As the title suggests, Why Nations Still Fight is a follow-on to Why Nations Fight, published in 2010.Footnote ¹ It is neither a revision nor an extension of that book but a new project. There are nevertheless conceptual and empirical connections. In both books I ask why wars occur and what, if anything, might be done to reduce their frequency. I approach these questions in different ways and in this project I focus more on context, ask additional and related questions, and use a new, more comprehensive, and extended data set that encompasses so-called interventions as well as wars. It includes all significant interstate uses of armed force from 1945 to the present. I reconfirm some of my earlier findings and offer new substantive and theoretical insights, especially into the frequency and causes of miscalculations so often responsible for war.

Why Nations Fight argued that war is on the whole no longer an effective means of advancing security, maximizing wealth, or gaining status. States that start wars are likely to lose them militarily, and even if they come out on top militarily, they are only infrequently able to achieve their political goals. States still resort to force. I proposed two explanations for these phenomena: leaders fail to carry out informed cost-calculi before acting and are highly motivated to believe their military ventures will succeed. These two sources of miscalculation are closely connected because failure to consider the risks and possible downsides of an initiative can be the result of the need to succeed and the corresponding need to avoid the anxiety a serious assessment might or would arouse.Footnote ² In this volume I explore the role of miscalculation in a larger and wider range of cases, probing its immediate and underlying causes and the synergy between them.

The overall frequency of war has declined over the centuries, which I attributed to learning. In this study I ask why some countries and leaders have learned that war is generally unproductive and others have not. I recognize that interventions and wars sometimes succeed, although only in a minority of cases. Great powers were the first to recognize that wars are not good for economic well-being but among the last to learn that they no longer confer standing, status, or prestige. Why is this so? Why is it that the very states that claim to be upholding order are most often the initiators of wars that threaten order? Going by the numbers, the US is the greatest offender, followed by Russia.

In Why Nations Fight, I focused on wars involving rising and great powers between 1648 and 2003. In this study I address all major uses of force since 1945 and categorize them by type. This permits a finer-grained analysis of the origins of interventions and wars and forecasts of their future frequency. These types are based on the origins and kind of conflict, the nature and goals of initiators, their direct or indirect involvement, the extent to which force is constrained to the initial adversaries or spreads, and in one instance the political strategy behind the use of force. My fifteen types cut across interventions and wars and each tends to have a different frequency, cause, success rate, and likelihood of decline. On the whole, I offer generalizations within types rather than about intervention or war in general. I find, however, that miscalculations responsible for the use of armed force and its failure are endemic to all types of conflict.

I study interventions and wars together because any distinction between them is conceptually empirically arbitrary. Both represent uses of force to achieve political goals. I argue in Chapter 2 that there is no effective way of distinguishing between interventions and wars as their level of violence, range of initiators’ goals, and duration overlap considerably. Intervention, I contend, is a political category, not an analytical one. It is best employed to study why political actors label some uses of armed force as wars and others as interventions.Footnote ³ Throughout the book I refer to interventions as well as wars purely as a convenient way of referring to uses of armed force.

War is a legal and political category. It is defensible as an analytical one, although not unproblematic to operationalize. According to international law, wars should be declared, but this requirement has often been ignored. Tanisha Fazal found that only half of the wars fought between 1915 and 1948 were declared and hardly any since.Footnote ⁴ Leaders nevertheless acknowledge international law by their very avoidance of the term “war” and parallel efforts to legitimize their behavior by calling it something else, like police action, humanitarian intervention, and special operation.Footnote ⁵ In Russia, calling the invasion of Ukraine a war has resulted in fifteen-year prison sentences.Footnote ⁶

What political leaders call interventions are often called wars by the media and analysts, as were the US and Soviet “interventions” in Vietnam and Afghanistan. In the former, American forces were invited into the country by the local government. In both, foreign armies fought against local guerilla forces. Wars are typically waged against states and their governments, not domestic opponents. Wars are generally assumed to produce more casualties than interventions, but this too is not always true. The postwar communist government of Vietnam estimated that some two million civilians and one million combatants died in the course of American intervention.Footnote ⁷ Some wars produce relatively few casualties and are settled quickly, as was the so-called Football War in 1969 between El Salvador and Honduras.Footnote ⁸

I investigate success and failure in interventions and wars collectively and across types. Unsuccessful interventions and wars, I find, almost invariably involve serious miscalculations by initiators. American military involvement in Indochina, Argentina’s invasion of the Falklands, Iraq’s occupation of Kuwait, and Russia’s invasion of Ukraine are cases in point. I identify different kinds of miscalculation, the stages at which they occur, their related short- and longer-term causes, the ways in which they are sometimes reinforcing, and their distribution across cases. These different kinds of miscalculation and their causes have varying implications for the frequency of war and the possibility of reducing it. My analysis further suggests that the principal focus on improving decision-making procedures to reduce miscalculations is misplaced. Rather, I emphasize the importance of critically addressing the assumptions policymakers bring to the table. They are often unspoken but determinant, and frequently inappropriate and dangerous.

My findings highlight the inadequacy of existing approaches to war and theories nested in them. Marxism explains foreign policies primarily with reference to economic motives. There were very few uses of armed force in my data set that had identifiable economic motives, and ever fewer where they were primary. By contrast, there were many uses of armed force that were costly in a material sense; they were expensive and limiting in access to markets, technology, and capital.

Anarchy is central to realist accounts of war. It is an ideal type that many realists mistakenly describe as a reality. International relations differ from domestic relations, but more in degree than in kind, and the thickness of society varies greatly across regions. One size does not fit all. Not surprisingly, my data indicate variation across regions in the frequency of the use of force. The thickness of society accounts for some of this variation. Other factors are equally important, among them domestic structures, great-power rivalries, and degrees of miscalculation. There is something to be said for reversing the arrow of causation and regarding the degree of anarchy in any region as more the result than the cause of armed conflicts.

Other realists stress the balance of power, or disagreements actors have about it.Footnote ⁹ Here too my data offer little support. A significant percentage of wars were started by weaker parties. Disagreements about the balance rest on the assumption that leaders recognize it as something important to which they should pay attention and do their best to calculate. In many of my cases there is no evidence that initiators were interested in the balance of power or tried to determine what it was. The balance of power is not as central to international relations as realists contend. It may discourage weak states from attacking very powerful ones but it is all but indeterminate in the absence of such obvious power imbalances.

The other popular paradigm in the study of war is rationalism. In Why Nations Fight I exposed the serious conceptual problems and even contradictions in the existing literature.Footnote ¹⁰ Many rationalists emphasize the balance of power, and disagreements about it, but also problems in signaling and the overall absence of information. As their name suggests, rationalist theories of war assume a high degree of rationality on the part of actors, or at least the ability of rational models to capture the dynamics of international conflict.

My data set and numerous case studies suggest that leaders are at their least rational when considering the use of force.Footnote ¹¹ There is accordingly no way rationalist theories can capture their behavior. Rationalist theories further err in their approach to information. They emphasize the lack of information that would permit accurate judgments about capabilities, resolve, and intentions. My cases suggest that such information is often available but ignored. Leaders fail to conduct any serious calculus of expected cost and gain before using force. To the extent this is so, we need an “irrationalist” paradigm with theories that start from the premise that leaders see what they want to see and ignore what is threatening to them, and that offers propositions to explain why this happens and when it is most likely to happen. My data set reveals that miscalculation occurs in nearly three-quarters of postwar uses of armed force. It should be seen as the norm, not as deviant behavior. If my data set were extended further back in time, judging by the case studies, it would show the same patterns as nineteenth- and early twentieth-century wars reveal serious miscalculation. My data set in Why Nations Fight found that since 1648 initiators have won only about half their wars militarily and achieved their political goals in only about a quarter of them.

Much of the appeal of Marxist, balance-of-power, and rationalist theories is their parsimony. They reduce complex patterns of interactions to a simple set of rules. In the process, they deny the importance of culture, history, path dependency, and agency. I do my best to bring them back into the picture, not because I am fascinated by them but because we cannot account for wars and so-called interventions without them. I reject the possibility of parsimonious theories and make the case instead for a kind of forecasting based on theoretical insights. Theory can help us organize our inquiry, and nothing more, because context is determining.

Although I emphasize the role of miscalculation, I reject the idea that there is any simple relationship between miscalculation and war. Misjudgments in the form of erroneous assessments of capability, resolve, and intentions are one source of miscalculation; so too is the belief that force can be used successfully, and often at low cost, to achieve political goals. Force is likely to remain an all-too-common practice as long as leaders mistakenly believe they have a good prospect of success.Footnote ¹² We should also note that miscalculation can make war less likely when it prompts caution because of exaggerated estimates of adversarial capability, resolve, intentions, or third-party support.

It is sometimes impossible to know about adversarial capabilities, intentions, and resolve with any certainty beforehand. Erroneous assessments become apparent in retrospect. As I was writing this book, there was an ongoing debate about whether Putin would use tactical nuclear weapons against Ukraine in response to NATO providing it with weapons capable of striking deep into Russia.Footnote ¹³ Some analysts dismissed it as a bluff; others took it seriously. We simply did not know. Many other misjudgments were avoidable on the basis of information available at the time and cannot be considered innocent. We cannot assume the opaqueness or transparency of the international environment. It is an empirical question.

My data set is sensitive to agency. I explore the concept, its applicability to individuals and states, and the empirical problem of determining how much freedom of action either possessed. Agency begins in the eye of the beholder. Leaders may believe that they possess freedom of action or are severely constrained, or anything in between. Leaders who feel compelled to act for political or national interests may be motivated to exaggerate their ability to do what they consider necessary, as arguably Mikhail Gorbachev was in seeking to end the Cold War.Footnote ¹⁴ Leaders who feel constrained may downgrade the severity of these threats. Context – or, more accurately, perceptions of it – helps shape but does not determine agency, just as agency can help shape but does not determine context.

Leader estimates of their freedom of action can differ from those of other political actors and observers. These varied understandings generally have contextual and personal explanations. I identify some of them and the consequences of over- and underestimation of agency for interventions and wars. Agency is important in another sense. Leaders differ in personality, confidence, self-esteem, experience, goals, decision-making styles, and openness to dissent and criticism. Some leaders are noticeably more ambitious and aggressive than others. Consider the counterfactuals that Idi Amin, Muammar Gaddafi, Saddam Hussein, and Vladimir Putin never came to power. We might have avoided nearly 10 percent of the interventions and wars in my data set. But then we must also imagine other leaders who might have achieved high office elsewhere and begun wars that never happened.Footnote ¹⁵

Agency is generally juxtaposed with structure. In a country like Pakistan, where presidents often cannot exercise effective control over either the military or intelligence services, and where independent militias and terrorist groups proliferate, it would have been difficult for any leader to prevent the kinds of provocations that led to crises, fire fights, and wars with India. Agents are also constrained by the political culture of the society in which they operate. Given the hold of the “Korean lesson” on the Chinese political elite since the Ming Dynasty, it seems unlikely that any communist leaders of that country would have stood aside when US forces attempted to occupy North Korea in 1950.Footnote ¹⁶

Agency and structure are both important. I identify multiple structural and ideological features of international and domestic political environments I believe to have been responsible for uses of force and some of the conditions that give rise to them. These include weak states with weak governments, colonial rebellions and the breakup of empires, postcolonial territorial disputes, divided nations, partitioned countries, regional rivalries, rump states with leaders intent on regaining lost territory and international standing, great-power and cultural arrogance, and exaggerated fears for credibility.

A couple of caveats are in order. I have argued elsewhere – and do so here too – that we cannot explain the use of force solely with reference to underlying conditions. It frequently requires appropriate catalysts or immediate causes. In their absence, force may not be used.Footnote ¹⁷ Catalysts can be causes in their own right, as the 1914 assassinations at Sarajevo were for World War I.Footnote ¹⁸ At the other end of the causal spectrum we must also step back from underlying causes to look at the political, economic, and social character of regional and international societies. The kinds of conflict and their frequency are very much a function of the society in which actors and their political units are embedded – or excluded. I will return to this last problem in Chapter 14.

Miscalculations can be a source of both war and peace. I explore only the former in this book. I argue with some confidence that many interventions and wars are due to miscalculation. I further find that these miscalculations are not for the most part attributable to a lack of information. They are far more likely to be willful or the product of false and even dangerous assumptions. Like miscalculation, a lack of information can sometimes be restraining, and too much information – one of the consequences of the information revolution – can also have divergent consequences. This is not a relationship that has been studied in the context of the use of force.

Optimism versus Pessimism

When I published Why Nations Fight in 2010, I was reasonably confident that interstate war was on the decline.Footnote ¹⁹ Wars had been fought throughout the postwar era, some fifty of them according to one data set, with casualties in the millions.Footnote ²⁰ These wars brought additional suffering in the form of malnutrition, disease, destroyed homes and infrastructure, family separations, and denied educational, economic, and marital opportunities. They also produced refugees: 108.4 million in 2024 according to the United Nations Refugee Agency. Some 76 percent of these refugees are “hosted” by low- and middle-income countries where they find it difficult to make new lives for themselves.Footnote ²¹

My optimism in the face of this suffering derived in the first instance from a longer view of warfare. The frequency of interstate violence had diminished over the last 400 years, although some of the twentieth-century wars were considerably costlier than their predecessors.Footnote ²² Using a data set of wars fought since 1495, Kalevi Holsti found that interstate war had consistently declined over the centuries and asserted that “the world is significantly safer today than in any previous period.”Footnote ²³ More striking still, since 1945 there had been no wars between or among great powers. For almost eighty years, Europe, the principal battleground of these wars, had been peaceful in a way that almost nobody in 1945 had expected.

My second reason for optimism was the rising cost of war in lives and money and growing awareness of these costs. Wars were traditionally fought to gain territory, wealth, status, or security. In Why Nations Fight, I argued that they were no longer effective means to these ends, and there seemed to be growing recognition of this political truth. The benefits of war in today’s world are few, if any, for an initiator. Before the era of nationalism, conquest generally required only the defeat of whatever military forces were mobilized against an invader. Now, an invader must subdue a people, who, if mobilized by appeals to nationalism or religion, as in Vietnam and Afghanistan, can make the cost of occupation all but intolerable. This is a near certainty when they are provided weapons and other forms of assistance from the outside.

Technology plays an equalizing role. It offers superior weapons to the stronger, invading state but often ways of offsetting these advantages to weaker, less wealthy states and non-state actors. Taliban deployment of improvised explosive devices (IEDs) against armored personnel carriers, and Ukrainian use of drones, jamming, and telephone monitoring, illustrate this truth.Footnote ²⁴ During the 2024 Gaza war, relatively weak Houthi rebels in Yemen used readily available and inexpensive drones to attack commercial shipping in the Red Sea, compelling major shipping companies to divert their vessels around the African continent.Footnote ²⁵ When major powers use force, it is by contrast disproportionately expensive, and often does not serve their ends.Footnote ²⁶ Another consideration here is the sharing of weapons, with weaker states and militias receiving modern weapons and the training necessary to use them. An early example was Washington’s supply of Stinger missiles to Afghani rebels, which proved very costly for the Soviet Union.Footnote ²⁷

Realists have long denied any real chance of eliminating war and constructing a viable international order. Some offer the recent wars in Ukraine and the Middle East as more evidence.Footnote ²⁸ The Middle East has been one of the most violent regions of the world since 1945 and has been the site of several costly wars, including the eight-year-long war between Iraq and Iran in the 1980s and the Syrian civil war, from 2011 to 2024.Footnote ²⁹ The latter conflict drew in Turkey, Iran, Russia, and the US.Footnote ³⁰ In 2023, a bloody cross-border terrorist attack by Hamas in Israel provoked region-wide violence that engulfed four states and three militias.

The Syrian civil war was largely ignored by the Western media but the Ukraine and Israel–Palestine conflicts received daily coverage. This made Western publics more fearful of the prospect of war than at any time since the hotter days of the Cold War.Footnote ³¹ A study conducted in Germany in the summer of 2023 found only 7.7 percent of those surveyed were not worried about the prospect of a nuclear war. Some concern was acknowledged by 45.7 percent and 46.6 percent admitted to acute fear of a nuclear war.Footnote ³² In a poll conducted by the American Psychological Association following Russia’s invasion of Ukraine, seven in ten Americans worried “that we are at the beginning stages of World War III.”Footnote ³³ Time magazine reported that American politicians were talking about the prospect of war.Footnote ³⁴ The Bulletin of the Atomic Scientists moved its minute hand closer to midnight, signifying a shift among its editors in the likelihood of war.Footnote ³⁵ In March 2024, Prime Minister Donald Tusk of Poland, a man known for his common sense and caution, warned that Europe has entered a “prewar era” and must dramatically increase its defense spending.Footnote ³⁶

I think much of this was an overreaction. It was also indicative of the tendency of people – including well-informed, intelligent people – to exaggerate the importance of the unexpected and to make predictions on that basis. An earlier case in point was the widely expressed American belief in response to the 9/11 terrorist attacks that the world had fundamentally changed. Should the Ukraine and Gaza wars usher in a more violent decade and confirm Donald Tusk’s characterization, pessimists will rightfully carry the day. If these wars are standalone events, in part because they discourage other would-be initiators, there is room for hope. All the more so if they strengthen international norms against invasion and conquest. Even if both wars have primarily structural causes, there would still be grounds for optimism if those structures are on the wane or their consequences overcome through learning and concerted political action. We must keep an open mind with respect to both possibilities.

The core distinction between optimists and pessimists is their understanding of social relations. Pessimists insist they are more or less unchanging across history and cultures because they reflect human nature. There will always be inequalities, with benefits achieved and maintained by means of power. Realists assert that international relations lack a leviathan and is accordingly a self-help environment in which states must ultimately rely on armed force. Globalization is no panacea. Europe’s economies were tightly integrated on the eve of World War I, realists rightly remind us, and did not reach this level again until the 1990s.Footnote ³⁷ Nor can new weapons transform the character of international relations, although they can make war more deadly. Realists note that ever since the invention of the crossbow in medieval Europe, people have consistently and mistakenly predicted that the increased lethality of war would compel restraint. Bayonets, rifles, machine guns, high explosives, aerial bombing, and nuclear weapons shocked the publics of their era and gave rise to such predictions.Footnote ³⁸ Victor Hugo warned that hot-air balloons had the potential to deliver devastating aerial attacks and urged their ban.Footnote ³⁹

Optimists believe that reflectivity and learning can enable people and their political units to escape from tragic and seemingly timeless scripts. John Mueller points to the success societies have had in outlawing slavery and dueling and moving closer in recent years to racial and gender equality.Footnote ⁴⁰ Michael Howard observes before World War I, “war was almost universally considered an acceptable, perhaps an inevitable and for many people a desirable way of settling international differences.”Footnote ⁴¹ Most people no longer feel this way. World Wars I and II brought about a seismic shift in opinion. Liberals claim that developed economies have made great strides in devising multilateral and supranational institutions that limit, mitigate, and overcome the consequences of periodic economic crises.Footnote ⁴² The European Union has many critics but most international relations scholars agree that a major war among western European countries is no more likely than one between the United States and Canada – or at least until Donald Trump became president and expressed his desire to annex Canada and make it a state. In 1957, Karl Deutsch developed the concept of a “pluralist security community,” a region populated by sovereign states in which war had become all but unthinkable. He described North America (the US and Canada) and Scandinavia as regions in which security communities had not only developed but were robust.Footnote ⁴³ By the end of the Cold War, liberal international relations scholars were claiming that the entire North Atlantic Community – as Deutsch had envisaged – had become a pluralistic security community, as had New Zealand and Australia, if not much of the Pacific rim.Footnote ⁴⁴

This debate encodes certain ironies. Pessimism about war often rests on a bed of optimism about human nature, while optimism about peace frequently invokes pessimism. The security dilemma, so central to realist understandings of international conflict, assumes that national leaders are rational actors capable of understanding and responding intelligently to the constraints and opportunities generated by the international environment.Footnote ⁴⁵ Indeed, instrumental reason coupled with fear for security is what leads states to act in ways that end up making them and others less secure. Reason also provides the incentive to develop and deploy ever-deadlier arsenals. Optimism about the ability of humans to reason is what generates deep pessimism about their ability to live in harmony with one another. There is nothing new about this negative understanding of the power of reason; it goes back to the ancient Greeks. The first stasimon of Sophocles’ Antigone sings the “wonders of man and how he tames nature with his cunning and contrivances but when prompted by restlessness and evil is horrifying to behold.” The Greek word for wondrous, deinon, also means frightening.Footnote ⁴⁶

Optimists mobilize pessimism as a catalyst for radical change. Some Marxists consider war inevitable and horrible but necessary to bring about socialist revolution. Arms controllers and environmentalists deploy a variant of this logic. They predict the worst possible outcomes if new weapons systems are not controlled and if humankind does nothing effective to control climate change. Their pessimism is intended to appeal to the emotions and reason of public opinion and policymakers.Footnote ⁴⁷ Like realists, optimists envisage the combination of fear and reason as a powerful, positive incentive for change.

There are good reasons for questioning both understandings of international relations. Human beings are not rational in the ways realists and rationalist theories and models assume. This truth is well documented in case studies and also statistically. It is abundantly clear from the Cold War and the present-day world that great powers greatly exaggerate threats to their security, overreact, and behave in ways that make them less secure.Footnote ⁴⁸ The democratic peace strikes its critics as more ideology than empirical truth.Footnote ⁴⁹ Even if democracies more or less avoid war with one another, they are the most frequent initiators of interventions and wars.Footnote ⁵⁰ As for globalization, it is hardly an unalloyed benefit and there is no evidence that it has promoted peace and harmony, as liberals predicted.Footnote ⁵¹ None of our theories have served us well. They are simplistic and the problem of war and peace is complex.

We currently live in an era of pessimism. Western publics, and young people especially, worry about their personal prospects and some about the planet more generally. To cite one study, six in ten Americans say that life for people like them is worse than it was fifty years ago.Footnote ⁵² This pessimism may have contributed to Donald Trump’s election in 2024. In many countries pessimism, frustration, and anger have led to a swing to the right politically or disengagement from politics. Emotional appeals that resonate are not to positive kinds of changes but to nationalism, keeping out or kicking out immigrants, and support for leaders who allege they are anti-system. As Aristotle knew, and neuroscience has documented, emotions are critical to good judgment but they are a double-edged sword. They can contribute to good judgment or enable the most horrendous crimes.Footnote ⁵³

Assumptions about the social world – optimistic, pessimistic, and other kinds – become the basis for theories. They too are Janus-faced in their consequences. They constitute useful, even necessary, tools for making sense of reality and gaining some insight into the future. However, they rarely succeed in predicting events or even trends in international relations and they blind those committed to them to other causes of events, outcomes, and patterns.Footnote ⁵⁴ How do we maximize the benefits of theories and minimize their drawbacks?

My approach is to use theory as a starting point for forecasts. Forecasts take the form of narratives that lead from the present to the future. They assume that much of what happens cannot be predicted because it is the product of context in the form of agency, path dependence, confluence, accident, and interactions among them. These determinants are all outside of any theory of international relations. Our forecasts accordingly have a significant probability of being wrong. We can and must monitor them carefully and change our storylines when new developments call into question the existing one. This updating is critical, distinguishes forecasts from predictions, and attempts as far as possible to take contextual changes into account.Footnote ⁵⁵ In lieu of making predictions in the conclusion, I offer competing forecasts, and with each stipulate what kind of information we need to have more or less confidence in any of them – or perhaps the need to create new ones.

Organization of This Book

The book is divided into four parts: description and defense of data set, types of interventions and wars, success and failure and its causes, and what has been and still needs to be learned about the largely counterproductive consequences of the use of force. The central findings of the study are that interventions and wars have a low rate of military and political success, and that the consequences of failure can be costly. The evidence belies those rationalist theories that explain the existence and persistence of war in terms of their positive payoffs. I turn to political culture, political need, and miscalculation for explanations. They help to account for the conundrum that interventions and wars continue despite their high rate of failure being no secret.

Chapter 2 describes my data set of eighty-eight cases of interventions and wars since 1945. I begin with a justification for including interventions and wars in the same data set and making no distinction between them. I describe the criteria I used to include or exclude uses of forces in the data set and the problems and sometimes difficult choices this entailed. I elaborate fifteen categories of armed force and explain why they are critical to my analytical goals. I discuss my coding rules, the problems to which they give rise, and how I addressed them. I include the entire data set, categories, and coding of individual cases, and explanations for these codings in Appendix B. Appendix A offers a table of results that summarize my findings by category of analysis.

The second part of the book, Chapters 3–10, analyzes different types of the use of force. Rather than addressing the problem of war generically, as most studies do, I break out interventions and wars into fifteen categories, all of which tap into different causes or types of the use of force. They are colonial, postcolonial, divided nations, partitioned countries, rump states, regional rivalries, proxies, great powers, border disputes, spillover, revenge, spheres of influence, quests for hegemony, civil wars with outside intervention, and leader efforts to gain recognition and greatness. Almost all these categories include what are commonly referred to as interventions and wars, which is another reason for analyzing them together. A number of cases lend themselves to multiple coding. Initially, I considered reducing the number of types but combining some or subsuming some to other types. I decided against this because telescoping categories and cases proved awkward and left a substantial number of cases without types in which to classify them. A list of fifteen types lacks parsimony but is worth it analytically. It not only captures all my cases but also allows me to link types due to shared causes, and to examine cases from multiple perspectives.

These fifteen categories of armed force have on the whole distinctive causes. I situate each type in its historical context, describe its most common underlying and immediate causes, evaluate its military and political success rate, and identify miscalculations and their nature. I ask what could be done to reduce their frequency. In this part I also include a chapter on ethical traps. It is less a type of war than a strategy pursued with some success by weaker adversaries.

The third part of the book, Chapters 11–13, consists of comparative analysis across cases. It addresses the degree of military and political success, the extent, reasons, and consequences of miscalculation and the reasons for it, and the motives of initiators. I begin by distinguishing between military and political success, with an emphasis on the latter. I evaluate both kinds of success in the short and longer term, with the longer term being the more meaningful category. I examine some of the ways in which short- and longer-term success and failure are related. I compare degrees of success across types of interventions and wars and analyze them by type of initiator, region, and decade. I also ask what kinds of goals are more and less likely to be achieved by force.

My data set indicates that miscalculation is common to uses of armed force. Leaders regularly misjudge the military capability, resolve, and intentions of states they attack, and sometimes their own country’s capability, resolve, and intentions. They routinely fail to think ahead, planning only for the initial stage of a conflict, and often have no “Plan B” if their initiative falters or fails. I posit two principal causes for this kind of behavior. The first has to do with the failure of leaders and their advisers to gather relevant information and conduct meaningful assessments of the likely costs and gains of using force. The second is the assumptions that undergird such analysis – if it is indeed carried out. These assumptions can be about oneself or an adversary, the utility of war, or its ability to achieve political goals. Some kinds of miscalculation have short-term causes and are context-specific. Others are institutional and cultural and are longer-term in the development and more general in their effects. I analyze both and some of the connections between them.

Chapter 13 on motives codes them at two levels: immediate goals and underlying goals. The former refer to the range of specific foreign and domestic objectives leaders seek. With regard to underlying motives, I turn to my Cultural Theory of International Relations and Why Nations Fight. In both books I follow Thucydides, Plato, and Aristotle to argue that foreign policy is motivated by fear, interest, and honor.Footnote ⁵⁶ Interest (appetite) and honor (thumos) are fundamental human drives. Fear is an emotion that becomes increasing pronounced when reason loses control over the appetite and thumos of some actors, making others feel threatened. I analyze both kinds of motives across types of conflict and actors and also inquire about the extent to which they are differentially associated with success and failure. Readers unfamiliar with my theory of international relations can look at either of these books or the short elaboration of motives provided in Chapter 11.

In the final part of the book, Chapter 14, I review my findings in comparative perspective, drawing inferences across categories of conflict, decades, regions, motives, and outcomes. This fine-grained analysis generates propositions about the future likelihood of each kind of armed force. My analysis stresses the importance of learning. The decline of war, I have argued, is in its most fundamental sense attributable to recognition that material well-being and security for the most part are better advanced by peaceful means. This is now happening as well for recognition, standing, and honor – all expressions of thumos. This is a more recent development and still underway. Many political leaders and foreign policy elites have learned that status is no longer achieved by means of warfare and have sought it by other means. The great powers on the whole have not yet recognized this political truth, a finding I document in Chapter 9. I offer thoughts about why this is so and what might be done to bring about a change in their beliefs and preferences.

The other great barrier to peace is the character of international society. It is structured to reward the powerful and punish the poor. This inequality and its substantive and psychological consequences are a powerful cause of conflict and disorder. Here too, great-power discourses, abetted by realist and liberal international relations theory, justify these inequalities and explain away their consequences. They generally attribute conflict to rebels who fail to recognize the value of the so-called rules-based order. In much of the developing world this is regarded as hypocrisy, and all the more since Donald Trump is blatantly violating these rules and doing his best to destroy the existing order. I will return to Trump in Chapter 14. His administration constitutes a serious challenge to this order and makes the future of international relations that much more unpredictable.

In Chapter 14, I discuss the implications of my study for international relations theory. I noted at the outset how my findings cannot be squared with either realism or rationalism. Building on Max Weber and my cultural theory of international relations, I highlight the importance of motive and the different rationalities associated with different motives and what they have to say about the use of armed force, the likelihood of miscalculation, and the chance of success. This understanding offers a more pessimistic take on the future of war.

My study is built on data, and my analysis is quantitative and qualitative. My data set of eighty-eight cases includes what I consider to be the universe – or close to it – of significant interstate uses of armed force since 1945. Readers may have quibbles about other uses force that might have been included in the data set. I doubt that many such examples will be proposed or, if included, would have affected my findings in any meaningful way. Accordingly, no statistical tests for significance are required to ascertain how representative my data set is, which would be necessary with a sample.

With eighty-eight cases, the data set is large enough to conduct some kinds of statistical tests, but all my analytical categories involve smaller numbers of cases because I generally analyze cases within categories. My most populated categories have under thirty cases, and most fewer. I report my results in numbers and percentages. Readers can decide for themselves how meaningful these numbers are. There are no established benchmarks for the frequency of causes, success, or miscalculation. All judgments are subjective and take on meaning in context. I make a number of inferences in this regard and readers can decide if they appear justified.

Many international relations books have standalone theory chapters. This one is an exception. It relies on theory for case selection, coding, and analysis and more general inferences about the future of war. Case selection and the concepts that guide it are discussed in the next chapter. The concepts are not derived from particular theories but devised de novo to critique existing data sets and provide the justification for this one. They address such questions as including interventions and wars in the same data set – a decision I justify further in Chapter 2, the value of violence as the criteria for case selection, and other criteria that might be relevant. The categories of conflict that constitute my data set were generated inductively from the cases, but of course most of them already exist in the literature. The novelty is twofold: some of these types (e.g., rump states, partitioned countries, revenge) are not commonly used in either quantitative or qualitative studies of war. And rarely, if ever, are categories compared and combined and analyzed separately and collectively.

Let me close by stepping back from my project to reflect further on the distinction between peace and war. For positivists, this is an empirical problem. It has an empirical component, to be sure, but it is primarily a conceptual and normative question. Peace and war are not abstract analytical categories but political terms that have cultural and legal origins and political and ethical meanings. Neither the ancient Greeks nor modern Europeans framed them as opposing binaries. They were closely connected and overlapping, and even in part co-constitutive.

From Immanuel Kant on, theorists have argued that peace means more than the absence of violence. For Kant, the extinction of war requires a moral transformation that is then reflected in individual and collective behavior at every level of social aggregation.Footnote ⁵⁷ For liberal thinkers, the end of war also involves a human transformation: appetite had to replace thumos as the dominant human drive and political systems had to become democratic and prioritize citizen desires for material well-being. For some of their critics, globalization and a liberal or democratic peace is not a real peace but a means of furthering exploitation of weaker states and peoples.Footnote ⁵⁸ Other opponents of the existing order focus on the nature of the order itself, particularly hierarchy, how it is imposed or enforced, and the lack of alternatives available for those who dissent.Footnote ⁵⁹ Given the current state system, Bertrand Badie insists, the absence of war is only a truce (une trêve).Footnote ⁶⁰

If we buy into even part of this critique – which I certainly do – it follows that the narrow empirical focus of war cannot address its real causes or make meaningful forecasts about its frequency. It is a scholarly smokescreen that obscures the real causes of war and, albeit unwittingly, legitimates many of the practices responsible for it.Footnote ⁶¹

How should we address this problem? I opt for two strategies. The first pertains to the operational definition of the use of force. It should be conceptual, not merely empirical. I will argue in the next chapter that separating war and peace by means of battle deaths – regardless of number – is a meaningless and even pernicious exercise. We must recognize that there is no simple metric that works, and that the reason for this is that intervention-war and peace are not distinct categories. They are fuzzy sets at the very least, so we need multiple markers that fit only in part. This is only a problem if one believes, as so many positivists do, that our categories can map neatly on to the social world. The difficulties that arise in including and excluding events from this – or any other – data set are not so much problems as they are opportunities. They have the potential to open our eyes to the limitations of our methods, and more importantly in this instance, to the many links between so-called peace and intervention and war that make it difficult to compartmentalize them conceptually.

This recognition prompts a second strategy, one that requires bridging the wall positivists are so desperate to build between the empirical and the normative. Analyses of the use of force and other international phenomena generally take the world as given. Realists in particular pride themselves on addressing the world as it is, not as we would like it to be. This is another false dichotomy, as prominent realists like Hans Morgenthau, E. H. Carr, and John Herz understood well.Footnote ⁶² Sophisticated realists have always recognized the extent to which war and peace are social categories that only take on meaning within social orders. Some social orders value the use of force and others constrain it. Still others do both.Footnote ⁶³ To understand the causes of interventions and wars, how they are fought, their consequences, and their frequency, we must situate our analysis within the social order in which they occur. To change any of these in any fundamental way, we must change the social order. For realists like Morgenthau, Carr, and Herz, the factual and conceptual, empirical and normative, and diagnosis and prognosis were inseparable. They sought to describe the world as it was but did so in the hope of providing justifications for this transformation.

In looking at the use of armed force in the present world, we must accordingly ask what features of the political, economic, and social order encourage and constrain the use of force. If we are interested in reducing it, we must also ask which of these features must be altered. I will return to this more general, normative question in Chapter 14.