The Anti-Realist Assumption

Realism has a long and distinguished intellectual lineage, which is another way of saying that there are many varieties of realisms: classical realism, neoclassical realism, post-classical realism, offensive realism, defensive realism, structural neorealism, etc.⁶

Realists share at least one common belief, however—Americans do not like their policy wares. Sixty years ago, classical realists grounded this assumption on two arguments. The first concern, which stretches back to Thucydides, was that democracies had a disadvantage in crafting foreign policy vis-à-vis authoritarian governments because of the need to appease a mass public that holds inchoate views about international relations.⁷ While this concern applies to all democratic publics, the United States has been considered an exemplar of this problem. Indeed, elite wariness of mass American attitudes towards foreign policy has been around since the days of Walter Lippmann. In The Public Philosophy he warned, “The unhappy truth is that the prevailing public opinion has been destructively wrong at the critical junctures. The people have imposed a veto upon the judgments of informed and responsible officials.”⁸

The first wave of American public opinion research led to the Almond-Lippmann consensus, which suggested that mass attitudes towards foreign policy were inconstant, irrational, and ill-considered.⁹

Because the mass public was so uninformed about foreign affairs, scholars argued that their reaction to current events would be based on emotion rather than reason. This leads to a public with erratic mood swings about foreign policy issues of the day. A recent variation of this concern can be seen in debates about the “CNN effect.”¹⁰ In 1993, George Kennan described this effect as the tendency for American foreign policy to be, “controlled by popular emotional impulses, and particularly ones provoked by the commercial television industry.”¹¹

The second concern was that Americans filter foreign policy decisions through a moralistic system of beliefs. This renders the mass public unable to digest the realist logic of a dispassionate, hard-headed national interest.¹²

Hans Morgenthau wrote that “the statesman must think in terms of the national interest, conceived as power among other powers. The popular mind, unaware of the fine distinctions of the statesman's thinking, reasons more often than not in the simple moralistic and legalistic terms of absolute good and absolute evil.”¹³ George Kennan warned that the “legalistic-moralistic approach to international problems” would resonate among the mass public.¹⁴ Henry Kissinger concludes, with more than a tinge of regret, that “it is above all to the drumbeat of Wilsonian idealism that American foreign policy has marched since his watershed presidency, and continues to march to this day….Wilson's historic achievement lies in his recognition that Americans cannot sustain major international obligations that are not justified by their moral faith.”¹⁵

These concerns have not abated as realist thought has adopted a more scientific cast.¹⁶

If anything, the reverse has been true—realists now assert that American political development has made the country and its citizens uniquely unable to accept the tenets of realism.¹⁷ Stephen D. Krasner concluded that “U.S. policy after the Second World War must be understood in terms of ideology: leaders were driven by a vision of what the global order should be like…. The vision itself was a manifestation of American liberalism.”¹⁸ John Mearsheimer is even more outspoken about the incompatibility between realist theory and American public opinion. Although Mearsheimer argues that offensive realism can explain U.S. foreign policy outputs, the theory cuts against American beliefs:

Scholars argue that the exceptionalist history of the United States allowed a liberal worldview of international relations to develop unfettered by realpolitik concerns. Separated from Europe, the United States emerged as a great power relatively unscathed by foreign invasion, military occupation, or economic deprivation. There was no need for realism when no great power could overcome the stopping power of water.²⁰

A liberal approach to foreign affairs was most symbiotic with the liberal tradition in domestic politics.²¹ Liberal internationalism is more ideationally consistent within America's self-image as “a shining city on a hill.” Democracy promotion in particular is rooted in America's national identity and national security strategy.²² As George W. Bush phrased it in his second inaugural address:

Bush's rhetoric underscores the fact that realists are not the only analysts who make this assumption. Michael Lind argues the ideas of realpolitik fundamentally conflict with the “American School” of foreign policy.²⁴

Walter Russell Mead argues that “continental realism” is ill-suited to the United States.²⁵ Within the academy, liberal institutionalists, constructivists, and public opinion analysts have echoed this line of argument as well.²⁶ Michael Doyle asserts that, “in the United States, and in other liberal states to a lesser degree, public policy derives its legitimacy from its concordance with liberal principles. Policies not rooted in liberal principles generally fail to sustain long term public support.”²⁷ Indeed, the norms variant of the democratic peace hypothesis is predicated on the idea that democratic publics hold different sets of values on foreign policy issues than realpolitik statesmen. More recently, neoconservatives have insisted that realism is antithetical to American values. Robert Kagan and William Kristol have argued, for example, that precisely because Americans do not think like power-maximizing realists, other states would not balance against the U.S.²⁸

It should also be noted that the anti-realist assumption is not limited to the United States. Ironically, America's most recent foreign adversaries have also made this assumption. In the past decade alone, Slobodan Milosevic, Saddam Hussein, and Osama bin Laden all assumed that the American public was reluctant to apply force in a ruthless, realpolitik manner. They all assumed that the mass public was inordinately sensitive to casualties, and that this sensitivity would constrain American leaders. The anti-realist assumption led each of them to underestimate the likelihood of American military action.²⁹

To be sure, this logic behind the anti-realist assumption is not universally accepted as a given. Numerous historians of American foreign policy have argued that multiple traditions have competed for dominance within American foreign policy, some of which are perfectly compatible with realism.³⁰

It should also be noted that the anti-realist assumption does not mean that realists believe their model cannot explain U.S. foreign policy.³¹ Mearsheimer, for example, asserts that “we act according to the dictates of realpolitik, but we justify our policies in terms of liberal ideologies. So what is going on here is that in many cases, elites speak one language [in public], and act according to a different logic and speak a different language behind closed doors.”³² At a minimum, however, there is a strong consensus across a broad spectrum of international relations theorists and public commentators: powerful ideological currents within American political development cause generations of Americans to repeatedly reject the logic of realpolitik.

Testing the Assumption

Realists share a consensus about how the structure of the international system works, but that consensus partially breaks down at the level of foreign policy recommendations.³³

Realists debate amongst themselves the importance of balancing, the importance of maximizing power or security, and the optimal grand strategy for American foreign policy.³⁴ Despite these conceptual disagreements, realists do share a core of policy recommendations that distinguishes them from other foreign policy paradigms. Realists stress that the anarchic world structure makes it impossible for governments to fully trust each other, forcing all states to be guided solely by national interest. Because all states can only count on their own resources and capabilities, realists are skeptical of the utility of international institutions to regulate world politics. Realists advise paying attention to relative gains when considering cooperation. The question, for realists, is not “will both of us gain?” but “who will gain more?”³⁵ The U.S. government should also be prepared to balance against rising great powers, before they amass sufficient power to become a peer competitor.³⁶ This paradigm also has a large number of negative policy prescriptions. Realism is uninterested in the domestic politics of other countries, and recommends against interventions to promote democracy or human rights. Realists are wary that this kind of policy activism could drain scarce resources or trigger the security dilemma.³⁷ They also caution against excessive economic interdependence, both because of relative gains concerns and fears of asymmetric vulnerability.

Table 1 lists the set of testable predictions that flow from the anti-realist assumption across three areas. The category of foreign policy priorities and worldviews considers how Americans views international politics, and how they would rank order particular objectives for the American government to achieve. To realists, the international system is a Hobbesian one, in which anarchy makes it impossible to fully trust other countries. The United States' top priority should therefore be the defense of American borders and territorial integrity. Alliances should be based on geopolitics and not regime type. The U.S. government should also be prepared to balance against rising great powers, before they amass sufficient power to become a peer competitor—again, regardless of regime type. Beyond those obligations, however, realists would not support other “internationalist” policy priorities—the protection of human rights, democracy promotion, or the strengthening of multilateral institutions.

The anti-realist assumption predicts that the American mass public should be more likely to trust other countries and prefer cooperation over self-help. They are therefore expected to place a greater reliance on international law and multilateral institutions to regulate world conflict. In terms of substantive preferences, Americans are predicted to advocate for the spread of liberal norms to other countries. The active promotion of democracy, human rights, and economic development to other countries cannot be defined as realist.³⁸

Policies that contravene liberal values—alliances with unsavory regimes, or tolerance of human rights abuses to advance the national interest—should register significant opposition. Strong American support for liberal objectives would substantiate the anti-realist assumption.

Realists would also be expected to differ with the mass American public on the justification and use of force. For realists, again, the top priority would be preserving the territorial integrity, homeland security and regional hegemony of the United States. Because realists place great value on state sovereignty, they would be expected to support the use of force to deter or repel a violation of that sovereignty. Similarly, realists would support military measures to contain rising powers. When war breaks out, realists can tolerate a heavy loss of men and material—provided the opponent suffers even greater costs.

The anti-realist assumption predicts that Americans would support the use of force for a much wider set of policy objectives, but not at a sustainable level. To ensure the spread of liberal norms in world politics Americans would be expected to support other types of military missions, such as humanitarian interventions, multilateral peacekeeping, and intervention in civil conflicts. It is safe to say that realists would disapprove of all of these activities.³⁹

Despite the support for a broader category of military actions, however, Americans are also simultaneously hypothesized to be squeamish about the costs of prolonged military commitments.⁴⁰ As Peter Feaver and Christopher Gelpi observe, “There is an implicit comparison between a naïve public and a ‘hard-nosed’ view of what it takes to be a great power (or the only superpower).”⁴¹ Americans are expected to sour quickly on uses of force that entail significant costs in the form of mounting casualties—regardless of the outcome on the battlefield. Consistent with the Almond-Lippmann consensus, realists believe that Americans will prefer cutting and running if the use of force fails to achieve victory within a short time span.

Foreign economic policy is the final significant policy arena where realists and the mass public are expected to diverge. Realists should be warier of trade liberalization and economic interdependence than most Americans. In an anarchic world system, realists argue that states must be concerned about the distribution of gains that come from economic cooperation.⁴²

Even if economic integration leads to a balanced distribution of gains, realists are wary of the interdependence that can emerge from a liberal economic order. For realists, “interdependence” is simply another word for “vulnerability,” and implies a loss of economic autonomy and heightened interstate frictions.⁴³

Because the United States is such a large economic power, the realist concern about interdependence should be relatively muted when thinking about trade with smaller countries. Relative gains concerns should be present, however, when thinking about economic exchange with other candidate great powers, such as China, Japan, and the European Union. The fear of rising economic powers should be present regardless of regime type.

The American public is predicted to be more comfortable with an open economic order. Mainstream economists argue that free trade is a win-win proposition, benefiting all countries that participate. Because Americans are predicted to care more about absolute gains than relative gains, there should be majority support for any liberalizing measures, given that the economic benefits outweigh the costs. In a similar vein, the broad mass of Americans are expected to be more comfortable with enhanced levels of economic interdependence than realists—in part because of the longstanding liberal argument that interdependence is a force for peace.⁴⁴

Two sources of polling data will be used to test the robustness of the anti-realist assumption. Extensive surveys have been conducted to test American attitudes towards the use of force, the expansion of trade, and the proper orientation of American foreign policy. The most prominent of these polls is the quadrennial Chicago Council on Global Affairs (CCGA) survey of public attitudes about American foreign policy.⁴⁵

If the anti-realist assumption is correct, then one should see low levels of support for realist policies and greater levels of support for liberal policies in these surveys.

It should be stressed that these data are far from perfect. The most obvious problem is that most survey questions are not tailored to isolate whether Americans are receptive to any particular paradigm of international relations theory. Many of the survey questions that have been consistently asked do not directly address realist doctrine. Furthermore, on some foreign policy questions there is extreme sensitivity to the framing of the questions. Nevertheless, public responses to real-world policy problems can serve as useful proxies for general attitudes towards foreign policy preferences.

The second set of data consists of experimental surveys designed to ascertain how Americans think about foreign policy in the abstract. The advantage of these experiments is that they allow the researcher to vary the scenario parameters, determining how sensitive public opinion is to changes in external circumstances. The disadvantage to this survey methodology is that, because these scenarios are hypothetical in nature, there is reason to question whether the respondents have fully thought out their answers. It is possible that respondents who are willing to support a policy in the abstract could change their minds when confronted with a real-world variant of the scenario.⁴⁶

One final observation: it is logical to expect that the easiest test for the anti-realist assumption should come from polling data taken during the interregnum between the end of the Cold War and the September 11 attacks. During the Cold War and the post-9/11 periods, the United States faced a clear, paramount, external threat. One would expect Americans to be more comfortable with realpolitik when the threat environment is high. During the nineties, American power was unrivaled and the threat environment was low. Polling data strongly suggests that Americans knew this fact.⁴⁷

If individuals rank foreign policy priorities like a Maslovian “hierarchy of needs,” one would expect Americans to be more enthusiastic about liberal principles during the nineties—once realist concerns were sated.

American Public Opinion on Foreign Policy Worldviews and Priorities

To determine the sources of American attitudes towards foreign policy attitudes in a post-9/11 world, Paul Brewer led a series of panel studies to determine how much Americans trusted other countries.⁴⁸

What they found was that Americans were cynical about the outside world. Over 70 percent of respondents agreed that “the U.S. can't be too careful” in dealing with other countries. And 65 percent of Americans rejected the idea that other countries try to help the U.S.; instead, they said that other countries were “just looking out for themselves.” Brewer et al. concluded that “most Americans see the realm of international relations as resembling the ‘state of nature’ described by Hobbes. Put more simply, they see it as a ‘dog-eat-dog’ world.”⁴⁹

One could argue that such a Hobbesian outlook was simply a post-9/11 anomaly. However, survey evidence shows that American attitudes about the international system were not all that different during the nineties. The Pew Research Center conducted surveys in 1994 and 1999 that asked Americans about trusting other countries. In both surveys, more than 70 percent of respondents believed that other countries took unfair advantage of the United States. The numbers were not significantly different between the two surveys. Brewer concludes that “trust appears to be the exception, rather than the rule,” during both the pre and post-9/11 worlds.⁵⁰

Public Agenda's foreign policy polling—conducted in summer 2005 and winter 2006—confirms the self-interested nature of the American public.⁵¹ In both surveys, more than 60 percent of respondents agreed that the United States was “already doing more than our share to help out less fortunate countries.”⁵² Like realists, an overwhelming majority of Americans view the world through a Hobbesian prism.

There is also evidence that Americans do not subscribe to key components of the liberal world view. When asked whether it was right, wrong but sometimes necessary, or always wrong for the U.S. to cooperate with “harsh, undemocratic governments” in order to fight terrorism, less than 15 percent said it would always be wrong; 64 percent said it was wrong but sometimes necessary. While a slim majority agreed with the contention that more democracies reduce global conflict and violence, a slim majority disagreed with the contention that less poverty in the world does the same.⁵³

Given bipartisan support among elites for both of these contentions in recent years, the weak level of mass public support is surprising. Contrary to an absolute gains perspective, Benjamin Page and Marshall Bouton find a majority of Americans oppose the idea of the European Union acquiring superpower status. They conclude that “just as ‘realist’ international relations theorists might hope (but not expect), most Americans are comfortable with the idea of the United States as the world's sole military superpower.”⁵⁴

What about foreign policy priorities? In the summer of 2004, the Council on Foreign Relations and Pew Research Center released a poll on American foreign policy attitudes that appeared to confirm the anti-realist assumption. Analyzing the data, Lee Feinstein, James Lindsay, and Max Boot concluded that “Realpolitik does not play well with the American public…. Americans overwhelmingly believe that morality should influence foreign policy decisions.”⁵⁵

Indeed, when asked what values should guide American foreign policy, 72 percent of respondents agreed that “moral principles” should be the guiding light in U.S. foreign policy.

However, when Americans were asked which issues should be thought of as “top priorities,” the responses suggest a public that is much more comfortable with realism than previously thought. Table 2 reveals that policy priorities conventionally categorized in the “legal-moral” tradition—promoting democracy, advocating for human rights, and strengthening the United Nations—are considered low-priority options. The only clearly liberal issue that earned more than 70 percent support was the prevention of AIDS and other epidemics.⁵⁶

In contrast, realist priorities—protecting against terrorist attacks, protecting American jobs, insuring adequate energy supplies—all scored better than 70 percent support. As table 2 shows, this preference ordering remains consistent, even when looking at polls taken prior to the September 11 terrorist attacks.

These two results hint at a pattern of American attitudes towards foreign policy priorities. Americans possess aspirations for liberal internationalism, but when asked to rank order priorities, realist principles come to the fore. Table 3 looks at the CCGA data on top foreign policy priorities over the past 30 years, and finds a similar result to the Pew/CFR surveys. Over the past few cycles, realist policy priorities, emphasizing security and autonomy, consistently earn more than 60 percent support from the mass public. Liberal policy priorities, emphasizing multilateralism, democracy and human rights, consistently earn less than 50 percent support. As figure 1 demonstrates, the top-ranking policy priorities for the past three decades—protecting American jobs, securing energy supplies—have a realist cast to them. In contrast, Americans have consistently placed less importance on liberal internationalist policy priorities—democracy promotion, human rights, and strengthening the United Nations. These results are stable across the three different time periods, despite variation in the level of international threat.⁵⁷

These results are also consistent across survey instruments.⁵⁸ Intriguingly, support for liberal policy priorities fell during the nineties—at exactly the point when the anti-realist assumption would have presumed the opposite to the true.

Sam McFarland and Melissa Matthews come to a similar conclusion about American attitudes towards human rights promotion. They find that Americans give strong support for abstract human rights principles—but when asked to rank order policy priorities, human rights came in 12th out of 15 possibilities. They conclude that, “although most Americans express agreement with the ideals of human rights, a willingness to commit American resources to promote and defend human rights is much weaker.”⁵⁹

In other words, while Americans aspire for liberal policy ends, realist considerations of national interest trump those aspirations.

The CCGA surveys do reveal one trend that would, at first glance, support the anti-realist assumption. Most Americans strongly support multilateral institutions like the United Nations and NATO. More than 70 percent of Americans supported American participation in the International Criminal Court and the Kyoto Protocol to protect global warming.⁶⁰

Healthy majorities of Americans—including conservatives and Republicans—support giving up America's veto in the UN Security Council if it meant a more effective global body. Similar majorities endorse giving the UN limited powers of taxation to raise a standing multilateral force. In questions about going to war or deploying American forces as peacekeepers, public support is highly contingent on whether there is multilateral support.

A second cut at the data, however, suggests a realist motivation behind the support for multilateral institutions. The major reason Americans support international organizations is because they can facilitate burden-sharing. In the 2002 CCGA survey, for example, by 71 percent to 17 percent, Americans preferred the U.S. to “do its share to solve international problems together with other countries” over “be[ing] the preeminent world leader to solve international problems.”⁶¹

That survey also showed that, in the Middle East, over 80 percent support “the European Union [to] be more involved in the negotiations while also bearing more of the political and economic costs.”⁶² Americans view multilateralism both a means of enhancing global governance and as a means of redistributing costs from the United States to other countries.⁶³

A closer examination of public attitudes towards the ICC and the Kyoto Protocol also shows that support for multilateralism in principle is soft. Although a majority of Americans endorse the ICC in the abstract, for example, they also oppose allowing U.S. soldiers to be tried in The Hague. This is in sharp contrast to France, Germany, and the U.K, where majorities supported having their soldiers tried by the ICC. Similarly, on global warming, a strong majority of Americans want poorer countries to bear as much of the burden in dealing with global warming as richer nations.⁶⁴

Sifting through the Pew data, Andrew Kohut and Bruce Stokes conclude that, “If asked to choose, Americans prefer proactive, assertive unilateral action to multilateral efforts beset by delay and compromise.”⁶⁵

A 2006 Bertlesmann Foundation poll crystallizes how Americans think about international cooperation.⁶⁶

When asked to choose between the best framework for ensuring peace and security, the populations of most major powers prefer “a system led by the United Nations” over either a balance of regional powers or a unipolar world. The United States was the only country in the survey where a majority supported the balance of regional powers over the UN. Whereas other countries value multilateralism as an intrinsic good, Americans view it instrumentally—as a means to advance American interests while reducing enforcement costs.⁶⁷

American Public Opinion on the Use of Force

The debate about mass public attitudes towards the use of force revolves around two kinds of survey questions. The first set asks Americans whether they would support a future use of force. These questions vary with the proscribed mission and the anticipated adversary. The second set of questions polls Americans on their support of imminent or ongoing military interventions—either wars or other kinds of troop deployments, such as peacekeeping missions. If the anti-realist assumption holds, we should expect to see public enthusiasm for military interventions into genocides, civil wars, and humanitarian relief operations. Support for these activities is expected to decline, however, as casualties begin to mount.

The responses to hypothetical scenarios strongly refute the anti-realist assumption. Richard Herrmann, Philip Tetlock, and Penny Visser found in their surveys a strong predisposition towards balancing behavior among the mass public:

Herrmann et al. achieved similar results when questions regarding nuclear proliferation and the prior behavior of possible adversaries were posed. Americans were more likely to prefer tougher responses to states that had acquired nuclear weapons and states with “revisionist motives” in the international system.⁶⁹

These responses are fully consistent with the balance-of-power logic stressed by realists.

Other experiments have yielded similar results. Peter Feaver and Christopher Gelpi asked military elites, civilian elites, and mass publics about their attitudes towards the use of force. They found that across the board, respondents ranked the realist goals as more important than the interventionist or humanitarian goals.⁷⁰

The mass public was also just as supportive of realpolitik missions—and just as tolerant of civilian casualties—as civilian elites.⁷¹ The CCGA surveys found Americans thinking about the use of force in the same way as offshore balancers. The mass public's strongest support for the use of force came in missions close to home—interdicting illicit drugs, combating terrorism, and armed interventions in Latin America.⁷²

Analyses of polling during real world applications of force also contradict the anti-realist assumption. The first, overriding metric that Americans use to gauge their support for war is whether the operation appears to be successful. Most scholars have found a “halo effect” that comes from successful uses of force.⁷³

As Richard Eichenberg summarizes this work, “quite simply, successful military operations enjoy high support, even when the objective is unpopular and casualties are suffered.”⁷⁴ This rationalist result is consistent with several foreign policy paradigms, including realism. The finding does undercut the Almond-Lippmann consensus, however—and the anti-realist assumption rests partly on that consensus.

Beyond the importance of perceived success, Americans appear to have realist instincts in placing a value of the uses of force. Table 4 demonstrates the mean level of support for American uses of force between 1981 and 1995, as collected by Richard Eichenberg.⁷⁵

Two facts stand out immediately. First, the only examples of military statecraft that garnered more than 60 percent support from the American people were realpolitik missions: retaliating against al Qaeda attacks, compelling Saddam Hussein to withdraw from Kuwait, and preventing Hussein from acquiring weapons of mass destruction (further discussion of the second Gulf War follows). Second, the only missions that garnered less than 50 percent support can be categorized as humanitarian or peacekeeping interventions: the uses of force in Liberia, Kosovo, Bosnia, Lebanon, and Haiti. This is fully consistent with David Burbach's conclusion that “the public is quite negative towards restoring order in countries where American lives are not at stake.”⁷⁶

The empirical literature on public responses to war and other uses of force also undercut the anti-realist assumption. Testing public reaction to thirty-eight uses of force during the Cold War era, John Oneal, Brad Lian, and James Joyner found that Americans were more likely to support the use of force to repel armed aggression than to intervene in civil conflicts. They concluded that “the public evaluates the use of force pragmatically and in a way consistent with fundamental international principles regarding sovereignty and self-determination.”⁷⁷

Eichenberg found similar results: Americans provided significantly more support for uses of force that involved restraining foreign aggressors than other kinds of military intervention.⁷⁸

One possible counterexample to these findings is the apparent contrast between public and realist attitudes about the invasion of Iraq in 2003. Most realists opposed the invasion of Iraq, while by March 2003 the American public largely supported the invasion—ostensibly in reaction to the Bush administration's public relations campaign.⁷⁹

This could be viewed as support for the anti-realist assumption. However, this case is at best ambiguous. First, public support for the war was highly contingent on the survey question. Philip Everts and Pierangelo Isernia point out that “when different alternatives to deal with Saddam's regime were offered, a plurality if not a majority of the American public would opt for the softer ones.”⁸⁰ Second, the Bush administration used realist rhetoric to make the case for invading Iraq. Mearsheimer characterized the Bush administration's case for eliminating Saddam Hussein's weapons of mass destruction program as a “hard-nosed realist argument.” Other security scholars concur, labeling the official rhetoric as, “a Realist logic focused on response to threat.”⁸¹ High-ranking officials might have had non-realist motivations for going to war, but they were also conscious that these motivations did not resonate with the mass public.⁸² If the administration used realist rhetoric and shied away from non-realist rhetoric to make its case for war, then Operation Iraqi Freedom provides no support for the anti-realist assumption.

Another prediction that flows from the anti-realist assumption is that Americans should not be able to stomach casualties during combat. However, multiple reviews of casualty data with public support for military deployments contradict this expectation. Eric Larson concludes that the public's aversion to losses of U.S. life in military interventions has less to do with intolerance for casualties than with the debatable merits of the operations themselves. Multivariate analyses shows that casualties in and of themselves do not cause an erosion of public support; Americans compare these costs to the possible benefits from victory.⁸³

Furthermore, process tracing of high-profile military reversals suggest that such events trigger rally-round-the-flag effects. For example, after the October 1993 “Black Hawk down” incident in Mogadishu, public support for the Somalia mission to temporarily increased.⁸⁴ Nevertheless, the sensitivity of democratic publics to casualties remains an open debate.⁸⁵

With regard to public opinion polling on different types of armed interventions, the results again support a realpolitik mass public. McFarland and Matthews found consistent results when polling Americans about the use of force to prevent genocide or ethnic cleansing. Seven years after the Rwandan genocide, they found only 26 percent of respondents supporting the use of U.S. troops to halt the genocide, while 41 percent preferred to “not become involved.” When asked whether the U.S. military should be used more or less often to stop either mass killings or ethnic cleansing, there was plurality support (38 percent to 15 percent) for curtailing activity. In June 1999, support for sending U.S. ground troops into Kosovo as part of a NATO operation to serve in a combat situation registered only 40 percent support.⁸⁶

McFarland and Matthews conclude, “Although most Americans applaud human rights principles, the American public en masse does not appear to care enough about human rights to invest significant American resources and troops to defending them, even in grave situations of genocide and ethnic cleansing.”⁸⁷

Turning back to the question of multilateralism, surveys conducted by the German Marshall Fund of the United States, the Program on International Policy Attitudes (PIPA), and CCGA show that multilateral backing also has a dramatic effect on American support for activities like armed interventions.⁸⁸

For example, the 2002 CCGA survey shows that American support for defending South Korea against an invasion from North Korea increases from 36 percent to 57 percent if the effort is UN sponsored. This would seem to be consistent with a liberal worldview.

Further examination, however, suggests a realist logic is at work. The increases in public support for defending vital allies came when the question specifically said that the United States would be “contributing forces together with other countries.”⁸⁹

Ceteris paribus, even academic realists evinced strong support for the use of force when such action receives multilateral backing.⁹⁰

The question is whether the mass public desires multilateral outcomes if all else is not equal. Compared to other countries, Americans still favor reserving the right to use preemptive military force to a much greater extent than Europeans. Kohut and Stokes conclude that “the American public looks primarily to its own military for its self-defense.”⁹¹

Bruce Jentleson and Rebecca Britton found that between 1992 and 1998, Americans preferred unilateral over multilateral operations when the mission was one of foreign policy restraint. At the same time, when the mission was perceived to be humanitarian in nature, there was a strong preference for multilateral participation.⁹² This result suggests that collectively, Americans have an intuition about applying realist principles to realist situations and liberal principles to liberal situations. When large-scale war is expected, Americans believe in self-help; when the mission is more consistent with liberal internationalism, however, Americans prefer using multilateralism as a way to redistribute costs.⁹³

American Public Opinion on Foreign Economic Policy

Since the beginning of modern polling, there has been public suspicion about the merits of free trade. This suspicion remains even during boom times. In 1953—a time when the U.S. was running a massive trade surplus—a plurality of Americans supported greater import restrictions over greater import expansion.⁹⁴

During the sixties, mass survey research revealed a strong protectionist bent among Americans.⁹⁵ In 1998—when the late-nineties boom was well under way—a majority of respondents agreed that, “foreign trade has been bad for the U.S. economy.”⁹⁶ Kenneth Scheve and Matthew Slaughter have catalogued a persistent fact throughout the late nineties; majorities of Americans repeatedly affirmed that a) the costs from more imports always outweighed the benefits of more imports; and b) the costs from more imports exceeded the benefits from more exports.⁹⁷ Polls conducted by the Chicago Council on Foreign Relations, the German Marshall Fund, and the Foreign Policy Association in the past decade all confirm that a strong majority of the American public is suspicious about trade liberalization.⁹⁸ One PIPA assessment observes that “questions that do not emphasize trade in principle, but rather how it has been practiced, find quite modest levels of [public] support.”⁹⁹

The debates over offshore outsourcing offer an even more pungent example of a mercantilist public at work.¹⁰⁰

Beginning in 2004, there was a marked increase in media coverage about this newer form of economic interdependence. Consistent with overall views about the costs and benefits of international trade, a majority of Americans stated that offshoring is a serious economic problem. An Associated Press poll in May 2004 found that 69 percent of Americans believe outsourcing hurts the U.S. economy. A Marlin Company poll found that 53 percent of Americans believed that it was unpatriotic for American firms to send jobs overseas.¹⁰¹ An Employment Law Alliance poll taken the same month found that 58 percent of American workers believed the federal government should penalize companies that send work offshore. A Watson Wyatt survey a few months later found 85 percent of American workers believing that the practice has a negative effect on the U.S. economy.¹⁰² These attitudes have persisted. A March 2006 Pew Research poll found 71 percent of Americans believed that outsourcing was bad for the U.S. economy.¹⁰³ In the wake of the 2006 midterms, pollsters argued that voters specifically rewarded politicians who espoused mercantilist themes.¹⁰⁴

What is the source of the mass hostility to trade expansion? A non-realist explanation is that people are responding out of self-interest rather than the national interest. Scheve and Slaughter argue that respondents with lower education and income status are more likely to project mercantilist opinions about trade.¹⁰⁵

However, this is only a partial explanation. For example, neither class nor educational status can explain why, between 1999 and 2004, public support for free trade declined across the board.¹⁰⁶ Similarly, this does not explain the intense opposition to offshore outsourcing. The Watson Wyatt survey, for example, found that while American workers were very concerned about its impact on the economy as a whole, they evinced little concern about how it affects their personal situation.¹⁰⁷

While individual economic factors undoubtedly play a role in determining opinions about trade, there is strong evidence that realpolitik considerations of national interest also play an important role. Ronald Inglehart, Neil Nevitte, and Miguel Basanez conclude that “free trade galvanizes concern for a much wider set of issues, such as cultural integrity and national identity.”¹⁰⁸

Other scholarly studies have found a similar correlation between support for mercantilism and national pride or concerns about American sovereignty.¹⁰⁹ As previously noted, the protection of American jobs has consistently ranked among the most important of foreign policy priorities.

A recent example of realism driving American policy attitudes towards foreign economic policy comes in the wake of the March 2006 Dubai Ports World controversy. A Pew survey found that 58 percent of Americans supported Congress acting to block Dubai Ports World from operating port terminals in six American cities. The framing of questions surrounding foreign direct investment revealed a key driver of mass public opposition on this issue is a loss of economic autonomy. On the one hand Americans approved of “companies in other countries investing in the U.S.” by 53 percent to 36 percent. On the other hand, Americans disapproved of “investors from other countries owning companies in the U.S.” by 53 percetn to 33 percent.¹¹⁰

The difference between investment and control relates to questions about national sovereignty—a realist theme.

Experimental survey results confirm the strong bias in favor of relative gains concern among the mass public of Americans. Informal surveys by Robert Reich found a high degree of relative gains concern vis-à-vis Japan in 1990—the peak of Japan's perceived threat to U.S. economic hegemony.¹¹¹

Richard Herrmann, Philip Tetlock, and Matthew Diascro asked whether respondents would support a foreign economic policy that benefited the United States more than its trading partner, benefited both countries equally, or benefited the trading partner more (in all cases, both countries received positive benefits). Among the mass sample, the distribution of gains significantly affected responses—64 percent of Americans supported a policy that benefited the United States more; when the partner benefited more, support fell to 38 percent.¹¹² Mass responses were also more protectionist when the trading partner was described as either wealthy or strong. Using real world countries, respondents were more likely to favor restricting trade against Japan than either England or India. Herrmann, Tetlock, and Diascro conclude that “a larger percentage of the general public than of the elite think about trade as if they were intuitive neorealists…. And more of them are sensitive to the factors neorealists say should affect policy calculations.”¹¹³ David Rousseau conducted a similar survey, and found majorities opposing trade agreements that resulted in small economic gains for the United States but significant economic gains by other major powers. Consistent with realism, opposition increased when the proposed trade partner was a country viewed as an economic challenger, such as China or Japan.¹¹⁴

Why Does the Anti-Realist Assumption Persist?

This paper concludes that the anti-realist assumption is not empirically valid. Americans do hold some liberal aspirations for their conduct across the globe, and believe that morality should play a role in foreign affairs—in the abstract. However, surveys about foreign policy world views and priorities, the use of force, and foreign economic policies all reveal a strong realist bent among the mass American public. The overwhelming majority of Americans possess a Hobbesian world view of international relations. Americans consistently place realist foreign policy objectives—the securing of energy supplies, homeland security—as top foreign policy priorities. Objectives associated with liberal internationalism—strengthening the United Nations, promoting democracy and human rights—rank near the bottom of the list. On the uses of force, experimental surveys reveal that Americans think like intuitive neorealists; they prefer balancing against aggressive and rising powers while remaining leery about liberal-style interventions. On foreign economic policy, Americans think of trade through a relative gains prism, particularly if the trading partner is viewed as a rising economic power. Surveys and polling do suggest that Americans like multilateral institutions, but they appear to like them for realist reasons—they are viewed as mechanisms for burden-sharing. Liberal internationalist attitudes did not resonate more during the nineties, the period when the anti-realist assumption should have had the greatest support. If anything, during that decade Americans appeared to adopt attitudes consistent with offshore balancing.

The implications of these results for the analysis of American foreign policy are significant. Realists have repeatedly asserted that U.S. foreign policy leaders act in a realist manner but disguise these actions through liberal rhetoric. The anti-realist assumption allows these scholars to assert that the only source of realpolitik behavior comes from the systemic level. If the American public holds realist views on certain foreign policy dimensions, then a unit-level causal mechanism exists that would also explain these policy outcomes.

These findings also lead to an important question. If American attitudes towards foreign policy have been consistent for decades, and those attitudes are receptive to a realist world view, then why does the anti-realist assumption persist within the academy and the policymaking worlds? One possible explanation is that, even at the elite level, misperceptions about public opinion are remarkably difficult to change. Public opinion scholars refer to this phenomenon as “pluralistic ignorance”—shared but incorrect beliefs about what other people think. Americans believe that a majority hold views similar to the actual outputs of U.S. foreign policy.¹²⁴

It is possible, then, that realists commit a double error in their analysis of American foreign policy. They assert that Americans take realist actions against the desires of a liberal public. The reverse might be true—because the United States pursues a liberal foreign policy, Americans assume that the majority supports these policies even if that is not actually the case.

Another possible explanation is that realist scholars believe the anti-realist assumption because they encounter hostility to realism on a daily basis. Hostility to realism resides in the elite public more than the mass public. Most survey research in foreign policy—including the Foreign Policy Leadership Project—separate out elite opinions from the mass public. The elite public is usually defined as people who are both knowledgeable about foreign affairs and have some access to foreign policy decision-makers. This includes high-ranking members of the executive branch, members of Congress and their staffs, lobbyists and interest group representatives, journalists, academics, and leaders of labor, business, and religion.¹²⁵

Researchers have identified persistent gaps in attitudes between the mass public than the elite public on a series of foreign policy issues.¹²⁶ Realists would predict this gap to exist; they assume that the mass public is hostile to hard-headed calculations, requiring the elite public to think in a realpolitik fashion behind closed doors.¹²⁷

However, many of the studies cited here also find that across a broad array of questions, the elite public is more liberal internationalist than the mass public.¹²⁸

During the mid-nineties, the mass public was significantly more pessimistic about world affairs than policymakers.¹²⁹ Elites adopted more positive attitudes towards multilateralism.¹³⁰ Elites prefer spending larger amounts of foreign aid than the mass public. Military elites are more casualty-phobic than the civilian mass public.¹³¹ After the end of the Cold War, the elite public was far more concerned about Russia's transition to democracy than the American public; the mass public was more concerned about Japan's economic ascent and the threat of international terrorism. On the use of force, the mass public was more willing to use military statecraft to intervene in the Western hemisphere, combat illegal drugs, and prevent terrorism than the elite public. In the months after the September 11 terrorist attacks, the elite public evinced a more cosmopolitan policy response than the mass public.¹³²

Foreign economic policy is where the most prominent gap between elite and mass public opinion exists. In their experimental surveys, Herrmann, Tetlock, and Diascro found that elite participants were far less sensitive to relative gains than most Americans: “even in the condition of a ten-to-one relative loss, more than two-thirds of elite Americans still pursue absolute gains and choose free trade.”¹³³

The CCGA surveys reveal a similar gap between the elite and mass publics on questions of trade, investment and immigration. Over the course of the eight surveys, the elite public has consistently favored freer trade, been less exercised about the trade deficit, and preferred higher levels of legal immigration. Other public opinion researchers have detected similar gaps between the elite and mass publics.¹³⁴

The gap between elites and the rest of America on trade policies have ramifications beyond foreign economic policy. Some commentators have hypothesized that in the current era of economic globalization, elite members of the American public would adopt a more cosmopolitan view of world politics, consistent with liberal internationalism.¹³⁵

Samuel Huntington derisively dubbed this the “Davos culture” hypothesis, but the notion has attracted serious scholarly interest.¹³⁶ Certainly a cosmopolitan elite would be less likely to embrace a “folk realism” borne out of inattention to world politics. Richard Herrmann and Jonathan Keller have found that among the American policy elite, a free trade orientation was strongly and positively correlated with supporting policies of engagement and a reluctance to use force. Not surprisingly, Herrmann and Keller found more than 60 percent support among policy elites for engaging both China and Iran. They further found that policy elites embraced the democratic peace hypothesis.¹³⁷

The relatively liberal attitude of American elites comes into particular focus when one considers precisely which elites strongly influence American foreign policy. Lawrence Jacobs and Benjamin Page conclude that the preferences of four elite groups had the strongest effect on the attitudes of policymakers: media, business, think tanks and other foreign policy organizations, and educators.¹³⁸

With the partial exception of think tanks, surveys of all of these different subgroups indicate a strong predilection towards cosmopolitanism or liberal internationalism.¹³⁹ The correlation of preferences between these groups is so strong that Jacobs and Page conclude that a “foreign policy establishment” still exists in the United States.¹⁴⁰

The existence of a liberal internationalist foreign policy establishment would help to explain an enduring puzzle in this area of research—why elites continually believe that the public is more isolationist than it actually is.¹⁴¹

It is entirely possible that either the elites or the polling questions are confusing wariness about American isolationism with liberal concerns about a realist mass public. Ironically, the fear that gripped Kennan and Morgenthau in the early fifties may be completely at odds with present-day reality: America now has a realist mass public governed by a liberal internationalist elite.

This leads to a final speculative hypothesis. The liberal internationalist trend is strong among the elites that realist scholars interact with the most—other international relations professors.¹⁴²

Survey research and analyses of journal articles reveal that most IR scholars believe realism to be on the wane as a tool for explaining international relations in terms of scholarship, teaching, and real-world relevance.¹⁴³ It is possible that realists believe that most Americans do not like realism because the Americans they interact with the most—their professional colleagues—are hostile to the paradigm.