This file is also available in Adobe Acrobat PDF format
The wrath of war has long been with humanity and even more so in the twentieth century. This book is an attempt to understand its causes by using the scientific method to delineate those factors that increase the probability of war. It does this by testing one of the major contemporary explanations of war—the steps to war. The explanation provides an analysis of why interstate wars occur in the modern global system and how a knowledge of these factors can help uncover the conditions of peace.
Today, war is still a pressing social scourge and an intractable intellectual problem. To explain it will require the best method available and this, we believe, is the scientific method. A variety of approaches and methods have their place in the analysis of war and greatly enrich each other, but we focus here on the use of statistical analyses of long-term interactions to delineate patterns. We do this in the context of an historical understanding of how different eras can have different patterns of behavior and in the context of insights we have derived from the numerous case studies of interstate wars and international crises that have been conducted in the last two decades. Nevertheless, we think that the systematic and rigorous analysis of replicable data that have been painstakingly collected provides a body of evidence, separate from historical and case-study analysis, that needs to be more fully incorporated into our theories and discussions of why war occurs. In addition, scientific analysis provides a way of thinking, a perspective, about social problems and the historical record that can be a source of new insights. It also provides a self-correcting mechanism—statistical testing—that allows the analyst to see whether theoretical explanations are empirically inaccurate and where they might be reformulated. Thus statistical analysis is a way of learning from evidence and building theory through a process of conjectures and refutations (Popper, 1963).
Because we adopt a scientific approach, the logic underlying the book is fairly straightforward. We begin with an explanation. We review some of the previous tests that have been conducted on it, and on the basis of these we construct a research design to test aspects of the explanation. We then present the various tests we have conducted and their implications for the explanation. We conclude with an overall assessment of the explanation’s empirical accuracy and the implications for the future study of the steps to war and for the prospects of peace.
Part I lays out the foundation on which the rest of the book is built. It presents the underlying theory of the analysis and the research design of the study. Chapter 1 presents the explanation and its logic. In brief, the explanation sees war as arising from a series of steps actors take to deal with the issues and grievances that separate them. Territorial issues, if handled in a certain way, are posited as the most war prone. Territorial issues are intrinsically salient and attract hard-line domestic constituencies, which makes compromise difficult. As disagreement persists, states resort to the threat or use of force. If militarized disputes repeat, this leads to a sense of rivalry between the contending parties. In the modern global system, realist diplomatic culture suggests that in the presence of security threats, states should increase their power by making alliances and/or building up their military. Because of the security dilemma, the making of alliances often leads to counter-alliances, and military buildups can produce arms races. These actions, instead of preventing war by peace through strength, as some realists claim, actually constitute a series of steps to war. The repetition of crises increases the number and influence of hard-liners in each side, which in turn tends to increase the level of conflict across crises and a sense of rivalry between disputants. Eventually a crisis comes along that escalates to war. While territorial issues are the most prone to this syndrome of power-politics behavior, any issue subject to it has an increased probability of going to war. All of these propositions hold, ceteris paribus. The details, as well as the logic connecting the steps, are discussed in the chapter.
Chapter 2 reviews the previous studies on the steps to war and how the research program has evolved in light of criticisms and in light of competing explanations. This chapter also sets out the research design and how the empirical tests conducted in the remainder of the book build on one another. It presents an overview of the data and measures used in the analysis.
Part II deals with the relationship between territory and war; in particular it deals with some of the main criticisms of previous research that shows that territory are highly war prone. Chapter 3 investigates the role of territorial disputes in increasing the probability of a militarized dispute escalating to war. A variety of tests (including a two-stage analysis that looks at possible selection bias) will be conducted. The chapter examines whether it is the mere presence of territorial disagreement or how territorial issues are handled that is more significant for bringing about war. The chapter addresses the issue of selection bias by seeing whether it is territorial disputes that give rise to war or the factors that produce territorial disputes in the first place that are responsible for war breaking out.
Chapter 4 looks at the joint role of contiguity and territory for increasing the probability of conflict and war. It also employs a two-stage analysis, that tests the counter-proposition that it is not territorial disputes that increase the probability of war, but the contiguity of states that accounts for war; that it is really the proximity of states and not territory per se that explains the onset of war. It also tests for other possible general selection effects that might be operating.
Part III examines the relationship between power politics and war. Chapters 5 and 6 examine the role of outside alliances, repeated disputes (including rivalry), and arms races as factors increasing the likelihood of both territorial and nonterritorial disputes escalating to war. Chapter 5 conducts this investigation by examining the long-term relations of pairs of states (e.g., Britain and Germany) during three historical eras (1816 –1945, 1946 –89, 1990– 2001) to see whether the probability of a pair of states having at least one war increases as those states make outside alliances, have repeated disputes, engage in arms racing and are contending primarily over territorial questions. Chapter 6 tests the same set of hypotheses during the same historical eras, but by looking at specific militarized disputes and their characteristics rather than the long-term relations of two states. Here we identify the risk factors that increase the probability that a militarized dispute will escalate to war. This design permits us to get at timing (e.g., which comes first, territorial disputes or outside alliances). Chapter 7 is the most technical chapter; it looks at statistical interaction effects for the propositions examined in the previous chapter. It also looks at why the post-1945 period is different from the classic 1816 –1945 international politics period, and whether the current era (post-1990) will more likely resemble the classic earlier period or the Cold War nuclear period.
Chapter 8 brings together the major results to see what we have learned about the probability of war and how well the steps-to-war explanation has captured the major patterns of interstate warfare from 1816 through 2001.It then looks at the implications of the study for future research in general, as well as within the steps-to-war research program. It closes by looking at some of the policy implications of the analysis for promoting peace in the future.
This book is the product of many years of collaboration. Its origins go back to 1998 when we were both at the same university. We applied for and received an NSF grant (SES-9818557) to test aspects of the steps-to-war explanation. We spent the next five years doing so, publishing a number of peer-reviewed studies. These studies, along with those we have published with others and those of our former students, constitute the larger research program on the steps to war. This book is the capstone of our NSF grant. It supersedes the earlier studies in that it is based on data going from 1816 through 2001, whereas the earlier studies went only through 1992. This enables us to look at three distinct historical periods, 1816–1945 (the classic international politics period), 1946 –89 (the Cold War period), and 1990–2001 (the post–Cold War period). These three eras permit us to make historical comparisons we could not do before. This book also extends the analysis into new areas, specifically an examination of interaction effects inherent in the steps to war.
One of the distinctive qualities of science is that it is a collective enterprise that has many scholars and students studying the same problem often using shared data and common conceptualizations. If the puzzle of war is ever to be solved, it will most likely be by such a community of scholars and not by a single individual working alone in a study. Today, there are more scholars working on the problem of war than at any other time in history.We have been fortunate to be working within a larger peace science community. This community of scholars, centered around the Peace Science Society (International), has been in the forefront of collecting data, developing new measures, debating concepts and methods, and encouraging the publication of rigorous research, all with the purpose of understanding war in hopes of promoting peace. We offer this book as our contribution to this larger effort.
Return to Book Description
File created: 6/16/2008
Questions and comments to: email@example.com
Princeton University Press