Roundtable 9-6 on Narrative and the Making of US National Security

narrative-and-the-making-of-us-national-security-coverAmerican Presidents are often maligned for their rhetoric. U.S. leaders’ words are not to be trusted: they will use whatever language allows them pander, to bluster, even to deceive their audiences. Yet as Ronald R. Krebs demonstrates in this impressive new work, presidential rhetoric is far from cheap. Indeed, it lies at the core of national security politics: under the right conditions, a leader’s rhetoric tells the public a story about national security, in the process creating a narrative that shapes the president’s understanding of the political world, and lays the foundation for grand strategy.

H-Diplo | ISSF Roundtable, Volume XI, No. 6 (2016)

H-Diplo/ISSF Editors:  Thomas Maddux and Diane Labrosse

H-Diplo/ISSF Roundtable and Web/Production Editor:  George Fujii
Commissioned by Thomas Maddux

Ronald R. Krebs.  Narrative and the Making of US National Security.  Cambridge:  Cambridge University Press, 2015.  ISBN:  9781107103955 (hardback $103.00); 9781107503991 (paperback, $35.99).

Published on 28 November 2016


© Copyright 2016 The Authors

Introduction by Stacie Goddard, Wellesley College

American Presidents are often maligned for their rhetoric. U.S. leaders’ words are not to be trusted: they will use whatever language allows them pander, to bluster, even to deceive their audiences. Yet as Ronald R. Krebs demonstrates in this impressive new work, presidential rhetoric is far from cheap. Indeed, it lies at the core of national security politics: under the right conditions, a leader’s rhetoric tells the public a story about national security, in the process creating a narrative that shapes the president’s understanding of the political world, and lays the foundation for grand strategy.

Presidents must create narratives because events cannot speak for themselves. Instead, as Krebs forcefully argues, they “acquire meaning only when people weave them into coherent stories” (4). After 9/11, for example, the George W. Bush administration told a story about the attacks that characterized the terrorists as evildoers, suggested the United States as an innocent victim, and named American counter-terrorism efforts a ‘Global War on Terror.’ This interpretation of 9/11 was not determined, and this story Bush told about the War on Terror would drive national security policy in the decade that followed. Likewise, Franklin Delano Roosevelt narrated World War II as an existential clash between democracies and fascists, a story that not only brought the United States into the war in Europe, but ultimately blamed the Nazis, and not the German people, for the catastrophic violence.  In both cases, presidential narratives crafted both the understanding of political crises, and the national security strategies that followed.

Narratives are thus essential to national security. Through narrative, presidents can persuade an audience of the urgency of a threat, and the necessity of a particular foreign policy. But Krebs cautions that crafting a compelling narrative does not come easily.  Conventional accounts stress presidential skill as the driver of success. From this perspective, presidents Woodrow Wilson, Roosevelt, Ronald Reagan, all masterful orators, were able to use their talents to sell their visions to the American public.  Yet as Krebs notes, such explanations are insufficient. Wilson, Roosevelt, and Reagan might have been formidable narrators, but all met resistance to their foreign policies, whereas far less skilled speakers, such as Bush, told stories that resonated profoundly with the American public.

For Krebs, the success of the narrative does not lie with the speaker alone, but in the intersection of presidential speech and discursive structures.  Crafting a new narrative of national security, Krebs cautions, is only possible when times are unsettled, when there is little coherence in a dominant discourse and “multiple narratives legitimately circulate in the public sphere” (5).  Under these conditions, presidents are able to introduce a new story of national security, one that is likely to move an audience to support the preferred grand strategy.  For this to happen, however, also requires agency: presidents must recognize the opportune moment, and forge ahead with a master narrative, rather than lapse into everyday policymaking.

Krebs tests his theory in case studies of U.S. national security, ranging from World War II to the War on Terror.  In the 1930s Roosevelt failed to persuade Americans to abandon isolationism.  This failure, Krebs argue, lay in his rhetorical mode: rather than build a narrative that explained why the United States must intervene against Nazi Germany, Roosevelt instead used mundane policy arguments that proved unable to move the public.  Likewise, Reagan may have been the Great Communicator, but his attempts to paint Nicaragua as an epic Cold War battle failed. Here the problem was not rhetoric, but structure: the Reagan administration encountered a settled narrative of Cold War threats, which even a master orator could not unsettle.  In contrast, Krebs shows how Roosevelt after Pearl Harbor, and Bush after 9/11, seized the opportunity to weave new tales of national security, and lay the basis for broad-reaching changes in foreign policy.

In Part II of the book, Krebs turns to the question of the lifecycle of security narratives: when they are born, whether they grow, and how they die. Here Krebs offers both theoretical and historical novelty.  Whereas most scholars believe policy failures should undercut narratives, Krebs argues that policy failure can strengthen them; in contrast, it is success, not failure, that leads to a narratives’ demise. And in what is perhaps Krebs’s most original historical contribution, much of the book traces the origins, stabilization, and demise of the Cold War consensus.  Relying on detailed content analysis of op-eds in the New York Times and Chicago Tribune, Krebs suggests that the consensus did not form until the Korean War, and lasted only until the 1962 Cuban Missile Crisis, a timeline at odds with much of historiography.

Krebs’s work is theoretically and historically ambitious, and the reviewers in this roundtable are largely impressed with the book’s analytical scope and empirical richness.  Brian Rathbun calls the book a “theoretical masterpiece.” Krebs has moved the field beyond the widespread claims that “talk matters,” to specify, as John Owen notes, the critical causal processes of “why one national-security narrative comes to dominate; why it persists over time; how it constrains and shapes debate and policy; and how and why it ultimately collapses and gets replaced.”  Both Owen and Sarah Graham praise the book’s empirical work, with Graham remarking on the deft way that Krebs’s approach weaves together agency, structure, and the logic of politics.  Hopf, the most critical of the reviewers here, notes the book’s “clever methodological moves,” such as Krebs’s decision to analyze not only the results of public opinion polling, but the appearance and absence of particular survey questions as evidence of narrative dominance.

Of course, any ambitious work will raise as many questions as it offers answers.  Three particular criticisms appear throughout the critical commentary in this roundtable.  First, the reviewers are skeptical of Krebs’s efforts to conceptualize and operationalize unsettled times, the structural driver of Krebs’s theory. As Owen observes, “the way in which times unsettle is under-theorized in the project.” Ultimately it is difficult, both he and Hopf suggest, to separate unsettled narratives from the critical junctures they produce, leading readers to wonder if there is any “unsettled time” that can be measured apart from the outcome it produces.  And if we cannot separate structure from outcome, there is a risk of tautology that even the most careful of empirical work cannot overcome.

Second, the reviewers also question the relationship between events and narrative, or the ‘objective’ and the ‘intersubjective,’ in Krebs’s account.  Krebs is clear, in theory, where he stands: there is no objective reality outside of narrative. Yet his position does not always come through in his case studies. At times, for example, existing narratives seem to collapse, not because of their inherent contradictions, but because of ‘exogenous’ crisies, such as the defeat of France in 1940.  As Hopf asks, is this an “objectivist argument? Did the narrative crumble under the pressure of external events, independent of interpretation?”  Likewise, Graham wonders whether Krebs has given ample consideration to more materialist alternative explanations of his cases.  Was the public driven to support a ‘Europe first’ strategy because of Roosevelt’s narrative, or because Britain stood isolated and imperilled in Europe?  Was the Morgenthau plan rejected because is conflicted with Roosevelt’s narrative, or because of bureaucratic politics?

Finally, the reviewers ask whether Krebs’s focus on American politics, and specifically presidential rhetoric, limits the generalizability of his theory to other cases.  There is much to be said for Krebs’s choice to focus on the rhetoric of American presidents. It is this choice—to narrow the case studies that in turn allows him to produce the careful and rich analyses—that lies at the center of the book.  Nevertheless, this choice might undercut what the theory can say about other cases.  For example, Rathburn asks whether all of this focus on rhetoric “Is not a uniquely American phenomenon to tell a story about Americans and how great the United States is?” Likewise, Hopf asks whether Krebs’s additional argument, that policy failure tends to create more consensus around a narrative, is not also a uniquely American phenomenon, if United States politics are not peculiarly structured in a way that punishes the politician who actually admits that she is wrong.

Krebs’s book raises hard questions for the roundtable participants here.  And this, the reviewers agree, is exactly what an important work of social science is supposed to do.  In the end, Krebs has written a work that makes clear the importance of narrative in world politics.  Its rich empirical work and careful methodology are likely to serve as a model for qualitative scholars, and its cases offer insights of interest to international relations scholars and historians alike.  It is, as Owen states, a signal achievement, one likely to set the standard for the study of rhetorical politics for years to come. In fact, Narrative and the Making of US National Security has received the 2016 Robert L. Jervis and Paul W. Schroeder Best Book Award in International History and Politics, and the 2016 Giovanni Sartori Book Award, for the best book developing or applying qualitative methods, both from the American Political Science Association. It has also been awarded an honorable mention for the 2016 Joseph S. Lepgold Book Prize, for an exceptional contribution to the study of international relations, from the Mortara Center for International Studies of Georgetown University.


Ronald R. Krebs is Beverly and Richard Fink Professor in the Liberal Arts and Professor of Political Science at the University of Minnesota. He is the author most recently of Narrative and the Making of US National Security (Cambridge University Press, 2015), which received the 2016 Robert L. Jervis and Paul W. Schroeder Best Book Award in International History and Politics and the 2016 Giovanni Sartori Book Award, for the best book developing or applying qualitative methods, from the American Political Science Association. Ron Krebs is co-editor of “Rhetoric & Grand Strategy,” a special issue of Security Studies (2015), co-editor of In War’s Wake: International Conflict and the Fate of Liberal Democracy (Cambridge University Press, 2010), and author of Fighting for Rights: Military Service and the Politics of Citizenship (Cornell University Press, 2006). For a complete listing of his articles on a wide range of topics in international relations in both leading scholarly journals and more popular outlets, see his website. Ron Krebs currently sits on the editorial boards of Security Studies and the Journal of Global Security Studies.

Stacie Goddard is the Jane Bishop ‘51 Associate Professor of Political Science at Wellesley College. Her research explores issues of identity, legitimacy, and conflict in international relations.  Her articles have appeared in International Organization, International Security, International Studies Quarterly, International Theory, and the European Journal of International Relations.  Her book, Indivisible Territory and the Politics of Legitimacy: Jerusalem and Northern Ireland, was published by Cambridge University Press in 2009. She is currently working on her second book, which explores legitimacy and decisions to accommodate rising powers in international politics.

Sarah Ellen Graham is a lecturer at the United States Studies Centre, University of Sydney, Australia. She has published articles in Diplomatic History, Diplomacy & Statecraft, Orbis, and International Studies Review, among others. Her work addresses the intersection of diplomatic history and IR, and takes an interest in the Cold War, public diplomacy and U.S.-India relations.

Ted Hopf is the Provost Chair Professor of Political Science at National University of Singapore. In addition to articles published in American Political Science Review, European Journal of International Relations, International Organization, Review of International Studies, Security Studies and International Security, and numerous book chapters, he has edited or authored six books. His most recent book, Reconstructing the Cold War: The Early Years, 1945-1958 (Oxford University Press, 2012), won the 2013 American Political Science Association Robert Jervis-Paul Schroeder Award for Best Book in International Relations and History and the 2013 Marshall D. Shulman Award, presented by the Association of Slavic, East European and Eurasian Studies for the best book published that year on the international politics of the former Soviet Union and Central Europe. His research has been supported by the Ministry of Education of Singapore, The Mershon Center, the Ford Foundation, the American Council for Learned Societies, and the Olin and Davis Centers at Harvard University. He, Bentley Allan, and Srdjan Vucetic are currently coordinating a project to create a national identity database for all great powers since 1810.

John M. Owen (Ph.D., Harvard) is Ambassador Henry J. and Mrs. Marion R. Taylor Professor of Politics, and a Faculty Fellow at the Institute for Advanced Studies in Culture, at the University of Virginia.  During 2015-16 he is a recipient of a Humboldt Research Award and is a visiting scholar at the Free University of Berlin and the Berlin Social Science Research Center (WZB).  He is author of Confronting Political Islam: Six Lessons from the West’s Past (Princeton University Press, 2015), The Clash of Ideas in World Politics: Transnational Networks, States, and Regime Change, 1510-2010 (Princeton University Press, 2010), and of Liberal Peace, Liberal War: American Politics and International Security (Cornell University Press, 1997), and co-editor of Religion, the Enlightenment, and the New Global Order (Columbia University Press, 2011).  He has published in the European Journal of International Relations, Foreign Affairs, International Organization, Internationale Politik und Gesellschaft, International Relations, International Security, International Studies Quarterly, Monkey Cage (Washington Post), National Interest, New York Times, Perspectives on Politics, and a number of edited volumes.  From 2011 through 2014 he was Editor-in-Chief of Security Studies.

Brian Rathbun is an Associate Professor at the School of International Relations at the University of Southern California. He is the author of three books—Partisan Interventions: European Party Politics and Peace Enforcement in the Balkans (Cornell University Press, 2004), Trust in International Cooperation: Domestic Politics and American Multilateralism (Cambridge University Press, 2012), and Diplomacy’s Value: Creating Security in 1920s Europe and the Contemporary Middle East (Cornell University Press, 2014)—as well as articles in International Organization, International Studies Quarterly, the Journal of Conflict Resolution, European Journal of International Relations, Security Studies, and the Journal of Politics among others. He is currently completing a new manuscript entitled Reason of State: Rationality, Realism and Romanticism in International Relations.


Review by Sarah Ellen Graham, University of Sydney

Under what conditions can presidents shape public opinion, silence opponents and introduce new doctrines of national security, and under what conditions will they fail? Ronald Krebs offers a compelling analysis of the rhetorical presidency and the politics of national security in this book. He demonstrates the power of the narrative rhetorical mode in U.S. national security when presidents effectively address the five elements Kenneth Burke identified within narratively-ordered texts: act, scene, agent, agency, and purpose (12).[1] Arguments about means and ends, and even about normative goals, are not sufficient to reconfigure the broad terms within which the nation debates national security; “fundamental change in national security policy—in its goals and basic orientation …hinges on change in narrative” (179-180). Storytelling stabilises identities—constructing images of self and others—as well as encapsulating the scene in which events take place and specifying their causal entailments.

Krebs’s foundational assumption is that events and material facts do not yield their meanings independently of representation. While Krebs acknowledges that there is a set of enduring discourses that have spanned different narrative phases in U.S. national security policy—American exceptionalism, though the concept of national security itself is arguably another[2]— his study is better placed to address the functions of rhetoric, not the discursive constitution of the American self and national security per se. The book highlights the role of presidential skill and the power of authority in establishing the narratives that generate policy change, offering both important data and a compelling analysis to support its claims. Krebs makes a second theoretical intervention, as well, in establishing of a set of expectations about how publicly-circulating national security narratives are affected by military success and failure.

In Part I Krebs offers four national-security case studies that use structured content analysis to identify the terms presidents have used to represent national security and to measure the narrative qualities of presidential speech at any given time. Krebs demonstrates how President Franklin D. Roosevelt in the wake of Pearl Harbour and President George W. Bush in the wake of 9/11 undertook effective storytelling from the bully pulpit, storytelling that addressed the five elements of a fully-spun narrative. These stories had far-reaching effects on public opinion, configuring how Americans understood their enemies and the setting of the conflict at hand, while offering critics few rhetorical openings to articulate alternative courses of action outside the the dominant narrative.

It is intuitive that these two shocking attacks on U.S. territory should have sparked significant departures in official thinking about U.S. national security, but Krebs argues that important policy implications flowed from the particular narrative depictions selected by each president. In the case of Roosevelt, his greater commitment initially to war in Europe over Asia, despite the fact that it was Japan that had attacked the U.S., and his insistence that the war was to be waged against the Nazi regime not the German people, were by no means the only possible courses of action available at the time. Roosevelt’s narrative choices also had important policy consequences later in the war. Krebs thus carefully places his analysis between realist accounts that explain national security policy with reference to material threats and those that place the drivers of policy largely in the realm of longstanding identity discourses. Scholars on both sides will productively engage with this argument.

Realists may question the finding that Roosevelt’s policy followed his narrative and not material interests, highlighting the strategic reasons for his ‘Europe first’ strategy such as Britain’s imperilled situation or the fact Germany’s threat to Latin America was a palpable one.[3] Krebs also situates Roosevelt’s decision not to adopt the punitive Morgenthau Plan for post-war Germany as a product of his post-Pearl Harbor narrative. In one of the most authoritative diplomatic histories on this subject, John Gaddis depicts this policy debate as the result of bureaucratic politics, alliance considerations, and, increasingly, suspicion about Soviet leader Joseph Stalin’s intentions, as well as Roosevelt’s characteristic of leaving his options open as long as possible. Both Roosevelt’s flirtation with the idea of a harsh peace, and his rejection of it, were dictated by the tenor of relations with Stalin.[4] Consequently, Krebs’s argument would seem to be on more solid ground in addressing the public presentation of the policy dilemma rather than the setting of national security policy itself. On the ideational side, historians have addressed the role of ideology and race during the Second World War,[5] which presents another element to consider in assessing Roosevelt’s war-narrative with respect to ‘Europe first’ and the rejected Morgenthau Plan, especially if we consider the implicit comparison with Japan. While the scope of Krebs’s work is impressively broad, how the “deeper identity narratives” (13) that Krebs signals in his introduction might have informed Roosevelt’s narration of the war are not tackled here, though they perhaps deserve more attention.

Scholars of the presidency and the history of U.S. foreign policy will be particularly interested in Krebs’s cases of narrative failure. Roosevelt’s failure to narrate a successful policy of intervention before 1941 and President Ronald Reagan’s failure to narrate a more substantial and open commitment to dislodging the Sandinista government in Nicaragua are puzzling because they involve presidents who were masterful orators. Although Roosevelt faced an unsettled narrative situation toward the end of period Krebs addresses (1940-1941)—a situation which normally provides presidents with an opportunity to use their authority to author a compelling new narrative—Roosevelt failed to make effective use of the narrative mode by failing to represent the international scene. The discussion also pinpoints the initially insurmountable barrier of the non-interventionist narrative, although Krebs is perhaps a little too brief in his account of the composition of the non-interventionist bloc—too frequently he writes of ‘they’ without defining whom exactly he is discussing —the role of partisanship and non-interventionism in Roosevelt’s own party, and the sources of Roosevelt’s opponents’ narrative capital.

The developing correspondence between the President’s narrative and events abroad is told well by Krebs, and it was ultimately the surprise attack at Pearl Harbour that dislodged the non-interventionist narrative and gave Roosevelt the opportunity to narrate the world in the way he wished. This again shows the deft way that his approach weaves together agency, structure, and the logic of politics. Krebs’s analysis of Roosevelt’s narrative failure rests on an assumption about his intentions: that the “U.S. entered the war far later than the president would have liked” (69), but some readers may take a different view. There is certainly little dispute among historians that Roosevelt found the strength of the non-interventionist bloc frustrating. Robert Dallek, though, characterises Roosevelt’s position as one of ‘reluctant neutrality,’ which entailed increasingly substantial material support to the Allies but not full-blown interventionism. Dallek’s portrait is of a President who was highly attentive to both the significant limits of American military power as well as his own limited political/rhetorical capital.[6] Effective narration, in this situation, would undoubtedly have been valuable to the President in helping him confront the political obstacles to aiding the allies more while remaining neutral. But the policy change that Roosevelt failed to accomplish by narrative would not be as substantial as Krebs’s account posits if we take on the foregoing view of Roosevelt’s intentions.

A similar (realist) objection with respect to Krebs’s analysis of Reagan’s failure to extract from Americans greater support for aid to the Contras follows recent histories of that case.[7] These emphasize the absence of a sufficient security threat to the U.S. from the Sandinistas. The Soviet Union had a limited and pragmatic policy on Central America when the U.S. first intervened, thus there was no fitting rhetorical strategy because the U.S. did not have a sufficiently compelling material interest in doing so. This is in contrast to the highly pragmatic logic and bipartisan support for aid to the Afghan Mujahedeen. Krebs offers a several intriguing points of comparison between the interventions in Nicaragua and Afghanistan that could provide an important test of his theory. Was acceptance of intervention in Afghanistan the result of greater bipartisan support and Congress’ greater narrative ‘ownership’ of the intervention, or the fact that the intervention garnered less public attention and rhetorical challenge and could be conducted more covertly? Or was U.S. aid to the Mujahedeen possible because it was unambiguously tied to the balance of power between the superpowers?

Krebs’s second theoretical intervention in this book is as significant and provocative as the first. In Part II there is another theoretical chapter, followed by lengthy analysis of the so-called ‘Cold War narrative.’ Here Krebs advances a set of propositions about how military success and failure ought to affect the national security narrative in circulation. He advances a new periodization for the phases of the Cold War struggle and a provocative rethinking of the impact of the Vietnam War on the politics of U.S. national security. The expectations of his theory of narrative change are measured against data from editorials in the Chicago Tribune and New York Times between 1945 and 1990, which show both the content and the degree of elite consensus on the terms of the Cold War national security narrative. His expectations are at first blush counter-intuitive: that victory unsettles the prevailing narrative and offers presidents an opportunity to change the trajectory of U.S. national security (although they sometimes fail because public demand for a new narrative is low), and that lack of military success creates an incentive to bolster the existing narrative by both opponents and supporters of a war.

By testing his theoretical insights against this data, Krebs offers findings that Cold War historians should take seriously. He argues that the Cold War narrative consensus was consolidated during the Korean War, not the late 1940s. The military setbacks of that conflict placed incentives on President Harry S. Truman’s critics to articulate their criticism of the conduct of the war from within the terms of the global Cold War narrative. President Dwight Eisenhower, whose foreign policy legacy is undergoing a new reappraisal by historians,[8] was the beneficiary. Krebs highlights the strong rhetorical foundations of the Eisenhower presidency in ways that will resonate with this agenda. This finding perhaps even offers an opening to situate the civil-rights reforms that historians have connected to the Cold War[9] as a further implication of the ascendant narrative of globalism that emerged from the Korean stalemate. A persistent question arising out of Part II concerns partisanship. While Krebs contends that the turning points in the Cold War consensus cannot be explained by partisan dynamics, could partisan dynamics be better integrated into the expectations about when and why presidents will seek to articulate new narratives? The analysis of the narrative context in the 1960s and 1970s addresses party dynamics in detail, begging the question as to whether there are generalizable propositions about divided government or the electoral cycle that should be incorporated into the theoretical model.

Narrative and U.S. National Security makes an important contribution to International Relations theory and I expect it will be read by scholars in the field for years to come. Krebs rigorously tests his argument against alternative IR-theoretical predictions about policy change. He addresses the agency/structure problem in a nuanced way. Krebs also demonstrates how it is possible to situate the logic of politics more centrally within IR theory. With its broad and provocative argument, the book richly deserves a readership among historians and American government officials as well.


Review by Ted Hopf, National University of Singapore

There are two broad audiences for this book, one that will take Krebs’s arguments as commonsensical, the other as heretical. The latter is the camp of International Relations (IR) specialists who believe in the objective world and have not made the sociological, let alone linguistic or practical, turns in their own intellectual universes. I will not be speaking to the latter. Instead, I offer a more ‘inside’ critique, as someone who believes in the autononmy of language, discourse analysis, and that the objective world does not speak for itself.

Krebs argues that in unsettled times, people crave narratives; if the President instead makes arguments, he will not be convincing. If in settled times, he tells stories, he will not be convincing. So President Franklin D. Roosevelt missed an opportunity before Pearl Harbor. Meanwhile President George W. Bush appropriately told a post-9/11 story to an unsettled America.

Second, Krebs argues that victories allow for foreign policy change, while defeats do not. So, “the Terror narrative persisted not despite the US military’s setbacks in Iraq, but because of them” (176). Meanwhile victories, supplying political capital to their authors, allow these leaders, if they wish, to present a new narrative with a greater chance of success, despite the fact there is no obvious need for it. The Korean War ‘loss’ gave rise to the Cold War consensus; the Cuban missile crisis ‘victory’ allowed for the collapse of the Cold War consensus; and the Vietnam War ‘defeat’ locked in that consensus.

Krebs writes that “facts acquire meaning only when people weave them into coherent stories,” or narratives. This is too strong a claim. Facts acquire meaning through any interpretation. As interpretivists have been telling us for decades, researchers should not impose a false coherence on messy empirical realities. Real people have contradictory beliefs, not coherent stories. We analysts fashion the coherent narratives only at the cost of destroying the meanings our subjects attribute to facts.

Cognitive science indeed shows that humans do not tolerate disorder. (10) But narrative is only one of myriad ways humans restore such order. Far less coherent and elaborate devices include scripts, schemas, frames, stereotypes, and other cognitive heuristics. These need not add up to a narrative. Ontological security can be maintained and restored without a narrative.

I am not convinced that Krebs has established a non-tautological method for establishing when a moment was settled or unsettled. For example, he writes that unsettled times are characterized by “multiple narratives legitimately circulat[ing] in the public sphere” (5). These are big meaningful words; how many is multiple? What constitutes legitimate? And what are the boundaries of the public sphere? Many scholars would argue that all times are always unsettled. Perhaps there is a matter of degree, but we would need to know how to measure the quantity of narratives, their legitimacy, and the boundaries of the public sphere. Krebs goes on to write that “such critical junctures are openings for narrative projects.” But if we are using multiple narratives as a measure for a critical juncture, how can we then say the critical juncture is allowing for multiple narratives?  Meanwhile, settled times are defined as periods of “routine, or settled, narrative situations.” (33) We need some unrelated measure of a critical juncture, or unsettled times to avoid a tautological specification.

I suppose many people reading this book will be surprised to find that the narrative terrain in the U.S. is so narrow. For example, Krebs argues on that those who opposed the surge in Iraq in 2007-2008 agreed with its supporters that the U.S. was in a war not of its choosing and that the U.S. had been struck on 9/11 through no fault on its own (6). That surely does not square with my opposition to the surge. But of course, I do not a narrative make. But a quick read of authors in the New York Review of Books and The Nation, among others, would reveal an opposition to the surge that does not share Krebs’s description of consensus.  By Krebs’s view, the over one thousand American IR scholars who signed on to a full-page New York Times advertisement “were the illegitimate dissenters” (34). This is a very narrow view of narrative consensus indeed. That does not make Krebs’s argument necessarily wrong. But it does make this book about something other than the narratives of U.S. national security. It concerns the narratives Krebs that has identified through his discourse analysis of the New York Times and Chicago Tribune.

Krebs’s sample may explain his ‘counterintuitive’ results, viz., that defeats like Vietnam do not inspire change in narratives, while victories like the Cuban Missile Crisis, do. It could very well be that the Times and Tribune provide exactly the evidence for Krebs to tell his counterintuitive story, but they might not be valid operationalizations or measurements of U.S. national security narratives, whose textual remains are distributed across myriad other media. Based on these sources, Krebs concludes that the Cold War consensus, as he defines it, emerged during the Korean War, rather than just after World War II, and it dissolved not during or after the Vietnam War, but before 1965.

This reader at least came away skeptical that all the historians and IR scholars that Krebs claims are wrong, are in fact wrong. I think what is going on is that all the scholars Krebs cites as wrong on the collapse of the Cold War consensus after Vietnam (191, note 1) conceptualize and measure ‘security narratives’ differently than Krebs, and Krebs is actually correct, but only if we accept his theorization and measurement of this issue. As he puts it, all these other scholars “lack systematic evidence regarding what the consensus was;” this, of course, puts great weight on how precisely Krebs himself measures that consensus. Krebs is probably right to assert that “one cannot classify a narrative as dominant if either the nation’s paper of record (New York Times) or the heartland’s spokesman (Chicago Tribune) deviated from it,” but that only shows that these sources are necessary, but not necessarily sufficient, to serve as a representative sample of U.S. national security narratives (198).

The methods by which Krebs derived his narratives also raise at least one concern. He writes that “based on existing catalogs of Cold War axioms, [he] constructed a fourteen-point questionnaire” to be asked of every editorial in each newspaper. He assures us that “[he is] not familiar with any elements that are commonly viewed as part of the Cold War consensus but do not appear on the questionnaire.” (196-197) Unfortunately, as soon as I read the questionnaire, which is helpfully reproduced in Appendix D, 315-318, I immediately thought of several elements of the Cold War consensus that I, at least, thought were missing: Munich, appeasement, deterrence, credibility, resolve, and reputation. Other IR scholars could plausibly come up with others.

Krebs writes that after Vietnam, “the 1970s was a time of surprising elite unity,” based on the two newspapers surveyed (257). This finding is surprising, given the fact that the historical consensus is that the U.S. was riven by Vietnam, which leads me to again suggest that by sampling these two papers, Krebs finds a middle ground that misleads. As Krebs put it in his conclusion, “The unique data I have collected revises our understanding” (282-283) of the history of American national security narratives. The problem may lie in its very uniqueness.

Half of Krebs’s argument rests on being able to distinguish between rhetoric characterized by storytelling versus rhetoric characterized by argument making. I came away unconvinced that it is possible to make such a neat distinction. Narratives can make arguments; and arguments can be stories. As important is the fact that I could not assess whether the coding software Krebs uses to distinguish between the two modes actually does what he claims it does. He writes that ‘Diction’ is coded in his book as storytelling texts that include spatial awareness, temporal awareness, motion, and human interest. I have no idea what any of this would mean. If I say ‘I should have gone to my office yesterday to review Krebs’s book,’ am I telling a story or have I presented an argument against procrastinating?

Krebs argues that Franklin Roosevelt mistakenly failed to use a storytelling narrative after 1939. But he then argues that “an adequate narrative would have made clear how the nation would be physically vulnerable to Axis predation and why the US could not remain a thriving city of liberty on its lonely hill” (96). I understand that statement to be an argument, not a story. I could tell a story that would make that argument, or I could make the argument that Krebs has just made.

Despite the fact that this book is very long, and even more of the book is on Krebs’s website, it would have been helpful if Krebs had provided an Appendix in the book that gave examples of textual excerpts and the coding results, so the readers could judge the reliability of the coding methods. As of 1 March 2016, the website,, offers the note “Replication files and appendices to come” when interested readers click on the URL that Krebs provides in this book. I hope that Krebs will include this information when he completes the online supplements so we can all make better, and more informed, judgments about his book’s arguments.

Krebs’s strong claim, which I share, that events do not speak for themselves, but have to be interpreted, sometimes, at least, runs afoul of some of Krebs’s own arguments. Explaining why the non-interventionist discourse began to erode in the U.S. from 1939, Krebs writes that “if the combined strength of Britain and France could not defeat Germany, an isolated Britain, without a continental toehold, could hardly defeat the Nazi war machine” (76). But is not this an objectivist argument? Did the narrative crumble under the pressure of external events, independent of interpretation?

One problem with the book is that in various ways Krebs’s argument is hard to evaluate. For example, there are citations that are impossible to track down for they lack page numbers. On page 35, note 15, in order to see whether, where, and how Antonio Gramsci, Jacques Derrida, Nicholas Dirks, and Raymond Williams make the claim that dominant narratives are always full of contradictions it would be necessary to have references to the pertinent pages on which they make their arguments in these rather lengthy works. Gramsci, for example, is often charged with making his theory of a dominant narrative, or ideological hegemony, too totalizing.[10]

As noted above when discussing Diction, without some sample of texts that have been coded as high on storytelling or argumentation, it is impossible to assess the validity of Krebs’s measurement of the degree of storytelling versus argumentation in any given text. Moreover, it is impossible to judge whether there is a distinguishable difference between the two.

I found it impossible to make any sense of Krebs’s Figures. Figure 3.1, on page 78 is dropped in without any explanation or interpretive guide. He indicates that it is derived from “Centering Resonance Analysis,” which is explained in Appendix A, (303-305). I found the explanation mostly incomprehensible.  Apparently, some nouns are more central than other nouns, but perhaps these appear in boxes, or with darker lines, or more lines, or are perhaps more centrally placed in the web. Frankly, I have no idea what to do with the data.

Again, in order to evaluate Crawdad, the program used to generate the diagrams, which very much remind me of those pictures of spider webs woven by arachnids on acid, far more than two pages of text are required. What we need, as we do with Diction, are examples of texts and how they were coded according to Crawdad. We also need to be walked through the diagrams in English, at least once. For example, we are told that Figures 3.1, 2, and 3 show the evolution of Roosevelt’s discourse, but we are not told how. Instructing us how to read Figure 3.5, Krebs reveals that the higher an element in the web, the more important, and the darker the line, the more tightly connected a word is to another word (102). So, Communist, Cuba, and Sandinista are key. But in terms of ‘altitude,’ several words are higher than Cuba, like El Salvador, which is higher than all three, or democratic and economic. Dark lines connect most of the web. In sum, it is difficult to understand Krebs’s interpretations.

There are easily two books here. The first would be on storytelling versus argumentation and the failures and successes of U.S. national security narratives. The second would be on the effects of victories and defeats on U.S. national security narratives. Two books would have provided the kind of data and guidance that readers need to evaluate Krebs’s arguments. In this book, Krebs repeatedly asks his readers to go to his website in order to understand his book more completely. Essential information ideally belongs in the text of the book, and not on a website link, which should contain additional, not essential, information.

Krebs’s book contains a number of very clever methodological moves. Instead of looking at the results of public opinion polls per se, Krebs looks at the questions that were (not) asked. His hunch is that if questions were not asked then the issues had already become settled as common sense. For example, he notes that after 9/11, questions about whether the U.S. was responsible for 9/11 mostly disappeared from mainstream surveys. The question of whether the U.S. should engage in a war on terror was replaced with how the U.S. should engage in such a war. (154) This approach is very insightful. I would have appreciated a much more systematic utilization of this technique with a variety of polls and (missing/disappearing) questions.

In the same vein of making absence matter, Krebs pays attention to what U.S. presidents do not argue against in their speeches. His hunch is that if opposition arguments are systematically ignored, they are not part of mainstream discourse; otherwise, the President would feel the need to counter them. As with the above, President Bush felt no need to address whether U.S. foreign policy had anything to do with 9/11 (155-157). As with the above, I would have welcomed a more systematic treatment of these absences, as well.

Throughout Chapter Four, Krebs uses the fact that Roosevelt never demonizes Germans as evidence that facts do not speak for themselves. After all, he could have, and did not and instead focused his venom on the Nazi regime. Of course, the same was not true of the Japanese. I think Krebs too quickly rejects the obvious rationalist explanation for this choice: the vast number of German and Italian Americans in the U.S. (and not just electorally, as mentioned, and rejected, on p. 133) but as a plurality in U.S. society, not to mention armed forces, universities, corporate boardrooms, etc.

I was also surprised that John Lewis Gaddis’ fabulous book, Strategies of Containment, was not systematically compared to Krebs’s findings.[11] Gaddis, with a single variable, administration non/support for Keynesianism, explains U.S. grand strategy for every president from Truman to Carter. What is more impressive, his argument applies very nicely to presidential cases beyond the publication of his book: Ronald Reagan, two Bushes, Jimmy Carter, and Barack Obama, as well. Essentially, willingness to tolerate deficit spending correlates perfectly with ‘symmetrical containment,’ or going after communism everywhere it rears its head. So, Republicans who were big spenders: Reagan and George W. Bush, for example, bloated the military budget, engaged in arms races, and fought the communist/terrorist enemy wherever they could. (This was true with John Kennedy and Lyndon B. Johnson as well.) Democratic fiscal conservatives, for example, Bill Clinton and Obama, picked their battles, ended wars, restrained military spending the best they could, and adopted ‘asymmetrical containment.’ (This was true with Dwight Eisenhower, Richard Nixon, and George H.W. Bush as well) In this sense Gaddis provides a nonpartisan rationalist materialist basis to U.S. security narratives: deficit hawks make peace; drunken sailors make war.

One wonders how unique it is to the U.S. that its foreign policy defeats lead to greater adherence to the consensus that produced the failure. Is post-war U.S. politics especially structured to punish politicians who question wartime leadership in the White House? Do the media and partisanship conspire to demonize critics by accusing them of providing “aid and comfort to the enemy” by opposing ongoing military and geopolitical disasters (185-6)?

Since Krebs claims that all the scholars he cites are wrong to argue that failures cause reflection and change, while success yields a non-reflective attachment to the status quo, one wonders whether the U.S. case Krebs describes is simply unique. He argues the “conventional wisdom” that defeats and failures spur change “is intuitive,” but wrong (181). But have not these many scholars he cites as wrong done more than offer a theory? Have not they empirically tested it as well? Dani Reiter, for example, cited by Krebs as an example of a scholar whose work is wrong, demonstrated in his first book, Crucible of Beliefs, that the pre-World War II alliance choices of European powers can be explained by whether their alliance choices in World War I were successful, or not.[12] This might be conventional wisdom and intuitive, but it is also empirically accurate.

There is a very strange disconnect between the changed national-security narrative after the Cuban Missile Crisis toward more accommodation with the Soviet Union, recognition of the Sino-Soviet split, and the need to cooperate on Berlin and nuclear arms control, and the launching of the Vietnam War just years later. The overthrow of Fulgencio Batista in Cuba, the UN General Assembly resolution that launched mass decolonization, especially in Africa, and Soviet leader Nikita Khrushchev’s declaration of ‘revolutionary warfare’ as just warfare all made 1960 the year that transferred the Cold War squarely to the developing world for the next thirty years.

Finally, Krebs’s normative preference for narratives over ontological insecurity is surely defensible, at least according to his terms; after all who would not prefer order to “a world of anomic individuals adrift without political community or identity” (28)? But one wonders if this is really the choice. Is not pragmatic prudence a possible alternative? A strategy of treating each issue on its own merits, and developing a sensible set of means to achieving satisfying ends might provide some ontological security, as the greatest geopolitical and humanitarian disaster in post-war history—the 2003 Iraq War—would certainly have been avoided, while small successes, such as Camp David, the peaceful transitions to democracy in Eastern Europe and Latin America, the integration of China into the world capitalist economy, and so on, might look pretty reassuring to most Americans. This is not, as Krebs characterizes it, “lurching from crisis to crisis,” (299) but offering ontological security through a narrative of steady success. I would have thought, having revealed the costs of American exceptionalism and the ‘lessons of Munich,’ Krebs’s book would have come down in favor of continual contestation over narrative predominance.


Review by John M. Owen, Berlin Social Science Center (WZB) / University of Virginia

Over the past several years Ronald Krebs has published important papers in the European Journal of International Relations and Security Studies[13] on narrative and American foreign policy. His project has culminated in his book Narrative and the Making of U.S. National Security. A signal achievement, Narrative makes as good a case as we have that leaders’ rhetoric is causal in foreign policy. We know that politicians and their staffs believe that the way leaders speak and the arguments they make can make or break a policy. Krebs delves into some twentieth-century cases of American foreign policy and argues persuasively that this belief is true. In particular, a leader who knows his or her nation’s narratives, and can figure out how to argue for a preferred policy, can move that nation.

Like many scholars, Krebs is skeptical that national-security strategy derives directly, without mediation, from the threats and opportunities presented by the international system, as structural realists would have it.[14] Nor are those mediating entities limited to domestic interests (e.g., import-competing producers, militaries), as neoclassical realists[15] and liberals would have it.[16] Instead, between system and strategy sit dominant socially held narratives or frameworks. A narrative sets the boundaries of legitimate discussion and interpretation of international events. Thus the 9/11 attacks were not a single brute fact, the meaning of and appropriate response to which were self-evident to all. A number of interpretations and responses to 9/11 were feasible; the ‘Global War on Terror’ was not the only possibility. If this is so, we need to know how and why one national-security narrative comes to dominate; why it persists over time; how it constrains and shapes debate and policy; and how and why it ultimately collapses and gets replaced.

A number of strands of social theory touch on these questions in various ways, and a number of IR scholars have built upon these. Krebs finds the arguments of Jeffrey Legro[17] the most adequate, but even those come up short. Whereas Legro argues that large failures cause a narrative to collapse, Krebs argues the (paradoxical) opposite:  policy failure can actually shore up a narrative, while success can lead to its collapse. For Krebs, it is precisely because a narrative is powerful that opponents of the policy are under pressure not to challenge it; if they do, they appear radical, and discredit themselves. Instead, they buttress the narrative by saying that the security challenge that everyone agrees on would be better served by alternative policies.

As to where narratives come from, Krebs argues that they emerge when (1) times are ‘unsettled,’ i.e., there is no predominant narrative; and (2) leaders engage in story-telling as opposed to analytical argument. A national security narrative, then, is a picture that makes sense of the new challenges. Rather than assume that presidents are always perfectly rational, Krebs acknowledges that they make mistakes in perception or execution. Sometimes a president fails to understand that unsettled times (for example, the 1930s) require storytelling; another president might not see that settled times (the 1950s, for example) require argument. Krebs makes a good case that President Franklin Roosevelt, for all of his rhetorical and political gifts, failed to persuade his country to oppose the Axis until Pearl Harbor because he did not tell a story about why it should do so. Instead, Roosevelt made dry arguments about cause and effect (68-96).  Another fabled communicator, President Ronald Reagan, made the converse error:  he wove a story about the Nicaraguan Contras during settled times, i.e., when his arguments about facts would have been more effective (97-121).

Making good on his claims requires that Krebs operationalize narrative and measure changes over time. The book is a very strong piece of empirical work. First, Krebs distinguishes a foreign policy consensus from its lack. Second, he conducts content analysis of hundreds of Cold War foreign policy editorials in America’s leading liberal (New York Times) and conservative (Chicago Tribune) newspapers at the time. Third, he carries out content analysis (using centering resonance analysis or CRA) and offers his own interpretations of elite opinion using the statements and positions of presidents. The descriptive results are novel, interesting, and deserving of note by historians. Contrary to received wisdom, Krebs finds that the Cold War consensus that the United States had to wage a global campaign to contain Communism did not take hold in the country until the North Korean invasion of South Korea in 1950, and that this consensus lasted only twelve years – collapsing with the hair-raising Cuban Missile Crisis of 1962. The ensuing decade saw no new consensus, but in the mid-1970s a new bipartisan consensus emerged not on anticommunism but on American exceptionalism.

One of Krebs’s most persuasive claims is that the Cuban Missile Crisis disrupted the containment consensus by demonstrating the success of containment and freeing President John Kennedy, Senator William Fulbright, et al. to state and implement their true convictions about how to handle the Soviet Union. The mechanism is similar to that through which British Prime Minister Winston Churchill lost the election in 1945 or President George H.W. Bush lost in 1992: the public essentially sent the message ‘thank you for solving that problem. Now let’s move on to the next one.’ Krebs gives various alternative explanations a fair hearing, and ably defends his use of counterfactual arguments. Narrative will be influential not only for its substantive argument but for its exemplary use of qualitative research methods.

As with any ambitious social-scientific work on this high level, Narrative does not satisfy on all points. Krebs makes clear that he believes that some narratives, such as the War on Terror, outlive their usefulness and lead countries astray. That implies an evaluative stance:  this is not ‘value-free’ social science. What, then, renders a national-security narrative inadequate? In certain places, such as his final conclusion, Krebs implies that the answer is a suboptimal provision of security – rationalism’s answer, which presupposes that there is an optimal amount of security that we analysts can locate (300). But when he discusses ontology in the book, the answer seems to be, roughly, a poor fit with “enduring identity narratives” (14). This sounds like “ideas all the way down” (8f, 13f), as if the material world does not impinge on the social. Of course, Krebs is far from the only IR scholar whose precise position at the rationalist-constructivist boundary is elusive.

Neither is it completely clear that Krebs’s distinction between settled and unsettled times leaves him able to falsify his claims about presidential success and failure (33). Are there markers of unsettled times independent of the thing to be explained? In Franklin Roosevelt’s case, for example (73), it appears that Krebs knows the 1930s were unsettled by the failure of the President’s arguments to rally the nation against Germany; if that is so, the argument (about Roosevelt’s case) is circular.  More generally, the way in which times unsettle is under-theorized in the project.  Finally, the Reagan-Nicaragua case is carefully researched but Krebs’s claim implies an implausible counterfactual, namely that Reagan would have succeeded had he stuck with the facts about the Contras and Sandinistas.

These criticisms, however, are a kind of tribute to Krebs’s book, for they are the type that only can emerge from a first-rate piece of social science that grapples with some of the most difficult theoretical and historical issues in the IR subfield. Scholars skeptical of any causal role for rhetoric in foreign policy, or skeptical that empirical social science can test propositions about such a role, need to read this book. So do people who wonder whether presidential rhetoric and national narratives affect foreign policy. After this book, it becomes harder to maintain that they do not.


Review by Brian Rathbun, University of Southern California

Ronald Krebs’s Narrative and the Making of U.S. National Security is a theoretical masterpiece. Krebs asks a big question – what role does national storytelling play in foreign-policymaking – and gives us a big answer – it is a function of the right conjuncture of structural circumstances and the way that American leaders talk to the U.S. population. I will not go into a summary of the book here, but will instead free-ride on the reviews of my colleagues. Rather, I will note just a few of the book’s main contributions as well as pick slightly and unfairly at some of the lingering questions it left in my mind.

Talk, rhetoric, speech, acts, etc., – most of us who are honest with ourselves about international relations know these are important for international relations. But who is intellectually brave enough to show us how? Most of those who have tried have confused us with their very words. It is a tremendously jargonistic field, with fuzzy concepts and no real causal discipline. (By the way, a narrative is simply a story; it took me about ten years to figure that out.) I have tried myself to delve into linguistic theory, critical theory, narrative theory; it is a tough slog. Krebs does this tough work for us. This is not ‘narrative theory for dummies’ though; it is polishing a very, very rough rock with some valuable gems inside if one can make them intelligible. Krebs’s own way with words helps tremendously in this regard. Who else in this field would use the word ‘register’ as a noun? Krebs studies prose; his scholarship is almost poetry.

Constructivists have long told us that much of what we know is inarticulated, common-sense, and reified, what most recently has become known as “practice theory.”[18] Krebs’s point is to tell us that we have national narratives, self-understandings of ourselves that we do not question in terms of where we have been and where we are going. These in turn structure debate, making some counterarguments about different paths to take or options to try impossible. The trick is in offering a theoretical framework that explains variation in this regard – what situations favor a new narrative – and the proper way to talk in order to change the story. This is what Krebs does.  He tells us a story about stories.

The next challenge, though, is to demonstrate this empirically. Here my positivist side kicks in. Krebs argues that no major grand strategy is possible without a story that justifies it. Implicit here is that any collective national enterprise requires a national buy-in. This is enormously difficult to prove. I think here of an empirical example not drawn from Krebs’s book – Prime Minister Winston Churchill’s personal rhetorical role in asking all the British to ‘do their bit’ in World War II. He has been called the greatest speaker that ever lived, and many have argued, more importantly, that without his great words the British would have folded. How could we ever possibly know if this is true?

We probably cannot but that does not mean we should not try. Yet that requires a different focus, not on the speaker but the ‘spoken to,’ the ears and not the lips. In other words, an argument like Krebs’s really needs what we like to call (well some of us do not like it, but I do) ‘microfoundations.’ Is it the case that the kind of rhetoric that Krebs regards as particularly narrative-making at the right time is actually necessary to get people moving?

One possibility is to examine rhetoric experimentally. We will never be able to go back in time, change the speeches, and see if society behaves differently. But we can do this in hypothetical situations with real people. Do certain types of rhetoric inspire, and better solve collective-action problems (of which war is a big one) than others? I am not asking Krebs to be an experimentalist; that is for others to do. My point is to identify certain questions that this book can never really answer empirically because of its largely elite focus and yet offer new avenues of research. What more can one ask for?

I also have some questions about generalizability. (Please, stay with me.) Is this not a uniquely American phenomenon to tell a story about Americans and how great the United States is? I suppose the French still do (and the English used to), but that is just another way of showing the exceptionality of the enterprise.  Does Costa Rica have a national narrative? Or even Canada? Is this a great-power thing?

Last, even if one talks in the right register at the right time, are there certain substantive constraints in a narrative? Do narratives, for instance, always have to be romantic in quality? I cannot think of a tragic national narrative, as this is not particularly inspiring, which it seems is the whole point of narration – getting a country to do something big that is challenging and daunting. Or if there are tragic narratives, there is always victory in the end, which is another way of saying that all narratives are romantic.

With that in mind, however, let me offer my own romantic understanding of the future of the study of narrative, rhetoric, and talk in IR. With Krebs’s book as the cornerstone, it is ever onward to ‘sunlit uplands.’ Talk is no longer cheap.


Author’s Response by Ronald R. Krebs, University of Minnesota

I am grateful to H-Diplo/ISSF for organizing this roundtable on Narrative and the Making of US National Security and to the distinguished reviewers for the care with which they have read and engaged this work. It is gratifying that such a theoretically diverse set of reviewers—comprised of a political psychologist, a liberal theorist, a constructivist, and a realist-inclined historian—believes this book’s arguments and evidence worth grappling with.

Narrative and the Making of US National Security argues that foundational, unquestioned narratives commonly shape debates over national security. Imparting meaning and order to global politics, these narratives constitute the boundaries of legitimate debate and thereby set out the limits of sustainable national security policy. The book’s two parts explore respectively how and when particular narratives of national security have become dominant in the United States and how and when these dominant narratives have come undone, and throughout it shows what impact a narrative’s rise to dominance, and its subsequent fall from that perch, have had on U.S. national security policy. Stacie Goddard in her introduction and John Owen in his review have ably summarized the theoretical answers Narrative offers to these questions, and I will not rehash them here. Placing its analytical wager on elite narratives and on the imperative to legitimate policy, the book’s targets include those who believe that international events have some natural, unmediated meaning. I argue, in contrast, that alleged brute facts acquire meaning only when people weave them into coherent stories. Its targets also include the many scholars who place little stock in legitimation—perhaps because they believe that where there is a will, there is a rhetorical way, or perhaps because they see foreign policy as the play of isolated statesmen, unencumbered by the need to explain themselves. I maintain, however, that the largest questions of national security require leaders to engage public audiences and thus to provide public justification for the policies they prefer and that that which cannot be legitimated cannot be pursued over the long haul.

The reviewers raise numerous important questions, and I must necessarily respond selectively. I begin with conceptual and theoretical concerns and then turn to empirical and methodological matters.

Let me start with some of the key conceptual turns Narrative makes. First, Ted Hopf is skeptical of narrative and story-telling as coherent and useful categories. On the one hand, he points to a number of seemingly “far less coherent and elaborate devices”—scripts, frames, metaphors, etc. However, these are, in my view, typically shorthand formulations that reference in truncated form fully-fleshed-out narratives that audiences know well (and, because they know them so well, need not be fully fleshed out). When Cold War presidents invoked ‘Munich,’ they were certainly making an analogy, but they were also invoking a well-worn tale of appeasement and its disastrous consequences, and audiences knew their meaning without the whole story being told. On the other hand, he also questions the very distinction between these rhetorical modes.[19] I agree that the distinction between these ideal types is not ‘neat’ in practice: “real speakers do not express themselves consistently in only a single mode: they do not just argue or tell stories all the time” (36). Nevertheless, as ideal types, arguments and stories differ in their purpose, their structure, and the depth of their presumptions (10-13, 36-41). Most important, while arguments make the case for particular policies, storytelling seeks to impart meaning, to explain a series of events, to offer an interpretation of the world. Hopf’s critical example—“I should have gone to my office yesterday to review Krebs’s book”—is on its own neither an argument (since Hopf does not explain why he should have gone to his office yesterday) nor a story (which at the very least involves plot, something that is lacking here); it is rather a statement of self-reproach or of a hortatory nature (if it urges Hopf to go to his office today). I am very much open to other ways of operationalizing these rhetorical modes, but let’s not jettison the conceptual distinction.

Finally, with respect to conceptual clarity, both Owen and Hopf are troubled by Narrative’s discussion of settled and unsettled narrative situations. They raise the specter of tautology. “Are there markers of unsettled times independent of the thing to be explained?” Owen asks. The reason for their disquiet is that I define settled and unsettled narrative situations in terms of narrative dominance, and narrative dominance, or its absence, is the book’s explanandum. A settled narrative situation is one in which there is a dominant narrative, and an unsettled narrative situation as one in which “multiple narratives swirl about the public sphere with more or less equal legitimacy” (33). Narrative situation, they rightly point out, resides on both sides of the causal equation in Part I. However, the argument would be tautological only if narrative dominance at t0 accounted for narrative dominance at t1. A à B is tautological when B is actually A, and therefore A à A. (A classic example is Gabriel Almond and Sidney Verba’s claim that a civic culture is the source of democracy, yet in their account we know a civic culture by virtue of its democratic institutions and practices.[20]) But I argue in Narrative, first, that an unsettled narrative situation at t0—that is, the absence of narrative dominance—is a condition of possibility for a settled narrative situation at t1—that is, narrative dominance. This is not A à A, but rather a version of ~A à A. But that is not quite right either because, second, the unsettled narrative situation (~A) is only part of the story. It makes possible subsequent narrative dominance only in conjunction with a matching rhetorical mode (storytelling) and an authoritative speaker. If this tripartite conjuncture is out of alignment, I argue in Narrative, the effects are rather different. Such an argument is not tautological.[21]

Hopf further challenges my understanding of narrative dominance by invoking the scope of debate over the Iraq War. He seems to suggest there was no dominant Terror narrative that deeply constrained national security debate in the United States in the 2000s. Yet the very example he cites reveals how constrained that debate was, how little even dissenting policy positions typically broke from the dominant narrative.[22] The October 2004 statement signed by over 850 “Security Scholars for a Sensible Foreign Policy”—which included Hopf, Goddard, and me among its signatories—was a model of critique from within the dominant narrative, as its chief contention was that the Iraq War did not serve the War on Terror, but rather undermined that more important campaign: “We judge that the current American policy centered around the war in Iraq … harms the cause of the struggle against extreme Islamist terrorists… [T]he early shift of U.S. focus to Iraq diverted U.S. resources, including special operations forces and intelligence capabilities, away from direct pursuit of the fight against the terrorists.” It put forward numerous criticisms of the Iraq War’s prosecution (“policy errors during the occupation and reconstruction of Iraq have created a situation in Iraq worse than it needed to be”), and it condemned the Bush administration for “its failure to engage sufficient U.S. troops to capture or kill the mass of al-Qaida fighters in the later stages of that [Afghanistan] war.” These were the rhetorical lines that Democrats would follow (albeit unevenly) during the 2004 campaign and with greater consistency in 2008. Although these creatures of the Ivory Tower called for a ‘fundamental reassessment,’ even they—myself included, mea culpa—could not break from the dominant Terror narrative.

Turning to theories of narrative dominance, the reviewers raise three issues to which I devote attention here. First, Brian Rathbun asks whether some narratives are substantively more likely to the rise to the top. Perhaps we can predict which narrative will triumph based on its content: perhaps certain narratives better satisfy individual psychological needs or conform better to rules of good storytelling. I am not opposed to this approach, but I have been unable to use it to generate satisfying theoretical claims (see 52-55). Although narratologists have identified elements of good storytelling (e.g. impeccable timing, rich detail, memorable images), I cannot, after immersing myself in this literature, specify these a priori or separate them from their alleged effects. (In other words, here lies tautology.) Moreover, what happens when two narratives are, in terms of the quality of their storytelling, roughly equally good? It is too convenient to presume always that only one narrative meets the substantive requirements. Rathbun’s specific proposal—that dominant narratives are always romantic, never tragic—is problematic for the same reason. Might not multiple romantic narratives—setting out different daunting challenges, visions of the heroic, and portraits of the end—compete for dominance?

Second, Owen contends that dominant narratives come undone in Narrative when they fail to provide the optimal level of security. Hopf similarly suggests that, on occasion, it appears that dominant narratives are unsettled by objective realities. This would be ironic, given Narrative’s statements to the contrary, and the fault is mine for seeming to suggest it in places. Dominant narratives are not unsettled because they fail to provide some objective, structurally optimal level of security. After all, as Part II of Narrative demonstrates, national security narratives often survive even the substantial failures that can reasonably be laid at their feet, including defeat on the battlefield. Dominant narratives certainly can fail tests, but these tests are established by the defenders of the dominant narrative themselves—not by the international system or by some putatively objective national interest. The tests to which, for instance, the noninterventionist narrative was subjected and found increasingly wanting in the late 1930s and early 1940s were more of the noninterventionists’ own making than of President Franklin D. Roosevelt’s. As I write, “Crises are, in this sense, constructed and endogenous” (35). Moreover, I argue in Part II, narrative unsettling can be the product of triumph—as after the Cuban Missile crisis and the end of the Cold War.

Third, Rathbun, and to a lesser extent Hopf, wonder how generalizable this account is. I am confident that the general question of narrative dominance travels well beyond the United States (for examples, see 6-7). Rathbun suggests that narratives of national security are a luxury that non-great-powers cannot and need not afford. I do not see why that should be the case. Seeking to maneuver through a world of complexity, smaller powers too impose meaning and order on a reality that lacks any inherent meaning and order. It may be even more incumbent on small powers to insist on narrative clarity given their limited resources. I am uncertain whether my particular theoretical arguments travel beyond the United States, especially in Part I of Narrative, which rests on a particular understanding of the presidency and its peculiarly authoritative narrative position (48-52, 283-284), and I hope future scholars will explore these dynamics in other national and institutional contexts. I have more confidence that the account in Part II is not particular to the United States or Cold War competition (for my reasons, see 266-267), but again welcome careful consideration of non-U.S. contexts.

Finally, turning to methodological and empirical questions, I am compelled to be especially selective. Three reviewers (Rathbun, Owen, and Sarah Graham) raise questions about various counterfactuals, and I agree with them that counterfactuals figure especially centrally in work that emphasizes agency, and they are tricky to evaluate (for my methods, such as they are, see 62-64). Rathbun is right that we cannot know whether the British “stiff upper lip” remained stiff thanks to Churchill’s inspirational rhetoric, and I do not make that claim. We can have more leverage on the question of whether Roosevelt could have done more before Pearl Harbor, and whether he needed to be a “reluctant neutral” (as Graham calls him, following Robert Dallek), because politically sophisticated contemporaries weighed in on the matter (see 88-91).

The reviewers also offer specific alternative arguments. For instance, Graham, channeling a realist perspective, suggests that President Ronald Reagan’s desired narrative regarding Nicaragua failed to take hold because the Sandinistas simply did not constitute a sufficient threat. However, this sort of realist account of the success and failure of narrative projects, resting on a world of objective and measurable threats, confuses cause and effect. The Sandinistas did not seem sufficiently threatening because Reagan’s rhetorical efforts fell short. On other occasions, successful narrative efforts projected threats that realists believed to be inflated. Indeed, during the early Cold War, realists like George Kennan and Walter Lipmann assailed the emerging Cold War consensus as misleading and misguided. If dominant narratives consistently reflected allegedly objective realities, realists’ narrative preferences would have routinely triumphed, and realists would not so often have found themselves on the outs and complained about how policy had been hijacked by domestic interests.

Meanwhile, Hopf proposes that there was little demonization of German- and Italian-Americans during World War II because these ethnic populations were so large and powerful in the United States—in contrast to Japanese-Americans. But there were not appreciably more German- or Italian-Americans in the United States in 1941 than in 1914. Yet that hardly stopped the Wilson administration from deploying crude nationalist and racialist propaganda against the ‘Huns.’ Nor were those German- and Italian-Americans more committed to their ethnic identities at the time of the Second World War, for German-American civil society had withered under the pressures of the First World War and German-Americans had run from their ethnic affiliation. It is true that ethnic identity shaped German-Americans’ sympathies in the 1930s and 1940s, as Adam Berinsky has shown,[23] but if fear of German-American retribution drove Roosevelt in World War II, why did it not have similar effects on U.S. politics during the Great War? If anything, the effects during the earlier war should have been greater.

Part II of Narrative advances a historically revisionist account of the Cold War consensus and a theoretically revisionist account of the relationship between success, failure, and change. It is not surprising that some reviewers would think its revisionism goes too far. Hopf in particular is critical of the content analysis at the heart of Part II and the inferences drawn from it. He repeatedly cites the fact that Narrative’s findings run contrary to conventional wisdom as if that were itself evidence that they are wrong and misleading. Yet his arguments are not persuasive. Hopf maintains that the “textual remains [of U.S. national security narratives] are distributed across myriad other media” besides the two large newspapers I examine and that “these sources are necessary, but not necessarily sufficient, to serve as a representative sample of U.S. national security narratives.” I agree, and I am quite open to the possibility that there were “respectable” views to the right of the Tribune and the left of the Times and that their foreign affairs editorials either endorsed key Cold War narrative propositions later or diverged from them earlier or never joined the consensus at all (as I note on 199). But, if this were the case, we would be hard-pressed to speak of any dominant Cold War narrative for any period of time—which would be even more at odds with conventional scholarship and which seems to be the opposite of Hopf’s intent. Hopf further suggests that the content analysis omits key elements of the Cold War consensus. Perhaps, but many of those he cites (revolving around the logic of deterrence and reputation) are implicit in the questionnaire (notably, question #4 on global politics and interconnectedness), and I would be very surprised if including these elements fundamentally changed my account of the dominant Cold War narrative’s content or periodization. Content analysis is a valuable tool when one is sensitive to the method’s limits (see 197-200), and I fully acknowledge the need for caution and modesty when challenging such intuitive and deeply rooted arguments about the relationship between policy outcome and ideational change (266). The narrative domain may be different from that explored in prior institutionalist accounts (181-183), and I am therefore not sure how these theoretical arguments extend to other domains (284). But I would like to see more work employing and extending Narrative’s methods, not less. Part II of Narrative shows what sorts of provocative arguments and findings can emerge when one productively combines content analysis, discourse analysis, and historical process-tracing.

I’m grateful to the reviewers not only for their serious critical engagement with Narrative and the Making of US National Security, but also for their thoughtfulness in considering how this research might be extended. I find appealing Rathbun’s suggestion to measure how individuals respond to storytelling and argument via experimental methods; Rathbun and Hopf’s implicit calls to explore, theoretically and empirically, whether Narrative’s arguments travel across space and time (echoing my call on 283-284); and Graham’s suggestion that partisan dynamics shape rhetorical strategy. Many of their critiques also suggest avenues for further research. I am particularly impressed by Graham’s invitation that we consider more deeply the relationship between national security narratives and deeper identity narratives; by Rathbun’s contention that I have not paid sufficient attention to the audience(s) listening and responding to leaders’ rhetoric; and by Hopf’s embrace of pragmatism and his implied plea for a more robust and systematic pragmatism in foreign policy (as David Edelstein and I similarly concluded in a recent article).[24]

I do not intend to take up these questions myself. But I hope others will. And I will consider Narrative and the Making of US National Security a great success if they do.



[1] See: Kenneth Burke, A Grammar of Motives (Prentice Hall: New York, 1952).

[2] The work of Andrew Bacevich, for example, suggests a reading of national security as discourse. See, Andrew Bacevich, ed. The Long War: A New History of U.S. National Security (New York: Columbia University Press, 2009).

[3]  On the implications of the German threat to Latin America see, eg. David G Haglund, Latin America and the Transformation of U.S. Strategic Thought, 1936-1940 (Albuquerque: University of New Mexico Press, 1984); Max Paul Friedman, Nazis and Good Neighbors: The United States Campaign Against the Germans of Latin America in World War II (Cambridge: Cambridge University Press, 2003).

[4] John Lewis Gaddis, The United States and the Origins of the Cold War, 1941-1947 (New York: Columbia University Press, 2000), chapter 4.

[5] John W. Dower, War Without Mercy: Race and Power in the Pacific War, (London: Faber, 1986); Gerald Horne, “Race from Power: U.S. Foreign Policy and the General Crisis of White Supremacy,” in Michael Hogan, ed. The Ambiguous Legacy: U.S. Foreign Relations in the “American Century” (New York: Cambridge University Press, 2000). The original account of ideologies of race was Michael Hunt, Ideology and U.S. Foreign Policy (New Haven: Yale University Press, 1987).

[6] Robert Dallek, Franklin D. Roosevelt and American Foreign Policy, 1932-1945 (Oxford: Oxford University Press, 1979), 199-232.

[7] An example of this revisionism, which also surveys the literature, is Danuta Paszyn, The Soviet Attitude to Political and Social Change in Central America, 1979-90 (Basingstoke: Palgrave Macmillan, 2000). On the broader question of Soviet vs. U.S. interventionism in Latin America, see Max Paul Friedman’s counterfactual in: Daniel Sargent, et. al. “Online Roundtable: Hal Brands’ Latin America’s Cold War,” Journal of American Studies 46:1 (2012), 11.

[8] Eisenhower has featured prominently in recent studies on the waging of the Cold War in the Third World. See, for example, Odd Arne Westad, The Global Cold War: Third World Interventions and the Making of Our Time (Cambridge: Cambridge University Press, 2007).

[9] See, for example, Mary Dudziak, Cold War Civil Rights: Race and the Image of American Democracy (Princeton: Princeton University Press, 2001).

[10] James Scott, Weapons of the Weak (New Haven: Yale University Press, 1985), 304-350, for example.

[11] John Lewis Gaddis, Strategies of Containment (New York: Oxford University Press, 1982).

[12] Dan Reiter, Crucible of Beliefs (Ithaca: Cornell University Press, 1996).

[13] Ronald R. Krebs and Patrick Thaddeus Jackson, “Twisting Tongues and Twisting Arms: The Power of Political Rhetoric,” European Journal of International Relations 13:1 (2007): 35-66; Krebs and Jennifer K. Lobasz, “Fixing the Meaning of 9/11: Hegemony, Coercion, and the Road to War in Iraq,” Security Studies 16:3 (2007): 409-451.

[14] E.g., John J. Mearsheimer, The Tragedy of Great Power Politics (New York: Norton, 2001).

[15] E.g., Steven E. Lobell, Norrin M. Ripsman, and Jeffrey W. Taliaferro, Neoclassical Realism, the State, and Foreign Policy (New York: Cambridge University Press, 2009).

[16] Andrew Moravcsik, “Taking Preferences Seriously: A Liberal Theory of International Politics,” International Organization 51:4 (1997): 513-553.

[17] Jeffrey W. Legro, Rethinking the World: Great Power Strategies and International Order (Ithaca: Cornell University Press, 2005).

[18] Vincent Pouliot, “The Logic of Practicality: A Theory of Practice of Security Communities.” International Organization 62:2 (2008): 257.

[19] I apologize that I had not completed and posted the promised online appendixes as I should have at the time of the book’s publication and then neglected to do so as other matters seemed more pressing (though many elements appeared on the site as replication files and appendixes to articles in International Organization and Security Studies). They are now available here.

[20] Gabriel Almond and Sidney Verba, The Civic Culture: Political Attitudes and Democracy in Five Nations (Princeton: Princeton University Press, 1963).

[21] Owen clearly means tautological in this proper sense of the word. Hopf may be using the term more loosely, as a synonym for ‘non-falsifiable.’ However, I discuss at some length the operationalization of ‘narrative situation’ using content analysis, opinion polls, contemporary observers’ assessments, and audience response to particular arguments and storylines (see pp.60-61, and in practice throughout the empirical chapters).

[22] Hopf references a full-page New York Times advertisement in opposition to the 2007 “surge” signed by over 1000 U.S.-based IR scholars. I’m not sure what advertisement Hopf is referring to, and my web searches did not turn it up. There was a famous advertisement against the Iraq War published in September 2002 on the Times’ op-ed page, signed by a relative handful of IR luminaries. There was also the October 2004 statement signed by over 850 “Security Scholars for a Sensible Foreign Policy.” I think Hopf is referring to this latter statement (which was not directed against the surge and which did not, to my knowledge, result in a New York Times advertisement), and my discussion proceeds on that premise.

[23] Adam J. Berinsky, In Time of War: Understanding American Public Opinion from World War II to Iraq (Chicago: University of Chicago Press, 2009).

[24] David M. Edelstein and Ronald R. Krebs, “Delusions of Grand Strategy: The Problem with Washington’s Planning Obsession,” Foreign Affairs 94:6 (November/December 2015): 109-116.