Embedded Counterfactuals and World War I
as an Unavoidable War

by Paul W. Schroeder

This essay, though it may seem to do so, does not take a determinist stand either on counterfactual reasoning and contingency in history in general or on the origins of World War I. On the war, though its conclusions differ from R. N. Lebow's argument elsewhere in this volume that it could readily have been avoided,(1) it agrees with Lebow's views on many points and concedes a large causal role in all great events in history, including World War I, to contingency, chance, and particular choices. It even argues that in a certain objective sense the war remained avoidable up to its very outbreak, and presents other grounds for considering it unavoidable.

It also broadly agrees with Philip Tetlock's and Aaron Belkin's views on the necessity, unavoidability, and potential utility of counterfactual reasoning in historical study.(2) The difference between my views and theirs lies in the practical rather than the theoretical realm. While in general accepting the Tetlock-Belkin analysis of the various types of counterfactual reasoning and the basic tests to use for them, I will suggest a different notion about how and where to apply counterfactual reasoning concretely to historical explanation, as a better way of showing historians the value of counterfactual reasoning for accomplishing their task. That task (here I agree with the historians who are skeptical about it) is not to speculate on what might have happened in history, but to shed light on what actually did happen, why it did, and what it means.

As for whether World War I ever became unavoidable, and if so, when and why, once again the point I wish to make is a practical more than theoretical one. Clearly there were contingent events and developments occurring and important choices being made in 1914 right up to end of the July Crisis. One can undoubtedly make a case, as Lebow has done, that some vital events in the causal chain, such as the assassination of Franz Ferdinand, could easily not have happened. It is more difficult, but not impossible, to argue that vital decisions could have been made differently. But this fact in itself does not make the war avoidable, because it does not cancel the practical limits to the avoidability of outcomes in human affairs. To term an outcome inevitable often means no more than to say that the time when it could have been avoided is past--that the kinds of decisions, actions, or chance developments required to avert it, though possible earlier, have become so unlikely or unthinkable as to rule out any plausible scenario for avoiding it.(3) This is an obvious point, yet it is not easy or commonplace to apply it to particular developments in history, especially major ones like the outbreak of great wars or revolutions. My hope is to use a particular application of counterfactual reasoning to history to show one specific way in which World War I by 1914 had become unavoidable.


I. The Case for "Embedded" Counterfactual Reasoning

The Tetlock-Belkin theses seem to assume that the way for historians or other scholars to apply counterfactual reasoning to historical exposition and explanation is to pose the question, "What if ?"--i.e., to imagine or conceive of a way in which a particular event or development could have unfolded differently, and to ask, "What if this had happened? What further changes would have resulted?"(4)

A working historian, however-even if, like me, he agrees that counterfactual elements are logically implied in all explanation, including historical explanation--may have serious qualms about this procedure. The reason is that though it is logically defensible to think up counterfactual questions with which to confront the historical record, the exercise seems pointless or at best of limited value from a practical standpoint because even so-called "easily imagined variations" introduced into the complex matrix of historical developments can change so many variables in so many unpredictable or incalculable ways, leading to so many varied and indeterminate consequences, that the procedure quickly becomes useless for helping us deduce or predict an alternative outcome. Tetlock and Belkin of course see the problem and deal with it in terms of abstract logic. Yet my feeling is that the procedure's practical limits and problems sensed by the working historian are not sufficiently grasped.

The difficulty, as noted, is that if the variation is really important, involving a central
component or variable in the historical equation, introducing it will alter so much of the complex web of history that the results of omitting or altering that crucial variable become incalculable. If the variation introduced, however, is minor, sufficiently precise and limited enough that its immediate consequences can be calculated with confidence, its implications for general historical explanation or for suggesting any important alternative outcome are unlikely to be important. A major counterfactual, in other words, will change too much, and a minor one too little, to help us explain what really did happen and why, and why alternative scenarios failed to emerge, the only sound reasons for using counterfactual reasoning. Thus using this kind of "What if?" counterfactual procedure might well have the perverse and ironic effect of confirming ordinary historians in their resistance to counterfactual reasoning and strengthening their tendency to see history as the result of pure contingency and chance-the well-known "Cleopatra's nose" view of history.

One might of course reply that if historians cannot or will not recognize the
presuppositions and assumptions involved in the explanations they offer, other scholars will have to identify and analyze them, and subject them to various tests, including those of counterfactual reasoning. Again I agree, up to a point. The alternative kind of counterfactual reasoning I will suggest might help historians get over the tendency toward a naive pragmatic empiricism. Yet it would be rash of scholars in other fields to suppose that because a particular historian fails to give compelling theoretical grounds for being dissatisfied with a particular counterfactual procedure, his or her concerns can be safely ignored. Practitioners may well sense from experience and schooled intuition that a plausible idea or theory will not work in their field, even if they have difficulty articulating theoretical reasons why. Moreover, the principal contention here, that some major easily imagined counterfactual variations in particular sections of history change so much that the whole subsequent development becomes incalculable, while other variations, just as easily imagined, make no significant difference at all, can readily be illustrated in any period of history.(5)

One can get round the difficulty, I suggest, by a different concept and method of counterfactual reasoning. It starts by conceiving of counterfactuals not as non-history (that is, imagined or virtual as opposed to real history, what might have happened rather than what did), but rather as real history, an integral part of history, embedded in history both in the actual experience of historical actors and in those constructions or reconstructions of history constantly made not only by scholars but also by everyone who reflects on the past. History, like life itself, is lived, acted, made, and relived and reconstructed in the face and presence of counterfactuals. Historical actors in all arenas of life constantly think, calculate, decide, and act in the face of uncertainty; they repeatedly ask the question, "What if?", try to answer it, and make decisions and act on that basis. Historians take this for granted. They know instinctively (or are quickly taught) that historical actors regularly face an uncertain, open future. Recognizing this, they must portray and analyze that situation and show why and how actors responded to the questions, choices and alternatives they faced as they did. If they are at all sophisticated, they also realize that carrying out this task requires not merely trying to discover and analyze the actors' thought worlds and the role played by their counterfactual questions and calculations, but also framing and posing their own counterfactual questions as to what might have happened had the actors answered their counterfactual questions differently. Thus in seeking to discover the real nature and results of the actual choices made by actors in the face of their uncertainty and their counterfactual questions, historians must use their advantages of hindsight and historical evidence to ask counterfactual questions of their own, such as: What other decisions and actions could the historical actors have made under the existing circumstances? To what extent did they recognize and consider these? What circumstances made these choices or alternative courses genuinely possible or merely specious and actually unreal? What might the alternative results of these choices have been? The real justification for the use of counterfactual reasoning in history and the best answer to those who reject it is the fact that historians cannot faithfully convey the real nature and results of historical decisions and actions simply by constructing a factual narrative of "what happened" without confronting the various counterfactuals, both those faced by the actors and those necessarily posed by the historian, integrally embedded in that story.(6)

This understanding of counterfactual reasoning not only justifies its use but also suggests how it ought to proceed. The first task is to discover and analyze the counterfactual questions actually seen and faced by the historical actors themselves. This part is so obvious, normal and ubiquitous an element of historical research that it needs no discussion here. The second step, less obvious but no less necessary, involves looking carefully at the reconstructions and explanations of historical events and outcomes offered by historians (especially oneself) with the specific aim of discovering and analyzing the implicit and explicit counterfactual questions and assumptions they contain. The next assignment is rigorously to test these counterfactual assumptions and scenarios that historians wittingly or unwittingly pose, by means of the kinds of tests and criteria Tetlock and Belkin suggest and the same types of historical evidence as are employed to construct the "factual" story. By ferreting out and analyzing the overt or concealed counterfactuals embedded in historians' reconstructions and explanations of history, I contend we can both better explain the actual course of historical events and better judge whether the counterfactual possibilities envisioned by the actors and those constructed and used by historians were sound or illusory. In short, it can help us better understand both what did happen in history and why this particular thing rather than some other possible thing occurred.(7)

This kind of counterfactual reasoning has other advantages as well. It represents something historians regularly do, whether or not they are fully aware of it, and thus is a method that, once understood, they can hardly reject. Most important, it seems to me to fit the nature of history, recognizing its openness and uncertainty for the actors themselves while insisting at the same time that history's outcomes, though not predetermined, can and must be explained by causes. It thereby takes seriously both the contingent and the determinate character of the past; respects both the extent and the limits of its range of possibilities. It depicts history as unfolding in an indeterminate way, the product of unpredictable human conduct and material circumstances, but not as kaleidoscopic chaos. It offers a way of distinguishing between genuine and specious counterfactual scenarios, showing that while much could have happened differently, not everything, including many of the things historical actors and later historians have thought were possible, could have happened. It fits our sense, learned from life as well as history, that at some point some things once indeterminate do become inevitable.


II Embedded Counterfactuals and World War I as an Unavoidable War

I now need to illustrate how a certain degree and kind of inevitability in history applies to the origins of World War I (not demonstrate it, which might be impossible and would certainly take too long).(8) The topic suits the purpose for several reasons. First, the story is very well known and does not need to be expounded in detail. Second, though the scholarly debate over the origins of the war which raged for decades after 1914 has not completely died out, a clear consensus view has emerged which denies that the war was inevitable and ascribes its origins to specific avoidable choices and actions taken by particular actors. Hence it represents a good challenge. Third, it illustrates particularly well the potential value of detecting and analyzing embedded counterfactuals, the surprising results it can lead to, and the dangers of failing to do so. It also connects with the broad theme of this volume, the origins and causes of European supremacy in the world. In asking whether this world war, a major source and cause of Europe's subsequent relative decline, was avoidable, it suggests a further major counterfactual question it will not attempt to answer, namely, whether absent World War I European world supremacy would have lasted a considerable while longer.

My treatment of this huge subject must be brief and sketchy, little more than an outline, and is bound to seem dogmatic in some places and trite in others. It will start by discussing the current prevailing view of the origins of the war, analyze and criticize the counterfactuals embedded in it, and from this develop a divergent view.

A. The Standard Explanation and Its Counterfactuals

By common agreement, the direct proximate cause of World War I was the German and Austro-Hungarian decision that Austria-Hungary issue an ultimatum to Serbia in July 1914 following the assassination of Archduke Franz Ferdinand by Bosnian nationalists with connections to Serbia. The German powers intended by this ultimatum to provoke a local war against Serbia and eliminate it as a political factor in the Balkans, thus shifting the balance there and in Europe generally in favor of themselves. Without necessarily intending to start a general war, the German powers consciously risked provoking one by this initiative, as actually happened.

Disagreement persists over the motives and attitudes prompting this go-for-broke gamble, with some historians emphasizing the fear and desperation felt by leaders in Germany and Austria-Hungary, others stressing their aggressive aims and their hopes that they could either get away with a successful local war or win a wider one. This disagreement makes little difference in deciding whether this war was avoidable, however, because everyone agrees that Germany and Austria-Hungary, whatever their reasons, chose to take this gamble; they were not forced into it. This belief implies a counterfactual: they had viable alternatives, could have chosen other ways to protect their interests without risking a great war. A similar consensus prevails that the other great powers, Russia, France, and Britain, reacted essentially defensively to the German-Austrian move and had little choice other than to do so in self-defense, given their vital interests and the unmistakable challenge presented them. The counterfactual scenario embedded in the consensus explanation thus ascribes to Germany and Austria-Hungary a choice of alternate strategy or strategies by which they could reasonably have hoped to protect their vital interests by peaceful means, while denying that the other major actors had practical alternatives for saving peace once the Central Powers launched their initiative.

To be sure, few attribute the German-Austro-Hungarian gamble purely to aggressive expansionism, militarism, and paranoia, or deny that the international situation was becoming increasingly unfavorable and dangerous for the Dual Alliance in 1914. The consensus view, in fact, uses this to help explain the Austro-German action, while denying that this justifies it or renders it necessary, claiming among other things that Germany and Austria-Hungary had themselves largely created the dangers threatening them by failing to reform internally while pursuing unrealistic, aggressive policies abroad. Again this involves an implied or stated counterfactual: even as late as 1914 the Central Powers could have changed their policies and thereby made themselves more secure within the existing international system without overthrowing it.

Consensus historians recognize further that Germany, already in 1914 largely isolated diplomatically and threatened with encirclement by the Triple Entente, faced an imminent future threat, that once Russia had completed its announced plans for military expansion, scheduled for completion by 1917, the German army would be numerically as decisively inferior to those of its opponents as the German navy already was on the sea. But the consensus view claims that Germany had largely created this perilous situation for itself by the aggressive world policy it had followed ever since Bismarck's fall in 1890. Its naval race with Britain, its restless quest for colonies, bases, and spheres of influence around the globe, and its frequent resort to bullying and threats, all designed to give Germany hegemony over Europe and a world position competitive with those of Britain, Russia, and the United States, provoked the alliances, ententes, and armaments races, first at sea and then on land, by which Germany now felt encircled and threatened. Since these dangers arose primarily from Germany's policies and actions (here comes another important counterfactual), different German policies could over time have reduced or eliminated them. Even as late as 1914, had Germany realized that none of its neighbors intended to attack it or violate its rights and had it decided to give up its drive for world power, pursuing instead a sensible, moderate policy focused on economic expansion, it had good chances to enjoy a reasonably secure, prosperous, and honorable place in the European and world international system. In fact, prominent historians have argued that Germany's economic dynamism was so great that it needed only a prolonged period of peace to achieve mastery in Europe.(9)

Almost everyone also recognizes that Austria-Hungary faced even graver dangers than Germany, and that these were less obviously the result of its own actions, at least in the international arena. The Habsburg Monarchy before 1914 was growing steadily more isolated politically and diplomatically and losing its great power status and reputation. Two allies, Italy and Rumania, were unreliable and hostile, the latter to the point of open defection. Its most important ally, Germany, was the Monarchy's most serious economic rival, especially in the Balkans where Austro-Hungarian interests were concentrated, and the Germans tended both to dominate Austria-Hungary politically and strategically and to ignore its vital interests. Austria- Hungary's military security against a host of possible or probable enemies (Serbia, Russia, Italy, Russia's ally France, and even Rumania) depended totally on receiving major, timely help from Germany in case of war. Yet given Germany's own threatened military position facing a likely two-front war, Germany's gambling offensive strategy for fighting it (the Schlieffen Plan), and the fact that the Dual Alliance lacked a military convention or an agreed and coordinated military strategy, how much actual military help Germany would provide its ally was anyone's guess. Meanwhile Austria-Hungary's long-standing security problem had been further worsened by the disastrous outcome of the two Balkan Wars in 1912-13. The Peace of Bucharest in August 1913 left Austria-Hungary with no reliable partner in the last region, the Balkans, where it still counted as a great power and had its most vital interests. The Ottoman Empire was virtually expelled from Europe, while Bulgaria, which the Austrians counted on to check Serbia, was defeated and exhausted, Rumania alienated, the new Kingdom of Albania a basket case and albatross around Austria-Hungary's neck, and Italy an active rival in Albania and the Adriatic with irredentist claims on Austrian territory. Even Germany had not given its ally steady support during the prolonged crisis, but had held Austria-Hungary back in order to preserve general peace and pursue its own particular aims. Meanwhile Austria-Hungary's worst rivals and enemies, Russia, Serbia, and Montenegro, had emerged from the Balkan Wars stronger, more confident, and more hostile, and Russia, aided by its ally France, seemed poised to consolidate its dominance over the entire region by expanding the Balkan League it had earlier sponsored and thereby promoted the Balkan Wars in the first place. The decline in Austria-Hungary's strength and status, obvious to everyone, enabled other powers to ignore its interests, to exploit its internal problems, especially the nationalities conflicts, to raise irredentist claims on the Monarchy's territory, and in Serbia's case to wage a cold war of propaganda and a guerilla war of terrorist subversion against it. They further spurred dissatisfied nationalities and groups within Austria-Hungary to demand concessions from the Austrian and Hungarian governments, sometimes soliciting foreign support for them, thus exacerbating the already grave problems of governance in both halves of the Monarchy and inducing anger and hopelessness in those who remained Habsburgtreu. Most important, Austria-Hungary, far more than Germany, had fallen hopelessly behind in the land arms race then reaching a crucial stage among the great powers in Europe (only Italy was worse off, and Italy was only a would-be great power). Given Austria-Hungary's limited economic and fiscal resources and the restrictions imposed on its military exertions by the parliamentary system in Austria and the autonomy enjoyed by Hungary, there was simply no hope for it to catch up. It thus faced the prospect of fighting a great war against several foes with only doubtful German support and under conditions of hopeless inferiority.(10)

No historian to my knowledge denies the gravity of Austria-Hungary's situation; various factors in it are regularly invoked to explain its go-for-broke gamble in 1914. Yet many also contend, as they do with Germany, that Austria-Hungary had largely brought this on itself. For decades or generations it had failed or refused to solve its own internal problems, especially the nationalities conflicts, and thus exposed itself to irredentist subversion and external threats. It had made this worse by a stubborn, aggressive defense of outworn positions and untenable claims in foreign policy (its so-called Pig War against Serbia before 1908, its annexation of Bosnia in 1908, the subsequent humiliation of Russia in 1909, its refusal to reach reasonable compromises with Serbia, Montenegro, and Italy during the Balkan Wars of 1912-13, the hopeless attempt in 1913-14 to create a viable new Balkan satellite in Albania). Once more the consensus verdict, by implication more than explicitly, posits a major counterfactual: though by 1914 the hour was late, the Monarchy's one remaining chance to survive its crisis was not the use of force, either internal or external, but reform in the direction of federalism, turning itself into a more free and democratic union of peoples and recognizing the interests of the other nationalities besides those of the master races, the Germans, Hungarians, and Italians.(11)

Thus in both cases the supposedly counterproductive and dangerous foreign policies of Germany and Austria-Hungary culminating in their gamble in 1914 are linked to a wider problem and at least partly explained by it: the failure or refusal of their regimes to reform and modernize in order to meet their internal political and social problems. Instead these regimes chose to stay in power, preserve their existing social order and the interests of their respective elites, and manage their internal social and political divisions and problems through an assertive, expansionist foreign policy (a resort to so-called secondary integration and social imperialism).

These explanations, in assigning Germany and Austria-Hungary the primary responsibility for causing the threats against which they decided to act in 1914 and explaining their policies as directed as much against internal problems as external dangers, add (as noted) further counterfactuals to the original counterfactual thesis, that these two powers had means and choices for protecting their legitimate interests in 1914 other than aiming for a local war and risking a general one. To lay these out for Germany: (a) Had Germany not conducted a reckless, aggressive pursuit of world power for decades before 1914, its general interests and position in world politics would not have been threatened as they were or were perceived to be in 1914. (b) Had Germany pursued political reform and social integration rather than manipulated social imperialism and secondary integration at home, its government would not have needed to pursue a reckless, aggressive foreign policy for domestic-political reasons. (c) A more democratic, liberal, and well-integrated Germany using peaceful, normal ways of protecting its legitimate interests would not have encountered enmity and opposition from the other great powers, especially from Britain and France as fellow democracies, but would have been welcomed as a partner for peace and prosperity in Europe and the world.(12) Somewhat similar counterfactuals apply to Austria-Hungary. A reformed, more progressive and democratic Monarchy pursuing wiser policies toward its nationalities and a more conciliatory foreign policy could have solved or managed its internal and external problems to such an extent that it would have both been less vulnerable to pressures and threats from its opponents and have encountered fewer such threats, thus eliminating the need for the suicidal gamble of 1914.(13)

B. The Counter-Argument and Its Counterfactuals

First, a logical and methodological point: If, as I claim, these counterfactuals are embedded in the consensus scenario and logically implied by it, then those who advance this view have an obligation to back them up, showing by research, analysis, and evidence that these counterfactual propositions are at least reasonable, more probable than not. The burden of proof lies on them to do this, not on others to disprove them. By and large this has not been done. Historians have by and large devoted close attention first to determining the facts on the origins of the war, both immediate and long-term, and then to linking the outbreak of the war to the German-Austro-Hungarian initiative in July 1914, both by connecting that initiative to their particular situation and aims in 1914 and by trying to show how their general situation and aims derived from their previous foreign and domestic policies and actions. In other words, starting from a correct initial premise that the German powers' initiative was the immediate proximate cause of the war, they have gone to great pains to construct a plausible case that this initiative derived from and was caused by a general situation which also primarily resulted from German and Austro-Hungarian actions and policies over a much longer term. The null hypothesis stated or implied in this argument, however, has not been systematically laid out and examined, nor have the counterfactuals embedded in it been analyzed and researched in detail. No serious attempt has been made to back up the (hidden, implied, unarticulated, but real and logically necessary) claim that absent those supposedly decisive German and Austro-Hungarian policies and actions, the general situation in 1914 would have been different in the ways the consensus view contends.

Whatever the reasons for this disparity, so long as the counterfactuals clearly, logically, and necessarily implied in the consensus argument have not been researched and analyzed with the same care as the other so-called facts in the case, both the argument making the German- Austro-Hungarian initiative the main cause rather than merely the occasion for war and thus making these powers primarily responsible for it, and the argument that the war was inherently avoidable, the result of particular decisions that could have been made differently, remain unproved. Absent this analysis, we do not know whether in fact the leaders of Germany and Austria-Hungary had any real freedom to act otherwise than they did, or what difference it would have made (ex hypothesi) had they done so.

The argument could stop here, with the claim that the consensus case for World War I as an avoidable conflict remains unproved and a call for more research. This might be prudent, but would be inconclusive and not very interesting. Instead I will try to do three things: First, show that the counterfactual assumptions and implications of the consensus view are not merely largely unexamined and unproved, but also highly improbable and in some instances untenable. Second, lay out an alternate set of counterfactual requirements necessary for averting war both in 1914 and for some indefinite but significant period thereafter. Finally, argue that this set of conditions and actions required ex hypothesi to avoid a general war in that era not only was not recognized, accepted, and carried out by the various actors at this time, but also that the existing international system, i.e., the circumstances, political culture, and rules and practices which then prevailed in international power politics, worked to make it very unlikely that the necessary steps would or could have been taken. This makes it highly improbable that any plausible counterfactual historical scenario by which the war could have been avoided can be constructed, and justifies calling the war inevitable.(14)

Obviously this is a tall order. The counter-argument against the consensus view will
have to be as bare-bones in character as the previous exposition of it, or more so.

The case starts with conceding (in fact, insisting) that the standard view is basically correct about the immediate origins of the war, i.e., that the German-Austro-Hungarian initiative of July 1914 aimed at a local war and risked a general war with the aim of reversing the prevailing trends in international politics by violence, that this launched the crisis resulting in general war, and that during the July Crisis the other great powers were primarily reacting to the Central Powers' initiative. Hence it can be argued that without this particular Austro-German initiative no local or general war would have developed at this particular time. But this has always been obvious. The key question regarding both responsibility for the war and its avoidability is the counterfactual one already shown to be implicit in the consensus case: the question of whether other choices were available to the Central Powers at that time which, under the existing rules and conditions of the game, offered them the opportunity to satisfy their security needs without risking a major war. If so, they chose war when it was avoidable--if not, then not.

The related question of whether they were also responsible for creating the insecurity which prompted them to take this gamble is not strictly speaking relevant to the question of inevitability, though it is to that of their ultimate responsibility for the war. That is, they could themselves have created a general situation in which their only hope for survival was this desperate gamble. In this case, one might conclude that war in 1914 was in a sense unavoidable-they were bound soon to do something desperate that would touch it off-but also that they were responsible for creating this necessity. Yet though this question is less central for our purposes than the first, the two are so closely related that even a prima facie case against the consensus argument, to be coherent, must deal with both. Therefore I will deny both sets of counterfactuals. That is, in addition to denying that they had viable alternatives in 1914, I will also briefly state some reasons why Germany and Austria-Hungary were not chiefly responsible for creating the critical security challenges they faced in 1914, why different policies on their part would not have substantially changed their situation, and why the existing international system precluded other reasonable peaceful alternatives for meeting the threats they faced..

I start with an assertion that will sound deliberately provocative, even outrageous, but that in my view represents a reasonable, almost self-evident interpretation of the evidence. In the period from about 1890 to 1914, the international policies and actions of Germany and Austria- Hungary, as distinct from their aims, attitudes, gestures, language, and ambitions (especially those of Germany) were actually more restrained and moderate than those of any other great power. One cannot point to specific German or Austro-Hungarian actions between 1890 and July 1914 that were as aggressive, expansionist, imperialist, law-and-precedent-breaking, and belligerent as many of those taken during this same period by every other major power--Russia in East Asia and Central Asia, Britain in East and South Africa, Southeast Asia, and Central Asia, France in West, Central, and North Africa and Southeast Asia, Italy in North and East Africa and the Eastern Mediterranean, the United States in Central America, the Caribbean and the Western Pacific, and Japan in East Asia. The same point holds, mutatis mutandis, for a number of small powers, notably Serbia, Greece, Bulgaria, and Montenegro.

This of course does not make Germany and Austria-Hungary, especially the former, peace-loving defensive status quo powers. Germany was as active a participant in the colonialist- imperialist scramble of the era as it could be, while Austria-Hungary would have liked to participate, tried to do so on an informal basis, and did join half-heartedly in the open imperialist scramble toward the end, but never had the means to pursue it seriously. Both powers had active foreign policies and pursued aims by no means limited to preserving the status quo. Germany in particular constantly sought gains and repeatedly made attempts at achieving them--seldom, however, pursuing its initiatives consistently or very far or succeeding in doing more than arouse fear, resentment, and opposition from other states. Behind its various restless impulses lay the overall goals of Weltpolitik. This meant for Germany essentially a policy of maintaining its security in continental Europe (which, given Germany's central location, required at least half- hegemony there) while simultaneously making gains in world power and position (colonies, bases, markets, a formidable navy, and alliances) that would make it competitive in the twentieth century with Britain, Russia, and the United States. Both goals were to be achieved with the aid of Germany's military and economic power, but mainly by means of shrewd diplomacy and power politics--using Germany's key position in Europe and the free hand it supposedly gave her to exploit what Germans supposed were irreconcilable rivalries between Britain, Russia, and France, so that Germany could reach favorable deals and arrangements especially with Britain. Austria-Hungary's main aims were necessarily more defensive--to preserve its territorial integrity, independence, and great power status against many serious challenges and threats, particularly in the one area where it still had vital great-power interests and some imperialist ambitions, the Balkans and Near East. Its policies, toward the other great powers if not lesser ones, were correspondingly more conciliatory.

Yet to dwell, as most historians do in explaining the origins of World War I, on what the Central Powers wanted and tried to do is largely beside the point. The salient fact is that throughout 1890-1914 their various initiatives, regardless of their nature and intent, regularly failed--failed either relatively in the sense of yielding them only limited gains at high long-term costs (e.g., Austria-Hungary's annexation of Bosnia in 1908 or Germany's Berlin-to-Bagdad Railway project), or absolutely in the sense of ending in defeat and greater insecurity for one or both (e.g., the two Moroccan Crises and the two Balkan Wars).

Equally striking is the contrast in this regard between their experience in this regard and that of the other great powers. The latter were able to gamble, commit serious blunders, provoke wars, experience serious setbacks and defeats, and not only survive their gambles and failures but often reap long-term profit from them. The French, though they were humiliated by Britain at Fashoda, escaped unscathed from this foolish gamble and eventually gained the colonial deal and entente with Britain they wanted. Two overt, dangerous French challenges to Germany in Morocco launched serious crises, but ended by improving France's colonial and European positions. Britain used the threat of war successfully to compel France to back down over the Sudan and Egypt, got away with an aggressive, badly-run war in South Africa, and forced the Germans to accept their terms in Persia and Mesopotamia. The Russian government pursued an especially reckless imperialist policy almost everywhere, especially in the Far East, and yet not only survived the disastrous war and the crippling revolution in 1904-06 its policies had brought upon it, but by 1914 was not only pursuing its old imperialist goals in the Balkans and the Turkish Straits more boldly than ever and exploiting its new accord with Britain to encroach on Persia, and even laid the foundation for reviving its Far Eastern expansion as well. Italy's reckless adventure in Ethiopia in 1896 led to a humiliating defeat-but also to a rapprochement with France that enabled Italy thereafter to play off both sides in the European alliance system for the benefit of Italian interests in Africa, the Mediterranean, and the Balkans. Eventually this policy emboldened Italy to commit what was arguably the most cynical and dangerous act of imperialist aggression in the whole prewar period, condemned by everyone--its attack on the Ottoman Empire in Libya and the Dodecanese in 1911-12, an act directly linked with the two Balkan wars and World War I itself. Yet Italy emerged from this adventure with no concrete losses and handsome territorial gains. Japan's risky, all-out gamble in 1904-05 in launching a preventive war against Russia paid off handsomely. The United States' war against Spain in 1898, a war against a state that posed no threat to the United States and was thus surely avoidable even if in some respects justified, paid off even more handsomely at almost no risk.

This will doubtless be seen as an argument drawn from a familiar exculpatory tradition: the contention that Germany and Austria-Hungary were not as imperialist, reckless, or aggressive in the prewar era as other powers-to which the obvious answer is that they were imperialist, reckless, and aggressive where and when it really counted, in Europe in 1914. Let me say emphatically (I have the impression that this is an instance where one must shout in order to be heard) that this is not the point. The argument has nothing whatever to do with the character of German and Austro-Hungarian policy as compared to those of other powers. It has to do with who was really controlling the system, making the rules, and running the show. It establishes a fact highly relevant to the consensus case that makes the German powers primarily responsible for the security threats they faced in 1914, and contends that they could have warded off these threats by peaceful means: the fact that Germany and Austria-Hungary were not in control of the international system, but being restrained and controlled by it. The initiative and leadership in European politics from 1890 to 1914 always lay with their opponents, increasingly so as time went on. The standard reply, that the Central Powers lost control because of their own blunders and provocative acts, simply breaks down on many counts. It is circular; it begs the question; it smacks of the ethic of success; it ignores the patent evidence that Germany's and Austria- Hungary's policies and initiatives regularly failed regardless of whether they were aggressive and provocative or moderate and conciliatory;(15) it fails to specify in concrete detail what different policies could have led to success, or explain how and why they could have. In more theoretical terms, it ignores a fundamental argument advanced by realists in international relations theory, an argument not always valid but here supported by powerful evidence: that systemic factors, the distribution of power, vulnerability, and opportunities within the system, account for the major power-political patterns and outcomes of international politics more than do the character and aims of the individual actors. Logically and methodologically it errs in applying its principle of the primacy of domestic influences and interests not only to explain decisions in foreign policy (which is always in principle legitimate), but also to account for outcomes in international relations, where systemic factors must be taken into account. Finally, it errs by applying this dubious principle of the primacy of domestic politics one-sidedly, to the Central Powers far more than to the Entente.

The moral of all this is simple: to understand international outcomes from 1890 to 1914, one must stop looking first and foremost at what Germany and Austria-Hungary were doing, and concentrate on the powers who held the initiative in world affairs, basically running the system and making it work for them. One must further assume, barring evidence to the contrary, that their policies were primarily internally motivated, driven essentially by their own needs, purposes, and interests, and that Germany and Austria-Hungary, who could not and did not control events, were reacting to what the other powers were doing more than the other way round. Research on the policy of the various Entente powers done from this standpoint serves to confirm this judgment and produces a picture very different from the standard one.(16)

The distortions produced by focusing on Germany and Austria-Hungary as the prime movers in the international system are not remedied but made worse by stressing the domestic pressures and unsolved internal problems supposedly driving their foreign policies. Regardless of the extent to which this explanation may be justified (obviously their foreign and domestic policies were inextricably interwoven; in the case of Austria-Hungary, the distinction between foreign and domestic policy virtually breaks down), such a concentration on their internal problems in explaining their policies and motives simply reinforces the error of making these powers the prime movers within the system. The key to explaining the German powers' policies lies not in what their governments and their constituent interest groups and elites would have liked to do, but what they found themselves compelled to do. It makes better sense to analyze British, French, Russian, American, Italian, and Japanese policy in terms of domestic pressures and influences, for each of these governments had more effective choices and room to translate its desires into some kind of action.(17) It also bears remembering that the foreign policy/domestic politics nexus works both ways. Domestic pressures influence and shape foreign policy, but success or failure in international politics and foreign policy also strongly influence domestic politics. This was obviously the case in prewar Germany, where the government's perceived failures in foreign policy promoted dissatisfaction throughout the political spectrum, with right- radical groups and special interests especially calling for strong action to defend the country's interests.(18) If the danger to the regime and governing elites arising from foreign policy failure was serious for Germany, it was life-threatening for Austria-Hungary. The steady erosion of the state's independence and international prestige not only encouraged dissident elements to press their claims and weakened the attachment of the loyal and dominant ones, but at the same time encouraged foreign governments, especially Russia, to promote internal discontent and subversion within Austria-Hungary and, in the Serbian case, to support terrorist resistance within it,(19) and encouraged almost every government to ignore or oppose its interests in international crises. To contend that the internal problems allegedly motivating the Central Powers' aggressive, dangerous foreign policies should have been handled instead by internal reforms is to ignore the extent to which, especially for Austria-Hungary, developments in the international arena contributed to those internal problems and made them unmanageable without foreign policy success.

This is (to repeat) not an attempt to blame their opponents for the failure of German and Austro-Hungarian statecraft that terminated in their July 1914 gamble.(20) It is instead an attempt to get beyond the old, tired blame game by showing that the root cause lies deeper than the policies of either the Central Powers or their rivals. It derives from the overall character of the international game being waged and the fundamentally unfavorable geopolitical position Germany and Austria-Hungary occupied within it.

That game requires at least a thumb-nail description here. It comprised two simultaneous contests, inextricably intertwined and interdependent but with differing characteristics, stakes, and rules. The first was that of the old European balance of power. By 1914 this had evolved into an extremely competitive, zero-sum contest played for very high stakes (national survival) and at great risk (general war among populous industrialized states possessing mass armies); yet until just before 1914 certain minimal restraints or norms of international conduct left over from the Vienna era still prevailed. These norms, combined with prudence derived above all from fear of a general war, kept the competition in Europe between individual powers and rival alliance systems one waged for relative advantage rather than decisive victory; the powers' aims were to ensure victory in case of war and to gain an upper hand in imperialist competition, but not to conquer or eliminate one's rivals. The notion of preserving a balance of power, still widely held as an ideal though each power defined the desired balance differently and pursued it in incompatible ways, rested on a general recognition that even a victorious great war would be terribly risky and costly and might prove counterproductive, creating new international dangers by destroying the existing balance or eliminating essential actors. Thus the game resembled high- stakes poker played by heavily armed men bent on winning but reluctant to raise the stakes too high, both to avoid losing themselves and to avoid provoking others who faced bankruptcy into kicking over the table and starting a gunfight. As a result, there was a certain unspoken, consensual limit on the size of the bets and a general assumption that over major issues some compromise involving a minimal level of satisfaction for everyone, or at least all the great powers, should emerge. This last remnant of the old European Concert principle remained alive, though barely so, in the two Moroccan Crises, the Bosnian Crisis, and the diplomacy of the Balkan Wars.(21)

Another game was played alongside European balance-of-power politics, however, called imperialism or world politics (different names for the same thing). Its stakes were shares in the economic, military, political, and territorial control and exploitation of the non-European world; its goals and rules resembled the board game Monopoly; and it evolved differently from its companion game. The nineteenth-century European balance game began in 1814-15 with conservative monarchical cooperation against war, revolution, and territorial change and gradually evolved by 1914 into almost unrestrained zero-sum competition. Imperialism, always present throughout the nineteenth century but only taking center stage after about 1870, started out then as an individualistic scramble, carried on initially more by individuals and firms than governments, for goods supposedly free for the taking. This made imperialism at first a win-win contest for governments, less dangerous and more cooperative than the European balance game, in some ways a safe outlet for drives and energies too dangerous to be employed in Europe. Hence late nineteenth and early twentieth century imperialism sometimes led to confrontations but seldom to wars between European states (even the wars that occurred between European powers and the colonized peoples and states were usually small scale affairs)(22) and often to deals dividing the spoils between certain claimants.(23)

Yet in the end European imperialism was more rapacious than ordinary balance-of-power politics, quite apart from the rapacity exhibited to colonized peoples and territories. Unlike the European game, its primary aim was not security and relative advantage, but gains and acquisitions, which as time went on increasingly drove states to seek clear-cut control of particular areas, shutting others out. To be sure, sharing-out agreements continued to be made up to and through the Great War-consortia to build railways, carry on commercial activities, or exploit mineral resources in China or the Ottoman Empire, agreements to permit other powers commercial access to one's own colonies, international or bilateral deals over Egyptian, Ottoman, or Chinese customs, etc. Yet not only were these agreements often a pis aller necessary to avoid dangerous conflicts or to share prizes too expensive or troublesome to exploit exclusively. They were also usually monopolistic or semi-monopolistic in character, dividing up regions so as to exclude others and enable each partner to monopolize its own sphere. Moreover, even this element of cooperation tended increasingly to break down into confrontation or open conflict. New Imperialism tended inexorably toward exclusive paramountcy and control. Witness the aggressive extension of the American Monroe Doctrine in the Western Hemisphere (Cuba and the Caribbean, the Panama Canal, Venezuela, Brazil) and the Pacific, extending even to the Philippines; the British version of their own Monroe Doctrine, informal but effective, in much of Africa, India, and elsewhere; France's preemptive extension of its exclusive control from Algeria to Tunisia and Morocco; Russia's version of exclusive empire in Central Asia, tried less successfully in Manchuria, North China, and Korea, where Japan countered this with its own program. The Anglo-Russian Convention of 1907, which tried to avoid conflict and ensure cooperation by dividing Persia and Central and South Asia into clear-cut spheres, led to far more friction than cooperation.

This points to further crucial differences. European balance-of-power politics before 1914, even at its most competitive as in its rival security alliances, was supposed to keep all the necessary players in the game and to last indefinitely with no decisive end-point. The players had established, relatively fixed, legally recognized positions and well-known, comparable assets and opportunities, making the idea of regulating competition among them by an equilibrium of forces thinkable, though not necessarily feasible. European imperialist politics, in contrast, was designed to keep some players in the game while driving others out. Its conclusion, with winners and losers, would come when all the available world spoils were divided up, promoting a dominant spirit of Torschlusspanik--panic at the closing of the gates-from early on. Finally, the players started from very different starting points with vastly different, almost incommensurable and non-comparable assets, liabilities, and opportunities. When the serious game began after 1870, Great Britain started it with a vast empire and many opportunities for further expansion, but at the same time facing a new formidable challenge. Its vast, far-flung possessions and the informal character of its paramount position in much of Africa and Asia, both stemming from a period in which it had no serious rival in naval, industrial, and commercial terms, made the British Empire now vulnerable and hard to defend against new competitors at a time when Britain was losing its industrial lead, and the efforts necessary to defend it might undermine the very commercial strength and prosperity on which the Empire ultimately depended and which it was supposed to promote. Two other powers, Russia and the United States, had extensive empires that were mainly continental and hence less vulnerable, giving them both considerable security and potential for further expansion. France had a substantial colonial empire and numerous opportunities for expansion, but relatively little power and capital to expend on them. Other actors (Spain, Portugal, and the Netherlands) had residual empires they were determined to retain and exploit but could not defend against serious challenge. Finally, new players having no prior stake and widely varying capacities nonetheless entered determined to play (Germany, Japan, Italy, the King of Belgium, and toward the end Austria-Hungary). Even this does not exhaust the roster of players. Those who became targets of imperialism-China, the Ottoman Empire, various African states and empires-did not simply react passively, but developed their own programs, sometimes expansionist-imperialist ones (Great Serbia, Great Bulgaria, the Greek Megale Idea, Pan-Turanianism, and the like.)

All this insured that the imperialist game, unlike the European one, could not be played according to more or less rational rules and calculations leading to some sort of balanced power and satisfactions, but would end in clear winners and losers. Moreover, while the high stakes of the European balance game, the fact that the survival of the nation was at risk in any general war, made for caution, the high stakes in imperialist politics, based on the general conviction that a nation's future survival and prosperity in the coming century depended on acquiring world power and position, had an opposite effect. Since the immediate danger of a great war breaking out over imperialist quarrels seemed small, imperialist competition encouraged strategic and tactical boldness, going for broke.

This relates to our main question, because in order to judge whether Germany and Austria-Hungary had alternative policies available by which they might have averted the threats they faced in 1914 and eliminated any need to gamble, one must appraise how much intrinsic chance they had to succeed in these two interlocking games from an earlier point-say, 1890--at which both games became more seriously competitive. In the European balance game, in my judgment, their basic starting positions, strengths, and liabilities gave neither much chance for significant gains and Austria-Hungary in particular would be hard-pressed to hold its own regardless of what it did. One of the worst miscalculations German leaders made was to expect Germany's central position in Europe to help it exploit rivalries between other powers and make itself indispensable to both sides, at a profit. That geographical position (as Bismarck had known-it gave him his nightmare of coalitions) was instead a huge handicap, forcing both powers always to reckon with the likelihood of a two-front war (in Austria-Hungary's case a multi-front one), increasing their vulnerability and limiting their freedom of maneuver and alliance capability. Centrally located as they were, they could make firm commitments only to each other or to weaker states needing their support, such as Italy and Rumania. They also had an additional liability often ignored in the literature: unlike all the other important powers save the Ottoman Empire, they had territories other states and/or peoples coveted and in certain instances claimed by right. In Germany's case, this meant France (Alsace-Lorraine), Denmark (North Schleswig), and the Poles (Polish Prussia). Austria's case was far worse: Italian nationalists claimed the Veneto and Trentino, Istria, and parts of Dalmatia, Russian nationalists and leaders, including the Tsar, wanted to solve the Ukrainian problem by annexing East Galicia and the Bukovina, Serbia claimed all the Austro-Hungarian territories populated by Serbs or Croats, and Rumania had its eyes on Transylvania.

To be sure, these claims and velléités did not immediately threaten Germany's and
Austria-Hungary's territorial integrity. Like other notions some Russians had about the Turkish Straits or East Prussia and Prussian Poland or some British and French had about Germany's colonies, these aims would come into play, and did, once war broke out, but no one save radical Serbian nationalists and their backers in the Serbian military and in extreme Panslav circles in Russia wanted a war to achieve them. Throughout the prewar period the Central Powers, especially Germany, remained too strong for other powers to challenge them too directly in Europe. In other words, their basic situation was unfavorable but not disastrous; it was likely that they would lose in terms of relative security and advantage, but unlikely that they would forfeit their positions as European great powers. This pretty much sums up the outcome of European great power politics from 1890 to 1910. Given their basic situation at the outset, it was natural and normal if not strictly predictable, readily understandable without invoking particular blunders or provocations on their part. Nor is there ground to suppose that other policies on their part would necessarily have changed this result very much.

If one asked about their basic chances for success in the imperialist-world politics game from about 1890 on, the answer could well be the bon mot on their situations in 1917: in Berlin the situation was serious but not hopeless, in Vienna it was hopeless but not serious. Almost everything in their geopolitical situation worked against their succeeding in imperialist expansion: no initial foundation in terms of colonies, overseas trade, bases, and readily projected naval or military power; an unfavorable geographic location with only limited access to one ocean, easily blocked by rivals in case of war; an exposed position in Europe which forced them to limit their commitments and be risk-averse in the world game, making them unattractive as imperialist partners and tempting as targets; and internal divisions and weaknesses hampering both, especially Austria-Hungary. Germany had only one of the requirements for success in the imperialist world game, a vibrant growing economy, and Austria-Hungary, though growing economically, did not enjoy even that.

Just as important as these liabilities in insuring their defeat were the rules of the imperialist game and the way the other powers played it. The dominant fact--obvious yet somehow frequently overlooked or, if noticed, not taken seriously-is that the other imperialist great powers, Britain above all, but also Russia, France, Italy, and to some extent the United States and later Japan, played to make Germany and Austria-Hungary lose, as part of their strategy to win. The common German charge that the Triple Entente deliberately encircled Germany in Europe was false, at least so far as Britain was concerned, but it is true that Germany and Austria-Hungary were deliberately circled out of world politics as much as possible. The Anglo-French Entente Cordiale in 1904 was intended to keep Germany from interfering with exclusive British and French control in Egypt and Morocco. British efforts from 1890 on to reach an agreement with Russia over the Middle East, culminating in their 1907 convention on Persia and Central Asia, were designed to prevent German penetration of this region-an aim Russia shared. Much of British foreign policy on South Africa was directed at keeping Germany from interfering there at all, whether as a partner or as an opponent. France deliberately set out to do the same vis-à-vis Germany in both Moroccan Crises, violating international agreements of 1880, 1906, and 1909 in the process. The United States worked with Britain in Latin America, the South Seas, and the Far East to limit German influence. The British and Russians collaborated against Germany on the Bagdad Railway and fought especially hard against German influence at Constantinople and in Mesopotamia. Russia, encouraged by Britain and aided by France, worked from 1907 on to check Austro-Hungarian influence in the Balkans and especially after 1911 to eliminate that influence entirely. Russia, Britain, Japan, and the United States all tried to check German economic and political expansion in China and the Far East. In the prewar scramble for concessions in Asiatic Turkey, all the other powers, including Italy and Germany, worked against Austria-Hungary.

Of course this is not evidence of a sinister anti-German or anti-Austro-Hungarian conspiracy. These tactics broke no rules because these were the rules, the way to play the imperialist game for fun and profit. One no more needs to invoke an anti-German or anti-Austro- Hungarian conspiracy to account for this pattern than one needs to talk of conspiracies to account for monopolistic and oligopolistic combinations and strategies in the business world, or to explain how these often target particular firms and sometimes drive them out of business. Everyone recognizes these tactics as part of the game.(24) The pattern, however, does further undermine the view that German and Austro-Hungarian policies were primarily responsible for the threats to their interests and security, and that Germany could have achieved its needed place in the sun had it followed less aggressive and provocative policies. This is like arguing that Nissan or Apple could have avoided their troubles by following less aggressive and provocative policies toward their competitors. It ignores both concrete evidence to the contrary and the basic rules and nature of the game.

A similar unrealism afflicts the related argument that Germany, even if lost the contest in power politics, would nonetheless have survived and prospered simply by continuing its current rate of economic growth, becoming in a few more years of peace an economic hegemon too powerful for the others to challenge. This argument, on the surface plausible, seems to ignore certain facts, such as the precarious nature of Germany's economic achievements and prosperity in an age of intense competition and frequent booms and busts (German economic growth was being surpassed by the United States at the same time and much the same rate as Germany was surpassing Great Britain, and Russia had the fastest rate of industrial growth before 1914); or the fact that the more German trade and exports grew, the more dependent the German economy became on external markets and imports for further growth and survival, and the more vulnerable it became to military threats to these. Since British leaders calculated that they could easily destroy German overseas commerce and ruin Germany's economy by a naval blockade, it seems reasonable that Germans would feel vulnerable to this regardless of how much economic power and wealth they amassed. But even apart from this, the most important thing here is to understand the counterfactual question. It is not, "How would the European and world economies have developed had peace continued for some years after 1914?" That question is both too loaded with indeterminate contingent variables to be answered, and not relevant here. The question is rather, "In the real world of 1914, could Germany's leaders and public have been expected to rely for their security against foes already superior to Germany on the sea and expected shortly to achieve superiority also on land on the prospect or possibility that if peace lasted long enough, Germany's economic dynamism would protect it?"

To do so, German leaders would have had to be confident not merely that Germany would win the current economic competition, but that a generally free, liberal world economic order with open access to international trade, especially overseas, would continue indefinitely, regardless of developments on the European and world strategic and military stage and regardless of whether Germany could if necessary support its economic interests with political and military weapons. This assumption flies in the face of facts. It assumes that Smithian free-market liberalism by 1914 had decisively triumphed over neo-mercantilism, protectionism, and economic imperialism, when in fact all the major powers save Britain believed in protectionism and mercantilism rather than free trade, and most states were more protectionist than Germany. It assumes that Germany's rivals would have peacefully come to terms with Germany's economic domination, when in fact they were already worried by Germany's economic progress and took active measures before the war, especially in Russia and France, to avoid becoming economically dependent on Germany.(25) It assumes that the prewar international economic system operated largely independently of European high politics and military strategy and would continue to do so, when in fact everyone believed that a strong state and a strong economy required each other and that it was the government's duty to bring the nation's political, military, and economic resources together to promote its national interests. Tariff wars, discrimination against foreign goods and enterprises, and attempts by governments to promote their nationals' economic interests or to use these interests to promote their political and strategic ends were central to the age of imperialism. Even the British, who still adhered to free trade principles, relied on their naval supremacy and empire as a hedge against dangerous competition or decline. In other words, this counterfactual holds that the Germans, of all people, should have believed and trusted in the message of Norman Angell's prewar book The Great Illusion: that growing interdependence in the modern capitalist economy had rendered war obsolete, counterproductive, and unthinkable. True, Angell was right in criticizing the reigning neo-mercantilist, protectionist, and militarist doctrines of his day, but he also ignored power political realities and their connection to economics then and since.(26)

The counterfactual argument that Germany could have broken up or loosened the alliances or quasi-alliances against it by more moderate, patient policies and conduct has similar problems. Granted, Germany's opponents genuinely perceived Germany as unpredictable and dangerous, and were acting partly to counter that threat. But this does not mean that Germany could have removed that perception and changed its opponents' policies simply by becoming somehow more moderate and conciliatory in its behavior. Germany posed a threat particularly to Russia and France mainly because of where it was located and the power it possessed rather than by its policies, and their alliances and ententes were intended to meet this objective, structural threat by giving the Entente powers a margin of military preponderance over Germany. Any signs of German restraint would and did serve as proof that these alignments were working and should be continued. Besides, as already noted, these combinations had important uses in world politics. Their central value for Britain was to help preserve the British Empire by maintaining Britain's good relations with France and Russia so as to curb both their colonial rivalries with Britain and German competition. Preserving the so-called balance of power in Europe was part and parcel of this policy. In other words, the anti-German alliances and ententes were so intrinsically valuable for the Entente powers for both their security in Europe and their world-imperialist purposes that German good behavior would not have made them give them up, and that German attempts to undermine or loosen them, or even join them, served as more proofs that Germany was treacherous and dangerous. The history of prewar politics shows this. The cognitive biases apparent in the consensus view-the ascription of more freedom of choice to one side (in this case Germany) than to the other, and the belief that it could easily have changed its policy and thereby have induced the other side to change its--are familiar to political psychologists.

All this concerns only the German problem-on which most historians concentrate, ignoring thereby the more immediate and pressing half, the Austro-Hungarian problem.(27) The counterfactuals embedded in the consensus view involving Austria-Hungary are even stranger than those for Germany, and receive less scrutiny. But in a way this is not surprising, for the consensus case in regard to Austria-Hungary with its embedded counterfactuals rests on assumptions so unwarranted as hardly to deserve discussion. A good example is the notion that internal reforms could have solved the nationalities disputes within Austria-Hungary and thus given it the needed power and cohesion to survive the ruthless competition of European and world politics. This assumes two things: that nationalities conflicts of the kind that have troubled the Habsburg Monarchy and other multi-national states in modern times are soluble by any means,(28) and that internal reforms, if they succeed in promoting greater domestic harmony, also make a state stronger for foreign policy purposes. Neither assumption has much to support it in theory or evidence. In fact, the Austrian government launched many reforms between 1867 and 1914. These helped make the Monarchy a progressive, modernizing state in important respects--a thriving culture, growing economy, advanced educational system, and a political system that, though riddled with conflict and tensions, respected civil rights and included democratic features. But these reforms also, inevitably, hampered the Monarchy's efforts to conduct a strong foreign policy. The more freedom its many peoples, factions, and parties enjoyed to contend for their particular rights, status, and share of power within the Monarchy, and the more parliamentary (and thereby more chaotic) its politics became, the less chance there was to unite everyone on a single foreign policy agenda, or to raise the taxes needed to keep Austria-Hungary competitive in the European arms race, or to prevent foreign governments and groups from intervening in the Monarchy's nationalities conflicts, and the nationalities themselves from exploiting this.

Even more implausible is the suggestion that successful internal reforms, whatever these might have been, would have lessened the hostility or changed the aims of its opponents abroad. Russian nationalists and the Russian government were not interested in protecting the rights of Ruthenians in East Galicia; their concern was to prevent Ruthenian (i.e., Ukrainian) nationalism from spreading from East Galicia and the Bukovina to Imperial Russia, and the ideal solution was to annex these territories to Russia. Much the same holds for Italian nationalists and their irredentist claims, as well as Rumanian nationalists, to say nothing of the Serbs. This has nothing to do with the question of whether Austria-Hungary should have done something more or different to meet its internal problems; it means only that meeting its internal problems would not have significantly changed the attitudes or actions of its opponents.

The central weakness in the counterfactual case on Austria-Hungary, however, parallels the one in regard to Germany: it ignores the basic rules and nature of the game. Austria- Hungary's competitors and opponents were acting in regard to the Monarchy essentially in behalf of their own interests and aims, not in reaction to what it did. Austria-Hungary could have prevented this only by changing the nature and stakes of the game to make this unprofitable-- which was what it was trying to do by its July 1914 gamble.

One more feature of the standard counterfactual scenario deserves mention: that it leaves the two sides of its case, the German and Austro-Hungarian aspects, unconnected when they are in fact tightly interwoven. It suggests a counterfactual solution for Germany's security problem, namely, that it show greater restraint, moderation, and patience toward its opponents and accept some temporary military and strategic insecurity while seeking its future security in relaxed tensions in Europe and German economic growth. For Austria-Hungary it suggests domestic reforms to strengthen it politically and militarily so that it could better defend its interests against external challenges. Leave aside for the moment the inherent weaknesses in these proposals, already discussed, and ask simply how they fit and work together. The answer is, "They do not- they contradict each other." Suppose per impossibile that Austria-Hungary could before 1914 have achieved the internal cohesion and economic strength to keep up with the others in the arms race; how would that have fit in with a simultaneous effort by Germany to try to cool the arms race? It would have been obviously and directly contrary to it-the main reason being that the fixed policy of all three Entente powers was to consider Austria-Hungary as simply Germany's subordinate ally, no matter how desperately the Austrians pleaded that they were pursuing an independent policy, so that a stronger, more confident and assertive Austria-Hungary automatically meant in St. Petersburg, Paris, and London a stronger, more dangerous Germany.(29) Or consider the impact of German efforts to conciliate its opponents on Austria-Hungary's security problem. The historical evidence is that such efforts by Germany made Austria- Hungary's problem worse. What Russia wanted as proof of German moderation and cooperation, also demanded by Britain, was that Germany restrain Austria-Hungary in the Balkans. Germany's refusal in Russia's eyes to restrain Austria-Hungary in 1909 was the source of massive, permanent Russian resentment. Germany did, however, restrain its ally from 1910 to 1913, thereby helping prevent a general war and temporarily improving Russo-German and Anglo-German relations-and also helping to undermine Austria-Hungary's position and fuel frustration and despair among its decision-makers. Or consider the suggestion that Germany should have relied on peaceful economic expansion for its future security. One of Germany's most important economic targets before 1914 was the Balkans and the Ottoman Empire. German economic expansion there directly threatened Austria-Hungary's trade, prosperity and independence more than those of any other state, serving to encourage Austria-Hungary's opponents, Serbia in particular, and push Austria-Hungary toward violent countermeasures.

These points are important not just as further instances of the internal contradictions in the consensus scenario and the ways it neglects the Austro-Hungarian problem, but as evidence of a complete misunderstanding of the German problem as well. Those who insist that Germany was mainly responsible for the Central Powers' gambling strategy in 1914, even though Austria- Hungary conceived that strategy, demanded German support for it, and finally launched it, argue that Austria-Hungary could not possibly have acted without German help, and that since Germany gave its ally a blank check, subsequently pressed Austria-Hungary forward, and never really tried to restrain it, Germany was chiefly responsible. Once again the embedded counterfactual assumptions demand examination. There are at least two, closely related: first, that the German government, if it genuinely wanted peace, could have rejected Vienna's demand for support, regardless of Austro-Hungarian warnings that a denial of support would critically affect the alliance, future Austrian policy, and the survival of the Monarchy as a great power; second, that Germany, in the interest of general peace, could and should have detached its security and great-power status from Austria-Hungary's survival as a great power-a survival that Germans, like everyone in Europe, including especially the Austrians, considered genuinely threatened.

I cannot see how these assumptions could possibly be defended. They seem to me to contradict everything we know about the history of German and Austrian relations in Central Europe, the connection between this problem and the wider problem of relations with Russia in Eastern Europe, the equally intertwined Near-Eastern-Balkan problem, and the nature of European international politics, both political and military. Above all they strike me as an absurd way of promoting a durable European peace. In a more peaceful and stable earlier era, Bismarck recognized and acted upon an insight fully confirmed by history since 1914: that the breakup of the Habsburg Monarchy or its disappearance as a great power, regardless of how it happened or whether or not Austria deserved it, must have revolutionary consequences for the security of Germany and Europe as a whole. That insight is here ignored. It is not at all surprising that Germany chose in July 1914 to respond to Austria-Hungary's plea for strong action and back it in its desperate gamble. It had pretty well decided to do this earlier in 1914. It is instead surprising that the German government had earlier tried for so long to ignore its ally's problems or sweep them under the rug, and had in various ways helped make them worse, so that only now did it seriously reckon with the consequences for Germany of Austria-Hungary's continued decline and potential collapse or defection as an ally.

In other words, the crowning anomaly in the consensus view and its counterfactuals lies in its ignoring precisely what the July Crisis most clearly proves: that Germany could not ignore the Austro-Hungarian problem even though it wished to, because the German and Austro- Hungarian problems were Siamese twins, and part of still wider and more complex Central and East European and Near Eastern problems, so that an attempt to solve or manage the German question and the question of European peace without seriously dealing with the Austro- Hungarian problem was an attempt to present Hamlet without the Prince of Denmark. The failure of scholars to see this strikes me as invincible incomprehension.


III. An Account of the Origins of the War with Different Embedded Counterfactuals

My argument thus far seems to suggest that the war was inevitable for the following

a) The nature of the European power game and of Germany's and Austria-Hungary's respective positions within it made its actual outcome by 1914, namely, relative loss, frustration, and looming danger for Germany and even worse decline and immediate peril for Austria- Hungary, likely from the outset.

b) Similarly, the nature of the imperialist game and of Germany's and Austria's positions in it made its actual outcome, that Germany would clearly lose relative to its rivals and Austria- Hungary lose entirely, even more likely.

c) These unfavorable outcomes and trends were probably not enough individually and by themselves to make the two powers risk a general war in order to reverse them. This is indicated by the fact that on several occasions previously (1904-05, 1908-09, and 1912-13) they passed up opportunities for war when their chances for success were better than they were in 1914. Nevertheless, given the facts that these two games were tied together both in reality and in the public perception, and that the Central Powers, like others, believed that both contests were critical to their ultimate survival, security, and prosperity as great powers, their belief that they were losing and declining in both made some risky action on their part to reverse the trend likely at some point. An immediate challenge and threat to the independence, integrity, and great-power status of one or both of them, such as arose on June 28, 1914, would increase that likelihood dramatically.

d) Since their rivals shared their assumptions regarding the nature, rules, and stakes of the combined European-world politics game and were therefore equally determined to maintain their favorable positions or improve them, any German-Austro-Hungarian initiative to reverse the existing trends of the game was almost certain to meet strong resistance and produce a direct collision between the two sides. The tense, crisis-laden atmosphere of prewar politics, with many vital issues unresolved and major developments in flux, made it virtually certain that occasions for confrontations and clashes of interest would arise. Given the nature of the stakes for both sides and the absence of any mutually acceptable way to satisfy their irreconcilable purposes, one must consider a general war as inevitable sooner or later.

This view comes close but is too determinist, or determinist in the wrong way. It makes the determining element the nature, rules, and stakes of the prevailing game of international politics and the objective conditions under which the various actors began it and played it out. The version I propose locates the determining element elsewhere-not in the international game itself, which still could conceivably have continued for some time without general war and without radical changes in its rules (here my view approaches Lebow's), but in the political culture of the era and in certain dominant beliefs about the prevailing game.

Let me try to show the difference by an analogy, inevitably inexact but perhaps useful for illustration. Compare World War I to a train collision involving five trains, all in a race to reach the station first or at least to avoid coming in last. The strict determinist view just outlined holds that they collided because all five were on intersecting tracks, the only way to avoid an accident was for at least one or two of them to give way to the others, thereby losing the race, and none considered this outcome acceptable. An indeterminist view would hold that the trains, though they were running on unsafe tracks at dangerously high speeds with obsolete equipment operated in certain instances by reckless engineers, were not running on intersecting tracks but parallel ones set dangerously close together. Hence a collision was not inevitable but could only arise by accident (say, if one of the trains left the tracks or swayed into another one) or by deliberate recklessness. The latter caused the actual collision. My version holds that while all five trains were involved in the race and running together closely enough that all would be involved in any accident, only three of the five were on a collision course. These three, however, had been in similar races over this same terrain a number of times before, and knew how an accident could be avoided-when to slow down, what signals to give, what switches or side-tracks to take, etc.- actions that involved some active coordination between themselves and at least passive cooperation from the other two trains in the race. What caused the collision in this instance was a refusal by the engineers on all five trains at critical moments to take the steps known from experience to be needed to avoid an accident. This failure to act derived from a shared conviction that such actions were no longer part of the game, had become futile and counterproductive, would cause them to lose the race, and were in any case not their particular responsibility. This collective mentality and fixed attitude made the collision unavoidable.

Notice that this last version shifts the focus from, "Who or what caused the train wreck?" to "Who or what caused the failure to avoid it?" Applied to World War I, the focus is changed from, "Who or what caused the outbreak of war?" to "Who or what caused the breakdown of peace?" I am convinced, for many reasons impossible to discuss here, that this should be the prime emphasis for the study of war in general-that, to put it simply, scholars of international relations need to give more attention to explaining peace than explaining war. But regardless of the merits of this claim as a broad principle, I hope to show at least in a prima facie way that the distinction is meaningful and that World War I is best explained as a case of the breakdown of peace rather than the outbreak of war. Such an analysis reveals that the war had become unavoidable not because the factors driving the different powers toward it had become irresistible, but because the actions needed to avoid it had become unthinkable. Consistently with the theme of this volume, the argument involves counterfactual reasoning.

The strict determinist argument sketched out earlier holds that under the circumstances prevailing by 1914 the German powers were bound sooner or later to try to reverse the prevailing trends which promised an unacceptable outcome, using violent means that risked general war if necessary, while the others were equally certain to resist this strongly, leading to war. To see why this comes close but misses the target, one needs to ask two closely related counterfactual questions. First, what plausible circumstances might have led Germany and Austria-Hungary in 1914 to decide once again, as they had done several times before, to try other ways of defending their security and vital interests? Second, what actions, plausible under the circumstances, might the other powers have decided to take before or during the July Crisis suitable to deter and/or dissuade the German powers from risking war?

These counterfactual questions seem to give the game away to the indeterminists, opening the door wide to many suggestions and alternative scenarios commonly encountered in the literature. Things would have been different had the assassination attempt failed, or had the Austro-Hungarians stopped in Belgrade, or had Russia given the Serbian government different advice, or had Britain given Germany a clear warning that it would enter the war on France's side, etc. However, in light of the argument made earlier, that all the major actors were fundamentally driven by long term concerns based on shared assumptions about the nature, rules, and stakes of the game and a shared understanding of where that game was headed, the door remains closed to that kind of speculative contingency. If one side by 1914 was determined to reverse that trend and avert that ultimate outcome even at the grave risk of war, and the other side was equally ready to accept war rather than let that happen, then different individual events and actions in the July Crisis would only alter the occasion, timing, and form of the final collision, not avert it, unless they changed these shared assumptions, beliefs, and expectations. The real questions therefore are, first, whether such a change in these reigning collective European mindsets and understandings about international politics was possible at all before or during the July Crisis, and second, what alternative policies, decisions, and actions conceivable in terms of the minimal-rewrite rule and of historical evidence might have effected this change, i.e., altered the reigning perceptions of current and future trends sufficiently on both sides, especially on the Central Powers, to change their views of what could and must be done. The strong determinist position denies that any such shift was possible; the indeterminist one denies that any was necessary. My view is that objectively speaking, strategies and tactics were still available to the great powers that might have averted a collision by changing crucial prevailing mindsets, but that subjectively, in terms of what the actors considered conceivable and feasible, they were not- hence the inevitability of war.

The first step in testing this is establishing just what needed to change in the mindsets of what particular actors. The consensus view holds that only German and Austro-Hungarian attitudes needed to change; that view, as we have seen, will not do. But did virtually everything in the whole situation have to change? The determinist view fits the common impression that Europe by 1914 was a tinderbox filled with explosive material waiting for a spark, so that war could have broken out over any one or any combination of many issues or causes. That picture is also misleading. Actually, Europe in June 1914 was near general war, as it had been repeatedly since 1908, but it was not yet at the brink or certain to go over it, and most of the conflicts which divided the great powers were not such as to set off a war. In fact (here again comes counterfactual reasoning) no convincing scenario can be constructed by which most of the contentious issues would or could have become a casus belli, alone or even in combination. One could compile a long list of issues--Anglo-German naval rivalry, Alsace-Lorraine and other irredentist territorial claims, military threats, colonial and commercial rivalries, historic national hatreds, ethnic and racial animosities-that were serious, sufficient to create hostility and tension, but were not questions over which any great power wanted or intended to fight, or for which it could plausibly start a war. In contrast, there is only a short list of great powers who were ready under any circumstances to start a general war--Russia, Germany, and Austria- Hungary--and they had only limited and specific grounds for doing so. Russia would fight for two reasons: to prevent any other power from gaining control of the Turkish Straits (witness its willingness to use force to prevent its own allies and associates, Bulgaria and Greece, from seizing Constantinople in the first Balkan War, and its strong stand over the Liman von Sanders affair in early 1914); and to avert what the Russian government, driven by a nationalist press and so-called public opinion, viewed as another humiliation like that of 1908-09 in the Balkans at the hands of the German powers. Germany would choose war rather than allow its army to become decisively inferior to those of its foes, either through Russia's successful completion of its armaments program or by Austria-Hungary's collapse or defection, or both. Austria-Hungary's will for war was the most desperate and dangerous of all. Although it feared general war more than any other great power, with good reason, its leaders had already concluded by early 1914 that it could not tolerate any further decline in its great-power status and its Balkan position, and particularly any more challenges and provocations from Serbia. These were the only issues that that could have caused general war in 1914, and the ones that did cause it. The question of the avoidability of war therefore rests neither on whether some sweeping change in the whole international situation was possible, nor on whether particular contingent events in July 1914 could have gone differently, but on the specific question of whether these particular great powers could have been deterred and/or persuaded from risking general war for these particular reasons.

The answer is "Yes." It arises not from theory or speculation, but solid historical evidence. The first thing to recognize is that these problems were not new to these powers, but old and familiar, almost standard; they had caused wars or threatened to cause them before. (This is the point, in my analogy of the train wreck, of noting that the engineers of the three trains had been over this terrain previously and knew what caused wrecks and what needed to be done to avoid them). Twice in the previous century (1809 and 1859) Austria had gone to war rather than accept a further decline in its great-power status and position and more threats to its prestige and rights. In 1756 Prussia had deliberately launched a preventive war against Austria and Russia rather than wait for an overwhelming coalition to jell against it.(30) Russia had been ready in the Bulgarian Crisis of 1884-87 to fight Austria rather than accept another supposed humiliation at its hands. Not only were the essential dangers in 1914 familiar, almost commonplace; so were the theater, the terrain and the three players. Ever since 1763 at the end of the Seven Years War, when Russia and Prussia had fully emerged as recognized great powers, these three states had dominated Central and Eastern Europe, competing over territory, interests, influence, leadership, and security. This area, even during the Napoleonic Wars, had been constantly the main focus and center of European politics. The issues that dominated the Austro-German-Russian relationship and threatened the peace in 1914 had mutatis mutandis been vital for them the whole time.

But if the issues and dangers were familiar, so were the remedies. The astonishing fact (astonishing both in itself and in its being so widely ignored) is that the 150 years of Austro- German-Russian relations after 1763 represent a story not of constant rivalry, conflicts of interests, struggles for power and influence, and frequent tensions and crises leading to war, but of constant rivalry, conflicts of interests, struggles for power and influence, and frequent tensions and crises resulting in peace. Between 1740 and 1914, Austria and Russia, always rivals in the Balkans, often rivals elsewhere as well, frequently at swords' points, never fought each other, except for two occasions in 1809 and 1812 when they were dragged by Napoleon into half- hearted campaigns that they would never have entered on their own. The same is true between 1762 and 1914 for Prussia-Germany and Russia. Austria and Prussia fought two short wars over Germany, to be sure-one indecisive in 1778-79, the other decisive in 1866. Yet within thirteen years of the latter they were again allies, as they had been most of the fifty years before 1866- without ever ceasing to be rivals. In the same way Austria and Russia and Germany and Russia were frequently allies though always rivals.

Thus the central story in European international history from 1763 to 1914 is this remarkable Austro-German-Russian peace. 1914 must be seen first and foremost not simply and generically as the outbreak of general European war, but as the breakdown of that specific long peace. To explain the war, scholars must first explain it, understand what maintained and revived it so long, often against improbable odds, and then, having done this, ask themselves whether the measures and devices that had previously served to maintain this Austro-German-Russian peace no longer would work in 1914, or whether (as I believe) they simply were not tried.

To attempt any such serious analysis here would stretch the already elastic bounds of this essay beyond the breaking point. I will therefore merely make some general points, more by assertion than by argument. First, the procedures and principles of European diplomacy used for dealing with such problems as these, especially those of the European Concert, were well known. Where seriously tried, they still worked even in 1914. One vital issue capable of causing war in 1914, that of the Turkish Straits, was actually handled successfully in this way. When Russia warned Bulgaria and Greece to stay away from the Straits and succeeded with British and French support and restraint in getting Germany and the Ottoman Empire to back down on the Liman affair under cover of a face-saving formula without using force against the Turks, this all had precedent in previous Concert diplomacy over the Eastern question. The underlying principles were clear: Russia had special interests in the Straits and could not allow others besides Turkey to control them, but could rely on diplomacy and the Concert to defend its interests and was not allowed to act unilaterally or by force. True, the other fighting issue for Russia, that it would not tolerate another humiliation in the Balkans at the hands of Austria-Hungary and Germany, and the corresponding fighting issue for Austria-Hungary, that it could not endure any further undermining of its great power position in the Balkans or challenges from Russia and its client Serbia, were far more difficult to handle, not merely because mutually incompatible perceptions and enflamed public opinion were involved on all sides, but also because (in my view at least) Russian perceptions were here remarkably one-sided. The conviction that Russia's rights had repeatedly been violated, its prestige and honor challenged, and its security and historic mission in the Balkans threatened ever since 1908 by the German powers contrasts glaringly with the undeniable facts that in 1908-09 as in 1904-06 Russia had been fortunate to escape from dangers of its own making, dangers that the Central Powers had not exploited, and that since then Russia had been winning and getting away with a very bold offensive policy. Yet this was far from the first time that Russia had blamed difficulties of its own making on Germany and Austria (witness the Eastern Crisis of 1875-78 and the Bulgarian Crisis of 1884-87) or that Austria had seen Russian pressure as an alp that it had to shake off at almost any cost (the Crimean War). Historically, there were tested ways of handling such problems short of war.(31)

As for Germany's fear of Russia, here again one must distinguish between what was irrational-the fear of being overrun by barbarian hordes from the East and the notion that a great Teuton-Slav struggle for mastery in Europe was ineluctably approaching-and what was concrete and rational, the fear of being outmanned by 1917 by the combined Russo-French armies. Diplomacy could do nothing directly about the former, but it could have done something to manage the latter, even within the existing alliance structure. For instance, there could have been some informal equivalent for Germany of Bismarck's Reinsurance Treaty with Russia in 1887, assuring Germany that France and Britain would not support a Russian attack on Germany, as Bismarck had reassured Russia that Germany would not support a British-Austrian offensive against it.

The other requirement for simultaneously deterring and reassuring Germany concerns Austria-Hungary, and brings us to the heart of the problem. One can hardly emphasize too strongly the destabilizing effect of the conviction German leaders themselves had reached by 1914, one promoted by the Austrians themselves, that Germany must now use its ally or lose it- by it now at any risk and cost, or expect shortly to have to fight without it because of Austria-Hungary's defection, paralysis, or breakup. It is difficult enough to imagine a counterfactual scenario by which Germany with its powerful, irresponsible military, its erratic, impulsive monarch, and its semi-authoritarian, deeply divided government and society would have calmly stood by while Russia and France completed their efforts to achieve military superiority over it. It is quite impossible to imagine Germany doing so while it simultaneously was losing its last remaining useful ally. The implication is clear: one indispensable key to restraining Germany and in general to preventing a major war was stabilizing Austria-Hungary's international status by doing something serious about the Austro-Hungarian problem.

The ready reply to this assertion, or rather, easy dismissal of it, is that Austria-Hungary's decline was irreversible and the result of its internal decay and that this could not be solved or arrested by international politics and diplomacy. As already indicated, I deny the premise, as do many scholars more expert on Austro-Hungarian internal affairs than I. Austria-Hungary's problems and weaknesses were real and would not go away or be cured, but they were not of themselves destroying it or even keeping it from being a working political entity, functioning far more soundly, for example, than Russia or its Balkan neighbors or Italy. It was the combination of internal and external pressures and their deliberate exploitation by other states within the cauldron of European international competition that was ruining Austria-Hungary internationally. It makes no sense to argue that European international politics could do nothing about this or about preventing a war that might arise out of it. It would amount to claiming that the balance of power principle, still invoked as the operating principle of European politics, had nothing to do with preserving the independence of states, especially essential actors, or contending that since international diplomacy could not arrest the internal decline and breakup of another essential actor, the Ottoman Empire, the European Concert had throughout the nineteenth century simply been forced to let the Eastern Question run its course. The Austro-Hungarian problem in its international dimensions and repercussions was precisely the sort of question with which the international system was supposed to deal, and could have dealt. There were many historical precedents for how Europe could manage the problems presented by declining and threatened vital units. They varied widely, of course, from brutal measures like planned partitions and more or less balanced compensations (Poland and the German Empire in the late eighteenth and early nineteenth centuries) through less brutal, more controlled management (the Ottoman Empire in the Balkans and North Africa) through measures of joint guarantee and protection (Belgium, Switzerland, Denmark). There is little or no precedent for what actually happened before and during 1914--the absence of any collective European response to the prospective downfall or disappearance of a central actor like Austria-Hungary, a contingency long and widely foreseen and predicted. That was a great, astonishing departure from tradition.

What could and should have been done, of course, is controversial, as is whether any feasible action would have been effective. But it is not hard to propose measures, at least plausible on their face, for a European intervention following the assassination of Franz Ferdinand to stop the incident from escalating into a dangerous confrontation and war. Something surely could have been attempted to satisfy Austria-Hungary's prestige and honor and to compel Serbia to conform at least outwardly to its international commitments to act as a good neighbor.(32) Since the particular steps Austria-Hungary demanded-a serious investigation of the ties between the assassination plot and Serbia's government, its nationalist organizations, and its military intelligence, followed by concrete measures to prevent future provocations, would not and probably could not be carried out by Serbia no matter what its government promised, and since the Russian attitude meant that they also could not be undertaken by Austria-Hungary without provoking an international crisis, the obvious corollary is that Europe acting in concert would have to ask Austria-Hungary to turn its cause and demands over to them, and then carry through seriously on an investigation and any required sanctions. There was ample precedent in the nineteenth century for international action to force smaller states, however innocent they might claim to be or however righteous their cause, to cease challenging great powers and causing international crises, just as there were for requiring great powers to act through the international community and not take the law into their own hands. Greece and other Balkan states, for example, were repeatedly compelled by joint great-power intervention to stop irredentist campaigns against the Ottoman Empire, and the Ottomans prevented from taking revenge on their rebels and enemies. Russia was more than once required in the nineteenth century to turn its cause and national honor in the Balkans over to the European Concert to defend. The fact that this procedure did not always work or was not always tried makes no difference. It was there, it could and did sometimes work, and in some instances like this one it was the only thing that could have worked (the only means, for example, that could have prevented the Crimean War, and almost did).

Yet in a way this discussion is irrelevant and distracting, for the obvious, overriding fact is that before and in 1914 nothing of this sort was tried or seriously considered. The danger of war steadily increased, the European powers were quite aware of the crucial specific source of that danger in the Austro-Hungarian problem, they knew about the kinds of measures used in the past and still available to meet them, and collectively they did nothing. This inaction is the most important development in prewar diplomacy and in the July Crisis (and a striking illustration both of how counterfactual reasoning can serve the purpose not of telling us what might have happened, but of illuminating what really did happen, and of why one must see 1914 not as the outbreak of war but as the breakdown of peace.) Every account of the July Crisis discusses the delay between July 5, when Austria-Hungary received Germany's support for its ultimatum to Serbia, and July 23 when the ultimatum was actually delivered. Some have speculated that the delay was fateful in allowing the initial shock of the assassination to wear off (which is doubtful-the Serbian and Russian reactions, the decisive ones, would have been the same earlier). Another delay, more fateful and inexplicable, is hardly mentioned or discussed in the vast literature. For a full month after the assassination, the powers did absolutely nothing in concert to prepare for or deal with the possible or likely consequences of this sensational incident. Everyone knew that Austria-Hungary and Serbia were mortal enemies, that they had gone to the brink of war at least four times in the past five years, three of them in the past year, and that Russia was Serbia's ally and protector and Austria-Hungary's main enemy. Yet when something occurred that anyone could see might set off this long-envisioned war, the Entente powers averted their eyes, went about their other business, waited for whatever Austria-Hungary and Germany might do, and hoped for the best. And of course Austria-Hungary and Germany took the very action that did set off the war.

This argument seems paradoxically to prove the precise opposite of what was promised and intended: that the war was avoidable. If the means for a serious attempt at avoiding it were known and available, then the root cause of the war was contingent, a collective failure to apply them.

But of course that collective inaction was neither inexplicable nor really contingent.
Behind the failure to act lay precisely those shared assumptions and convictions about the nature, stakes, and reigning course of the international contest earlier cited as the reasons why determinists consider the war objectively unavoidable, by virtue of the force des choses. I contend only that the pressure of events did not make war objectively so, by making peaceful choices impossible in the face of hard realities like security threats, alliance commitments, and arms races, but subjectively so, by fatally constricting what all the actors would entertain as a conceivable, rational course of action in the face of this crisis or any like it. If we look for the particular reasons why the various powers did not even consider taking any of the steps mentioned above to anticipate a crisis and manage it collectively, the answers are familiar and obvious. Austria-Hungary and Germany were determined to reverse the existing trend considered fatal to them, and saw in this crisis a good, possibly final, chance to do so. The Entente power equally saw in this crisis a danger to the existing trend and were equally determined not to allow it to be reversed. Russian policy, seen by Russians as a defense against German and Austro- Hungarian aggression, was resolutely determined to maintain and extend Russia's control over the Balkans. French policy was rigidly fixed on maintaining the existing alliances and therefore doing nothing to weaken the Franco-Russian one.(33) Britain's was fixed on maintaining its ententes, both in order to check Germany in Europe and avoid threats to the British Empire-the latter aim, the primary one, requiring maintaining the entente with Russia at all cost.(34) But behind these familiar positive reasons for failure to act collectively, there was a still more fundamental negative one. No one believed that a sane, rational policy allowed any longer for this kind of collective response. Anyone who tried to suspend the rules of power politics, of "every man and every alliance for himself, and the devil take the hindmost," was a fool and would earn the fool's reward. Hence to ask any British, French, Russian, Italian, or even German leader to sacrifice or subordinate particular interests and opportunities of theirs for the sake of some sort of collective action to stabilize the international position of Austria-Hungary so as to lessen the chances of a general war was to ask the impossible and absurd-to ask them to commit political suicide at home and to be laughed at and swindled abroad. Stabilizing Austria- Hungary's position was really not anyone's business except that of Austrians and Hungarians, or perhaps Germans if they wished to do so for their own power-political reasons. This profound practical indifference to the survival of a vital actor such as the Habsburg Monarchy was, to repeat, a break with tradition-not normal Realpolitik, but a different concrete definition of it, a different collective attitude toward international politics.(35) The power whose final break with the Concert principle proved decisive, Austria-Hungary, was also the last and most reluctant to abandon it, because it was the one most dependent it and on collective international support and restraint to survive. This is evident before 1914. One of Foreign Minister Aehrenthal's chief aims in 1908 had been to revive the old Three-Emperor's League and the moribund Austro-Russian entente in the Balkans by a deal with Russia over Bosnia and the Straits. Even in 1914 this idea was far from dead. The original Austrian proposal for reversing the current disastrous trends in the Balkans called for political rather than military action and was changed only in the wake of the assassination (though how much difference this would have made is debatable.)(36) During the July Crisis itself Austro-Hungarian leaders hoped against hope that Russia might let it get away with a local war against Serbia, and if Russia did, they intended to use the opportunity to seek a fundamental rapprochement with Russia through negotiations for a joint solution to both the Balkan and the Ukrainian problems.(37) There is a tragic appropriateness about Austria-Hungary's breaking at last with the Concert principle and thereby destroying itself and Europe with it, like the blinded Samson pulling down the pillars of the temple, just as there is about Tsarist Russia's acting upon the shibboleths of its honor and its alleged historic mission of protecting the Balkan Slavs rather than its true state interests, thereby signing its own death warrant.(38)

To argue for the inevitability of World War I on this ground is, to repeat, not to blame Britain, Russia, and France for it while exonerating Germany and Austria-Hungary, or to characterize the former as more blind and reckless than the latter. It is an attempt to root the disaster deep in a political culture which all shared, which all had helped to develop, and upon which all acted in 1914, Germany and Austria-Hungary precipitating the final descent into the maelstrom. It is to see the origins of the war as finally a tragedy more than a crime, inevitable by reason of wrong beliefs, hubris, and folly too broadly and deeply anchored in the reigning political culture to be recognized, much less examined and changed. The tragedy of its origins thus connects with the tragedy of the war itself in its hyperbolic protraction and destruction, evoking, like Shakespeare's Romeo and Juliet, the verdict, "All are punished."


Afterthoughts on this Essay's Implications for Counterfactual Reasoning

The editors have asked contributors for brief comments and reactions on the three guidelines originally posed for this counterfactual project now that it has been completed. The first and second guidelines (the need for explicitness in choosing key points, establishing criteria for doing so, and explicitly examining the underlying assumptions behind inferences, and the need to make a strong case based on historical evidence for any counterfactual scenario and to recognize how quickly they become improbable in embellishment, I heartily endorse and have tried to put into practice. The third, somewhat vaguer, calls for identifying what if anything contributors have learned in working on the project, and especially what may have surprised them.

I must say that though I have learned much from the other contributions, I have not learned much new from my work on it or really been surprised by anything. I have, however, had previously held convictions on certain points deepened and sharpened. The first is a feeling that major historical events and developments have deep roots and that their component elements and causes are densely interconnected. I may here be at odds with many contributors and the editors, but it now seems to me even harder and riskier than I had thought before to make a single small cut in Clio's garment and insert a new patch without tearing the whole thing apart. Put another way, I have become more convinced that if one investigates counterfactual scenarios with the same rigor, skepticism, and insistence on reliable evidence as one demands for testing normal factual narratives, one usually discovers that it requires increasingly large and arbitrary alterations in the factual scenario to make even apparently probable alternative scenarios work.

Another point that struck me (this was something of a surprise) is that a common impression about counterfactuals is unsound, namely, that one can choose a particular spot to insert a counterfactual element into history and then trace the changes it might have made in developments subsequent to that point, without altering or affecting what had gone before. In other words, in counterfactual reasoning the path of history antecedent to the counterfactual stays the same; only the consequent future path is altered. This notion now seems to me untrue even for reasonable, plausible counterfactuals. To make them work, one has to change their antecedents as well as their consequents. Examples: Can one envision a plausible scenario in which Napoleon sometime in his career-say, in 1805-decides to stop his course of conquest and settles down to establish a durable system of French hegemony in Europe? Certainly one can imagine this; many of his associates tried hard to persuade him to do so, and I have argued elsewhere that objectively a durable French hegemony was entirely feasible. The only trouble is that for this counterfactual to work, Napoleon would have had to cease being Napoleon, and if he had been a person capable of thus transforming himself, it is impossible to see how he could never have reached a point in 1805 or another time at which a stable French hegemony in Europe became possible. Napoleon, in other words, comes to us historically in one piece. To change what he was capable of becoming and doing after 1805 is to change what he was and was capable of doing before then. The same point, that counterfactual alterations change the past as well as the future, can easily be illustrated by other examples. Could Nazi Germany have defeated the Soviet Union in 1941, destroying the regime and replacing it with German satellite regimes? Quite possibly, by exploiting the nationalities' and peasants' discontent, posing as liberators, giving the peasants back their land, etc. But in order to do that, Hitler and the Nazis would have had to cease being Hitler and the Nazis, and abandon not only the goals for which they had invaded the Soviet Union, but those for which they had seized power. The whole Nazi past as well as the future is changed.

This leads to another conviction deepened by this experiment: the clear necessity of explaining historical developments by causes, and the need for great care, sensitivity, and scrupulousness in deciding what acts as a cause in human affairs, and what suffices as evidence of it. At the same time, it has reminded me again of the difference between the scientific and the ordinary human (and thus historical) notion of cause-the latter being vastly wider, looser, more various, and harder to define and control than the former. Thus while this exercise has not, I believe, made me any more a determinist, it has made me even more an opponent of post- modernism or anything else that reduces history to pure kaleidoscopic contingency and chance- the "chaostory" that Niall Ferguson and others talk about-and more sure than ever of the proposition (is it Aristotle's?) that revolutions (meaning great developments) may have trivial occasions, but have profound causes.

The last observation is that the effort to justify counterfactual reasoning in history tends to give hindsight a bad rap. Without trying to judge the psychological experiments that allegedly demonstrate a "hindsight bias" that distorts historical interpretation in the direction of determinism, I would say only that if the lesson drawn from these experiments is that hindsight is bad and should not be used or must be guarded against, or that it induces biases only in the direction of determinism at the expense of contingency, then I disagree. Not only is it impossible to do history without using hindsight; historians should use it. Hindsight and evidence are the twin, inseparable and indispensable resources available to us not only for telling the story, but telling why what happened did happen rather than something else. Nor is it true, I think, that hindsight necessarily biases us in the direction of determinism, a belief that things had to turn out as they did and in no other way. Hindsight can work equally well to undermine determinist bias and make us recognize contingency-see both that wholly unexpected things can happen, and that things we thought had happened routinely and by necessity actually could have gone the other way. Everyone expected Hitler's legions quickly to conquer the Soviet Union. Hindsight taught us otherwise, and helps tell us why. Ordinary life and history books alike are full of examples of this.

My sense is that these somewhat random observations fit and support the call for
examining embedded counterfactuals contained in this essay, but I will let others judge whether
this is true.

Return to WWI Resource Centre Index