The Weary Titan: Britain and the Experience of Relative Decline, 1895-1905

International Relations Theory and the Assessment of National Power

Historians who live in democratic ages are not only prone to attribute each happening to a great cause but also are led to link facts together to make a system. . . . As it becomes extremely difficult to discern and analyze the reasons which, acting separately on the will of each citizen, concur in the end to produce movement in the whole mass, one is tempted to believe that this movement is not voluntary and that societies unconsciously obey some superior dominating force. . . . Thus historians who live in democratic times do not only refuse to admit that some citizens may influence the destiny of a people, but also take away from the peoples themselves the faculty of modifying their own lot.

—Tocqueville, Democracy in America

How do statesmen think about power and, in particular, how do they seek to measure the relative power of the nations they lead? How do individuals and entire governments become aware of changes (and especially unfavorable ones) in the relative power of their own country? How do nations seek to adapt to such shifts? In other words, what is the relationship between changing power and changing assessments, on the one hand, and shifting assessments and evolving state policies, on the other?

These would appear to be questions of considerable theoretical and practical significance, yet they are generally little studied and poorly understood. The first section of this chapter will be devoted to an examination of several of the major existing bodies of theoretical literature on international relations. These writings either ignore the question of how relative power is actually measured, gloss over it, or highlight the problem without resolving it. Despite a general lack of direct attention, there are nevertheless two models of the processes of assessment and adaptation implicit in the existing literature, and these will be discussed in the second major portion of this chapter. Questions of method will be addressed in a final section in which the overall structure of this study and the organization of its four central chapters will be laid out and explained.

Theories of International Relations

Structural Realism

Perhaps the most influential approach to the study of international relations in recent years has been that referred to as structural realism or neorealism.¹ Adherents of this school believe that world politics can best be understood at what they call the systemic level of analysis.² The most important characteristic of an international system in this view is its structure, and that, in the formulation put forward by Kenneth Waltz, is determined by the principle on which the system is organized, the functions of the units of which it is composed, and the distribution of capabilities among those units.³

General definitions aside, the units that make up (modern) international systems are nation-states, and the principle that governs them is that of anarchy. In practice, therefore, the structure of such anarchic systems is determined by the way in which power is distributed among states. Structural realists hold that the shape of a particular system will strongly determine the behavior of the states that make it up; structure (like anatomy), in this view, is akin to destiny. Changes in structure will also, quite logically, lead to generally predictable changes in behavior.

The appeal of structural theories lies in their parsimony and their promise of predictive power. Conversely, the weakness of such systems of explanation is their tendency to collapse into a materially driven determinism. In the words of one noted political scientist: It has been a peculiar tendency of recent social science to disparage the importance of learning, cognitive ideas, and understanding. Few theories incorporate these concepts as major explanatory variables.⁴ This observation applies particularly to structural realism, which has little or nothing to say about how statesmen think and, in particular, how they think about power.

Waltz presents balance-of-power theory as an expression of the importance of structure in world politics. Under conditions of anarchy (and assuming a desire for self-preservation) the major states that define the structure of an international system will tend to act so as to counterbalance one another. If one state behaves aggressively or begins to augment its power relative to the others, some of them will usually form an alliance to defend themselves. This inclination is a strong one and can bring together countries that would otherwise have nothing in common. As Waltz puts it: We find states forming balances of power whether or not they wish to.⁵

But is balancing inevitable? There would not appear to be any reason, either logical or empirical, for believing that this must be the case. As Arnold Wolfers observes in his criticism of the more traditional forms of automatic or self-regulating balance-of-power theories:

Some weak countries seek safety by getting on the bandwagon of an ascending power, hoping somehow to escape complete subjugation once their powerful friend has gained supremacy. Other countries are so absorbed with their internal affairs or so unheeding of national power that the effects of their policies on the distribution of power, whether helping to preserve or upset the balance, are purely accidental.⁶

Waltz tends to downplay the possibility of what he calls band-wagoning,⁷ but he does not go so far as to suggest that balances must form or that the behavior of any particular state will be completely determined by the shape of the system in which it operates. Ultimately Waltz is careful to avoid the trap of structural determinism. Balance-of-power theory, he argues, explains the constraints that confine all states. The clear perception of constraints provides many clues to the expected reactions of states, but by itself the theory cannot explain those reactions. They depend not only on international constraints but also on the characteristics of states.⁸

This seems both sensible and familiar, but it leaves a number of important unanswered questions. What are the internal characteristics that determine how a state will respond to external pressures? And who is possessed of the clear perception of constraints to which Waltz refers? Is it the remote, omniscient analyst or the statesman forced to make decisions without benefit of hindsight or perfect information?

Shifts in the distribution of power within an international system may be real in some sense, but they may fail to have any impact unless and until they are perceived. A nation whose leadership does not realize that its power is declining relative to that of another country will probably not feel compelled to enter into protective alliances with third parties. A knowledge of how such conclusions are reached and acted upon would appear to be essential to any complete balance-of-power theory, regardless of the language in which it is couched.

Similar criticisms can be made of the cluster of theories that forms another variant of structural realism. In 1973 Charles Kindleberger advanced the notion that the stability and openness of an international economic system is critically dependent on its structure. Specifically, Kindleberger suggested that stability requires hegemony. Only when one country is both willing and able to take on the role of leadership will a lasting free-trade regime be possible. If the hegemonic power loses either its desire or its capacity to manage the system, then, barring the emergence of a successor, protectionism and general economic disorder will result.⁹

Kindleberger’s initial hypothesis has been elaborated into what has become known as the theory of hegemonic stability by political scientists concerned with understanding the effects of what they see as the recent loss of American world economic leadership.¹⁰ Attention to the general problem of hegemonic decline has also sparked renewed interest in cyclical theories of global politics. Robert Gilpin, for example, writes that all of history can be understood in terms of the rise and fall of successive hegemons. As one state grows and matures it begins to face challenges from newer competitors. Eventually the question of world leadership will be resolved by war. But, Gilpin points out,

the conclusion of one hegemonic war is the beginning of another cycle of growth, expansion, and eventual decline. The law of uneven growth continues to redistribute power, thus undermining the status quo established by the last hegemonic struggle. Disequilibrium replaces equilibrium, and the world moves toward a new round of hegemonic conflict.¹¹

An earlier and less developed version of this idea was advanced as the theory of the power transition by A. F. K. Organski. Again, differential rates of economic growth are held to be the motive force that drives a simple but seemingly inescapable mechanism:

At any given moment the single most powerful nation on earth heads an international order. . . . In the present period, the most powerful nation has always been an industrial nation. . . . As new nations industrialized, the old leader was challenged. . . . Ordinarily, such challenges by newcomers result in war. . . . One could almost say that the rise of such a challenger guarantees a major war.¹²

What both the economic and the more broadly political theories of hegemonic stability and decline have in common is their strong deterministic quality.¹³ It is often claimed that a change in the structure of the international system (i.e., in the distribution of power within it) must produce certain specified consequences, whether the onset of protectionism or the outbreak of a major war. The intervening mechanisms of perception, analysis, and decision are usually either overlooked or their outcomes are considered to be preordained.

Once again, there would seem to be strong logical and historical reasons for questioning the explanatory and predictive power of theories that move directly from international structures to state behavior. What reason is there for assuming, for example, that the rulers of a present or prospective world economic leader will be able to detect the shifts in industrial power that (in theory) must compel their nation to abandon or adopt a managerial role? (Why then did Britain continue to advocate free trade even after the nation had lost its position of industrial primacy at the end of the nineteenth century, and why was the United States so slow to take up Britain’s mantle during the 1920s and 1930s?) If shifts in military capabilities are readily detectable, why do declining powers often fail to launch preemptive attacks on their rising competitors? When unfavorable shifts are quickly identified, will war be the inevitable result? (How then were England and America able to avoid a war of hegemonic succession during the early years of the twentieth century?)

Structural considerations provide a useful point from which to begin analysis of international politics rather than a place at which to end it. Even if one acknowledges that structures exist and are important, there is still the question of how statesmen grasp their contours from the inside, so to speak, of whether, and if so how, they are able to determine where they stand in terms of relative national power at any given point in history. It would appear that, right or wrong, such estimates will go a long way toward shaping state behavior and, in particular, toward determining national responses to structural change.

Mathematical Realism

Where structural realism essentially ignores the problem of assessment, another strand of contemporary thinking about world politics glosses it over with simplifying assumptions. Efforts to construct mathematical models of international interactions, such as competitions in armaments or alliance formation or war, usually substitute retrospective, objective measures of relative national power for the judgments made by contemporary statesmen.

The most fully developed example involves the attempt of students of arms races to establish the existence of a direct, predictable, quantifiable relationship between the defensive efforts of pairs of potentially hostile states.¹⁴ For modeling purposes, theorists generally assume that there is some single, observable, and mutually agreeable surrogate for military power (whether spending or equipment stocks) on which both sides will focus and to which each participant must respond. In other words, the effort to evaluate enemy capabilities, which is at least implicitly identified as central to the process under investigation, is assumed to be a simple and direct procedure. It seems clear, however, that the inner workings of competing states are not nearly so straightforward or transparent as this picture suggests. Figures on foreign military spending are notoriously controversial and unreliable, and physical indicators of capability can sometimes be even less dependable. This may be so under the best of circumstances, but it is especially likely to be the case when the enemy is doing what he can to conceal his own preparations for war. The apparent complexity of the assessment process in real life may help to explain some of the failings of existing arms race models.

Calculations of relative capability must also, presumably, play a part in the processes through which leaders decide whether or not to enter into an alliance or begin a war.¹⁵ Here again, attempts at modeling typically substitute crude surrogates for the actual assessments of decision-makers.¹⁶ Although these may reflect accurately the thinking of national leaders, there is no a priori reason for believing that they will always do so. The more modelers differ from statesmen in their assumptions about the best way to measure power, the less likely it is that their models will accurately capture reality.

Classical Realism

The centrality of the problem of assessing national power is a familiar theme in the work of more traditional, realist writers on international relations. Indeed, the formulation of accurate assessments is widely held to be the single most important prerequisite of successful statesmanship. In praising the wartime leadership of Pericles, for example, Thucydides reports as his primary virtue that he appears to have accurately estimated what the power of Athens was.¹⁷ Similarly, after discussing what he calls the elements of national power,¹⁸ Hans Morgenthau advises that it is the task of those responsible for the foreign policy of a nation. . . . to evaluate correctly the bearing of these factors upon the power of their own nation and of other nations as well, and this task must be performed for both the present and the future.¹⁹

Correct assessments are assumed by realists to be a first step in the critical process through which ends are brought into alignment with means. Thus Henry Kissinger writes in A World Restored that the test of a statesman . . . is his ability to recognize the real relationship of forces and to make this knowledge serve his ends.²⁰ Along the same lines, Walter Lippmann observed in 1943 that a foreign policy consists in bringing into balance, with a comfortable surplus of power in reserve, the nation’s commitments and the nation’s power. The constant preoccupation of the true statesmen is to achieve and maintain this balance.²¹

Power calculations may be important not only to the decisionmakers of each individual state but to the successful functioning of the system of states as a whole. In the view of many realists, it is only when national leaders can adequately appreciate existing and potential power relationships that they will act so as to preserve a balance among themselves. Edward Gulick, in his study Europe’s Classical Balance of Power, lists the presumed existence of a rational system of estimating power as one of the foundations upon which the eighteenth- and nineteenth-century European state system was based.²² As Gulick puts it, The estimation of power was, and is, one of the common mental processes of balance-of-power statesmen.²³

But how (and how easily) are these common operations performed? Realists have traditionally assumed that assessment is not only important but also difficult. It is an ideal task, Morgenthau concludes, and hence incapable of achievement.²⁴ Erroneous power estimates have plagued nations at all times and have been the cause of many national calamities, writes Arnold Wolfers. Neither the difficulties nor the importance of accuracy in the estimates of power can be exaggerated.²⁵

All this still leaves the question of means and methods unresolved. When it comes to explaining how statesmen actually do their difficult job, the classical realists appear to lose interest and move on to other, more tractable subjects. For the most part, the complexities of the problem are noted and the matter is simply dropped. Thus Hans Morgenthau concludes that

the task of assessing the relative power of nations for the present and for the future resolves itself into a series of hunches, of which some will certainly turn out to be wrong, while others may be proved by subsequent events to have been correct.²⁶

In a similar vein, Gulick writes that

statesmen, whether accurate in their estimates or not, must measure power, regardless of the primitive character of the scales at their disposal. . . . One may, then, dismiss the question of power with the admission, on the one hand, that it is impossibly complex and the assertion, on the other hand, that the statesmen themselves were reduced to making guesses. These guesses have themselves become historical facts.²⁷

Even if they were substantially correct, these statements would not be very satisfying, if only because all forms of guessing are not equally imprecise. It is possible to concede that judgments about changing power relationships may represent imperfect knowledge and even that they may be based on guesses or hunches without abandoning interest in understanding how they come to be formulated. In fact, because the weighing of relative power appears to be so central to foreign policy decision making, those concerned with developing theories of international politics should not be content to dismiss assessment as easily as has often been done in the past.

Two Models of Assessment and Adaptation

There has, of course, been a tremendous amount written about power and its measurement. Much of the literature on this subject is abstract and analyzes the relationships between pairs of unspecified actors.²⁸ More recently, efforts have also been made to look at the problem from a specifically international point of view.²⁹ What is striking about virtually all of this work is that it aims to demonstrate how power should be evaluated by detached, rational observers rather than to determine how it has been or is being weighed by residents of the real political world.

Despite this lack of direct attention, there are nevertheless two schematic sketches of the related processes of assessment and adaptation embedded in the various theoretical works discussed above. In fact, one or the other of these implicit models can be found at the core of most major theories of international relations. As will be discussed, however, while they provide a set of useful reference points, both alternative accounts are, in important ways, inadequate.

Calculative Model

National leaders are often assumed to think of power as a stock of one or more commodities, in much the same way as do the mathematical modelers and many of the international power theorists. The number, variety, and weighting of items included on the list of resources varies from one analyst to the next. Some concentrate only on tangible quantities like state area, population, government revenue, defense spending, trade, and the size of a nation’s armed forces.³⁰ Others try to capture the intangible elements that they consider to be important and that, by implication, either are or ought to be included in the calculations of real statesmen.³¹

An assessment process based on calculation would presumably be uncomplicated and might, indeed, be fairly mechanical. Governments would keep track of their own critical resources and could be expected at all times to know such things as the size of their army (or navy or population) and the capacity of their national economy. By keeping similar statistics for all other countries (or at least for those of importance either as actual or potential allies and enemies), a state’s leaders could be reasonably certain of knowing their relative standing at any given moment. Assessment would then be simply a matter of counting and comparison.³²

Dynamic measurements of changes in the distribution of power would be only slightly more complex. Through periodic or continuous updating of statistics, shifts in relative capabilities could be readily detected. And so, according to one author, the problem of estimating changing relative national power should be not much different than estimation of market share for the firm.³³ Decisive shifts would, in this view, be readily recognizable.

The clear implication of the calculative model is that assessment of changes in relative power can, should, and therefore probably will be, accompanied by appropriate adjustments in national policy. This assumption is sometimes stated directly. Quincy Wright, for example, notes that most balance-of-power theories rest on the belief that statesmen in pursuing a balance-of-power policy do so intelligently—that they measure the factors involved . . . accurately and guide their behavior by these calculations.³⁴

Elsewhere, as in most structural realist formulations, assessment through rational calculation plays the part of a reliable but invisible transmission belt connecting objective change to adaptive behavior. Here, as Robert Keohane points out, whether the proponents of structural realism acknowledge it or not,

the link between system structure and actor behavior is forged by the rationality assumption, which enables the theorist to predict that leaders will respond to the incentives and constraints imposed by their environments. Taking rationality as constant permits one to attribute variations in state behavior to various characteristics of the international system. Otherwise, state behavior might have to be accounted for by variations in the calculating ability of states.³⁵

In any case, the calculative model suggests that adaptation to changes in relative power should be essentially continuous. The path described by state policy is therefore likely to be a smooth and straight line with no lags or zigzags.³⁶

Perceptual Model

Classical realism, with its emphasis on the intellectual problems of statesmanship and on the sometimes tragic quality of international political life, contains within it an alternative to the calculative model of assessment. Writers like Morgenthau and Wolfers point out the difficulties of measuring power in practice and suggest that errors in judgment are not only possible but perhaps even likely. Such mistakes can have severe consequences, as when they cause a state to fail to form a balancing alliance or when they result in its mistakenly waging war against a stronger opponent. One author has even asserted that it is the problem of accurately measuring the relative power of nations which goes far to explain why wars occur.³⁷

The essence of this view is captured in a remark by the great English statesman Lord Bolingbroke:

The precise point at which the scales of power turn . . . is imperceptible to common observation . . . some progress must be made in the new direction, before the change is perceived. They who are in the sinking scale . . . do not easily come off from the habitual prejudices of superior wealth, or power, or skill, or courage, nor from the confidence that these prejudices inspire. They who are in the rising scale do not immediately feel their strength, nor assume that confidence in it which successful experience gives them afterwards. They who are the most concerned to watch the variations of this balance, misjudge often in the same manner, and from the same prejudices. They continue to dread a power no longer able to hurt them, or they continue to have no apprehensions of a power that grows daily more formidable.³⁸

Whereas in the calculative model facts and figures are the raw materials of assessment, here statesmen are seen to deal in less precise but more lingering images, both of other countries and of their own.

The manner in which decision-makers come to form such perceptions has been the subject of considerable analytical attention.³⁹ Not surprisingly, much of what has been written in this area concentrates on the question of how statesmen form beliefs about the intentions and capabilities of others.⁴⁰ In contrast, as one scholar has recently pointed out, national self images have received a good deal less direct scrutiny.⁴¹

A belief in the relative weakness of one’s own country or in the strength of another is presumably at least partly the result of past events. Kenneth Boulding has speculated that the citizens of a country as a whole may come to share a historical national image that extends backward into a supposedly recorded or perhaps mythological past and forward into an imagined future. It is "the consciousness of shared events and experiences like war, writes Boulding, that is of the utmost importance . . . in the creation and sustenance of the national image."⁴²

The literature on perceptions suggests that, however they come to be formed, the beliefs of national leaders (including, presumably, their beliefs about the relative power of states in the international system) are slow to change. Boulding argues that such adjustments occur rarely, if at all,⁴³ while John Stoessinger asserts that change is possible only as the consequence of some monumental disaster.⁴⁴

Students of cognitive processes have focused on crises or dramatic events as the most likely agents of attitude change. Drawing on evidence from behavioral psychology, Robert Jervis concludes that individuals are able to dismiss or absorb bits of discrepant information that might call an accepted belief into question if they arrive slowly and one at a time. Conversely, bad news is usually hardest to handle when it comes in large batches. For this reason Jervis believes that in politics, sudden events influence images more than do slow developments.⁴⁵

This view has a certain intuitive appeal, but it is by no means undisputed. In a study of public opinion that may also have some relevance to the analysis of elite attitudes, Karl Deutsch and Richard Merritt conclude that even spectacular events usually do not result in massive or permanent shifts in collective beliefs. Cumulative events tend to have a larger influence over long periods of time (perhaps two decades or more), but, the authors find, often it takes the replacement of one generation by another to let the impact of external changes take its full effect.⁴⁶

Similarly, in their study Conflict Among Nations, Glenn Snyder and Paul Diesing assert that the immediate images that national leaders have of their opponents in a crisis may fluctuate without affecting background images, the ways in which elites view each other’s countries more generally. These deeper beliefs usually do not change, and then only marginally during the course of a crisis.⁴⁷ Over the short term, according to Snyder and Diesing, change in background image for a government usually results from a change of regime, or a shift in the balance of power within a regime, not from individuals changing their minds.⁴⁸ In general, adjustment of background images occurs through changes of personnel.⁴⁹

Although there are obviously important differences regarding the means by which change can occur (whether through sudden shocks, or changes in personnel, or even in entire generations), there does seem to be agreement that central beliefs or images are remarkably resistant to modification.⁵⁰ If assessment is a matter of images and self-images and if beliefs about relative national power are like others that have been studied, then there is good reason to expect that changes in assessments will lag behind real shifts in the distribution of power. Adaptation is therefore also likely to be delayed and to come in discrete chunks rather than as a continuous stream of minor corrections. Where the calculative model postulates steady adjustment to change, a model based on perception predicts periods of quiescence punctuated by bursts of activity. Instead of a straight line, in this view the path of policy is more likely to resemble a step-function.

Method

Structure of the Study

Taken together, the next four chapters are intended to constitute a single case study of assessment and adaptation under conditions of relative decline. At best this investigation can shed light on only one-half of the larger problem of reactions to shifts in international position. Responses to relative increases in national power will have to be considered elsewhere.⁵¹ Moreover, the case examined here is one in which there appears in retrospect to have been both an objective erosion in power and a reasonably open, contemporary debate over what was occurring.⁵² Cases in which real change went essentially unnoticed, or in which erosion was feared but seems not to have occurred, are also worth considering.⁵³

Within these boundaries there are obvious limits to the generalizations that can be derived from the study of any single, unique group of events.⁵⁴ The four central chapters that follow lay out one set of specific answers to the broad questions with which this chapter began. In chapter 6 these will be compared with the two sets of hypothetical answers provided by the calculative and perceptual models. This comparison illuminates the inadequacies of each and points toward a more refined general understanding of the processes of assessment and adaptation.

Put briefly, an investigation of the British case suggests the importance of organizational, intellectual, and domestic political factors in explaining how nations respond to shifts in relative power. There is strong reason to expect that in modern states the assessment process will be decentralized rather than concentrated in one agency or individual. Instead of a single discussion about some aggregated notion of national power, there are likely to be several parallel debates about the different components of that power going on simultaneously. Each of these will involve elements of both rational calculation and perception or belief. Specifically, this study highlights the central role of simplifying, but sometimes misleading, indicators in both public and intragovernmental discussions of the various forms of national power.

Official estimates of where a country stands may change neither continuously nor in single large jumps but, rather, disjointedly and unevenly. For example, widely shared concern over relative military decline may precede worries about eroding national economic power. In each field, shifts are likely to result not simply from exogenous shocks or random personnel changes but rather from a prolonged process of bureaucratic discussion and, sometimes, public political debate. This process is driven by gradual developments in the thinking of change agents, middle- and upper-level officials whose views begin to deviate from the norm and who are able to receive a wider hearing only at moments of intense crisis.

National adaptation is a function both of the overall pattern of official assessments and of the division of domestic political power. Recognition of decline inside the government and agreement on how best to respond may come more quickly in some areas than in others. Even in those sectors where an internal consensus can be reached a national government may lack the political power to implement its preferred policies. The net result is likely to be a mixture of action in some areas and inaction in others. Instead of either failing to adjust at all or responding with a coherent national strategy, Britain reacted to the early evidence of its relative decline in a fragmented and only partially coordinated way. As will be argued in chapter 6, there are reasons to believe that what was true of England at the turn of the century may be true as well of other countries at other times. The validity of such generalizations will have to be established through an examination of additional case studies.

Structure of Chapters 2–5

At the turn of the twentieth century (and, in particular, during the period 1895 to 1905),⁵⁵ Britain’s leaders talked a good deal about national power in general. More specifically, they were concerned with what I have called economic, financial, sea, and military⁵⁶ power—Britain’s relative position in the international economy, its capacity to pay for an adequate defense, its control of the seas, and its ability to defend the land borders of the empire. Although these elements were separated to a considerable degree, none of them was treated in complete isolation from the others. Accepting a division that contemporary statesmen themselves employed, I will consider each kind of power in turn (rather than following a simple chronological scheme), while trying to show how interconnections were made.⁵⁷

Each of the next four chapters is organized in roughly the same fashion. All are centered around a debate or decision that occurred during the years 1895–1905, although, as will become apparent, the intellectual and historical background to these events often extended many years into the past. It was in the course of these discussions over practical policies that the assumptions of the participants came most clearly to the surface. After analyzing each specific debate, every chapter will conclude with a return to the three central questions about power measurement, changing assessments, and national adaptation raised at the opening of this chapter.

In studying a series of controversies that took place over eighty years ago I have found myself confronted by a tension between the demands of objectivity and judgment. On the one hand, I have tried to treat my subjects fairly. By this I mean simply that I have attempted to put myself in their shoes, trying as best I could to understand how they thought and how the problems they defined for themselves must have appeared at the time. On the other hand, I cannot and do not seek to avoid evaluating the performance of those whom I am describing. For this reason I have not hesitated to judge them against the standard of a future that was necessarily hidden from their view. At the close of this book, in chapter 7, I will have some general remarks about the overall quality of Britain’s response to relative decline. In addition, each of the next four chapters is aimed in part at solving a retrospective puzzle: Why were Britain’s leaders unable or unwilling to do anything about the relative decline in its economic power? Why, after the turn of the century, did they come to the incorrect conclusion that the limits of the nation’s financial resources had been reached? Why, despite the obvious gravity of that decision, did they surrender worldwide sea supremacy with so little anguish? Why, in the face of clear warnings, did they refuse to consider realistically the enormous problems posed by the emerging threat of large-scale land warfare?

No one who thinks carefully about these questions should come away with the soothing conclusion that Britain’s leaders at the turn of the century were nothing more than a pack of short-sighted blunderers. In the face of enormous economic, technical, military, and political changes, they struggled to preserve their nation’s place in the world. Perhaps they could have done better. They could almost certainly have done worse. In judging them, modern critics might well ask what they should have done differently.

¹ For a discussion of the concept see Keohane, Theory of World Politics: Structural Realism and Beyond, in Finifter, ed., Political Science: The State of the Discipline, pp. 503–40. See also the essays in Keohane, ed., Neorealism and Its Critics.

² On the idea of levels of analysis see Singer, The Level-of-Analysis Problem in International Relations, pp. 77–92.

³The definition is from Waltz, Theory of International Politics, pp. 100–101.

⁴ Krasner, Regimes and the Limits of Realism: Regimes as Autonomous Variables, in Krasner, ed., International Regimes, p. 368. On the problem of determinism in modern political science see Almond and Genco, Clouds, Clocks, and the Study of Politics, pp. 489–522.

⁵ Waltz, Theory of International Politics, p. 125.

⁶ Wolfers, Discord and Collaboration, p. 124. Hedley Bull makes a similar point: states threatened by a potential dominant power have the option of failing to counterbalance it (The Anarchical Society, p. 104).

⁷ Waltz, Theory of International Politics, p. 126. For the argument that balancing is more likely than bandwagoning see Walt, The Origins of Alliances, esp. pp. 17–49, 262–85.

⁸ Ibid., p. 122.

⁹ Kindleberger, The World in Depression, 1929–1939, esp. pp. 19–30, 291–308.

¹⁰See, for example, Krasner, State Power and the Structure of International Trade, pp. 317–43; Keohane, The Theory of Hegemonic Stability and Changes in International Economic Regimes, 1967–1977, in Holsti, ed., Change and the International System, pp. 131–62; McKeown, Tariffs and Hegemonic Stability Theory, pp. 73–91; Stein, The Hegemon’s Dilemma: Great Britain, the United States, and the International Economic Order, pp. 355–86; Russett, The Mysterious Case of Vanishing Hegemony, pp. 207–31.

¹¹ Gilpin, War and Change in World Politics, p. 210.

¹² Organski, The Power Transi tion, in Rosenau, ed., International Politics and Foreign Policy, p. 370. This excerpt is drawn from Organ- ski’s book World Politics, which was originally published in 1958.

¹³ For a critique of the determinism of basic force versions of the hegemonic stability theory see Keohane, After Hegemony: Cooperation and Discord in the World Political Economy, esp. pp. 31–46.

¹⁴ This relationship is typically expressed as a pair of equations:

Δ X = kY - aX + g

Δ Y = lX - bY + h

where X and Y are either the military expenditures or the arms stocks of the two states, —a and —b are fatigue coefficients that represent the burden of earlier decisions to buy arms, and g and h are some measure of the hostility that each state feels toward the other. Clearly, as Y’s spending (or level of preparedness) increases, X’s will tend to do so also. For reviews of the mathematical arms-race literature see Busch, app., Mathematical Models of Arms Races, in Russett, What Price Vigilance? The Burdens of National Defense, pp. 193–233; Luterbacher, Arms Race Models: Where Do We Stand? pp. 199–217; Rattinger, Econometrics and Arms Races: A Critical Review and Some Extensions, pp. 421–39; Russett, International Interactions and Processes, in Finifter, ed., Political Science, pp. 543–53. Another more recent and perhaps more promising branch of the arms-race literature uses deductive, game theoretic techniques to model military competitions and to assess the impact of complicating factors like imperfect intelligence (Downs, Rocke, and Siverson, Arms Races and Cooperation, pp. 118–46).

¹⁵ For a discussion of the assessment problem as it relates to theories of alliance formation see Holsti, Hopmann, and Sullivan, Unity and Disintegration in International Alliances: Comparative Studies, pp. 4–14.

¹⁶ One study measures total population, iron production, and military personnel to create a Composite Index of National Capability (CINC). This index is then tested as a predictor of war-pi oneness. No mention is made of whether real statesmen actually use this index or of what the causal link might be between their calculation of capabilities and their inclination to make war (Bremer, National Capabilities and War Proneness, in Singer, ed., The Correlates of War 2 : 57–82).

¹⁷ Thucydides, The Peloponnesian War, p. 163.

¹⁸ Geography, natural resources, industrial capacity, military preparedness, population, national character, national morale, and the quality of diplomacy and government (Morgenthau, Politics Among Nations, pp. 112–49).

¹⁹ Ibid., p. 150.

²⁰ Kissinger, A World Restored, p. 325.

²¹ Lippmann, U.S. Foreign Policy: Shield of the Republic, p. 9.

²² Gulick, Europe’s Classical Balance of Power, p. 24.

²³ Ibid., p. 28.

²⁴ Morgenthau, Politics Among Na tions, p. 153.

²⁵ Wolfers, Discord and Collaboration, p. 112.

²⁶ Morgenthau, Politics Among Na tions, p. 154.

²⁷ Gulick, Europe’s Classical Balance of Power, p. 28.

²⁸ Among the important articles are Simon, Notes on the Observation and Measurement of Political Power, pp. 500–516; Dahl, The Concept of Power, pp. 201–15; Harsanyi, Measurement of Social Power, Opportunity Costs, and the Theory of Two-Person Bargaining Games, pp. 67–79; Nagel, Some Questions About the Concept of Power, pp. 120–37.

²⁹ See, for example, Knorr, Power and Wealth: The Political Economy of International Power; Hart, Three Ap proaches to the Measurement of Power in International Relations, pp. 289–305; Keohane and Nye, Power and Inter dependence: World Politics in Transition; Baldwin, Power Analysis and World Politics, pp. 161–94; Goldmann and Sjostedt, eds., Power, Capabilities, Interdependence: Problems in the Study of International Influence. Perhaps the most interesting distinction to emerge both here and in the literature cited above is that between power as resources and power as control.

³⁰ See Ferris, The Power Capabilities of Nation-States. Another author uses a single index that combines military resources (numbers of men under arms and defense expenditures as a proportion of total world military capabilities), industrial power (share of world pig iron or ingot steel production and share of world industrial fuel consumption), and human potential (share of a given international system’s total population and urban population) (Bueno de Mesquita, The War Trap, pp. 101–109).

³¹ Thus Hans Morgenthau includes national character and national morale on his list (see note 18, above). Ray Cline has suggested that power can be expressed as an equation in which P = (C + E + M) x (S + W). Here C (critical mass) equals population plus territory; E and M are economic and military capability; and S and W are some impressionistic measure of strategic purpose and national will (Cline, World Power Assessment, 1977, p. 34).

³² The logical result of such calculations may be a table of the sort Cline compiles in which he gives every country in the world a power score, from 3 (Cuba) to 523 (USSR) (Cline, World Power Assessment, pp. 173–74).

³³ Doran, War and Power Dynamics: Economic Underpinnings, p. 422.

³⁴ Wright, A Study of War 2 : 754.

³⁵ Keohane, Theory of World Politics, in Keohane, ed., Neorealism and its Critics, p. 167.

³⁶The overall picture here is of the state as a cybernetic mechanism, reacting to stimuli reaching it along certain specified channels in ways intended to ensure a stable relationship between entity and environment. See Deutsch, The Nerves of Government; also Steinbrunner, The Cybernetic Theory of Decision: New Dimensions of Political Analysis; Rosenau, The Study of Political Adaptation.

³⁷ Blainey, The Causes of War, p. 114.

³⁸ Quoted in Gulick, Europe’s Classical Balance of Power, p. 29.

³⁹ Best and probably most influential is the work of Jervis, especially Percep tion and Misperception in International Politics. For a discussion of the litera ture and an analysis of recent findings see George, Presidential Decisionmaking in Foreign Policy: The Effective Use of In formation and Advice, pp. 25–80. See also Jonsson, ed., Cognitive Dynamics and International Politics.

⁴⁰ See, for example, the essays by William Scott and Ralph White in Kelman, ed., International Behavior: A Social-Psychological Analysis. See also articles in Image and Reality in World Politics, a special issue of the Journal of International Affairs; Holsti, Cognitive Dynamics and Images of the Enemy: Dulles and Russia, in Finlay, Holsti, and Fagen, eds., Enemies in Politics, pp. 25–96; Stoessinger, Nations in Darkness: China, Russia, America.

⁴¹ Lebow, Between Peace and War, p. 195.

⁴² Boulding, National Images and International Systems, pp. 122–23.

⁴³Boulding, The Learning and Reality-Testing Process in the International System, p. 10.

⁴⁴ Most national leaders, he writes, will not examine their prejudices and stereotypes until they are shaken and shattered into doing so (Stoessinger, Nations in Darkness, p. 194).

⁴⁵ Jervis, Perception and Misperceptions, p. 308. On the gradual, onion-peeling process of attitude change see Jervis, pp. 288–315.

⁴⁶ Karl Deutsch and Richard Merritt, Effects of Events on National and International Images, in Kelman, ed., International Behavior, pp. 182–83.

⁴⁷ Snyder and Diesing, Conflict Among Nations: Bargaining, Decision Making, and System Structure in Inter national Crises, p. 329.

⁴⁸ Ibid.

⁴⁹ Ibid., p. 332.

⁵⁰ Alexander George summarizes the findings from the psychological litera ture by pointing out that "while . . . be liefs can change, what is