|
|
|
|
Aggression or Humanitarian Intervention: International Rules and the Domestic Politics of Threat Perception
University of California
Los Angeles
Department of Political Science
International Studies Association
March 1998
Rules and the Domestic Politics of Threat Perception 1
Greg J. Rasmussen 2
The scenario recurs frequently in world politics. One country militarily attacks another and the two sides appeal to competing norms. The victim and its allies call the attack an aggressive violation of sovereignty and territorial integrity while the attackers and their allies call it a defense of some broader international norm or value, such as the human rights of the attacked country's citizens, the credibility of international legal institutions or prohibitions on weapons of mass destruction, or the re-establishment of the territorial status quo. One side's "aggression" is, as an example, the other side's "humanitarian intervention."
A key feature of the Westphalian legacy of interstate anarchy is not decentralized military capabilities but the difficulty of agreeing upon standards of acceptable conduct. These debates on normative issues often lead many to despair that they are "sound and fury, signifying nothing." Perhaps the apparent anarchy of moral meaning condemns us to a world where only might makes right. I take a different approach.
Careful attention to international norms and rules can yield significant benefits for a state's grand strategy interests. General compliance with the norm of multilateral consultation prior to using military force to settle a dispute, while involving serious short-term sacrifices, provides long-term benefits with respect to peaceful management of future crises. My research suggests that we can explain the outcome of a certain category of international crises only by considering the effects of these dynamics on the domestic politics of threat perception. My focus falls within what Lebow (1981) calls "spin-off" crises, particularly virulent crises that emerge between two parties because one of them has become caught up in crisis or war with an actor important to the other. Even when great powers are caught up in crises arising out of one's attack on the other's ally, international norms and rules can play a decisive role.
The implications extend in two directions. First, it has some immediate and obvious ramifications regarding how the U.S. government might avoid diplomatic isolation as it pursues its militaristic confrontations with Iraq. Conceivable but admittedly improbable future regional disputes might generate great power predicaments with similar dynamics, whether in East or Central Asia, the Middle East or the Balkans.
Second, and more importantly, this research potentially reshapes our thinking about the Westphalian legacy. International norms and rules influence state's security strategies, even in conditions of high stakes, external pressures for unitary state action, easy fungibility of asymmetrical cooperative gains into military power, etc. When it comes to high-stakes security interaction, most current schools of thought - from neorealists to neoliberal institutionalists - portray institutions as irrelevant. Successful reform of the sovereign state system requires knowing when states expect international norms and institutions to help in protecting their security. A rival great power's attack on an important ally severely tests this trust.
The paper moves towards an explanation of crisis outcomes, first examining the geopolitics of the state system, then threat perception, then domestic politics of coalition-building, then the political use of international norms and rules by state elites.
Focus: Resolvable Spin-Off Crises Between Great Powers
This project seeks to explain the outcomes of crisis, not in terms of which side gains the most but in terms of whether peace or war occurs. Rather than attempt to explain all crises, I focus on a particular type, one with important theoretical and policy ramifications. This section describes and illustrates the subset of crises.
This project focuses on what I call great power protection crises : a great power crisis arising when the ally of a great power suffers attack from another great power, when neither great power yet knows that the dispute can be settled peacefully, and when neither great power seeks crisis maintenance for its own sake. Fig. #1 summarizes these stipulations. Some require special comment.
Great Power Crises
There is a readily available, widely accepted definition of the major powers. Singer and Small (1980:47-50) show nine nations have been major powers at some point since 1816. 3
This study defines interstate crisis as follows:
A situation involving a pair of states that (1) arises out of recent unexpected events associated with a conflict or disagreement between the states, (2) involves a restricted "amount of time available for response" before the situation becomes transformed, 4 (3) threatens state elites' definition of their state's high-priority goals, and (4) involves "a fairly distinct possibility" of war with the other state. 5
This definition follows much of the scholarly literature on interstate crises. 6
Resolvable Spin-Off Crises
Here I borrow a term from Lebow (1981), reserving it to mean a crisis between one pair of countries which emerges out of a crisis between a different pair.
By geopolitical reconcilability, I refer to the potential for peaceful settlement of the dispute, independently of domestic and perceptual factors. A dispute can be reconciled if there exists a feasible settlement which both sides prefer to war. On this point, I follow the traditions of classic realism and neorealism, to define the actors as nation-states and their interests as maximizing the long-term probability of national survival.
Recent formal theoretic neorealism provides a good conceptual understanding of what constitutes an irreconcilable dispute. 7 Essentially, five factors work in additive fashion to expand the set of feasible peaceful settlements: the expected net costs of war, 8 the net benefits of potential settlement(s) independently of the mode of their achievement, aversion towards risky options, the technical feasibility of the potential settlements, and the military enforceability of settlements. For example, if one party gets greater geopolitical benefit through war regardless of what concessions can be accrued through diplomacy, the dispute lacks geopolitical reconcilability. For example, if each of the conceivable settlements results in military force structures (distribution, placement, readiness, etc.) such that neither party can credibly commit to an agreement once reached, the dispute lacks what I call geopolitical reconcilability. 9 As long as the set of settlements mutually preferable to war remains non-empty, the dispute has geopolitical reconcilability.
I also stipulate that the disputants do not know whether the dispute has reconcilability. Perhaps they do not know a party's war costs, the value one of the parties places on the potential peaceful settlements, the risk-proneness of one of the parties, the technical feasibility or military enforceability of settlements, or perhaps even of the existence of certain settlements. The incompleteness of their information about these factors adds up to them not knowing whether the dispute can be reconciled, if viewed in geopolitical terms. Also, here I go beyond geopolitical reconcilability. Their ignorance may stem from domestic or perceptual or other factors. For example, they might not know whether the rival governing elite understands or values its nation's geopolitical benefits. This condition - unknown resolvability - could arise in a number of ways and still fall within the scope of this project.
I also stipulate that both country's state elites prefer the crisis to end sooner rather than later. I do not specify what reasons they might have, except that it be independent of the risk or level of war.
This set of crises focuses research towards a particular path to war, leaving other crises - such as those with extreme intrinsic war gains or extreme offense-dominance - for future investigation. A number of hypothetical future scenarios fit this scheme. Russia develops strong alliances and relevant military capabilities with respect to Iraq, Afghanistan, or the Balkans and then the U.S., France, or Britain attacks one of these areas. Either the US or China drops or launches bombs into N. Korea or Taiwan, whether on grounds of sending signals to potential weapons proliferators or on grounds of sending a signal to domestic secessionist movements. France and Russia intervene in a war between Israel and Syria, in a fashion that threatens U.S. interests. Disorder in central Asia generates incentives for both China and Russia to send in military occupational forces.
This type of crisis - resolvable great power spin-off crises with incomplete information - has the advantage of focusing our attention on the issue of cooperation under high stakes. Lebow (1981) extensively studied crisis and developed a partial typology to describe the dynamics out of which the crises arose. He says that spin-off crises differ from justification of hostility crises in that "neither side really desires a confrontation let alone the war which might result from it." Nonetheless, because the initiator is at war or expects to be at war with one state, it "feels compelled, usually by perceived dictates of national security," to adopt policies which lead to confrontation with a third party. Hence, the label "spin-off." 10 As Lebow says, spin-off crises combine a keen desire to avoid war with a very real clash of important interests. 11 Crises falling within his sample and my theory include Germany-Britain, July 1914 and perhaps four German-American U-boat crises, 1915-17. Of course, we could add Britain-France, 1830, Austria-Russia, 1908-14, Germany-Russia, 1908-1914, Germany-Britain, 1939, US-China, 1950, and many more.
An Historical Puzzle
From these cases, a curious puzzle arises. Why did Britain and Russia avoid war in 1878 in a crisis over the same issues that led to war in 1853? In important ways, the Holy Places dispute of the 1850s and the Balkan uprisings of the 1870s resemble conflicts and potential crises in the contemporary era. They were crises between states who desired a peaceful settlement, arising from a regional dispute arising out of ethnic, religious and nationalist conflict, in a multipolar period without rigid alliances or intense animosity. We would expect a good theory of crisis outcomes to resolve this puzzle.
In many respects, both disputes followed a similar trajectory. 12 Christian subjects within the unstable Ottoman Empire demanded greater rights. Complex, arduous negotiations failed to resolve the intra-Ottoman dispute and failed to prevent the dispute from escalating into military confrontations between great powers. Domestic relations within the Ottoman Empire became the terrain for great power competition. In both cases, Russia violently occupied large tracts of Ottoman territory to coerce the Porte to succumb to Russian demands, threatening to turn an important British ally into a Russian dependency.
In both cases, Russia attacked despite explicit prior disapproval from the other great powers. Turkey requested other great powers, especially Britain, to intervene against Russia. Russia took several steps indicating an intent to exclude Britain from Turkish affairs. Turkey went to war against Russia, despite long odds and without firm assurances of allied help. The subsequent Russo-Turkish wars cost the two countries dearly. In both cases, the other great powers tried to settle the Russo-Turkish dispute in a way that secured Turkish independence, secured their own separate interests, and avoided permanent disruption of the equilibrium of forces and interests. In neither case did any great power seek war between great powers but in both cases several great powers considered the issues potentially worth war. The great powers used threats of military force to curb one another's ambitions, causing several crises and leading in both cases to concerns about a possible general European conflagration.
Many background conditions were similar: a multipolar distribution of power (with a single dominant power), defense-dominance regarding homeland attacks, offense-dominance regarding intervention in the immediate regional locale, and little significant uncertainty about relevant military capabilities.
Both disputants had capabilities for general war against one another and for intervening decisively in the regional dispute. In both cases, a reforming parliamentary aristocracy confronted a reforming autocratic aristocracy. In both cases, domestic opposition groups within the defending state clamored for military confrontation.
One dispute ended in great power war. One did not. The Crimean war of 1854-6 imposed significant costs on both countries and ran significant risks of general European conflagration. An unexpected development helped avoid opening of a Baltic front, intervention by Austria and Prussia, and lifting the reactionary lid off the social turmoil that had sparked Europe-wide revolution just five years before. While the western armadas were sailing eastward, - after the war declarations and before combat - Russia unexpectedly withdrew from the Principalities, ending the occupation that had triggered the entire great power crisis.
The dissimilarities between cases do not seem able to explain the outcome. Relations were more amicable prior to the crisis that ended in great power war, than in the other dispute. In the latter case, the defending great power depended more heavily on its allies - both the attacked ally and third party great powers. In the case without great power war, the challenger's military actions posed a much starker threat to the defender's geopolitical interests. So did the long-term geopolitical implications of the challenger's actions: potential exclusion from the direct partitioning of all of Ottoman Europe. However, after the denouement of the latter case, the defending state achieved a more favorable settlement - geopolitically and diplomatically - than they achieved in the former case.
What we know about international norms and rules does not seem to help much, either. The Crimean War of 1854-6 has been treated by political scientists and some historians as the end of the Concert of Europe, or a case of failed crisis management under the Concert. 13 While this depends on one's definition, clearly the norms and practices of a great power concert lingered - and failed to prevent war. The invocation of international norms and rules appears remarkably inconsistent. In 1853, when Russian troops stopped at the Danube, many called it an aggressive violation of European public law and ignored Russia's attempted defense of the human rights of a religious communities within the Ottoman empire. In 1878, when Russian troops stopped only at the gates of Constantinople, many called it a liberation of Balkan peoples and overlooked the aggressive violation of Turkey's territorial integrity. Ambiguity of norms and their self-interested application poses a severe challenge for theories that assign causal weight to international norms and rules, to security regimes or to great power concerts.
We can presume that one difference played a critical causal role. People in the second dispute had historical memories of the first. This certainly played some role in their successful management of the second crisis. We can not simply treat the two disputes as simultaneous runs of two roughly similar experiments in two roughly similar universes. Human behavior flows in part from human beliefs, which are shaped partly by historical experience. The cases must be treated historically. It would also help, at some point, to treat the cases interpretively. If people learned something which enabled them to settle the second dispute peacefully, an historical and interpretive approach becomes all the more urgent. What idea, what principle or proposition, did they learn? How did they discover, debate, convey and apply that idea rather than its competitors? For example, they may have concluded that the Crimean disaster was due to idiosyncratic policy choices or personal incompetence, or perhaps an unavoidable result of the insecurity of international anarchy, or perhaps from excessive popular participation in policy making. Which lessons did they learn and why? Comparing and interpreting these roughly similar cases will help us to share in their ideas, to learn from their learning.
Geopolitics Of The State System
Deadlock
As explained earlier, Deadlock, when it arises, means that there exists no conceivable settlement to the dispute which benefit both sides (in standard realist geopolitical terms) more than war. "In Deadlock," say Snyder and Diesing (1977:174), "the conflict of interest is so deep that no settlement is possible; any conceivable settlement would be unacceptable to at least one of the parties, i.e., it is considered worse than war, so war is the inevitable outcome." Snyder and Diesing have made this a popular theory among students of crisis diplomacy. 14 Crisis situations allegedly have a fixed underlying structure at the outset. During the preference revelation phase of a crisis, this structure emerges. If the initial structure is Deadlock, the outcome will be war.
In 1853, given traditional realist geopolitical analysis of the information available to state elites, several available settlements would probably have been expected to advance British and Russian geopolitical interests more than Russo-British war:
I believe it would be very difficult to argue that none of these would have benefitted Russia and Britain, in geopolitical terms, more than Russo-British war. A host of complicated issues swirl around each of these proposals and their consequences. I limit myself to five general conclusions which seem central to these debates. First, historians today and participants and observers then have generally claimed that the dispute was reconcilable. Second, Russia probably lacked the material basis for making an imperialist challenge to important British interests in central Asia and the Mediterranean. 15 Third, the process of the Vienna conference from June through October gave Britain reason to expect that she could get the four powers to cooperate in confronting Russia if Russia abused even an extensive religious protectorate to exercise undue influence over the Ottoman Empire or if, in war with Turkey, Russia threatened to seize or control Constantinople and the Straits. Fourth, any problems in securing great power cooperation in peacetime would also exist in securing their cooperation in intervening or staying neutral during a Russo-British war. The costs of war (escalation, lost relative power, etc.) might therefore be quite high. Fifth, as to the security of Britain's special alliance with Turkey, the unique net benefits of supporting Turkey in a war to oust Russia from the Danubian principalities were not so significant.
Other arguments ought to be raised - such as how action through great power concert could help protect Britain's alliance with Turkey and help deter Russia from encroachments - but they stand outside the scope of traditional realist geopolitical analysis. My only point is that a realist theory of deadlock can not explain the different outcomes in these two cases.
Shifting Distribution of Power
A number of theorists expect increased probability of war when a country - especially the leading power - faces declining relative power. This underlies the explanation of war in Gilpin (1980), Organski and Kugler (1980), Doran and Parsons (1980). However, the cases do not fit this theory's predictions.
Britain's relative power had been rising in the decades prior to the 1853 crisis but declining in the decades before the latter crisis. This shows up in the quantitative work of Spiezio (1990). A variety of other factors changed the relative power and interests of Russia and Britain by the second case, such as the declining importance of trade with Turkey and much less French interest in eastern affairs. Non-geopolitical factors also reduced Britain's incentive to support Turkey in a war against Russia. Turkey had lost the high moral ground because of the Bosnian uprisings, the Bulgarian atrocities, the more advanced state of Balkan nationalism and administrative ability, and the persistent failure to implement agreements which the Great Powers had helped mediate between the rebels and the Ottoman government. Russia had taken the higher moral ground because of the abolition of serfdom and extensive and patient multilateral consultation with the Great Powers before engaging in attack.
The anomaly persists if we look at a shorter time period - say, months. Russia's rapid approach on Constantinople in December 1877 and January 1878 were much more substantial losses for Britain than unofficial interpretation of the Vienna Note in September 1853.
Britain's greater relative decline in the second case did not cause war. States adapted. If shifting power leads one state to raise its ambitions, we would expect it to lead its rival to lower its ambitions. The terms of the dispute shift but not the mode of dispute: peace or war. Indeed precisely this happened. As Russia advanced in 1877, Britain tolerated far more than they did in 1853. In 1853 the terms of dispute dealt with whether and when Russia would withdraw from the Principalities. In 1877 the terms of dispute dealt with whether and when Russia would withdraw from the vicinity of Constantinople, end its occupation of Bulgaria, and end its occupation of key fortresses in Asia. In 1853 the terms of the dispute dealt with Russia's legal right to protect Ottoman Christians while in 1877 they dealt with the autonomy and independence of many Ottoman Christians, the existence of Ottoman Europe, and the independence of Turkey. This is as theorists of power and anarchy would expect. For example, Blainey (1988) says states adjust their ends to their means. This shifting of terms is what causes capabilities and interests by themselves to have little or no impact on whether the dispute is settled peacefully.
We could expect the war-threatening mode of a dispute to persist even while the terms of dispute shift. For example, in 1878, crisis erupted once Russia reached the limits of British tolerance by marching to the undefended gates of Constantinople and threatening to exclude Britain from the negotiation of peace terms. While the terms shifted, they did not shift sufficiently to avert crises. So, power transitions might help explain where the two sides drew the line, took a stand, made their offers, etc. They do not help explain why the two sides were or were not able to reach an agreement.
In both theory and practice, some factors may impede this smooth adjustment process. One example is risk-acceptance in the domain of losses, coupled with a tendency to update one's reference point more quickly after making gains. Stein (1990:87-95), Levy (1994b) and others have posited this dynamic, using the prospect theory of Tversky and Kahneman (1981). I reserve that avenue of investigation for another paper. Another factor stems from uncertainty or information costs associated with power transitions. I deal with this in a later section.
Entrapment
The next systemic theory illustrates aspects of deadlock path to war. Some argue that escalation to war between unitary rational nation-states can occur through entrapment, being pulled into war for third parties whose interests they do not share (Waltz, 1979; G. Snyder, 1984, 1991; Christensen and J. Snyder, 1990). The fear of future abandonment - being left to one's own devices by an ally, during a crisis, confrontation or war - can force a country to defend the interests of an ally, even in a crisis during which they do not share the ally's interests. To illustrate the mutual entrapment theory, imagine a crisis between a currently neutral Patron and a Rival who is at war with the patron's Client. If the Patron took military measures to persuade the Rival to cease the war, the Client would recognize this effective increase in its bargaining leverage with the Rival. Hence it would raise its demands, thereby raising the risk of prolonged war. The Patron would be forced to back up these demands, for fear of losing an ally. This invokes the familiar chain-gang dynamic. 16 When the Patron depends highly on the Client's military powers, the Patron must follow the Client into war for unshared interests. There is little doubt that this dynamic plays some important role in the causation of some wars. Lebow (1981) notes that spin-off crises - those in which one of the rivals is at war with an ally of the other - are the least likely to be resolved peacefully. Waltz (1979) uses it to explain World War I. 17 One implication is that the frequency and extent of wars from deadlock are independent of whether the great power system is bipolar or multipolar, weakening a central claim of Waltz (1979), borrowed by Mearsheimer (1990), G. Snyder (1984, 1991), Layne (1993), and Christensen and J. Snyder (1990), etc.
However, Fearon's neorealist theory of reconcilability of interests requires a closer look for the actual causal link. If there is full knowledge in this situation, then the Rival will recognize the dynamic between Patron and Client. The threat of the Patron being pulled into the war raises the cost to the Rival of continued war with the Client. Realizing this, the Rival has every incentive to adjust its wartime demands upon the Client, lowering them sufficiently to overcome whatever new leverage or incentive the Client acquired from the Patron's influence in the dispute. At some point in this adjustment process, the Patron could remain neutral without destroying its alliance with the Client. The Client would recognize the gains it had made through the shift in the terms of dispute, and credit these gains to its Patron's threatened intervention.
So, the fear of abandonment does not, by itself, explain war resulting from a spin-off crisis. Even in these most intense of crises, there exist mutually beneficial settlements, even as the original war rages. While this may seem to weaken Waltz' argument for bipolarity, his argument also includes uncertainty.
Now let's assess the theory empirically. Mutual entrapment is not a good interpretation of French and British interaction in 1853. Prior to August, France favored confrontation more than Britain but France was unable to pull Britain into strong anti-Russian action. After August, France actually tried to restrain Britain, trying to pull Britain into coordination with Austria and Prussia. So, when the critical decisions were made, it was France restraining Britain. 18 Had fear of being abandoned by France led Britain to move closer to France's policy preferences, this would have reduced the risk of war.
Aggregate Overconfidence
A large number of authors believe that inaccurate predictions of war outcomes (which includes escalation) play an important causative role in war causation, from Hawtrey (1952) to Waltz (1979), from Quester (1982) to Doran and Parsons (1980). 19 Levy (1983:89) offers one argument along these lines:
"...accurate perceptions of resolve would lead any set of rational actors to reach a negotiated settlement. In terms of bargaining theory, each would recognize that the credibility of the other's firmness was higher than its own critical risk, thus leading to compromise."
These arguments share the premise of reconcilability of interests. If actors fully and objectively understood and followed their situation and their interests, war would not occur. Most specific work on the subject traditionally presumes that, under most conditions, effective deterrent signalling strategies can generate this full information, leading to peace.
War occurs when the summation of two parties' estimates of the probability of war success is too high. 20 For simplicity, let us call this mutual or aggregate overconfidence. Now States have an incentive to avoid this but, as Fearon persuasively shows, private information and the incentive to conceal it can prevent them from taking steps to correct the problem of mutual overconfidence. By this theory, war results when one side unknowingly becomes committed to a position which its rival expects to be worse than war and when its rival can not profitably send the deterrent signals necessary to prevent or force a correction in the error. Though the rival seeks to avoid war through signalling, they may lack the military capability, have a military incentive for secrecy, or be unable to send consistent signals because of the quirks of domestic politics.
So war could occur with signal scarcity, due to the absence of capabilities to send clear costly and nonprovocative signals of resolve. Levy (1983) offers two direct linkages to war from misperception: underestimating a rival's military capabilities 21 and resolve 22 and overestimating a rival's intentions to go to war. 23 So signal scarcity can lead to either inadvertent provocation or deterrence failure.
Signal scarcity can cause war through deterrence failure: failure to convince one's rival that their offer would yield one less utility than war. A State may go to war on the mistaken belief that it will win because its rival will capitulate during the war. This mistaken belief arises because of incomplete information about the rival's walkaway point and the unavailability of signals that distinguish between States who bluff and those who do not.
Unless bargaining costs are quite high, however, informational changes fail to explain outcomes, since clearer signals of relative resolve and capabilities will lead the parties to revise their offers. The new information creates a shift in the terms of dispute and the bargaining range, while the mode of dispute remains indeterminate. With bargaining costs, the situation resembles a take-it-or-leave-it model which has a determinate outcome under conditions of full information. The last offer will be placed just below the total of two quantities: the other's net gains from its minimally acceptable share and the other's net gains from one more round of bargaining.
In these two crises, there were only three decisive false predictions. Russia in Spring 1853 did not predict - and no one could have - that Britain in late Winter would choose war rather than accept Russia's demands on Turkey, though much diluted by winter. Britain in general, after October 1853, actually overstated Russian resolve, understating Russian desire to end the crisis peacefully and overstated Austrian and Prussian willingness to support Russia in the dispute. Better deterrent signals could not have avoided these false British predictions. Stronger Russian signals of resolve and capability would have hastened Britain's rush to war, as the Sinope incident demonstrates.
The right mix of reassurance and deterrence has eluded theorists. In summarizing their findings on bargaining strategies and tactics, Snyder and Diesing (1977) suggest a particular temporal allocation of coercive and accommodative strategies. Peacefully resolved crises follow a definite pattern. The parties "adopt primarily coercive strategies (offensive or defensive)," clarifying comparative resolve and the potential for war and enabling the parties to know how much to reduce their initial claims. "Premature accommodative attempts," by contrast, hold the danger of "leading the adversary to underestimate the risks of firmness" or delaying resolution of the dispute. 24 If an "initial period of mutual firmness" clarifies the structure of the crisis, 25 the outcome then depends on whether the dispute has irreconcilable conflicts of interest. They conclude that 14 of their 16 cases ended in peace, for two simple reasons. First, in these cases "the realization that the adversary was committed to war, or might be if its claims were not met, came soon enough to permit a settlement before non-bargaining elements took the crisis out of control." Second, once relative interests and resolve were clarified, the conflict of interest was not irreconcilable. 26
Since bargaining strategies signal relative resolve, Snyder and Diesing suggest that the best way to overcome the mutual overconfidence that leads crises to escalate to war is early firmness and later accommodation. Their discussion of world war I illustrates. In this case, the non-bargaining element which took the crisis out of control was military planning and doctrine. Given this, "the causal finger points to misperception or, more precisely, the tardy correction of misperception." 27 In large part this was due to Britain's failure to make their intentions clear. 28 Perhaps Britain's equivocation was due to a split in the cabinet, as they argue, 29 but let us focus on Britain's policy and its consequences: "Premature accommodative attempts, like England's attempted mediation in 1914" apparently led "the adversary to underestimate the risks of firmness." 30 Britain in 1914, by proposing mediation of the Austro-Serbian dispute through conference diplomacy, signalled their weakness and lack of resolve on the issue. By making this move early in the crisis, Snyder and Diesing argue, Britain led Germany into miscalculation.
This "coercion first" theory garners logical support from Schelling's precommitment strategy: early on, commit oneself strongly and publicly to a position, signalling (and even enhancing) one's costs of backing down. Ideally, the signals involve costs to the sender which it would not be willing to pay unless it valued its preferred outcomes in the dispute very strongly, enough to fight for them. Ideally, the signals should come early in the dispute, to prevent the possibility that rival elites form a false image which distorts their perception of signals sent later in the crisis.
Did the actors follow this strategy of early costly signals more closely in 1877-8 than in 1853? Russia in 1853 signalled its seriousness in the Menshikov affair by backing his demands with the threat of military force and then invading the Danubian principalities. This exceedingly hostile act involved some military costs, publicly committed the Tsar to Menshikov's position, and came very early in the crisis, before France and Britain had fully formulated their positions, long before non-bargaining elements threatened to take the crisis out of control.
Though a strong early signal, the strategy failed. Britain and France recognized the commitment strategy, noting that Russia had placed its prestige on the line. On balance, though, the effect was counterproductive. It triggered greater French and British hostility. It even led to a counter-balancing coalition - as Austria and Prussia accepted France's call for a European initiative, the Vienna conference. Some in Britain took advantage of the Russian action to argue for delivering a strong blow to Russian prestige. While the story tempts one to conclude that Russia blundered, 31 Britain and France had not yet decided whether the wider Menshikov demands warranted joining a possible Turkish war against Russia.
In the latter case, in contrast, Russia sent conciliatory signals long before it sent what Snyder and Diesing call signals of resolve. From the start of the Bosnian uprising in late summer 1875, Russia insisted on working for great power consensus through multilateral European fora. It did not engage in military action until Turkey had defied Europe's collective requests. Russia did the opposite of signalling theory's advice. Rather than precommitment and early confrontation first, they tried reassurance first. Something about their strategy apparently worked. They isolated Turkey and avoided European intervention.
British policy in the earlier case seems more mixed. They signalled firmness early - sending Stratford to Constantinople to deal with Menshikov, letting Stratford interfere with Menshikov's pressure to get Turkish agreement to a protectorate, and sending the fleet to Besika Bay when Russia threatened occupying the Principalities. After the occupation, though, they accepted four-power mediation of the Russo-Turkish dispute - conference diplomacy. 32 Their policy took more hard-line steps as the crisis progressed. Their confrontation in the early stages was much less than Russia's and a little less than France's. In later stages, it was more hard-line than either France or Russia. Russia's later conciliatory moves - accepting the Vienna Note in late August and then in October virtually repudiating the Nesselrode interpretation of that note - failed to produce corresponding concessions by Britain.
In the 1878 crisis, Britain did the opposite of Snyder and Diesing's recommendations. It followed fairly conciliatory early policies and then more confrontational ones later. In early stages, they did not send the fleet or occupy Gallipoli, as many proposed. In later stages, it sent the fleet; then recalled it. Later, it did the same: sending and recalling. The third time, though, they stayed with the fleet move. This does not constitute early, strong, consistent signals of resolve. 33
So in these cases, with three of four actors, we have the opposite of what a signal-based theory of crisis management would predict. Oddly, success in altering a rival's policies seems to be more correlated with early conciliation followed by vacillation, flaccidity and some late strong signal-sending than strong early precommitment with strong consistent signals of resolve, followed by later conciliation. In one of the four cases, we see an ironic consequence of signalling. As the target of a signal responds favorably, the sender learns it can feasibly demand more, leading to a shift in the terms but not the mode of dispute. This expresses a logical extension of signalling theory's underlying premises. In sum, deterrent signalling theory fails to explain the difference in outcomes.
Provocation
Signal scarcity can cause war through provocation: convincing a rival that further negotiation is futile since the bargaining range has closed. Some States are so concerned about their need to signal resolve that they take signaling too far. Because of the unavailability of credible signals of reassurance (or, in other words, because of the uncontrolled and unlimited character of signals of resolve), they inadvertently convince their rival that further diplomacy is futile and that attacking now is preferable to waiting to be attacked. Provocation could occur even if we assume that States understand the notion of neorealist reconcilability of interests; States may not know whether their rival has negative war costs. In game theoretic terms, this end process would be as if one actor created a situation where the other player perceives itself as no longer in a game situation. The provoked player concludes that its rival has already chosen war. Hence the provoked player sees its payoff as not depending on a rival strategy but on its own strategy. It believes it can no longer influence whether the outcome will be war or peace; it can only influence which side will be the first to attack, the first to finish preparations for attack or the first to find an opportunity to attack. If one side concludes the other has chosen war, it finds preparation for war necessary whether or not bargaining will continue or is costly.
At first blush, this seems to fit the cases. Britain in general, after October 1853, actually overstated Russian resolve, understating Russian desire to end the crisis peacefully and overstated Austrian and Prussian willingness to support Russia in the dispute.
Russia's reassurance signals in 1853 weren't overwhelming. Russia did send some signals of reassurance: acceptance of the Vienna Note, willingness to essentially repudiate the Nesselrode interpretation, and frequent assertions of good faith and willingness to withdraw from the principalities. While Russia in 1853 did not offer much in the way of reassurance, it was more than Russian did in Dec. 1877 and Jan.-Feb. 1878. They refused meaningful British mediation, refused to halt their armies' advance until after Turkey accepted the preliminary bases of peace, kept these preliminary peace terms secret from Britain, advanced upon Constantinople after reporting an armistice had been signed, forced Turkey to abandon the defensive fortifications of Constantinople, threatened to take Constantinople, and was unwilling to withdraw its armies from the vicinity of Constantinople until Britain withdrew its fleets a comparable distance. Provocation very nearly caused Russo-British war in Feb. 1878.
They did not take costly actions of military character to distinguish their policy from that of states with unappeasable ambitions. This might have included limiting their own offensive capabilities (i.e. by giving the Turk army control over Danubian bridges), partial unilateral force withdrawal, etc.
However, in the end, Russia did not. The case with the least reassurance and the most provocative and threatening military moves ended in peace.
Misperception
These system-centered theories do not explain why a country would appear to falsely conclude that the dispute had become deadlocked - as Britain did in late 1853 but not in early 1878. For this, we need theories of misperception.
The "objective" external situation does not absolutely constrain the choice of strategy. According to Snyder and Diesing, (1977) with all foreign policy decision makers, the "first few messages or incidents in a crisis are interpreted by deduction from one's beliefs and initial perceptions of crisis circumstances - there is, after all, no other basis available for interpretation..." 34 Often, the initial immediate images are false. After all, three-fourths of the time, in their cases, the initial images of the adversary are substantially different from the adversary's self-image. 35 They uncover a large number of incorrect interpretations of rival messages and moves in the early stages of crises. 36
If the objective situation determined perceptions, we would certainly expect it to do so regarding what neorealists call the fundamental factor in the state system: military technology. However, in the crisis of July 1914, most state elites perceived exactly the opposite systemic military constraint from what would have operated in reality. Van Evera (1984) shows that European leaders believed military technology was offense-dominant when in actuality it was defense-dominant. Further, the misperception held for virtually all the European great powers, meaning the system provided no negative feedback mechanism to correct the misperception. So, if there can be misperception of this depth, breadth and importance at the start of a crisis, we would expect the objective situation to be less than a determinate constraint. A fundamental factor in crises is rival intention. If they believe their rival has limited intent, they can postpone escalation and easily fend off criticisms of appeasement. 37 However, this judgement is difficult, given the paucity of adequate information at the outset of a crisis. So they rely on prior background images, as Snyder and Diesing show.
Perceptual effects played a key role in explaining the outcome in these two crises. For some reason, the British government found it easier to trust Russia in the second case, overlooking Russian miscues. In both October 1853 and December 1877 Britain received intelligence of potential secret Eastern Court collusion for an exclusive partition plan. Only in the former case did this become decisive (contributing to rejection of the concessions Russia actually made at Olmütz).
A critical stage in the 1853 was Britain's response to the Nesselrode interpretation. Russia had accepted the Vienna Note, much to the delight of European leaders. Then, rather unexpectedly, Turkey refused to sign without substantial modifications. This put the British leadership in a tight jam, since their ally now appeared to be flaunting the will of the European powers. Then, Russia allowed to be released a potentially unofficial interpretation (Examen) of the Vienna Note.
The British interpretation of the Examen was not obviously supported by the text. It requires considerable examination of the issue to even find the reasoning behind the government's position. Even then, further scrutiny undermines their position. Russia's central position at the time seems reasonable: if Turkey amended the Vienna Note, Russia should be able to retract its earlier acceptance of the Note without modification. As for Russia's specific resistance to Turkey's amendments, they stemmed from proposals Russia had made prior to the Vienna Note and of which the British government was fully aware. The Russian position did not rule out further European mediation; though it did request that Europe press Turkey to accept the Note unmodified. They made no ultimatums to Europe; they simply said that if this failed, they would retract their earlier acceptance. As for proof that the Vienna process had been a device to dupe Britain and France, this seems unreasonable if Britain had already known of the gist of the position implied by the Nesselrode Examen. Also the Examen was unofficial; Britain did not follow the Austrian and French example and advice of giving Russia a chance to repudiate or clarify it. At worst, it should have been treated as an unreasonable proposal. Instead, British state elites interpreted it as a condemnation of the Vienna process and of the diplomatic approach in general, and encouraged its more hawkish cabinet members and the press to do likewise. They then condoned a misleading interpretation of the Olmütz conference as involving a secret eastern alliance to exclude Britain. They allowed their interpretation to be influenced by an image assuming implacable Russian hostility. This then enhanced their popularity and kept them in office. Clearly, this was an image of misperception. They soon concluded war was inevitable. As Richardson (1994:275) says: "Overestimates of Russian hostility, by Britain in 1853 and Japan in 1903, led both to accept war, as Levy hypothesizes..." Clearly, perceptions played a key role.
Now we must ask how these perceptions arise. I prefer to conceive of imagery and perceptions as social phenomena. If individuals hold a faulty image, others can take steps to weaken their authority. Also we often encounter rhetorical justifications which go far beyond our credulity as scholars. They do not fit policy and they do not fit private statements and behavior. Also, at later and decisive phases of crises, we often witness misperception occurring across an entire government. So, to understand the origins of misperception we ought to look at domestic politics.
Domestic Politics
There are a variety of theories here.
Degree of Democracy
This can not explain the different outcomes. Both Russia and Britain moved in a democratic direction, roughly speaking, between the two crises, but the difference was very slight and unlikely to have an impact on crisis outcomes. In Russia, serfdom was abolished. Nonetheless, the freed serfs were unable to win any control over foreign policy. In Britain, the number of eligible voters was doubled or tripled in the early 1860s. It went from 3% of the adult population to about 5%. This is hardly democratic.
Popular Support.
Popular support for confrontational policy played a critical role, but only for one of the four crisis actors - Britain in 1853 - and only for the later stages of the crisis, after the British cabinet had essentially rejected the Vienna Note and the Vienna mediation process. Not until the November Sinope incident December cabinet crisis did this arguably prevent conciliatory policy. As late as mid-September, the government strongly considered resisting popular pressures and siding with Russia, joining with the great powers to abandon Turkey.
Public opinion by itself did not cause either of the two key British escalatory decisions in the 1853 case: the June decision to send the fleets to Besika Bay and the mid-September to abandon a collective four-power guarantee of the Vienna Note. Clarendon, in both periods, mentions public opinion as influencing his decisions but we have little direct evidence of popular opinion itself, as opposed to press opinion. Nonetheless, the balance of evidence suggests the cabinet, though concerned about popular or press opinion, was far from its captive. In late June, despite calls by some papers for the impeachment of Aberdeen and Clarendon their assistance to Russia, 38 Aberdeen argued for the merits of resisting popular opinion in cases such as this. 39 Their policy response to the Menshikov demands and the Russian ultimatum seems a good response to their continuing uncertainty about Russia's intentions: signal seriousness in a non-provocative way while demanding further assurances. The cabinet agreed on the need for firm measures, risking war but only if Russia's intentions turned out to be aggressive. 40
In mid-August, after the prorogation of Parliament, the situation looked good for the government. Of course, on the last day of debate in each house, there had been much criticism of government policy. Nonetheless, the government ended up parliament in what Aberdeen called "a triumphant close." Aberdeen was receiving much praise for having achieved peace by resisting public opinion to achieve peace. 41 There was a lull, after Parliament closed, in press attacks on the government. 42
Clarendon, in early September, as quite aware of Turcophilic opinion within Britain. He felt in considerable anxiety that it pushed him in one direction and the Tsar's immediate acceptance of the Vienna Note pushed him in the other. Despite the anticipated resignation of Stratford, he was ready on Sept. 15 to join in a four-power initiative to press Turkey to sign an unmodified Vienna Note. Schroeder (1972) suggests that a diplomatic approach relying on great power concert - i.e. the Vienna Note - would have satisfied popular opinion.
Press opinion is partly a reflection of, rather than a driving force in, intra-elite competition. In both cases, political elites (in and out of government) manipulated the press for their advantage, to a considerable extent, 43 though the 1878 cabinet was probably more successful, partly through slightly better control over public releases of diplomatic correspondence. In sum, details aside, we still do not have direct evidence that public opinion was yanking a divided cabinet into war.
Propaganda Boomerang
A number of theories and historians have explored a phenomenon I'll call propaganda boomerang. The imagery and myths that leaders use at one stage to legitimate their policies come back to thwart their policy aims at later stages. Snyder (1991), Peterson (1996) and Cortell and Davis (1996) all show how imagery used by state elites to build a support coalition at the outset of a crisis often restricts their political options in later stages of the crisis. As one example, Lebow (1981) showed how early images shape how subsequent external events get interpreted by politicized security bureaucracies and by individuals, making it hard for leaders to justify quick alterations in policy even if they wish to. As another, more topical example, legitimations used early in the crisis (usually by the government) gain currency and thus become more easily usable to criticize government policy in later phases of the crisis.
Existing research shows a number of the features of this process. To begin, usually these initial expectations are inferred from actors' prior beliefs. Snyder and Diesing's (1977:324) describe how initial expectations are formed at the outset of a crisis. The initial message or pattern of moves "is interpreted on the basis of the background image {of the rival} and the specific circumstances...to produce a whole diagnosis of the crisis situation (immediate image) and a whole set of expectations: the opponent's aim is x, his strategy is y, this move y 1 is intended to accomplish z, if we do A he will probably respond y 2, and if we do B he will probably respond y 3." These expectations, their research demonstrates, set the bargaining problem in the crisis and the crisis strategy. 44 "These initial interpretations are very important since they produce a whole diagnosis and set of expectations. And, since initial images are substantially different three-fourths of the time, these initial misinterpretations produce the initial difference of expectations that sets the bargaining problem." 45 Hence, how state elites respond to the initial window of crisis opportunity dramatically shapes crisis outcomes.
This process of coalescing around initial images reflects hastily negotiated social consensus among state elites, rather than something which can be called "perceptions." Richardson's (1994:258-9, 261-3) findings on the decision making process of coming to consensus during crises helps clarify the dynamics. Individuals differ in their perceptions at the outset of crises. This point, supported by many other researchers, 46 should be enough to convince us that external events do not automatically alter state's beliefs by providing them with new unambiguous information; rather, politicians interpret external events in diverse ways. Often, these differing perceptions arise from different prior beliefs. Richardson shows that differences are not settled by calm, deliberate use of the available time and information. Rather they are settled hastily and emotionally whether by authority or the exclusion of dissenters.
The immediate implications are that information channels will likely become politicized, as people recognize that the politicians involved have already staked out their positions. Additionally, once a ruling coalition has taken a position, its members (and their subordinates) are likely to sense strong pressures to toe the party line. So, when viewing the aggregate policy of the country, this process looks much like cognitive closure, but at the level of state officials as a group. Snyder (1991), Lebow (1981) and Cumings (1990) show a number of social mechanisms which contribute to a policy consensus based on faulty perceptions during crises. Cortell and Davis (1996) show how frames endure during militarized crisis bargaining - even to the extent of constraining state elite choices - because domestic opponents use the frame as a standard to evaluate state policy. In many situations, when the governing coalition takes a hardline stance, it becomes trapped in its own propaganda. It can not make concessions for fear of losing office. Such crises will follow a dynamic that would look to Snyder and Diesing (1977) like a process of preference revelation uncovering irreconcilable interests. In reality, there would be available settlements which would serve both countries' interests more than war, but one or more ruling coalition's legitimating frame would make those settlements incompatible with its continued hold on office. Such crises would look to Lebow (1981) like premature cognitive closure, depending on the availability of historical evidence. New disclosures of information about rival interests, about resolve and capability and about policy success or failure would tend to be rationalized, in leaders' internal semi-private and external public communications, as evidence for continuation of the existing confrontational policy and as evidence against negotiation and compromise. Politicization of the bureaucracy would produce insensitivity to incoming information. In actuality, the process can be described more fairly as premature social closure. Lebow (1981) describes these sociopolitical dynamics in the U.S. regarding China in 1950 and in India regarding China in 1962.
This domestic propaganda boomerang argument poses a strong challenge to the advice offered by Snyder and Diesing - making confrontational steps early with conciliatory steps later. The early confrontation requires justificatory rhetoric. However, this can be used by domestic opponents to criticize government policy when it turns toward conciliation. So, in other words, even if Snyder and Diesing's advice works internationally (which it didn't in this case), it may fail domestically. When the rival offers some concessions, the opposition use of the confrontational rhetoric may force the government to refuse the concessions it has won - sometimes even to raise its demands. So if we examine Snyder and Diesing's advice from the perspective of decision making elite governing a divided country whose future decisions are unknown, it appears dangerous. The 1853 case shows the difficulty of preventing this, even with a fairly flexible initial imagery.
Aberdeen's imagery in mid-June was explicitly conditional upon Russian intentions. Aberdeen agreed that the Menshikov demands of 5 May were excessive and unreasonable. The question centered around their consistency with Russia's past promises. The demands, said Aberdeen, "'ought to be resisted. But I cannot yet believe that it will be necessary to do so by war if the Emperor should hitherto have been acting in good faith; if his whole conduct should have been a cheat, the case is altered.'" 47 This can not be characterized as rigidly hawkish or dovish; he recognized the need for flexibility arising out of incomplete information. When Palmerston advocated sending fleets to Constantinople rather than simply to Besika Bay, Aberdeen called this a half-measure. It would be enough to provoke Russia (and violate the 1841 treaty, as others argued) but not enough to protect Constantinople. When the time came, the fleets should be sent right into the Black Sea, he said.Ê 48
When the Nesselrode Examen was leaked to the German press and became known to the British cabinet in mid-September 1853, the cabinet consulted closely for several days before deciding on a new course of action. Their policy was not an immediate response to an unambiguous event. Rather their policy was the resultant of internal discussion, with the circulation of long memoranda briefing and rehearsing arguments which might play well if the coalition fell apart and the debate became public. Aberdeen's imagery had a conceptual category ready for evidence of Russian hostility. He could now defend the consistency of earlier support for the Vienna conference process with its abandonment now. The government, especially as Schroeder explains, abandoned the Vienna process of mediation. 49 The press subsequently joined in the uproar against Russia.
Once the government had committed itself to a premise of Russian hostility, moving the fleet into the Black Sea became only a matter of time. The Sinope incident - a normal type of operation during the course of war - became known as 'the Sinope massacre' and formed a legitimation for even more confrontational orders to the fleets.
In 1878, Russia provided the British cabinet with similar opportunities to whip up pro-war fervor. Indeed, the government used this to some degree to justify a vote for war credits, enabling them to prepare for war against Russia, in case it was necessary. Many feared the government would not be able to control this fervor. The objective situation gave ample room to call for confrontational policy: Austria-Germany-Russia were alleged to have settled the issue exclusively (indeed Austria and Russia did have a secret convention resembling not partition but spheres of exclusive action. There were arguments that Russia was going to seize critical points on the trade routes to Asia, that Russia was going to force Turkey to treat bilaterally with Russia, that Russia was going to make Turkey into a Russian dependence. At one point, London learned that the armistice had been signed - with Russian troops within sight of Constantinople. Shortly thereafter, they learned that Russian troops were advancing on Constantinople, in apparent violation of the armistice. The same holds with Gallipoli, a site that would have made it difficult or impossible for British fleets to pass through the Dardanelles. The drums of war beat hard - struggling to keep pace with the pounding of Layard's voluminous and Russophobic dispatches from Constantinople as they hit the Foreign Office desk in London. Russia made some miscues to foster these escalatory pressures: particularly not releasing the terms of the armistice. As it turned out, for some reason, the propaganda boomerang did not occur. The government's initial frame enabled them to have a more flexible policy.
In 1877, the Disraeli government framed things differently. They also had an explicitly conditional imagery and policy. Britain would not declare neutrality but would intervene if important interests were threatened. It also protested the Russian failure to secure multilateral consensus before attacking Turkey. It also announced the narrowness of British interests in the Balkans. Their initial legitimating frame, therefore, focused on the potential threat to Britain's interests and the unilateral character of Russia's actions. These objections imply a criteria for subsequent interpretations of Russian intent. More so than in 1853, Russia easily met these criteria to demonstrate its "absence of aggressive intent." It offered the "little Bulgaria" proposal as a way to meet Britain's security interests and it repeatedly promised European participation in the final settlement of the Russo-Turkish dispute. The cabinet thus used a concert norm - multilateral great power consultation and coordination aimed at dispute settlement, with no references to obligations to join in collective security action against aggressors and with the construction of balancing coalitions left implicit in the political context. They did not label Russia's actions aggressive, which would have required Russia retract its actions in order to demonstrate non-aggressive intent. Throughout the remainder of 1877 and up through the actual holding of the Berlin congress in June 1878, they used this in 1878 as a way to weaken domestic opposition, considered abandoning it, but stayed with it throughout the crisis. These contrasting choices in initial legitimating frames were critical to crisis outcomes in both cases.
So, in summary, propaganda boomerangs do occur, even with conditional imagery more sophisticated than one normally finds theorized in this and some of the psychological literature on crisis decision making. However, we must now ask why they arise in some cases and not others. What led the state elites in the latter case to frame the Russian attack on Turkey in a less confrontational fashion?
State Structure
Three theorists embed the propaganda boomerang idea within broader theories of domestic politics in an effort to explain confrontational policies not warranted by national interest. Snyder (1991), Peterson (1996) and Cortell and Davis (1996) all present theories partly conditioning the propaganda boomerang effect on the character of state structure. This hypothesis undoubtedly warrants further research. For now I simply wish to point out that internal cabinet divisions plagued Britain in both cases.
Of course, there was sharp disagreement within the 1853 British cabinet. The Aberdeen government, lacking majority support in Parliament, formed an amalgam of several distinct factions, including Whigs and Peelites. The cabinet's internal cleavages have led many to call British decision-making itself divided. 50 The coalition nearly collapsed in December with Palmerston's resignation. 51 Up through September, however, the coalition was fairly strong, despite frequent policy differences. "In March, Clarendon said 'he never saw anything like the union and harmony of the cabinet. No people could get on better.'" 52 Gladstone agreed. 53 In late May, when the cabinet took a policy characterized by many as an unhappy medium of sending the fleets to Besika Bay but not to Constantinople, the cabinet was unified on essentials. As Conacher (1968:149-151) and Goldfrank (1994:173) note, even for Aberdeen, resistance by war would be warranted if the Tsar turned out to have deceitful.
The cabinet's coalitional character did not prevent policy success. It succeeded more than most governments in carrying out its 1853 parliamentary agenda, in domestic and international issues. On the eastern crisis, it appeared in general to contemporary observers to have stood together. 54 Indeed, when Russia signed the Vienna Note, it looked as if the crisis had ended. Observers credited Aberdeen with having withstood the public clamor for war. Prior to the news that Turkey would reject the Vienna Note, the coalition appeared strong and united. As Aberdeen put it, "...although a coalition of very different materials, we have adhered well together." 55
The British cabinet in 1877-8 contained intense divisions about what to do regarding the Russian threat. The Queen and the Prime Minister wanted war-risking confrontational escalation while Salisbury and Derby thought a more conciliatory policy appropriate. There were intense cabinet struggles in January and February. Salisbury began moving more towards Disraeli, which left Derby increasingly isolated. Carnarvon resigned with a bitter little row. Derby submitted his resignation, thinking this to be a permanent outcome. However Disraeli later convinced him to stay, all the while privately telling the Queen that Derby would eventually resign anyway. After February, Derby became more passive but the major foundation of the policy toward Russia had already been laid. Several crises had passed peacefully before policy solidified behind the so-called 'Salisbury note' of early April.
Practice of prior multilateral consultation
Previous sections have shown the importance of propaganda - the images or legitimating frames used by state elites to interpret unexpected and potentially threatening external developments, describing the situation and the rival's intentions. If the government fails to coalesce around a particular legitimating principle, or coalesces around an extremely dovish or extremely hawkish legitimating principle, it has less freedom of maneuver in subsequent crisis negotiation. Domestic opposition groups use pre-existing frames to criticize subsequent government policy. Anticipating the loss of office, the government may be forced by this effect into interpreting subsequent events in extreme fashion, resulting in unwisely confrontational or unwisely conciliatory policies. In other cases, the government coalition coalesces around legitimating frames which preserve both their unity relative to the opposition and their ability to maneuver flexibly during the crisis.
Historically, in both of our cases here, defender state elites wished to avoid war with the challenger, if necessary at the expense of requiring their ally to make significant concessions. Britain, for example, asked Turkey not to declare war after the occupation in 1853. In both cases, the state elites struggled to find a suitable depiction of the situation and their policies that would enable them to withstand clamoring for greater confrontation while providing policy flexibility until the challenger's objectives became more clear. In both cases, they had conditional imagery but in only one case did that imagery allow the government a moderately firm, but flexible policy. In both cases, there were pressures for crisis escalation and threats of coalition collapse. Only in one case was the coalition trapped into unwise confrontation.
Our burden now is to show what enables defender state elites to coalesce around flexible legitimating frames. The difference rests in diplomacy prior to the attack. Of course, in both cases, the challenger was equally motivated to isolate the victim of its intended attack and pursued policy initiatives with this aim. In one case, the challenger had followed the practice of multilateral consultation with the great powers. In the other, it had not. In the former case, Russia had packaged the Menshikov mission in a threatening and militaristic fashion, kept Europe in the dark regarding his broader objectives, and resisted a solution involving European participation in protecting Ottoman Christians. In the latter case, Russia had carefully followed the norm of multilateral consultation since the early stages of the Bosnian uprising years before. Russia made very thorough and patient efforts to exhaust all the multilateral dispute resolution mechanisms before taking military action, rather than imposing a fait accompli.
Russia's pre-war diplomacy achieved much more success in 1878, despite their attack on Turkey lacking specific sanction by the European great powers. The process of how this occurred deserves some examination, to see if it truly had an impact on British policy. Britain protested the violation of voice.
In late April 1877, Britain decided to declare conditional neutrality. The Cabinet discussed its response to the Russian circular justifying its decision to enter war. Two memos emerged in early May, distributed to Europe. One listed a framework of conditions under which Britain would no longer maintain neutrality. These gave Russia great leeway. Nonetheless, it refused to promise neutrality, preserving for itself the option of going to war. The other used the principle of great power voice to preserve Britain's future options to declare the illegality of the Russian war. This memo responded to Russia's justification for ordering its armies into Turkey, disputing Russia's reasoning in an earlier declaration to Europe that it was acting in the interests of Europe. Britain insisted that Russia had violated its obligation under the European concert to consult with the great powers:
"In taking action against Turkey on his own part, and having recourse to arms without further consultation with his allies, the Emperor of Russia has separated himself from the European concert hitherto maintained, and has at the same time departed from the rule to which he himself had solemnly recorded his consent.
"It is impossible to foresee the consequences of such an act. Her Majesty's Government would willingly have refrained from making any observations in regard to it; but, as Prince Gortchakow seems to assume, in a declaration addressed to all the Governments of Europe, that Russia is acting in the interest of Great Britain and that of the other Powers, they feel bound to state in a manner equally formal and public that the decision of the Russian Government is not one which can have their concurrence or approval." 56
Though a violation of the norm had occurred, Russia's reaffirmation of the norm enabled British state elites to use a concert-oriented standard of assessment for Russia's subsequent policy, when portraying their subsequent interpretations of Russia's intentions. This enabled them to give less hawkish interpretations of Russian miscues during subsequent phases of the crisis.
It was not simply the challenger's policy during the crisis itself. Policy across the two crises did not differ much, in terms of resistance to a European role, whether voice or mediation. When Russia accepted the Vienna Note in 1853, they accepted Europe's role. The Nesselrode Interpretation did not rescind that, in reality. And if it had, it was repudiated in this respect by Nicholas at Olmütz. In 1878, Russia repeatedly declared that the final settlement of their dispute with Turkey would be settled with the participation of Europe. However, those are only words. Their policy during the conduct of the war was to exclude all attempts at mediation. Then, as the war was winding to a close, their behavior and policy statements gave grave reason to doubt their commitment to multilateral practices and norms. A government trapped in hawkish imagery could have easily found reasons for confrontation. As explained earlier, the government spin on crisis developments followed from its earlier interpretations of the situation.
A key point in the pre-crisis developments was Britain's willingness to adopt a great power position visa-vis Turkey that included references to enforcement measures. As Britain considered this position - the London Protocol, the deference that had been paid to British opinion since the Berlin Memorandum in May 1876 was an important factor reassuring Prime Minister Disraeli, Lord Beaconsfield, to accept the Russian proposal.
In 1878, British state elites were able to coalesce around a principle of voice. British interests would be protected because Russia and the European powers had agreed to Britain's right to participate in the final settlement of the Russo-Turkish dispute. This principle gave the cabinet significant domestic advantages. The opposition could not unite behind a solid criticism of government policy as being too hard-line or being too soft-line. The voice principle gave the government the flexibility to shift its policy according to the subtle to-and-fro of crisis negotiation (as well as domestic politics). Both mildly accommodating and mildly confrontational policies appeared consistent with the government's frame of the situation. This policy flexibility prevented the opposition from uniting behind a particular policy direction in the crisis. This frame choice prevented the opposition from using the frame to seriously threaten the governing coalition.
Governing coalitions in both cases withstood popular clamors for war, for a while. In the 1853 case, the opposition had succeeded in stirring up significant anti-Russian sentiment, as Snyder (1991) shows. Many in the government, including the foreign secretary, were nervous and desirous of capitalizing on this sentiment to weaken the opposition challenge and to strengthen its own internal stability. The cabinet had withstood the domestic political costs of conciliating Russia for months. In mid-September, it was on the verge of joining the European powers in an effort to press Turkey to abandon its modifications, holding out the carrot of a European interpretation of the Note that interpreted it in a Turkish sense and the stick of leaving them to fend against Russia by themselves. As it turned out, they fell captive to an imagery that implied radical shifts in policy depending on small shifts in evidence of Russian hostility. The slightest miscue - Nesselrode's impolitic portrayal of Russia's motives for rejecting the Turkish modifications - let the government off the hook. They were thus able to coalesce around the image of Russia having betrayed and deceived Britain. This image served them well domestically. It was consistent with their past mildly pro-Russian policy and their acceptance of the Vienna process. They were able to portray the upcoming Olmütz meeting of the Eastern leaders as a potential attempt at partitioning European Turkey, at Britain's expense. They were able to delegitimize the Vienna process.
While this strategy served them well domestically, it failed them internationally. They were subsequently unable to conceive or portray any subsequent Russian concessions emerging out of the Vienna process as sincere. Their image of Russian intentions led the opposition and the press to converge on the impression that further diplomacy was futile. By mid-October, the government derived popular support by making decisions likely to lead to war with Russia. By late October, they were planning for war and conducting diplomacy with the objective of attracting allies. They thus precluded available peaceful settlements that would have been better for the country than war with Russia.
In 1878, however, the state elites were able to dampen their policy differences while resisting popular clamors for war. Occasionally the coalition keeping the cabinet in power almost collapsed. Instead the hawk majority and the dove minority coalesced around a principle of voice. The government then used its informational and other domestic resources to regulate the rising popular clamor for war. This despite all the factors listed above which would lead us to expect war. As in 1853, Russia made several diplomatic miscues. The objective situation gave ample room to call for confrontation, as shown above. As it was, the government interpreted this evidence through the imagery suggested by the principle of voice.
So, in sum, when Russia practiced extensive multilateral consultation prior to attack, this enabled state elites to coalesce around a moderate frame of the attack on Turkey. This enabled the state elites to withstand popular pressures for escalation, even when later Russian diplomatic moves appeared to violate the trust which Britain had given them. In the situation where Russia engaged in attack without extensive prior consultation, state elites had fewer framing options, with the result that they had less domestic maneuver room as the crisis developed.
Conclusion
Patient preparatory multilateralism reshapes subsequent crisis dynamics by reshaping the images countries hold of one another's intent and thus reshaping their strategies. That seems to be the most important general lesson which statesmen learned between these two crises. Given the many other similarities between the two cases, it remains difficult, within our current theoretical understanding, to argue that they chose different strategies because they were fundamentally different disputes (an alternative hypothesis of self-selection). More probably, peoples' historical understandings contributed to their ability to improve their relations. If the claim presented here is correct, they learned to manage crises more peacefully by postponing military action until after they had engaged in extensive multilateral consultation with the other great powers.
We are now in a position to specify more fully the specific claim of this paper:
Whether an expansionist great power engages in extensive multilateral consultation prior to attacking the close ally of another great power strongly influences whether or not the subsequent great power crisis ends in great power war. Under broad conditions, the practice of extensive prior consultation has a decisive influence on peace or war.
We now understand a little better the dynamics of resolvable great power spin-off crises. Existing research takes us some distance in narrowing the focus and the mystery of crisis outcomes. The thesis offered here has the potential to improve on that research.
A critical element of threat perception is an assessment of rival intentions and that, in turn, is influenced by the rival's attitudes and previous policies towards international norms and rules. Normative context influences states' threat perception and thus their crisis strategies. By shaping that context in the present, we might be able to reshape international relations in the future.
Resolvable Non-Brinkmanship Great-Power Spin-Off Crises
Notes
Note 1: Paper prepared for presentation at the International Studies Association annual convention, Minneapolis, Minnesota, March 17-21, 1998. The author intends to significantly revise this very preliminary work in progress and would greatly appreciate any and all comments. Please do not cite without permission Back.
Note 2: Dept. of Political Science, University of California Los Angeles, Los Angeles CA 90024; rasmus@ucla.edu Back.
Note 3: For example, today's list would include: USA, Russia, Japan, Germany, Britain, France, China. Back.
Note 4: Hermann 1972, 13. Back.
Note 5: Lebow 1981, 9 adds this characteristic and creates the category of 'acute crisis'. Back.
Note 6: Brecher 1993, 2-8, 15-25, offers an excellent discussion of definitions of interstate crisis. Back.
Note 8: This includes the actual net gains from war itself plus the benefits to be derived from the diplomatic settlements which could emerge from war, each discounted by their probability of non-occurrence. Back.
Note 9: Fearon, 1995, distinguishes this from irreconcilability. However, since it has effects similar to issue lumpiness in restricting the set of solutions within the bargaining space, I include it within this category. Alternatively, nonenforceability and lumpiness could be treated as questions of degree, raising the costs of each particular settlement, with the same analytic effect. I do not refer to international diplomatic or domestic political consequences.< Back.
Note 10: Lebow 1981:41-56. Back.
Note 11: Lebow 1981:41-56. Back.
Note 12: On the 1852-54 dispute, see Anderson 1966, Goldfrank 1994, Gooch 1967, Gouttman 1995, Guichen 1936, Rich 1985, Saab 1977, Schroeder 1972, Temperley 1936/1964, Wetzel 1985, Conacher 1968, Chamberlain 1983, Curtiss 1979. On the 1875-78 dispute, see Anderson 1966, DiNardo 1985, Florescu 1985, Harris 1936, Leathers 1963, Millman 1979, Seton-Watson 1935/1962, Stojanovic 1939, Sumner 1962, Tyler 1925, Wirthwein 1935. Back.
Note 13: Schroeder 1972, Miller 1992b:5, n. 13, Kupchan and Kupchan 1991:122 n. 23. Back.
Note 14: For example, Lebow, 1981, and Richardson, 1994, accept it in parts. Back.
Note 15: Snyder 1991:169, Curtiss 1979:35-6. Back.
Note 16: Christensen and Snyder 1990. Back.
Note 17: Snyder and Diesing 1977:501-2, give it importance but more heavily stress Germany's misperception of British intentions. Back.
Note 18: There are exceptions, but the general rule holds. Back.
Note 19: Hawtrey in Gilpin, 1980, pp. 31-2. Waltz, 1979, pp. 165, 168, 171-2. Quester, 1982, pp. 2-10. Doran and Parsons, 1980, pp. 949-952. Jervis, 1976, p. 52. Blainey, 1988. Levy, 1983. Organski and Kugler, 1980, p. 21. Gilpin, 1980, p. 202. Back.
Note 20: Fearon 1995:391, 398-9. Back.
Note 23: 87-9. Other types of misperception are only indirectly related to war, he says. See also Stein 1982 and Richardson 1994:266. Back.
Note 31: Most historians describe the Russian move as a blunder. The Tsar himself seems to accept this view. He said had he known that his legal authority according to the Treaty of Kutchuk Kainardji was not as great as he had thought, he would not have pressed his case so strongly. Back.
Note 32: As we will see later, whether this constituted confrontationor accommodation involved a serviceable ambiguity for British state elites. Back.
Note 33: They were consistent in other respects. Throughout the crisis, they insisted on Russia not occupying Constantinople - though they took little military action to back this demand until very late in the game. Back.
Note 34: Snyder and Diesing 1977, 334, 337. Back.
Note 35: Ibid., 292, 324. Back.
Note 36: Ibid., 316, 324. Back.
Note 37: On the importance of perceptions of rival intent, see Jervis 1976, 1985; Snyder and Diesing 1977; Cohen 1979; Lebow 1981; Walt 1987; Jervis and Snyder 1991, etc. Back.
Note 38: Conacher 1968:155. Back.
Note 39: Conacher 1968:157. Back.
Note 40: Conacher 1968:149-151, Goldfrank 1994:173. Back.
Note 41: Conacher 1968:173. Back.
Note 42: Conacher 1968:186. Back.
Note 43: For example, Conacher 1968:186. Back.
Note 44: Snyder and Diesing 1977:324, 329. Back.
Note 45: Snyder and Diesing 1977:324. Back.
Note 46: Snyder and Diesing, 1977:297-310 lay out the hawks/doves division "ordinarily" arising within governments at the outset of crises. Young, 1968:196-7; George and Smoke, 1974:433-5, and George, 1980:70-1, use this dichotomy to describe the division within the Kennedy administration in 1961 at the outset of the Berlin crisis. Back.
Note 47: Conacher 1968:150-1. Also, Chamberlain 1983:481. Back.
Note 48: Conacher 1968:149. Back.
Note 49: The fleet move was based on a variety of motives and not followed up after Stratford's initial refusal to summon the fleet. Back.
Note 50: Richardson 1994. Back.
Note 51: He rejoined the cabinet shortly thereafter. Back.
Note 52: Temperley 1936/1964:338. Back.
Note 53: Temperley 1936/1964:473. Back.
Note 54: Parliamentary debates on the last day of the session were an important exception. Back.
Note 55: Conacher 1968:172-3. Back.
Note 56: British Foreign and State Papers, 1868:841-3. Back.
Bibliography
Anderson, M. S. (Matthew Smith), 1966. The Eastern question, 1774-1923: a study in international relations (London: Macmillan; New York, St. Martin's).
Anderson, Paul A., 1981. "Justifications and Precedents as Constraints in Foreign Policy Decision-Making," American Journal of Political Science 25:4 (November) 738-761.
Blainey, Geoffrey, 1988. The Causes of War 3d ed. (South Melbourne: Macmillan Press).
Brecher, Michael, 1993. Crises in World Politics: Theory and Reality (Oxford: Pergamon Press).
Chamberlain, Muriel E., 1983. Lord Aberdeen: A Political Biography (Longman: London).
Christensen, Thomas J., and Jack Snyder, 1990. "Chain gangs and passed bucks: predicting alliance patterns in multipolarity," International Organization 44:2 (Spring) 137-168.
Cohen, Raymond, 1979. Threat Perception in International Crisis (Madison: University of Wisconsin Press).
Conacher, J.B., 1968. The Aberdeen Coalition, 1852-1855: A Study in Mid-Nineteenth Century Party Politics (Cambridge: Cambridge University Press).
Cortell, Andrew P. and James W. Davis, Jr., 1996. "How Do International Institutions Matter? The Domestic Impact of International Rules and Norms," International Studies Quarterly 40:4 (December) 451-478.
Cumings, Bruce, 1990. The Origins of the Korean War, Vol. II: The Roaring of the Cataract, 1947-1950 (Princeton Univ. Press: Princeton).
Curtiss, John Shelton, 1979. Russia's Crimean War (Duke University Press: Durham, N.C.).
DiNardo, Richard L., 1985. "Russian Military Operations in 1877-78," in Bela K. Kiraly and Gale Stokes, eds., 1985. Insurrections, Wars, and the Eastern Crisis in the 1870s (Columbia University Press: New York), pp. 125-140.
Doran, Charles F. and Wes Parsons, "War and the Cycle of Relative Power," American Political Science Review 74:4 (December 1980) 947-965.
Fearon, James D., 1995. "Rationalist explanations for war," International Organization 49:3 (Summer) pp. 379-414.
Florescu, Radu R., 1985. "The Russo-Turkish War of 1877-78: Diplomatic and Military Preparations," in Bela K. Kiraly and Gale Stokes, eds., 1985. Insurrections, Wars, and the Eastern Crisis in the 1870s (Columbia University Press: New York), pp. 93-104.
Gilpin, Robert, 1981. War and Change in World Politics (New York: Cambridge University Press).
Goldfrank, David M., 1994. The Origins of the Crimean War (Harlow, Essex: Longman).
Gooch, ed. Origins of Crimean War 1969
Gouttman, Alain, 1995. La guerre de Crimée, 1853-1856 (Paris: S. P. M.).
Guichen, Vicomte de, 1936. La guerre de Crimée (1854-1856) et l'attitude des Puissances Européennes (Paris: Pedone).
Hagan, Joe A., 1987. "Regimes, Political Oppositions, and the Comparative Analysis of Foreign Policy," in Charles Hermann, Charles W. Kegley, Jr. and James N. Rosenau, eds., New Directions in the Study of Foreign Policy (Boston: Allen & Unwin), pp. 339-365.
Hagan, Joe D., 1994. "Domestic Political Systems and War Proneness," Mershon International Studies Review 38:Suppl. 2 (Oct.) 183-207.
Harris, David, 1936. A Diplomatic History of the Balkan Crisis of 1875-1878: The First Year (Stanford: Stanford Univ. Press).
Hermann, Charles F., ed., 1972. International Crises (New York: Free Press).
Hermann, Charles F., 1969. "International Crisis as a Situational Variable," in James N. Rosenau, ed., International Politics and Foreign Policy: A Reader in Research and Theory (London: Collier Macmillan), pp. 409-421.
Holsti, Ole R., 1967. "Cognitive Dynamics and Images of the Enemy: Dulles and Russia," in David J. Finlay, Ole R. Holsti, and Richard R. Fagen, eds., Enemies in Politics (Chicago: Rand McNally and Co.,), pp. 25-96.
Jervis, Robert, 1983. "Security regimes," in Stephen D. Krasner, ed., International Regimes (Ithaca, New York: Cornell University Press, 1983), pp. 173-194.
Jervis, Robert, 1986. "From Balance to Concert: A Study of International Cooperation," in Kenneth A. Oye, ed. Cooperation under Anarchy (Princeton: Princeton University Press), pp. 58-79.
Kaufmann, Chaim D., 1994. "Out of the Lab and into the Archives: A Method for Testing Psychological Explanations of Political Decision Making," International Studies Quarterly 38:4 (December) 557-586.
Keohane, Robert, 1984. After Hegemony: Cooperation and Discord in the World Political Economy (Princeton: Princeton University Press).
Keohane, Robert O., 1989. International Institutions and State Power: Essays in International Relations Theory (Boulder: Westview).
Kupchan, Charles A. and Clifford A. Kupchan, 1991. "Concerts, Collective Security, and the Future of Europe," International Security 16:1 (Summer) 114-161.
Layne, Christopher, 1993. "The Unipolar Illusion: Why New Great Powers Will Rise," International Security 17:4 (Spring) 5-51.
Leathers, Noel, 1963. France and The Balkans, 1871-1879 {??:}(PhD Dissertation, University of Oklahoma; University Microfilms 63-6607).
Lebow, Richard Ned, 1981. Between Peace and War: the Nature of International Crises (Baltimore: Johns Hopkins Univ. Press).
Levy, Jack S., 1983. "Misperception and the Causes of War: Theoretical Linkages and Analytical Problems," World Politics 36 (October): 76-99.
Levy, Jack S., 1987. "Declining Power and the Preventive Motivation for War," World Politics 40 (Oct.) 82-107.
Levy, Jack S., 1994. "Prospect Theory and International Relations: Theoretical Applications and Analytical Problems," in Barbara Farnham, ed., Avoiding Losses/Taking Risks: Prospect Theory and International Conflict (Ann Arbor: University of Michigan Press).
Mearsheimer, John J., 1990. "Back to the Future: Instability in Europe after the Cold War," International Security 15:1 (Summer) 5-56.
Miller, Benjamin, 1992b. "Explaining Great Power Cooperation in Conflict Management," World Politics 45:{} (October) 1-46.
Millman, Richard, 1979. Britain and the Eastern Question, 1875-1878 (Oxford: Clarendon Press).
Müller, Harald, 1993. "The Internalization of Principles, Norms, and Rules by Governments: The Case of Security Regimes," in Volker Rittberger, ed., Regime Theory and International Relations (Oxford: Clarendon Press), pp. 361-388.
Organski, A. F. K., 1958. World Politics (New York: Knopf).
Organski, A. F. K. and Jacek Kugler, 1980. The War Ledger (Chicago: University of Chicago Press).
Peterson, Susan, 1996. Crisis Bargaining and the State: The Domestic Politics of International Conflict (Ann Arbor: University of Michigan Press).
Quester, George H., 1982. "Six Causes of War," The Jerusalem Journal of International Relations 6:1 1-23.
Rich, Norman, 1985. Why the Crimean War? A Cautionary Tale (Hanover: University Press of New England).
Richardson, James L., 1994. Crisis Diplomacy: The Great Powers Since the Mid-Nineteenth Century (Cambridge: University Press).
Rosecrance, Richard, 1992. "A New Concert of Powers," Foreign Affairs 71:2 (Spring) 64-82.
Saab, Ann Pottinger, 1977. Origins of Crimean Alliance.
Schroeder, Paul W., 1972. Austria, Great Britain and the Crimean War: The Destruction of the European Concert (Ithaca: Cornell University Press).
Small, Melvin and J. David Singer, 1980. Resort to Arms: International and Civil Wars, 1816-1980 (Beverly Hills: Sage Publications).
Snyder, Glenn H., 1984. "The Security Dilemma in Alliance Politics," World Politics 36 (July) 461-495.
Snyder, Glenn H., 1991. "Alliance Threats: A Neorealist First Cut," in Robert L. Rothstein, ed, The Evolution of Theory in International Relations (Columbia: University of South Carolina Press), pp. 83-103.
Snyder, Glen H. and Paul Diesing, 1977. Conflict Among Nations: Bargaining, Decision Making, and System Structure in International Crises (Princeton: Princeton University Press).
Snyder, Jack, 1991. Myths of Empire: Domestic Politics and Strategic Ideology (Ithaca: Cornell University Press).
Spiezio, K. Edward, 1990. "British Hegemony and Major Power War, 1815-1939: An Empirical Test of Gilpin's Model of Hegemonic Governance," International Studies Quarterly 34:2 (June) 165-181.
Stein, Arthur A., 1982. "When Misperception Matters," World Politics 34:4 (July) 505-526.
Stein, Arthur A., 1990. Why Nations Cooperate: Circumstance and Choice in International Relations (Ithaca: Cornell University Press).
Stojanovic ('), Mihailo D., 1939. The Great Powers and the Balkans, 1875-1878 (Cambridge: University Press).
Sumner, B. H., 1962. Russia and the Balkans, 1870-1880. (Hamden: Archon Books).
Tversky, Amos and Daniel Kahneman, 1981. "The Framing of Decisions and the Psychology of Choice," Science 211 453-458, reprinted in George Wright, ed., Behavioral Decision Making (New York: Plenum Press, 1985), pp. 25-41.
Tyler, Mason Whiting, 1925. The European Powers and the Near East, 1875-1908: Studies in the Social Sciences, No. 17 (Minneapolis: University of Minnesota).
Van Evera, Stephen, 1984. "The Cult of the Offensive and the Origins of the First World War," International Security 9:1 (Summer) 58-107.
Waltz, Kenneth N., 1979. Theory of International Politics (Reading, Mass.: Addison-Wesley).
Wetzel, David, 1985. The Crimean War: A Diplomatic History (Boulder: East European Monographs, distributed New York: Columbia University Press).
Wirthwein, Walter G., 1935. Britain and the Balkan Crisis, 1875-1878 (New York: Columbia University Press).
Young, Oran, 1968. The Politics of Force: Bargaining During International Crisis (Princeton, N.J.: Princeton University Press).Seton- Watson, 1935/1962. Disraeli, Gladstone and the Eastern Question.
Email this citation