (2007). Dispelling casualty myths: Citizens’ knowledge and news media’s coverage about casualties. Presented at the Annual Meeting of the Midwest Political
Science Association, Chicago, Illinois.
Feaver, P. D., & Gelpi, C. (2004). Choosing your battles: American civil-military relations
and the use of force. Princeton, NJ: Princeton University Press.
Gaines, B. J., Kuklinski, J. H., Quirk, P. J., Peyton, B., & Verkuilen, J. (2007). Same facts,
different interpretations: Partisan motivation and opinion on Iraq. The Journal of
Politics, 69(4), 957–974.
Gartner, S. S., & Segura, G. M. (1997). Appearances can be deceiving: Self-selection,
social group identification, and political mobilization. Rationality and Society, 9(2),
133–161.
Gartner, S. S., & Segura, G. M. (2000). Race, casualties, and opinion in the Vietnam
War. The Journal of Politics, 62, 115–146.
Gartner, S. S., Segura, G. M., & Wilkerning, M. (1997). All politics are local: Local
losses and individual attitudes toward the Vietnam War. The Journal of Conflict
Resolution, 41(5), 669–694.
Gelpi, C., Feaver, P. D., & Reifler, J. (2005/2006). Casualty sensitivity and the War in
Iraq. International Security, 30(3), 7–46.
Gilens, M. (2001). Political ignorance and collective policy preferences. The American
Political Science Review, 95(2), 379–396.
Holsti, O. R. (2004). Public opinion and American foreign policy. Ann Arbor, MI: University of Michigan Press.
Hurwitz, J., & Peffley, M. (1987). How are foreign policy attitudes structured? A hierarchical model. American Political Science Review, 81, 1099–1120.
Jentleson, B. W. (1992). The pretty prudent public: Post post-Vietnam American opinion on the use of military force. International Studies Quarterly, 36, 49–73.
Jentleson, B. W., & Britton, R. L. (1998). Still pretty prudent: Post-Cold War American public opinion on the use of military force. Journal of Conflict Resolution, 42,
395–417.
Klarevas, L. (2002). The ‘essential domino’ of military operations: American public
opinion and the use of force. International Studies Perspectives, 3(4), 417–37.
Kull, S., & Destler, I. M. (1999). Misreading the public: They myth of a new isolationism.
Washington, DC: Brookings.
Kull, S., & Ramsay, C. (2001). The myth of the reactive public: American public attitudes on military fatalities in the post-Cold War period. In P. Everts & P. Isneria
(Eds.), Public opinion and the international use of force. London, England: Routledge.
Larson, E. V. (1996). Casualties and consensus: The historical role of casualties in domestic
support for US Wars and military operations. Santa Monica, CA: The RAND Corporation, MR-726-RC.
Legro, J. W. (2000). Whence American internationalism? International Organization,
54(2), 253–289.
Mueller, J. E. (1973). War, Presidents, and public opinion. New York, NY: John Wiley &
Sons, Inc.
Page, B. I., & Shapiro, R. Y. (1992). The rational public: Fifty years of trends in Americans’
policy preferences. Chicago, IL: University of Chicago Press.

Public Opinion and International Conflict

15

Schultz, K. (2001). Democracy and coercive diplomacy. Cambridge, England: Cambridge
University Press.
Wong, C. (2007). ‘Little’ and ‘big’ pictures in our heads: Race, local context and innumeracy about racial groups in the U.S. Public Opinion Quarterly, 71, 392–412.
Zaller, J. R. (1992). The nature and origins of mass opinion. New York, NY: Cambridge
University Press.
Zaller, J. R. (1994). Elite leadership of mass opinion: New evidence from the Gulf
War. In L. Bennett & D. Paletz (Eds.), Taken by storm: The media, public opinion, and
U.S. foreign policy in the Gulf War. Chicago, IL: University of Chicago Press.

ADAM J. BERINSKY SHORT BIOGRAPHY
Adam J. Berinsky is a Professor of Political Science at MIT. Berinsky received
his PhD from the University of Michigan in 2000. He is a specialist in the fields
of political behavior and public opinion. He is the author of In Time of War:
Understanding American Public Opinion from World War II to Iraq (University
of Chicago Press, 2009) and Silent Voices: Public Opinion and Political Participation in America (Princeton University Press, 2004) and has published articles
in many scholarly journals. He has won several scholarly awards, is the recipient of grants from the National Science Foundation, and was a fellow at the
Center for Advanced Study in the Behavioral Sciences.
Webpage: http://web.mit.edu/berinsky/www/
RELATED ESSAYS
Mediation in International Conflicts (Political Science), Kyle Beardsley and
Nathan Danneman
Interdependence, Development, and Interstate Conflict (Political Science),
Erik Gartzke
Genetic Foundations of Attitude Formation (Political Science), Christian
Kandler et al.
Reconciliation and Peace-Making: Insights from Studies on Nonhuman
Animals (Anthropology), Sonja E. Koski
Why Do States Sign Alliances? (Political Science), Brett Ashley Leeds
Attitude: Construction versus Disposition (Psychology), Charles G. Lord
Political Psychology and International Conflict (Political Science), Rose
McDermott
Implicit Attitude Measures (Psychology), Gregory Mitchell and Philip E.
Tetlock
War and Social Movements (Political Science), Sidney Tarrow
Trends in the Analysis of Interstate Rivalries (Political Science), William R.
Thompson



Media

Public Opinion and International Conflict