Phil Taylor's papers
|
BACK TO : Public Opinion
When the Going Gets Rough: Does the Public Support the Use of Military Force? by P Everts 
When the Going Gets Rough: Does the Public Support the Use of Military Force? World Affairs, Wntr, 2000, by Philip Everts When democracy was admitted to, or rather forced itself upon the battlefield, war ceased to be a gentleman's game," the young Winston Churchill wrote--probably with some regret--in the diary that he kept as a war correspondent in the Boer War in South Africa.(1) More recently, it has not been the pressure of public opinion and the massive protest against nuclear weapons of the early 1980s that have forced governments to be defensive but the contradictory demands of the same public around issues such as the conflicts in the former Yugoslavia, UN peacekeeping in general, and the use of force to punish Iraq for its noncompliance with international demands. Governments are under constant public pressure, on one hand, to do something or to intervene in situations where fundamental human fights are being violated, but on the other, they seem to have reason to fear that their citizens will not have the endurance to suffer the consequences and the casualties that action might require, thus forcing them to make U-turns in their policies, which may detract from their credibility. The Boer War (also known at the time as la guerre de la Bourse contre les Boers) was a good example of the popular forces that may constrain the hands of governments in modern societies, but it was not the first. After the French Revolution introduced compulsory military service and the mass army, and as the industrial potential of nations became a primary source of power, citizens became directly involved in the preparation and conduct of war. In the nineteenth century, the major ideologies of nationalism, liberalism, and socialism--from which we still suffer today--also made their appearance, as a consequence of which civilians and their beliefs became legitimate targets in wartime. Since then, war has been democratized both passively and actively, and this has made the relationship between war and democracy, between the use of military force and public support for such use, into a problem of the first order of magnitude. Although much less frequent and much changed in character, war is still considered to be the ultimum remedium in international affairs. Warfare is not possible, however, when taxpayers are not willing to foot the bill for maintaining armies, and people (men in particular) are not prepared on a large scale to risk their own lives as well as to kill unknown others on command. (Remember the slogan of the German peace movement: "Stell dir vor, es gabe Krieg und keiner ginge hin" [Imagine there was a war and no one went].) VOX POPULI The public is always involved in wars--their preparation, conduct, or prevention--in one way or another, as participants or observers. That fact makes public opinion a major factor in understanding foreign policy and international politics. Sometimes beginning a war is very popular. In 1914, World War I, the troops went singing to the front carrying flowers bestowed on them by the elated public, the Germans to seek the place under the sun that was due them, the French to take revenge for the defeat of 1870. But that is not the only example of war frenzy. Others include the Spanish-American War of 1898, which was more or less forced on a reluctant American government by an aroused public opinion, and the more recent example of the (initial) frenzy of the people on both sides in the Falklands/Malvinas war of 1982. At the beginning of the Second World War, on the other hand, there was no war enthusiasm, at least not on the Allied side. Not a nationalist fervor, but indignation and ideological enmity in the case of Germany and racism in the case of Japan were decisive, providing the arguments to dehumanize the opponent, to insist on unconditional surrender, and to smother the bad conscience of those who questioned whether killing tens of thousands of innocent civilians in Dresden and Hiroshima could ever be justified. One other element in the "ideologization" of modern war was that civilians working in the arms factories were increasingly seen as targets that were as legitimate as soldiers on the battlefield, a fact that led to the indiscriminate bombing policies of World War II. The elimination of the traditional distinction between combatants and noncombatants set a precedent for the ideological and nationalist conflicts in the period since 1945: the wars of liberation from colonialism in Africa and Asia, Vietnam, and the Gulf wars. In ideological battles there are no innocent bystanders. The importance of ideology in overcoming the resistance of public opinion to war was shown particularly in the cold war, and in yet another way. Intensive ideological education and propaganda were necessary to get and maintain public acceptance for the strategy of nuclear deterrence and the accompanying armaments. How could the threat of total destruction of a civilian population ever be justified if the opponent did not represent the incarnation of evil ("the evil empire" as President Reagan said)? The Vietnam War also sharpened our awareness of the significance of public support on the home front for the ability of countries to carry on a war in yet a different way: by showing the difficulty of sustaining such support under the condition of direct and full media coverage. More than anything, the importance of public opinion, positive or negative, was illustrated in the Gulf War of 1991. To raise and maintain public support in the democratic countries that had joined the allied coalition, information was roughly manipulated, first by portraying Saddam Hussein, the West's good friend of earlier years, as the new Adolf Hitler and later by presenting the war itself as no more than a series of sanitary "surgical strikes" carrying no risk of bloodshed. The live television broadcasting of the war was a new phenomenon requiring drastic censorship. Because the war ended in a very short time, and successfully--at least for the members of the international coalition--the popularity of the American government and other governments involved increased rapidly.(2) But it was a short-term effect: President Bush was not reelected and little was heard afterward of the New World Order he had so triumphantly announced at the time. ZONES OF PEACE AND ZONES OF WAR Since the end of the cold war it has become rather popular to portray a general movement in the world toward a dichotomous situation in which there is a "zone of peace" and a "zone of turmoil."(3) Leaving aside the question of whether this is an accurate or overly ideological image, it is clear that this division also has important consequences for public attitudes on war and peace. It is argued that history as the straggle between ideas has come to an end, in the words of Fukuyama. With the victory of liberal democracy and the end of ideology, wars come to be seen as unlikely, as things of a horrid past, and the idea of engaging in actions in which lives could be at stake becomes ever more difficult to argue.(4) The peace movement appears to have had more success in this connection than it ever anticipated. The public in the zone of peace, horrified by violations of human rights, may clamor for action, but the military will say that they were not hired to fight "in faraway countries of which we know virtually nothing" (to use Chamberlain's phrase of 1938). If the ministers of defense of the countries concerned send troops into action anyway, they stress that the risks are acceptable in that no one will be killed, rather than that casualties will be unavoidable and the price to be paid for a worthy cause. These actions usually take the military outside the zone of peace and into the zone of turmoil, where conflicts continue to be seen as a fact of life, and the public's willingness to bear the consequences has been less affected, and where people are not asked for their opinions anyway. In most cases, the zones of turmoil are in countries or societies in which the constraining effects of democratic systems are not at work. After the initial enthusiasm of the early post-cold war years, the willingness of the international community to intervene in the zone of turmoil and carry the accompanying burden seems on the decrease, however. The policy of neointerventionism of the early post-cold war period may turn out to have been a very short-lived affair. The idea seems rather to have become that international efforts to contain or reduce violence are inherently risky, that little can be done anyway, and that the people concerned should be left to fend for themselves. The people of Bosnia and Kosovo and of the area of the Great Lakes in Africa were the first to suffer the consequences of that new reluctance. There has been little research into this neopacifism, its sources and implications for foreign and security policy. Singer and Wildavsky argue, among other things, that one of the causes of the alleged unwillingness to fight may be that "people in such wealthy economies are shielded from extensive experience with the violence of nature and from sustained physical adversity and hardship."(5) Fearing negative public reactions, governments in Western democracies have moved quickly to a policy of risk avoidance. When they claim or demand strong and decisive action internationally, they often do so in the safe knowledge that they will not be asked to honor such commitments for lack of support by other states. A POLICY OF RISK AVOIDANCE The first American president to be faced with this new climate of opinion, that is, a reluctance to take risks in humanitarian interventions, was Bill Clinton. He had great difficulties in enlisting the country's support for new military engagements, such as those in Somalia, Haiti, Bosnia, and later Kosovo, which were perceived as risky and not in the vital American interest. This became evident in the case of the conflict in Bosnia-Herzegovina. During his election campaign Clinton stressed that the United States should not tolerate aggression and acts of "ethnic cleansing," but once elected, he recoiled at the prospect of having to commit ground troops with the concomitant danger of entrapment and a domestic backlash. The fear of casualties, which rapidly became a major consideration, could also take a different and opposite form, as in the new Powell doctrine, which states that if the question of engaging American military forces arises, the United States should either stay out or enter with full force to guarantee few casualties and success before public fatigue has a chance to set in. That doctrine seemed a neat summary of the lessons that were drawn by many in the military from the conflicts in Vietnam, the Gulf, and Somalia, the main argument being that the American people would not sustain support for long and risky military operations. Indeed, military actions with the level of killing of World War II seem inconceivable today. One remarkable aspect of the new developments is that the terms of the debate seem to have shifted considerably in this respect, and roles have been reversed, as new, tacit, ideological alliances have come into being. The debate used to focus on the opposition between those who accepted war as the minor evil and, on the other side, pacifists, both nuclear and total. Today, there is a new debate with new participants: On the one side there are the neointerventionists, those who emphasize the duty not to stand idly by when justice and vital human rights are being violated, even on the other side of the globe. A part of the traditional peace movement has joined this camp, which used to consist mainly of those who always saw a role for the military to act as the world's police force. On the other side there are not only the traditional pacifists with their morally based opposition to killing but also many among the military and their supporters, who could be called the neopacifists. The latter include those who have become skeptical of, or reject outright, the use of military force for internationalist goals. The result is a confusing situation. As one observer put it recently. The confused nature of many of the recent conflicts has resulted in a curious reversal of political roles. It is now the military that often claims that little can be done to douse conflicts around the world, while those who traditionally argued that force solves nothing in international relations demand instant military action.(6) THE CASUALTY HYPOTHESIS Public opinion on questions of foreign and defense policy--or what passes for it--usually has primarily a legitimizing function, and governments usually enjoy the freedom of a permissive consensus. They also face certain clear limitations with respect to their freedom to commit their armed forces, especially when the costs and risks involved are not small. In this connection, references are often made to the alleged existence of a new general phenomenon, the "body bag syndrome," a tendency for public opinion to recoil from calls to use force when the prospect of casualties appears. The "casualty hypothesis" states that public support for the use of military force in high-risk operations in circumstances other than direct threats to the country's security will not be forthcoming, and that initial support in such cases will dwindle as soon as casualties are incurred. For example, it has been said that "We [Americans] have grown ever more sensitive about casualties--our own military casualties, opponent and neutral civilian casualties, and even enemy military casualties--and we seek to avoid them."(7) Two events in particular may have contributed to the reluctance to enter into risky military operations, the killings of 241 American marines in Lebanon in 1983, and 19 Americans in Somalia in 1993 in hapless circumstances, both in military interventions in civil conflicts. The body bag syndrome became observable not only in the United States, where the memories of the Vietnam War were most vivid, but also in European countries, where the risks to life in the military had been largely theoretical during the cold war. Those governments began to worry also whether after the cold war there remained any support for the actual use of the armed forces beyond protection of immediate and narrowly defined national interests. Moreover, they questioned why they should be the ones to carry the risks in international military operations, or, as in Bosnia-Herzegovina, why they should be providing the troops on the ground, while the United States reserved for itself the sexier but safer role of demonstrating its air power. The "zero dead doctrine" is gaining popularity in many places.(8) In the case of participation by the Netherlands in UNPROFOR and other UN operations, Dutch politicians have regularly warned that casualties might lead to a sudden reversal in public support. One might ask, however, whether they were saying so out of democratic deference or because they were seeking an alibi for their own hesitations. Various reasons have been given for the phenomenon of increased risk avoidance for fear of casualties.(9) In the American case it is suggested that there is a declining willingness to invest lives in hegemonic power.(10) A more general, if possibly fanciful, explanation for the decline of big wars has been offered by Luttwak, who suggests that the reduced size of families makes it less attractive for poor families to contemplate, as in the past, shifting the burden of feeding one or more children to the army, with the concomitant likelihood that one or more sons would lose their lives in wars. Today's small families are also characterized by greater emotional investment in fewer children.(11) The argument is not convincing, however, if only because the decline in family size is not a recent phenomenon, dating as it does from the latter part of the nineteenth century. By 1914, large families, which, it might be argued, could "afford" to lose some sons, had already become the exception. It is much more likely that, apart from decreasing nationalism in many countries, the decline in the willingness to obey uncritically the demands of one's superiors is at play here. In spite of the occasional strikes and mutinies, what stands out today about World War I is the inexplicable willingness to accept the senseless suffering and slaughter of millions in that war. One important reason for that might be that early in this century the common man still usually "knew his place," and did as he was told. For those higher up, used to leadership in politics and wars, and who were not directly exposed to the risks of war, his life often did not count for much. That has greatly changed. Moreover, increased wealth and social changes have increased equality, especially since the 1960s, and mourir pour la patrie, let alone for another country, does not fit into today's pattern of expectation.(12) Another reason could be sought in the direct impact of the media on the framing of issues and policies. Wars are no longer waged far away but are visible in the living room through television. This increased visibility probably has a negative influence on willingness to sustain the consequences of wars. As the now-well-established thesis of the democratic peace has shown, it is probably also the consequence of a more democratic decision-making process and increased interaction between political and military elites. One other factor that clearly helps to shape attitudes toward the acceptability of casualties is the evolution of the mass armies of the first part of this century into "constabulary forces"(13) --armed forces that seek viable international relations rather than military victories over opponents. Interventions and crisis management are the catchwords used in this connection; there is an enormous difference between the two types.(14) Certainly relevant in this connection is the end of the draft (in those countries that had that system of recruitment) and the switch to all-volunteer forces. Usually, stricter norms are used with respect to the risks to which draftees can be exposed.(15) Finally, the technological profile of the armed forces tends to strengthen the belief that casualties are a thing of the past. Taken together, it is logical to assume that the changes in value patterns and lifestyles of post-modern societies, including changes between rulers and ruled and increased sensitivity to violence, have also affected the attitudes to, and the conduct of, international conflict and diminished the willingness to run risks in this respect. It remains debatable, however, which factors precisely determine public support for participation in high-risk military operations and for the use of armed forces in general, and in particular, whether, when, and to what extent public opinion will withstand the occurrence of casualties. THE EVIDENCE ON THE CASUALTY HYPOTHESIS Setting aside theoretical arguments, is there any empirical evidence for the casualty hypothesis? I will look briefly at the international literature and in some detail at the empirical evidence from one country in particular, the Netherlands. The most systematic effort to measure general support for the threat or use of military violence in the form of the Netherlands armed forces was a survey in 1990.(16) By sketching systematically a series of hypothetical situations, this particular survey tried to ascertain to what extent respondents supported and agreed to (a) threatening to use the armed forces or (b) actually using them "to secure national interests," such as safeguarding the supply of oil, and if so, whether they would stick to that if casualties were to fall, respectively, on their own or the opponent's side, among members of one's family, or if the respondent himself were to lose his life. The same set of questions was asked with respect to "the protection or restoration of international order" and the "safeguarding of the national independence." In that way six points of measurement were obtained for three different cases--eighteen in total.(17) Although an interview situation may be totally different from the question in real life, the answers to the questions nevertheless provide interesting insights on the phenomenon of general and yet case-specific decline of support for the use of force under varying conditions. The result is shown graphically in figure 1. [Figure 1 ILLUSTRATION OMITTED] When casualties occur, support drops off considerably. This is what one would expect; but more remarkable is the fact that even when the conditions are then made more stringent, support (or rather the lack of it) remains on the same level. Whether casualties are incurred on the other or one's own side does not seem to have a strong effect. That is understandable to some degree, but the figure also shows graphically the ambiguities in support for threatening the use of force or actually using it. In terms of absolute levels, the survey suggests that there would not even be majority support for defending the integrity of the country should it entail considerable casualties. Even in this case, support was reduced from two-thirds to about half or one-third respectively when the use of force was introduced in the question. This outcome is not unique to the Netherlands. Similar results were obtained, for instance, in Italy when respondents were presented with similar options with respects to an armed intervention in the war in Bosnia.(18) In the Netherlands in 1995, another effort was made to estimate the degree of support for military force and the use of the armed forces. Both the use of the armed forces under varying circumstances and the willingness to sustain casualties were explored.(19) Table 1 shows the results from a 1995 survey of the public's willingness to endure casualties. Respondents were asked if they thought it could be justifiable to (a) use military force against another country, (b) use military force when a great number of soldiers may be killed or wounded, and (c) use military force when a great number of civilians could be killed or wounded. As in the previous survey, the prospect of casualties immediately reduces the support for the use of force, but the general level of this support remains much higher than the data in figure 1, from 1990.(20) These polls addressed the use of force and the problem of casualties in general, but one may wonder whether the outcomes are different in specific cases. TABLE 1 Public Acceptance of the Use of Military Force Yes (%) No (%) Is military force justifiable? 58 25 Is military force justifiable if military casualties are possible? 52 30 Is military force justifiable if civilian casualties are possible? 46 38 Do Not Know/ Did Not Answer (%) Is military force justifiable? 17 Is military force justifiable if military casualties are possible? 18 Is military force justifiable if civilian casualties are possible? 16 Source: Studiecentrum Vredesvraagstukken, Catholic University of Nymegen, 1995. THE CASE OF BOSNIA A great number of surveys, in the Netherlands and elsewhere, have explored public support for intervention in the civil war in Bosnia. Most showed considerable support for such intervention if not outright pressure to "do something."(21) Although some surveys conducted in the Netherlands suggested a higher degree of support for military force in the case of Bosnia, compared with support in general, other data suggest the opposite.(22) One survey in 1995 suggested much less support for using the Dutch military forces in operations to enforce a settlement in Bosnia-Herzegovina than previous surveys had suggested. It found 47 percent agreeing to commit Dutch troops in combat operations to maintain a peace settlement (less than for using troops for humanitarian operations), while 32 percent disagreed, and 21 percent did not answer the question.(23) The hypothetical acceptability of casualties was also explored in this survey. No less than 42 percent of all respondents agreed that an action such as the one envisaged in Bosnia would not justify a single Dutch military casualty; 22 percent were prepared to accept up to ten casualties, and only the remaining one-third of the respondents were willing to accept a higher number if necessary. That outcome was more pronounced among women (with percentages of 47 and 17 percent, respectively). A question concerning willingness to accept casualties if one's own child or partner should be involved confirmed this outcome. One should not forget, however, the hypothetical character of the question, which is reflected in the high number of nonrespondents (26 percent). Again, those outcomes seem to be fairly representative for other countries as well.(24) The outcomes suggest considerable support for the existence of a body bag syndrome, a conclusion that seems logical. If that effect occurs in cases in which national security could be said to be involved, one should expect it when the national interest in the narrow sense is less directly involved, such as in UN peacekeeping operations. The participation of the Netherlands in UNPROFOR, the United Nations force charged with keeping the peace in Bosnia-Herzegovina, would seem to provide a good test case for this proposition. Although the Dutch had initially been strongly supportive of international intervention in Bosnia and of participating in UNPROFOR (the Netherlands had taken part with a fairly large contingent of some three thousand men since 1993), support began to diminish as the conflict dragged on. It eroded considerably in the course of 1994. In one year the number of supporters of Dutch participation decreased by 14 percent (see table 2). TABLE 2 The UN Operations in the Former Yugoslavia (%) Aug. April Dec. May Dec. 1992 1993 1993 1994 1994 Do you support participation in UNPROFOR? Yes 87 88 68 -- 54 Neither agree no disagree -- -- 11 -- 15 No 8 8 14 -- 26 Do not know/ no answer 5 4 8 -- 6 Should the UN stay in Bosnia or leave? Stay -- -- -- -- 44 Leave -- -- -- -- 42 Do not know/ no answer -- -- -- -- 14 What is the UN's best option in Bosnia? Continue present course, with emphasis on negotiations -- -- -- -- 31 Withdraw and leave ito to the parties themselves -- -- -- -- 20 Continue air strikes to enforce peace -- -- -- -- 17 Send ground troops to enforce peace -- -- -- 16 -- June 12 July Aug. Dec. I Dec. II 1995 1995 1995 1995 1995 Do you support participation in UNPROFOR? Yes 44 -- 62 69 71 Neither agree no disagree 22 -- 10 -- -- No 26 -- 18 12 12 Do not know/ no answer 12 -- 11 20 17 Should the UN stay in Bosnia or leave? Stay 36 33 50 -- -- Leave 38 57 28 -- -- Do not know/ no answer 26 10 23 -- -- What is the UN's best option in Bosnia? Continue present course, with emphasis on negotiations 30 20 -- -- -- Withdraw and leave ito to the parties themselves 31 33 -- -- -- Continue air strikes to enforce peace -- -- -- -- -- Send ground troops to enforce peace -- -- -- -- -- Source: August 1992, April 1993: AVRO/NIPO; December 1993, December 1994, June and December-I 1995: NIPO for Stichting Maatschappij en Krijgsmacht; May 1994: international comparative survey in the European Union by Harris Research Center; July 1995: Intomart for RTL television; August 1995: NIPO for Studiecentrum Vredesvraagstukken Nijmegen a.o.; December-II: NOS. By the end of 1994, it seemed that the patience and endurance to continue with military operations that were seen as risky, militarily senseless, and unlikely to succeed had begun to diminish rapidly. Naturally, that also affected the willingness to run military risks. Whereas at the end of 1993, 57 percent had felt the risks to be acceptable, the number had declined to 47 percent by the end of 1994 (see table 3).(25) TABLE 3 Tolerance of Risks (Casualties) with Respect to Intervention in Former Yugoslavia (%) Aug. Dec. April Dec. Dec. 1992 1992 1993 1993 1994 Risks/casualties acceptable Yes 76 66 80 57 48 No 17 20 14 29 39 Do not know/no answer 7 14 6 14 13 June July Sept. Dec. 1995 1995 1995 1995 Risks/casualties acceptable Yes 34 30 55 54 No 47 64 23 33 Do not know/no answer 19 6 22 13 Source: August 1992, April 1993: AVRO/NIPO; December 1992, December 1993, December 1994, June 1995, December 1995: SMK/NIPO; September 1995: Telepanel/Marktonderzoek. On the other hand, there were few indications that the Dutch public actively wanted the UN troops, including the contingent from the Netherlands, to be withdrawn. That the public would not be willing to sustain casualties was increasingly argued at the time by politicians; others stressed that that was also the reason why other European countries were not ready to commit troops to confronting what was increasingly seen as Serbian aggression. As a matter of fact, views on staying or leaving remained much divided, at least until the end of 1994, reflecting a fundamental uncertainty about what ought to be done with respect to the conflict, rather than a form of cowardice. By June 1995, opinions had become even more divided. On one hand, the number of those who wanted to leave it to the parties to solve their conflict had increased (from 31 to 33 percent), but so had the number who supported military intervention (from 25 to 28 percent). Accordingly, the number of supporters of the third option--to muddle through--had diminished from 30 to 20 percent (see also table 2).(26) Support for continuing participation by the Netherlands armed forces in UNPROFOR, which had already declined from 68 percent at the end of 1993 to 54 percent at the end 1994, had declined even further by the end of June 1995, to 44 percent, after the taking of hostages among the UNPROFOR soldiers and unarmed observers (including a number of Dutch) had ended.(27) The number of opponents of participation had increased since 1993 from 14 to 26 percent at the end of 1994 and stayed at that level until June 1995. At that time, the number of doubters had increased, however, to 25 percent of all respondents.(28) The percentage of those who considered the risks involved to be still acceptable had arrived at a minimum of 34 percent acceptable and 47 percent unacceptable, with 19 percent not answering that question (see table 3), a true reversal of opinion, and one that was even more pronounced among women. Compared with the end of 1994, the balance had also shifted with respect to staying or leaving. For the first time, the proponents of leaving had gained the upper hand (38 versus 36 percent, with the remainder undecided). Again, the large number of doubters is remarkable.(29) Support for staying decreased with age. The failure of the numerous ceasefires in the first half of 1995, the renewal of the fighting, the Serbian hostage taking in May 1995, and the generally bleak outlook for any end to the war helped to further erode public support for maintaining the Dutch troops in Bosnia. To sum up the data presented above, as the war in Yugoslavia dragged on, public opinion became both more cynical and more polarized, with less support for UNPROFOR and more for withdrawal on one hand, but also with increased support for strong military action on the other. Risks were increasingly felt to be unacceptable. Mission support and tolerance for risks/casualties clearly go hand in hand to a certain extent.(30) How support for UNPROFOR correlated with the acceptance of risks is shown also in figure 2. [Figure 2 ILLUSTRATION OMITTED] There is a correlation both at the aggregate and the individual level between tolerance of casualties and perceived likelihood of success.(31) The tolerance of casualties is not an independent variable but seems mediated by expected success. Without further information, however, we cannot draw conclusions about causality. Nonacceptance of the risk of casualties may have led to less support for the mission, but the reverse may also be the case. In fact, that is the core of the casualty hypothesis. SREBRENICA The change in climate was further reinforced by the dramatic events of July when the Bosnian Serbs forced the Netherlands' contingent, Dutchbat, to evacuate and surrender Srebrenica, one of the areas that the United Nations had declared to be a safe haven. The Dutch were not capable militarily of mounting an effective defense (neither did they have the mandate to do so), and they received no air support. They were thus forced to surrender to overwhelming Serbian forces. The primary concern of Dutchbat indeed became its own safety, and all other goals were subordinated to that, which meant that the casualty syndrome was indeed operative. In consequence, thousands of Muslim inhabitants of the area disappeared or were killed. Although not a perfect test of the casualty hypothesis because only one soldier was killed during the action, the debacle of Srebrenica could be considered the moral equivalent of the incidence of casualties. A poll taken immediately after the events, on 11 July 1995, measured the impact of the events in Srebrenica.(32) Two-thirds of all respondents felt at that time that the Dutch risks in Bosnia were no longer acceptable, an increase of 20 percentage points in only a few weeks. Only 29 percent thought that the risks were still acceptable. Two out of three people (64 percent) also felt that it would be unacceptable for more Dutch soldiers to die while trying to protect the Bosnian population from the violence of the war.(33) A firm majority, 57 percent, believed that the government should bring the troops home as soon as possible, with only 33 percent thinking that they should stay. The embarrassment with the situation was reflected in the fact that a majority also agreed, however, that force should be used to free Dutchbat and the population of Srebrenica; 31 percent disagreed. Proponents of the use of force were to be found especially among men and younger people. Those outcomes suggest that the events in Srebrenica were a perfect example of the thesis of diminishing support when the going gets rough. Support for the hypothesis--whether in its simplistic form or in the more refined form in which the perceived success or importance of the mission is an intervening variable--can mainly be found in experiences in earlier, mainly American wars, especially in Korea and Vietnam.(34) According to Mueller, who came to his conclusion on the basis of surveyed respondents' reactions to different scenarios, a similar simple effect was to be expected in the case of the Gulf War against Iraq in 1991. According to Mueller, levels of tolerance for American casualties in the Gulf War would probably have been much lower than they were during the Korean or Vietnam wars. The public was promised low casualty rates, and if perceived reality had been different, support would probably have declined even if the war itself had been successful. Mueller also concluded that support for the war would decrease as a logarithmic function of the number of casualties, as was the case in Korea and Vietnam.(35) Other data, too, seem to support this finding of a direct and negative impact of casualties on support.(36) CONTRADICTORY EVIDENCE The general conclusion that the public is not prepared to support the use of armed force if it leads, or appears to lead, to casualties seems to be at least premature, however. There is still nothing approaching proof (at least in the case of the Netherlands, but probably elsewhere also) that support for participation in military UN operations (or other international military actions) cannot be raised or sustained or, more precisely, that it would necessarily dwindle as soon as casualties were actually encountered. Indeed, the findings of Mueller and others have been contested before. Kull, for instance, argues that American casualties in the Gulf War--a relatively small number--had no noticeable effect on the degree of public support for the war.(37) Another question is whether one can really generalize from earlier experiences and from a very small number of sometimes rather different cases. What seems especially dubious is Mueller's argument that the duration of the conflict is less important than the number of casualties in explaining diminishing support. In addition, as already suggested above, one might ask whether the perceived success of the war or the importance of the goals is not also an important variable. The Korean War was a rather long war and militarily only a limited success. The same was even more true with-Vietnam, where the justification of the war was not persuasive for many. One might wonder if the same phenomenon can be observed in cases of long wars with many casualties that finally end successfully for the country committing its troops.(38) Other studies also suggest that it is difficult to generalize and necessary to differentiate in the relationship between casualties and support for military actions.(39) Larson argues that the effect of casualties is mediated by a means-end calculus and is thus rather indirect: perceived benefits, prospects of success, and political consensus.(40) He includes in his study not only post-cold war conflicts but World War II, Korea, and Vietnam. He shows that sensitivity to casualties is not a new phenomenon. In World War II, ends-means calculations also played a role in support for the war effort. Support was neither automatic, nor unconditional. That is not to say that there may not have been a reduction over time in general tolerance of costs, leading to much lower levels of both expected and tolerated casualties, as happened in the Gulf War.(41) One might also ask whether the long and costly Korean and Vietnam wars were not fundamentally different from the short and, in terms of human lives, relatively inexpensive peace operations on which attention is focused today. There could, moreover and finally, also be cultural differences among countries that could account for differences in tolerance of casualties. Garnham explored the connection between duration, casualties, and costs on one hand, and support for war aims on the other in three other wars: the French Indochina war (1946-54), the Algerian war (1954-62), and the Israeli war in Lebanon (1982-85).(42) Although the quality of the available opinion data in these cases leaves much to be desired, Garnham generally observed the same relationships. He also noted the role of other factors, especially those concerning war aims and political culture. That becomes clear if we compare the much more lasting support for the Israeli military operations in southern Lebanon (in spite of initially high levels of casualties) with the quick erosion of support among Americans for the action in that country, including a hasty military withdrawal after 241 marines were killed in Beirut in 1983.(43) Garnham concluded that the public becomes impatient and frustrated when the stakes of the war are low (as was the case with the French in Indochina) or when the goal appears elusive (Lebanon). "Under these conditions, any level of continued casualties is likely to undermine public support."(44) Burk also argues that the support of military force and the acceptance of casualties are shaped primarily by the degree to which the war is perceived as being effective and meaningful. He bases his conclusion on survey data about the American military actions in Lebanon (1982-84) and Somalia (1992-94), cases that are often cited to prove the existence of the body bag syndrome.(45) Burk stresses that the support for peace operations is indeed not unconditional: "But it is not conditioned by a knee-jerk reaction to casualties. Judging from the responses we have seen to Lebanon and Somalia, it is conditioned rather by the demand that casualties be incurred for some clear and worthy purpose."(46) After a careful analysis of available data, he concluded that it is hard to demonstrate that support declined because of the incidence of casualties. Certainly Americans were shocked by the casualties, but the incidents had little influence on support for the operations as such. Enthusiasm for the operation in Lebanon had always been fairly limited, and in the case of Somalia the public supported the keeping, but not the enforcing, of peace. Another fairly recent survey (1994) of the American public's support for participation in UN peacekeeping operations is equally destructive of the body bag hypothesis. Respondents were confronted with a number of hypothetical cases in which American soldiers were being killed in UN operations in Rwanda or Haiti. They were asked, for instance, how they would react if they saw the bodies of dead soldiers on television. They could choose among four options. In none of the cases posed to the respondents did more than 23 percent (average 20 percent) choose the option "Withdraw the troops." The most chosen option (34 percent, on average) was "Bring in additional troops to hit back in the future." An average of 24 percent opted for "hitting back immediately," and 18 percent wanted to continue the current policies.(47) These outcomes confirm the results of earlier research.(48) Above (figure 1), it was suggested that casualties as such are decisive and that respondents would not make a distinction between casualties on one's own side and ones among the opponents. Garnham concluded that a similar effect occurred in the cases of Algeria and Lebanon (the massacres at Sabra and Shatila) but not in the cases of Vietnam and the Gulf War.(49) A RESILIENT PUBLIC That it would also be premature to conclude that the body bag hypothesis was operative in the Netherlands in 1995 can be concluded from table 3 and other polls taken after the events of July 1995. Very quickly, rather different observations could be made, using a number of polls taken shortly after the events described above and in the second half of 1995. The first signs became visible in a poll at the end of August 1995, shortly before the NATO air strikes began.(50) The downward trend in support for participation in UN military actions, so evident up to July, had now been reversed again, with 62 percent supporting participation in UNPROFOR and even more (68 percent) supporting participation in UN action in general (and only 11 percent against), the same high level as was previously observed. Seventy percent supported taking part in peacekeeping operations outside the Balkans. One wonders what brought about that sudden reversal. Was it the expectation that NATO would enter a new phase of strong action and that this action would be successful? Was it a desire for a second chance or for revenge? Whatever the case, the public showed a great deal of resilience, and support reached across the whole political spectrum. Even among the conservatives, who (on the whole) were most critical of taking part in UN action, a majority of 58 percent supported a Dutch role in former Yugoslavia. Similar levels of support were seen among the supporters of the "Green Left," who are traditionally against military actions. Support for UNPROFOR was relatively even stronger among men, young people, and the well educated. Another poll held at about the same time provided additional evidence that there was still considerable support for humanitarian and military aid.(51) Those results were at odds with the picture presented in the media and with the political commotion in the months since the fall of Srebrenica, which suggested that the Netherlands was suffering from a terrible hangover. That, it was said, would probably imply that the societal basis for participation in future military operations organized or mandated by the UN would be eroded, perhaps permanently. That this was true only to a very limited degree (unless one would argue, skeptically, that the hangover was suppressed only temporarily) also appears from other data.(52) The findings of August 1995 were confirmed four months later. About 70 percent then continued to support the participation by the Dutch armed forces in UN military operations.(53) The escalation of external intervention, which began at the end of August 1995 in the form of NATO aircraft attacks against Bosnian Serbian targets, undertaken with a general mandate of file UN, also received strong public support in the Netherlands. Three out of four respondents were in favor of the NATO actions, and only 5 percent disagreed with them. Two-thirds also thought the air actions had increased the chances of ending the war. It is remarkable, though, in light of earlier outcomes suggesting quite different views (compare also table 2) that in September 1995, no less than 55 percent supported participation by the Netherlands' armed forces in actions entailing considerable risks of casualties. Only 23 percent did not consider that to be acceptable, and 22 percent took a neutral position.(54) The events around Srebrenica seem not to have had a strongly negative effect on the degree of support for such actions, the dip in support being not very deep and only short-lived.(55) The societal basis for supporting military actions in the framework of the UN had been completely restored by the end of August 1995 and showed itself strong enough to allow the country to accept new commitments, when the question of participation in IFOR (the NATO-led international Implementation Force) arrived on the political agenda in December 1995. IFOR was to supervise implementation of the military aspects of the Dayton agreements, which brought about an end to the fighting in Bosnia. In spite of the many doubts that one could have, considering the mandate of this force, it was decided by the government to participate in IFOR, and a very large majority in parliament supported that decision as did 68 percent of the general public. Fifty-eight percent also agreed that NATO should be using force if necessary to enforce compliance with the Dayton agreements (17 percent disagreed). Fifty-five percent agreed to the acceptability of casualties (dead or wounded) among the (NATO) military, even when their likelihood would be great (31 percent said not acceptable, 14 percent gave no opinion). The same question, but mentioning Dutch soldiers specifically, produced very few differences in this respect. The outcome mentioned above was confirmed in another survey, which even registered 71 percent support for participation in IFOR.(56) That the readiness for sacrifice may be considerable also with respect to peacekeeping operations and that it may differ from country to country are also suggested, for example, by data from France that suggest that there is hardly such a thing as the body bag phenomenon in the French case.(57) In the case of the Netherlands, the likelihood of success and the perceived chances of peace considerably lessened the impact of any fear of casualties in reducing support for the peacekeeping mission in Bosnia.(58) Finally, it should be remembered that data on other aspects of defense and military affairs suggest that the end of the cold war has affected neither the general belief in the necessity of the armed forces, nor the trust in their effectiveness. This is certainly true for the Netherlands as well as other West European countries.(59) In particular, there is convincing evidence that, whatever they are called, new tasks of the military are firmly backed in principle, even though the participation of others in such multinational missions is often seen as an essential condition.(60) In summary, analysis of the available data produces a considerably more complex picture than that suggested by the casualty hypothesis in its simple form. CAVEAT LECTOR What can we learn from this case and the other evidence presented here? First of all, we should constantly remind ourselves in studying public opinion to be careful in drawing conclusions on the basis of poll data. The shortcomings in validity and reliability are too obvious to be repeated here. The validity of conclusions based on opinion polls in which respondents are faced with hypothetical cases is further questionable. Interview situations differ considerably from real life cases. As an aside (but not, as I shall argue, an irrelevant aside), one should note that in terms of political relevance, real attitudes are less important than how public opinion is perceived. Thus, a recent study of foreign policy that the Netherlands government commissioned argues (without presenting any evidence, one should add): "In most [UN] member countries public opinion is not willing to sacrifice the lives of their own soldiers for the sake of international principles in countries where war is endemic. At least this is how opinion is perceived by politicians" (emphasis added).(61) The available data suggest, secondly, that unstructured public opinion at the mass level is not an independent variable that can be treated in isolation from the general political context. To be politically relevant, it has to be mobilized, expressed, and organized into a meaningful and visible format.(62) Content and saliency are equally relevant dimensions of attitudes. Neither is public opinion a static factor. The willingness to sustain casualties is also not a constant. The military and political results that are (or are not) being achieved by force play a role. Whether support for participation in risky UN operations will be forthcoming and stable also seems to depend on the credibility and persuasiveness, as well as the unanimity, of political leadership.(63) Parsons was probably correct in this connection in concluding after an analysis of the UN peacekeeping operations of the past, "I shall always believe that the governments concerned underestimated the steadfastness of their own electorates, and that the excessive timidity was unnecessary."(64) Committed and credible government action backed by convincing arguments may help to sustain or even create public support for risky operations. It may partly compensate for the downward pressures caused by casualties or the fear of them, at least temporarily. Conversely, inept government policies may destroy such support as there was. In that connection, one is struck by the facile way in which the body bag argument is used by politicians and the media. There is a tendency to parrot one another and to anticipate situations, which may indeed be caused by such talk. The frequent--and somewhat patronizing--statements of politicians and observers about a to-be-expected body bag effect on public support may be turn out to be self-fulfilling prophecies. If politicians use that argument, it will often be because they seek an alibi to escape from their own, in themselves quite reasonable, hesitations. That has become particularly evident in the case of IFOR/SFOR in Bosnia-Herzegovina and the political unwillingness to run such risks as might be involved in actively trying to prosecute suspected war criminals. THE RATIONAL PUBLIC Third, the available data suggest that with respect to the possible use of military force there is little reason for concern about the alleged "emotionality" of the general public, its inconsistency and its susceptibility to sudden impulses and shifts. There are, it is true, sometimes indications of such inconsistencies and of a certain capriciousness, but the perceived success and sensibility of a particular military action, as well as credible political leadership, appear to be important in determining whether public opinion turns out to be stable or volatile, a reliable guide or a force to be mistrusted, and whether it presents an obstacle to or a condition of, a sensible and effective foreign and military policy. Although the general label "peacekeeping" suggests otherwise, modern military operations differ greatly in character, and that, together with changes in the sensitivity to violence, makes predictions about the threshold of casualty tolerance and the degree of public support an unreliable business. We are apparently faced with a calculating public opinion, which will draw up a balance sheet in each individual case, in a process of mutual influence among it, the military, and the political elites. The fact that psychological risks are increasingly brought into the calculation fits in with increasing control of violence and awareness of vulnerabilities in complex modern societies.(65) If there really is a body bag syndrome, it is embedded in a wider set of considerations. The resilience of Dutch attitudes, so visible after the fiasco of Srebrenica in the second half of 1995, may have been reinforced in that the humiliating withdrawal also contributed to a stronger "we-feeling" and to a desire for a second chance. Compensating effects, which may occur in other cases, probably also include perceived success of the use of military force and the combination of dehumanizing the opponent and the need to take revenge. Those tend to contribute to a hardening of positions and a willingness to persevere in military conflict. Those are also major reasons why the willingness to accept victims of war, or even to become victims, was on the whole so much greater, for instance, in the countries engaged in World War II than in a number of conflicts since 1945. It is also true, especially in democracies, where popular sentiments can be neglected only at certain expense, that the general willingness to participate in the peacekeeping operations of the UN (and other bodies) is reduced by the effect of the free rider syndrome, or the perception of one by decision makers. In most countries, only very few want their country to take the lead irrespective of what other countries do. Everybody has a stake in the production of collective goods such as international stability or the protection of human rights. The good is wanted very much, but potential contributors to its production would prefer even more to get it free, with others carrying the burden, especially when it is to be expected that the good will be produced anyway. The consequence is that everybody is waiting for others to move first, unless agreements can be made and enforced on what is to be considered a reasonable share in the collective burden. Such agreements are important for maintaining domestic support for contributing to internationalist goals, whether international peacekeeping, reducing the gap between rich and poor, protecting the international environment, or preventing violations of human rights. DEMOCRACIES AND MODERN WARS Compared with nondemocratic states, democracies are relatively well qualified to wage wars that are total, with respect to war aims and the mobilization of resources, or on the other hand, wars in which the risks are small (for example, because of escalation dominance or the weakness of the opponent). Democracies face severe problems, however, when success is elusive or when the trade-off between costs (especially in terms of human lives) and interests is seen as unfavorable for other reasons. Moreover, in cases not having to do with immediate threats, particularly individual or collective self-defense, but rather with peacekeeping or collective security, a level of support in the order of two-thirds (which is the level we can observe for the Netherlands for peacekeeping in general) may look impressive but may turn out to be insufficient for cases of comparatively low perceived importance. From a political, normative perspective it is worrisome that those wars and uses of force that seem least difficult to justify, in terms of the humanitarian interests involved and because of the support of the international community (however elusive that concept may be), seem to fall into that second category. In those cases, the free rider temptation rears its head also, and the public may always ask why we should be the ones to mourir pour Danzig. It is common today to explain the lack of support for risky humanitarian or collective security operations in the well-to-do democracies of the West by citing general lack of interest, even callousness, and a false sense of being secure from bloodshed and anarchy in the zone of democratic peace. It is relatively easy to plead ignorance about complicated conflicts in faraway countries and to be fatalistic about them. There is a degree of truth in that argument, but it does not seem to be the whole story. True, indifference is not uncommon. Also, the ideological certainties that allow us to draw sharp lines between good and bad and the cold war distinctions between "them" and "us" may have gone. There is also the pervasive stream of constant information in the mass media, which fosters feelings of concern and heightens our sensitivity. If people fail to act on that concern, it may be due less to indifference than to a sense of powerlessness, before the apparent failure of international institutions (especially in countries that, unlike the United States, never have the option to go it alone). In the former colonial countries of the West the sense of guilt (the tendency to ascribe the present intrastate conflicts of today to the legacy of colonialism and imperialism) no longer exists. Nor is it possible any more to see the hand of communist expansionism behind every violent upheaval. Yet shame, compassion and genuine concern are also important elements in the reactions of many people. On one hand, sensitivity seems to have increased with better means of communication, but so has a sense of fatalism and skepticism about what can be accomplished by outsiders and the use of international military force. That feeling is combined with a growing moral abhorrence of violence, which grows easily in societies that have more to lose than to win. |