Anomalies of the Wilsonian (Monadic) Democratic Peace in the Nineteenth Century: What Can They Tell Us

The Wilsonian Monadic Peace argues that not only do joint democracies not fight each other, but they are inherently peaceful. They get involved in war primarily by being attacked and rarely initiate wars. The institutional explanation of this monadic democratic peace maintains that democracies have this pacific tendency because the people use the legislature to restrain the executive. This paper argues that the best way to assess the causal logic underlying the institutional explanation is by comparing specific cases where war is avoided and where it occurs to see if legislatures and the public restrain leaders. Three historical cases from the Nineteenth Century, which were uncovered as part of a larger project, are reviewed in detail: one that did not go to war and two that did. Cases are drawn from the most democratic states at the time—France, England, and the U.S. In each instance the casual process did not work the way it was expected. Instead, the legislature and the “public” were more prone to war hysteria than the executive. Each of these cases is regarded as anomalous for the democratic peace. The implications of these three anomalies are explored in detail.


Introduction *
I came to the European University Institute to work on my book, War and Diplomacy: Lessons for World Politics (co-authored with Andrew P. Owsiak). In the process of doing so I came across a small set of unexpected findings related to the democratic peace. These are the subject of this paper. Occasionally, in science, discoveries occur that are not the result of the application of the scientific method that tests hypotheses on the basis of a theoretical hypothesis, but of serendipity, of the observation of unexpected outcomes that are then followed up. Such serendipitous or accidental discoveries include the discovery of X-Rays in 1895 by Wilhelm Rontgen and the Microwave oven in 1945 by Percy Spencer.
The results discussed in this paper are of this sort. The three anomalies were noticed as part of a larger study that compared crises among major states that went to war with crises that did not. These three stood out because they all involved early democracies and seemed to contradict the institutional explanation of the democratic peace. These contradictions were so stark that it was thought that they should be brought to the attention of the larger field.
Since the early 1980's with the publications of Rummel (1983), Doyle (1986) and ultimately Russett (1993), the democratic peace has dominated quantitative studies on interstate war and peace. Most of these studies have focused on the fact that joint democracies never fight each other, although with the 1999 Kargil War it is probably more circumspect to say that they rarely fight wars with each other. To this can be added the finding that they rarely have militarized interstate disputes (MIDs) or armed crises with each other (Russett and Oneal 2001).
In addition, to this dyadic democratic peace, there has also been a monadic democratic peace. A monad, in contrast to a dyad, is the individual country. Thus, the claim here is that individual democratic states are inherently more peaceful. This was the position of Woodrow Wilson, and the modern monadic peace claims are therefore often referred to as Wilsonianism. This study focuses on the monadic democratic peace as opposed to its more frequently studied cousin, the dyadic democratic peace. Ray (2000: 302-303), reviewing the empirical studies, has found some evidence for the monadic democratic peace.
Theoretically, the democratic peace, both dyadic and monadic, is often discussed in terms of the institutional-structural or normative-cultural explanation (Maoz and Russett 1993). The latter has found support for both explanations, but the support found for the normative explanation was found to be more robust. More recent work by Russett and Oneal (2001) has backed away from seeing these as competing explanations, but more as complementary. The institutional explanation, which has been favoured by Bueno de Mesquita et al. (1999), explains the democratic peace in terms of the restraints the public places on its leaders (especially the executive) through democratic institutions. Most frequently this includes the legislature but also the courts and the system of checks and balances in general. In many ways the institutional explanation is a more natural explanation of the monadic democratic peace, because it expects democratic states to be peaceful regardless of the opponent. The normative explanation sees democracies as more peaceful because domestically these actors are more used to dealing with disagreement through non-violent means, and this norm is transferred to the foreign policy * My thanks to the European University Institute in Florence where I received a visiting fellowship while on sabbatical and did research on this paper. It was such a pleasure to be at a place where I could spend my time thinking and writing in a collegial atmosphere. My thanks to Ulrich Krotz of the Robert Schuman Centre, to Gabriella Unger of the Department of Political Science and Rich Mahler for making my stay productive and intellectually simulating. My thanks to Ulrich and Rich for comments, as well as Jonas Johannes Driedger, Oliver Westerwinter for suggestions. My thanks to Marie T. Henehan for her reading of the Spanish-American War case.
My thanks also to my undergraduate RA's Jackson B. Barnett and Monica Bielawiec for research assistance. Lastly, my thanks to Emma Benson for compiling the data in Appendix 1 and for the Aberdeen cartoon. The sole responsibility for the study remains mine alone. Robert Schuman Centre for Advanced Studies Working Papers arena. Although individual (i.e. monadic) democratic states always bring this normative stance to dealing with other states, they are more apt to be successful in doing so when their opponent is another democratic state.
This study focuses only on the institutional explanation of the monadic democratic peace. Its purpose is to examine the accuracy of the causal process underlying the institutional explanation of the monadic democratic peace through an examination of specific cases. Examining the internal dynamics of how democratic states successfully avoid war in a situation that is at great risk for war is best done by investigating decision making in historical cases. Ultimately, all the relevant cases should be investigated as Brecher and Wilkenfeld (1997) attempted in their ICB (International Crisis Behavior) project. This paper can be seen as beginning this large effort by looking at the early 19 th democratic states. Because these are only three cases, the results are intended to raise questions and be suggestive, and obviously not a systematic test or confirmation of an alternate explanation.

The institutional explanation of the monadic democratic peace
The institutional explanation of the democratic peace is founded on the assumption that the domestic structure of democracies that is based on the separation of powers and often checks and balances, constrains the executive from going to war without the approval of other parts of the government. This kind of limitation is not present in absolute monarchies or modern dictatorships. Of equal importance, however, is the role of the people and their power to constrain belligerent executives through their electoral capabilities. It is assumed by a variety of liberal philosophers-from Kant to Tolstoy to Wilson-that the people are pacific and generally opposed to war and that they will use the institutions of democracy to express this will. The executive, in turn, will realize this and will be less willing to go to war because of fear of electoral punishment down the road. Russett and Oneal (2001: 53) stress the institutional constraints democracy imposes on decision makers as a causal factor for their reluctance to go to war. Democracies are seen as rarely, if ever, intimating war and only becoming involved in war when they are attacked. Buneo de Mesquita et al. (1999) argue that the use of force is less attractive to democratic leaders because of its domestic cost.
This sort of rationale can be found as far back as Immanuel Kant (Second Section, First Definitive Article) who states: If, as is inevitably the case under the constitution, the consent of the citizen is required to decide whether or not war is to be declared, it is very natural that they will have great hesitation in embarking on so dangerous an enterprise.
And the reason the people are reluctant, according to Kant (Second Section, First Definitive Article), is: For this would mean calling down on themselves all the miseries of war, such as doing the fighting themselves, supplying [paying] the costs…[taking] upon themselves a burden of debt.
This kind of pacific attitude based on self-interest is also emphasized by the Christian pacifist and anarchist, Leo Tolstoy, who says that it is in the common interest of common people to avoid war: Whoever you may be-a Frenchman, Russian, Englishman…you must understand that all our real human interests, whatever they be …all these interests…in no way oppose the interests of the other nations (Tolstoy 1900).
The implication in both Kant and Tolstoy is that monarchs fight wars for their own interest and the interest of their states-not for the common people who must do the actual fighting, although monarchs often were on the battlefield or at the head of their troops. Of course, this was in part because they sought glory and honor.
In the twentieth century this line of thinking is crystalized by Woodrow Wilson. For Wilson it is not in the interest of common people to fight wars. He believed that the spread of education would make the people aware of their true interest. The spread of democracy would give them control of their government. The people would then use democracy to restrain their leaders. This sort of logic applies to democracies regardless of their adversaries. Thus, it more easily fits the monadic democratic peace.
On the whole, other than Maoz and Russett (1993), there have been few direct databased tests of the institutional explanation of the democratic peace. An important exception is that of Quackenbush and Rudy (2009: 268). They conclude that in their study, "There is little, if any, empirical support for the monadic democratic peace." While an important study, this and other possible large-N studies do not get directly at the causal logic underlying the institutional explanation. This paper attempts to do so by examining the decision-making that goes on within democratic states as war looms.

Research Design and Caveats
The institutional explanation emphasizes the people restraining the executive from going to a war that is made possible by the democratic structure of the government. Presumably, in non-democracies, particularly monarchies and dictatorships, the people cannot do this or have fewer avenues for exercising such influence on foreign policy. In testing an explanation, Popper (1963: 47-48) argues that one should specify in advance what evidence would falsify the theory. To get at the accuracy of the casual logic of the institutional explanation, it is reasonable to expect that democratic states that have legislatures would restrain the executive. It might also be expected that the public might use other institutions, like a free press, to restrain the executive or try to directly influence the executive through "public opinion"although this latter phenomena might be more difficult to observe. At any rate, clear falsification of the institutional explanation, would be 1) if the legislature urges the executive to go to war, 2) if the press and "public opinion" are seen as doing the same thing and the society as a whole is gripped by a "war fever," and 3) the executive rather than the legislature is reluctant to go to war and actually prevents war in the face of a more pro-war legislature and "public." Any one of these pieces of evidence would raise questions about the causal logic of the institutional explanation, and the presence of all three, especially the last, would be quite serious.
To investigate these questions, it is best to compare cases that go to war with those where war is clearly avoided. Since this study is derivative of a larger study that compared crises involving two or more major states from 1815-1962, only cases that involve such states are used and only those that occurred in the nineteenth century. Table 1 column two lists the major states according to the Correlates of War classification of what constitutes a major state (Sarkees and Wayman 210: 36) and how democratic these states were in the nineteenth century according to Polity IV scores (see column four).
The three cases involved in this study are the three most democratic in the nineteenth century-France under its citizen king-Louis Phillipe, Britain during the Crimean War, and the United States in 1898. A criticism of the study is that the first two cases, although they involve the most liberal states at the time, are still states that do not make the Polity score threshold of 6 or above. Quackenbush and Rudy (2009: 272) respond to this sort of criticism by pointing out that: "Kant's assertion that democracy produces peace was focused on an elective government, not more modern characteristics of democracy." Nonetheless, it is ultimately agreed that the claim of this paper has to be tested on cases that cross the Polity 6 score threshold. The three cases herein involve two that went to war and one where it was clearly avoided. This distribution is useful because it permits an examination of the causal logic when the outcome variable is present and when it is absent.
There are certain flaws in this design that make the inferences that can be drawn from this study limited. The main flaw, common to many case studies, is that there are only a few cases, so generalization is difficult. With only three cases, generalization is nearly impossible. An equally important flaw is that the list of included cases was not selected by a random sample, nor even a representative sample, but by what can be called a sample of convenience; namely, the cases appeared in a larger study and seemed worth pursuing. All of this means that any conclusions must be seen as 4 Robert Schuman Centre for Advanced Studies Working Papers suggestive. Hence, the subtitle: "what can they tell us?" What is theoretically interesting about the study is that all of the cases are anomalies and that they are opposite of what would be expected.
The analysis must be seen as an empirical probe whose main importance is to highlight an unexpected finding that should be investigated in more cases. Since these are not the product of a random sample, the proposal here is that all the cases involving democratic states be systematically analysed to see if this finding is sustained. This study has started that process by looking at the 19 th century democracies.
With these serious caveats in mind, the study can be informative in that the causal logic of the institutional explanation is inspected in three of the most democratic, nineteenth century states in situations that came close to or went to war. The three cases are France during the Eastern Crisis of 1839-1841, Britain during the Crimean War Crisis of 1853-1854, and the United States during the Spanish American War of 1898. If the institutional explanation is correct it would be expected that where war is avoided the legislature would act to restrain the executive and where war occurs the executive has its way with the legislature. Likewise, it would be expected that the "public" would be hostile to going to war. It must be kept in mind, however, that before survey research it is difficult to talk about public opinion. Typically, the elite press is taken as an indicator of it, but this is clearly not the same thing.

The 1839-1841 Eastern Crisis
By 1839 France has taken important moves toward being democratic. In the 1830 revolution, the legislature overturned the Bourbon King Louis XVIII that had been restored by the victors of the Napoleonic Wars. The King was accused of trying to rig the election to the Chamber of Deputies and was removed. A new King was then elected by the legislature, something unheard of. They turned to the rival Orleans family and elected Louis Phillipe, henceforth known as the "Citizen King." The new King was not a constitutional monarch by any means, but had been elected. France had other characteristics of a democratic state, namely a vigorous independent legislature that was elected in free elections with competitive parties running candidates. Polity gives it a score of -1 and characterizes it as a closed anocracy, but by 1870 at the start of the Franco-Prussian War it was characterized as "in transition" to democracy.
Much of this crisis involves a conflict between the Sultan of the Ottoman Empire and his rebellious vassal, Muhammad Ali, the powerful Pasha of Egypt. The crisis begins in May 1838 with Muhammad Ali stating that he will declare his independence because he is not being permitted to have hereditary rule over Syria. About a year later in April 1839 because of this rebellion, Sultan Mahmud II attacks Ali's army in Syria. By the end of June the Ottoman army is defeated. The Sultan dies of natural causes and is replaced by the sixteen-year old Abdul Medjid. About a week later the Ottoman navy defects to Ali.
With this the Concert of Europe, which had been concerned from the beginning, moves to intervene. Led by Lord Palmerston, the Foreign Secretary of Britain, an agreement among Britain, Austria, Russia, and Prussia is reached to support the Sultan with military force. France, which has been supporting Ali and encouraging the Concert to meet more of his demands, including hereditary rule over Syria, is not consulted about this particular action. The four states issue an ultimatum to Ali on August 16, 1840 (Richardson 1994: 47): 1. He is offered rule over Egypt and Acre for life.
2. If he does not accept this in ten days, then the offer on Acre is withdrawn.
3. In another ten days the Egyptian offer will be withdrawn and the British and Austrian navies will attack.
Ali does not reply and on September 9-10 British and Austrian naval forces bombard Beirut, and subsequently defeat his forces. The ultimatum and military action outrages Thiers, the Prime Minister of France, as well as King Louis Phillipe and the nation as a whole. France had not been consulted and now its ally lay in defeat. At first, Thiers thinks of sending the French fleet, but there is not enough time and the fleet is probably no match for the British fleet. Instead, he threatens war in the Rhineland, a vulnerable area where he has a military advantage and that had been formerly part of France.
The threat raises the stakes considerably. Germans everywhere react vehemently and nationalism on both sides of the Rhine is inflamed. A war between the poets ensues, (Richardson 1994: 48) while Prussia is now brought in centrally to the crisis with Austria at its side. Thiers takes military measures that escalates tensions.
The crisis reaches a height when Thiers writes a provocative and highly nationalistic speech for the King to deliver to the Chamber of Deputies. Given the emotion in the nation and the Chamber of Deputies, the King fears that if he gives such a speech, it would inflame the situation and lead to a rapidly spreading war fever that could not be held in check. He feels that he and the nation would be entrapped and there would be no turning back from going down the road to war. The King refuses to give the speech. This is the turning point of the crisis. Louis Phillipe has reined in Thiers, and he resigns, being replaced by Guizot, who follows the line of the King.
Louis Phillipe, although initially incensed by the action of his Concert allies, is not prepared to leave the Concert as would have happened with war on the Rhineland. The idea that France would again face a Grand Coalition of Prussia, Austria, Britain, and Russia over the goals of Muhammed Ali is not something he sees as in the national interest of France.
From the point of view of assessing the institutional explanation of peace, why is war avoided? Instead of the legislature restraining the executive the opposite happened. War is avoided because the accommodationist King reins in a hard-line Prime Minister. The legislature is gripped by a war fever as is the "public" at large. This is not consistent with the pacific portrait of the people painted by Kant and Tolstoy. The evidence of this case is just the opposite of what would be expected by the institutional explanation of democratic peace. Instead of insuring peace, democracy and its institutions actually works more like Morgenthau's (1978: 551-555 and especially 558) picture of democratic publics as moralistic and easily gripped by war fever. The 1839-1841 Eastern Crisis must be seen as an anomaly for the democratic peace.

The Crimean War 1853-1854
England is the second nineteenth century democratic state to be examined. Again, while it does not reach a Polity score of 6, it is one of the more democratic states of the period. It has Parliament that is freely elected in periodic elections with organized political parties, and unlike France, it has a constitutional monarchy. It also has a vigorous and free press. What it does not have is universal male suffrage. As a result it has a Polity score of 3 and is characterized as an open anocracy. Twenty-nine years later in 1882 when it intervenes in Egypt, it has a Polity score of 7, above the minimum conventional democratic level of 6.
The crisis that leads to the war is started by Napoleon III, new self-made Emperor of France in December 1852. He intervenes in a dispute over the Holy Places between the Greek Orthodox and Roman Catholic priests. He demands from the Sultan of the Ottoman Empire the right to protect the Latin clergy, and to back up his demands, he sends the war ship Charlemagne through the Straits in violation of the 1841 Convention that prohibits warships from going through the Straits while the Sultan is at peace. Napoleon III takes this action unilaterally without consulting the Concert of Europe. The action also challenges Russia, which has traditionally regarded itself as protector of Christians in the Ottoman Empire. It is also seen by Tsar Nicholas II as an intrusion into what he regards as his area of special interests. Insulted, the Tsar sends a special envoy, Menshikov, on the warship Thunderer, to Constantinople to make his own demands to reconstitute the Tsar's traditional rights, along with some secret clauses that will increase his influence at the Porte. Robert Schuman Centre for Advanced Studies Working Papers Menshikov arrives in early February, but the Turks drag out negotiations until the British and French ambassadors-Stratford de Redcliffe and De La Cour-arrive to give their advice, which stiffens the Sultan's resistance. After giving an ultimatum to the Sultan on May 5 he demands that it be signed sign in five days (Richardson 1994: 73), and when it is ultimately is rejected on May 17, Menshikov leaves. The Tsar then becomes intransient and sends another ultimatum indicating that he will occupy the provinces of Moldavia and Wallachia if the Sultan does not comply. With this the British Cabinet debates whether to send in the fleet. Palmerston, Home Secretary at the time, argues for sending the fleet through the Straits up to the Bosporus, which Aberdeen, the Prime Minster opposes. It is agreed that the fleet will go to Besika Bay, just outside the Straits. 1 In early July, the Russians occupy the two provinces, which leads to rioting in Constantinople and calls for war.
In the interim, the members of the Concert of Europe led by Austria try to mediate the dispute. Eventually, they come up with the Vienna Note that is accepted by all the major states, including the Tsar on August 5. It provides a solution that satisfies the interests of all the major states and makes a number of concessions to the Porte.
Stratford is ordered to get the Sultan to agree to the Vienna Note. He argues that he did, but many think that he also led the Sultan to believe that the British (or at least he as their ambassador) would support him if he did not accept the Note. In the end, the Sultan offers amendments. The Tsar is even willing to accept these, which indicates the extent to which he is willing to avoid war and end the crisis. However, Nesselrode, his foreign minister and chancellor, objects saying it would be beneath the Tsar's dignity to accept amendments from a minor state. Not to be outdone, the Austrians meet with the Tsar at Olmutz and come up with the Olmultz proposals whereby they make the amendments the Porte wanted. This should have ended the matter, but when the Concert's representatives meet again, the British Foreign Secretary, Clarendon, who had previously accepted the Vienna Note, now rejects the Olmutz Proposals, saying accepting them 'would render the position of Britain and France 'false and embarrassing'" (Richardson 1994: 77).
Clarendon's decision reflects not only his own doubts about the process that ensued after the Turks made amendments, but also the growing opinion within Britain against Russia and in favor of the Ottomans, who were portrayed as a small country being bullied by an aggressive major power. Cartoons in the Press often portray Russia as a blood-thirsty bear and the Turks as an innocent victim. Meanwhile, Palmerston drums up support within Parliament for more hard-line actions that he consistently advocates.
This sentiment in London, coupled with the active support of Stratford and de La Cour at Constantinople, emboldens the Porte. With domestic rioting over the occupation of the Moldavia and Wallachia demanding war, the Sultan declares war on October 4, 1853. Such a declaration by a minor state against a much stronger neighbor and a major state in the system, who in turn had been defeated it in a series of wars, most recently in 1828-1829, makes little sense from a rational choice perspective. It can be explained, however, as Buneo de Mesquita (1981: 70-71) does, by the calculation that the Sultan believed he would get substantial support from Britain and France, not an unreasonable assumption given the actions of the two ambassadors at the time. Such support provides the possibility of fighting a war against a long-term powerful rival and reversing some of the territorial losses going back to Peter the Great and Catherine the Great.
Turkey follows up the declaration with military action on October 23 by crossing the Danube at Kalafar. Russia meets this attack, but more importantly, responds with naval action against the Turkish fleet harbored at Sinope and sinks it, killing a number of sailors on November 30. Now there is 1 Normally the Home Secretary could not have much influence on foreign policy, but Palmerston was able to do so because he had previously been Foreign Secretary, during the Eastern Crisis, 1839-841 and most recently from 1846-1851. In addition, he had his own independent power base in Parliament in the coalition cabinet headed up by Aberdeen. tremendous pressure for war. France officially calls for it. Sinope creates hysteria in Britain. The Press demands war and calls England weak (Richardson 1994: 79). Palmerston resigns when war fails to be declared. Much of Parliament, especially that following Palmerston is up in arms. Aberdeen, the Prime Minister, is still opposed to war. An accommodationist throughout the crisis he refuses to be driven into a war and resists the hard-line actions being proposed by Palmerston and others (Richardson 1994: 84). His position is that reasonable demands by Russia should be met and if necessary Turkey should be coerced to accept them (Richardson 1994: 83). He sees Russia as having moderate aims and willing to abide by the Concert's recommendations (Richardson 1994: 83).
Palmerston's position is, in contrast, guided by a more geopolitical stance that he has held for a long time. He said, "if we do not stop the Russians on the Danube, we shall have to stop them on the Indus" (quoted in Richardson 1994: 83). In a memorandum he sends to the Cabinet on March 19, a few days before Britain and France declare war on Russia, he outlines his "beau ideal" and the war aims that should guide Britain: "Aaland and Finland restored to Sweden. Some of the German provinces of Russia on the Baltic ceded to Prussia. A substantive Kingdom of Poland re-established as a barrier between Germany and Russia. Wallachia and Moldavia and the mouth of the Danube given to Austria. Lombardy and Venice set free from Austrian rule and either made independent States or incorporated with Piedmont. The Crimea, Circassia and Georgia wrested from Russia, the Crimea and Georgia given to Turkey and Circassia either independent or connected with the Sultan as Suzerain. Such results it is true could be accomplished only by a combination of Sweden, Prussia and Austria with England, France and Turkey, and such result presuppose great defeats of Russia. But such results are not impossible and should not be wholly discarded from our thoughts" (quoted in Rich 1992: 113).
Such a vision shows that Palmerston, although clearly wanting to defend Turkey, has ambitions way beyond this and wants to weaken Russia as a rival to Britain by dismembering it.
Aberdeen resists the wave as much as he can. He writes in a private letter to Gladstone: The Turks with all their barbarism are cunning enough and see clearly the advantages of their situation.
Step by step they have drawn us into a position in which we are more or less committed to their support (quoted in Rich 1992: 111).
While he refers to the Turks, this quote actually applies to his own domestic hard-liners-Palmerston and Clarendon in the Cabinet, and Stratford on the ground.
Many analysts and historians consider the Crimean War an unnecessary war. Richardson (1994: 103-104) certainly does. The case for it being an unnecessary war is set out in detail in Rich (1985). Kissinger (1994: 92) in his sweeping book on Diplomacy, says: "…the Crimean War long condemned by historians as a senseless and utterly avoidable affair…" The Vienna Note satisfied all the major states and certainly the Olmutz Proposals should have satisfied the Sultan, and if he persisted in opposing peace, the British should have been able to coerce him by disabusing him of any idea that they and the French would go to war with Russia over negotiable disagreements. Robert Schuman Centre for Advanced Studies Working Papers If war could have been avoided, why does it occur? Why does Britain end up going to war when the Prime Minister is against it? Britain gets caught up in an elite-press-pro-hardline trap that makes it unable to pressure a weaker minor state that needed Britain to intervene to win the war, thereby making it miss two opportunities for peace. Figure 1 reprints a cartoon from Punch in 1854 that captures the prevailing press sentiment. It shows Aberdeen strenuously restraining the outraged British Lion that is trying to pursue the aggressive Russian Bear. Even though Aberdeen resists the war sentiment after Palmerston resigns, he eventually becomes isolated and fearing that his fragile coalition would fall apart, he goes along with the tidal wave.
What does this case tell us about the institutional explanation of peace? The evidence is in the opposite direction of the expectations of the monadic democratic peace. It is the executive that wants to avoid war, which makes it accommodationist, while it is Parliament and the press that is hard-line and belligerent. So in this case that goes to war, it is not the legislature that is overcome by the executive, but the opposite--Parliament and a war hysterical Press, claiming to represent the public, push an accommodationist Prime Minster into war, which many in retrospect consider unnecessary.
We have now looked at two cases, one where war is avoided and one where war occurs. Both are anomalies for the monadic democratic peace. Although they are only two cases, the underlying causal logic of each case consistently undermines the expected causal logic of the democratic peace. We turn now to a third nineteenth century democratic state-the United States in 1898.

The Spanish-American War 1898
Of the three liberal states in this study, the United States is the only one to reach a Polity score of 10. Certainly in this case there can be no explaining away of the findings on the basis that the "democracies" in questions are not real democracies because they do not have a Polity score above 6. In fact, early on many thought this was a war between two democracies since Spain had a Polity score above 6. David Lake (1992) was the most prominent scholar who made this assertion. James Lee Ray (1995: Ch. 5), however, in his extensive analysis of exceptions to the claim (see Miriam Fendius Elman 1997) that joint democracies never fight each other, dismissed Spain as a true democracy and that position has emerged as the general consensus. Whether this position is accurate is not of concern to this analysis, since this study is focused on whether a monadic democracy in a war has a legislature, press, and public trying to restrain an executive bent on war. The origins of Spanish American War lay with the Cuban insurrection against Spain, the harsh Spanish response that in turn raised sympathy in the U.S., and the moral concern among the American public and the press. By 1895 the insurrection captures widespread attention, and there is pressure in Congress and the press to do something. In February 1896, under the Presidency of Grover Cleveland, a Democrat, both the House and the Senate pass a resolution recognizing the Cuban belligerency. As pressure mounts for war, a Congressional delegation comes to the White House and tells the President that Congress is prepared to declare war. Cleveland replies, "There will be no war with Spain over Cuba while I am President." A member of Congress says that the Constitution gives Congress the power to declare war. Then Cleveland says the Constitution makes him Commander-in-Chief and he states, "I will not mobilize the troops." 2 That was the end of the matter for the rest of his Presidency.
Things stand as they are, until the election of a new President, William McKinley, a Republican. McKinley, however, is also opposed to going to war with Spain over Cuba. Nonetheless, pressure mounts in the Hearst press and that of Pulizer for war. This reaches a crescendo with the sinking of the U.S.S. Maine in Havana harbor. Many have argued that the pressure for war came primarily from these newspapers, but Ray (1995: 182) argues that the "yellow journalism" was an important factor in generating war sentiment, and that public opinion nationwide constituted a separate factor pressuring McKinley. Ray (1995: 182) cites several sources to this effect and quotes Trask (1981: xiii) that after the sinking of the U.S.S. Maine "popular insistence on immediate action…grew by leaps and bounds." On February 15, 1898 the Maine explodes and sinks, killing 260 American sailors (Small 1980: 134). The reason for the explosion is not discovered but Spain is widely blamed in the press.''Remember the Maine! To hell with Spain!'' becomes a popular slogan reflecting the mounting pressure for war. Even Republicans are among the jingoes. For instance, Bailey (1968: 460) reports that the bellicose Theodore Roosevelt is reported to have shouted, 'McKinley has no more backbone than a chocolate éclair!'" Fearing a loss at the polls (Bailey 1968: 461), enough Republicans begin to call for war that it seems that Congress might declare war without the President asking for it (Small 1980: 139). Another concern is that McKinley's likely 1900 Democrat opponent, Bryan, is adopting a "Free Cuba" platform to go along with his Free Silver platform of the last election (Bailey 1968: 460). According to Ray (1995: 181, citing Trask 1981, McKinley chooses war because he fears an electoral loss to the Democrats. Generally, the only sizeable segment of the public that is against going to war is the business community (see Ray 1995: 184-185), part of the natural base of the Republican Party.
Meanwhile, hard liners in Congress are taking action. On March 9, Congress "unanimously votes $50 million for war preparations" (Bailey 1968: 457). On March 17, Senator Proctor, highly respected and far from a jingoist (as Small 1908: 140 puts it), speaks to the Senate as to what he had seen on his tour of Cuba and says something had to be done to end Spanish repression of the Cubans. This pushes many of those who had been on the fence over to the war party.
McKinley is often seen as unable to hold back the Congress or the American people. Such a sentiment is reflected in a New York Journal's political cartoon. Here McKinley is seen as trying to sweep back a roaring sea of popular sentiment in favor of war. McKinley, who is concerned about elections (see Ray 1995: 181), relents and condemns Spain in a message to Congress on April 11 requesting powers to take action that would end the crisis in favor of the Cubans. McKinley orders a naval blockade and on April 24 Spain issues a Declaration of War. On April 25 Congress declares that a state of war had existed with Spain since April 21.
In terms of the institutional explanation of the democratic peace, once again its theoretical expectations are contradicted. Instead of democratic institutions restraining the President from going to war, these institutions push a reluctant President to go to war. This pattern holds for both a Democratic and a Republican President. The Congress, the press in particular, and the public are all driven by a 2 This anecdote along with the quotes are taken from Fisher (2013: 52). My thanks to Marie T. Henehan for bringing this incident and the cartoon to my attention. Robert Schuman Centre for Advanced Studies Working Papers moral repugnance to the events in Cuba. With the sinking of the Maine this becomes a war fever. The events leading up to the war contradict the causal process set out by the institutional explanation. Ray (1995: 182) concludes, "…the results of our [study] with respect to the role of public opinion do not support the democratic peace proposition." Instead of the legislature restraining the executive, we see the opposite. Ray (1995: 181) states, "Virtually all sources suggest that public opinion was overwhelming in favor of war with Spain and in fact was a key factor pushing reluctant leading policymakers into the conflict." The Three Cases-What Can They Tell Us?
The three cases tell us in each instance that the institutional explanation is not working in the ways one would expect from the perspective of the democratic peace. Democratic institutions do not work to restrain the executive or avoid war. In fact it seems that democracy is actually a factor in war hysteria and pressuring the executive into war.
All three of the cases reflect the same pattern-it is the executive, not the legislature, the press, or the public that is reluctant to go to war. In the one case that does not go to war-the 1839-1841 Eastern Crisis--the King (Louis-Philippe) reins in his Premier (Thiers) who is threatening war. He resigns and French foreign policy changes so war is avoided. In the Crimean War, Prime Minister Aberdeen consistently tries to avoid war and hard-line actions, but "step by step" is manipulated into supporting a decision to go to war by hard-liners in the Cabinet, Parliament, and the press, who are constantly rattling for war with a moralistic fervor. Lastly, McKinley faces a similar situation, but is not so much manipulated as he relents to the pressure to go to war exerted by the press, Congress, and the public.
Each of these cases provide no evidence to support the causal process outlined by the democratic peace that democratic institutions should place a restraint on the executive, which is seen as the most likely wanting war. Instead, the opposite hypothesis is supported that the public and the press is susceptible to moralistic sentiment that can lead to a war fever and then war hysteria.

Conclusion
The underlying logic of this study was to see if the specified causal processes of the monadic democratic peace held when specific cases were examined in terms of why they went to war or how war was avoided. The theoretical expectation is that war would be favored by the executive, and that if it was avoided, it would be because the legislature, the public and press put a restraint on that branch of government.
The study consisted of one case where war was avoided and two where war occurred. The cases involved the three major states that were most democratic in the nineteenth century. There are two main caveats of the study. The first is that there are only three cases. The second is this is not a random sample, but a set of serendipitous findings from a larger study of crisis diplomacy. These caveats limit the generalizability of the cases and will be discussed in more detailed at the end.
Despite these serious caveats, the study makes two contributions. First, a close examination of three cases to see if causal logic of institutional explanation is working shows that it is not. Hardly any studies have examined cases of the monadic democratic peace to see if the underlying causal processes work the way they are supposed to work. These cases find that they do not. Second, the findings suggest two new hypotheses: first, that democracies are prone to war hysteria that can lead to war, and second that the executive is more likely to restrain the legislature than vice versa.
In the end, however, these are only three cases. Before the results can be accepted a systematic study of all the relevant cases needs to be conducted. Such a study will be able to tell us whether these are truly anomalies or in fact the predominant pattern as implied by the suggested new hypotheses.
Future research should be of two sorts-first an examination of cases of democracies involved in wars to see whether it is the executive or the legislature that is primarily responsible for pushing the state into war, and second an examination of cases where democracies successfully avoid war to see whether it is the public using the legislature and press to restrain a bellicose executive or vice versa. The first set of cases are easily derived from existing data. The sample that would be most damaging to the spirit of the democratic peace would be those cases where democratic cases actually initiated war. These cases are listed in the Appendix. As one can see there are a considerable number, which is unexpected. Of course before these can be accepted as anomalies the cases must be closely examined to see if the initiator is actually responsible for the war or is drawn in by its adversary. Also in terms of the causal logic of the institutional explanation it must be determined whether it is the executive or the legislature and public that pushes the democracies into war.
The second sample on avoiding war is more difficult to research because data would need to be collected. It is suggested here that a sample of serious crisis (or MIDs) be collected that involve democracies. The ICB (International Crisis Behavior) data is an obvious source, although it only goes back to 1919. Once the sample is drawn, then in depth case studies could be conducted to assess who avoided war-the legislature or the executive-as was done herein in the Eastern Crisis of 1839-1841. This discussion of future research reflects a final contribution of this study; namely, that there is a robust role for qualitative case studies in assessing the causal processes underlying the institutional explanation of war. Robert Schuman Centre for Advanced Studies Working Papers