CENTER FOR AMERICAN POLITICS AND PUBLIC POLICY OCCASIONAL PAPER SERIES (93-1) Strategic Politicians, Public Opinion, and the Gulf Crisis John Zaller February 4, 1993 WORK IN PROGRESS: DO NOT CITE WITHOUT CHECKING WITH AUTHOR! Prepared for The Mass Media, Public Opinion, and the Politics of the Gulf War (tentative title), Lance Bennett and David Paletz (eds.), contracted with University of Chicago Press. * I am grateful to Lance Bennett, Stanley Heginbotham, and David Paletz for encouraging me in this research. I also thank the Social Science Research Council and the UCLA Center for American Politics and Public Policy for research support, and the Guggenheim Foundation for the time necessary to do the work. Center for American Politics and Public Policy University of California, Los Angeles 310 GSLIS Building, 405 Hilgard Avenue Los Angeles, California 90024 (310) 206-3109 1300 19th Street, NW, Suite 300 Washington, D.C. 20036 (202) 296-8226 Strategic Politicians, Public Opinion, and the Gulf War John R. Zaller Department of Political Science University of California, Los Angeles Los Angeles, California 90024-1472 310-825-4331 Abstract The paper argues that public opinion as measured in contemporary polls had relatively little effect on elite policy-making during the Gulf War, but that the judgments of elites concerning the effects of policy on future opinion had a large effect. Four decisions are examined: President Bush's decision in August, 1990 to commit himself to the liberation of Kuwait; the decision of Congressional leaders to strongly support the president's initial policies, especially the deployment of U.S. troops to Saudi Arabia; President Bush's decision in November to send enough additional troops to the Gulf to create an offensive capacity; and the Congressional vote to uphold the president's right to use "all necessary means" to evict the Iraqi army from Kuwait. Strategic politicians, public opinion, and the Gulf crisis John Zaller University of California, Los Angeles "I know whose backside is at stake and rightfully so ..." -- George Bush, on prospects for U.S. victory in the Gulf War Past studies of the mass media and public opinion, strongly corroborated by the experience of the Gulf War, make two key points: At least in the domain of foreign affairs, the media normally take cues from government officials, "indexing" coverage to the range of opinions that exist within the government.[1] Further, mass opinion tends to follow elite opinion, with the most politically attentive members of the public following elites most closely.[2] These results imply a stark political world in which elites lead, masses follow, and the press does the bidding of the government. This top-down model of political influence can be defended as a useful first approximation of what occurs in foreign affairs policy-making, especially in foreign policy crises.[3] Nonetheless, reality is always quite a bit messier than social scientific models make it out to be. The major limitation of the top-down model is that, even in situations in which elites appear to be firmly leading mass opinion, the public can have substantial influence over its leaders. This influence arises from fact that political leaders, most of whom are under threat of electoral retribution, take great care to lead public opinion only toward goals that the public will, in retrospect if not always in prospect, applaud. Thus, politicians attempt to be responsive to future opinion at the same time they are trying to shape current opinion. An illuminating example comes from the Vietnam war. There is no evidence that the American public in 1964 was eager for a fight with communist guerrillas in Vietnam. But President Lyndon Johnson feared that if he failed to prevent a communist victory, the public would repudiate him and his party for "losing Vietnam" Hence, Johnson followed public opinion by leading it into a war that neither the public nor he wanted. V. O. Key, Jr. gave a general account of such cases in his classic study of "Public Opinion and American Democracy." The only public opinion that really counts in American politics, he contended, is the public opinion that politicians hope they might be able to create by their own actions, or fear might be created by the attacks of opponents in the next election. As a result, politicians' estimates of possible states of future opinion have more effect on policy-making than current opinion, which often evaporates by the time of the next election.[4] The events of the Gulf crisis, as we shall see in this chapter, provide several opportunities to glimpse the effects of anticipated states of future opinion on current policy. The crisis also affords instances in which current opinion affected decision-makers, but these effects appear less important than those of anticipated opinion. METHOD The method of this study is to examine four key decisions in the process by which the United States was led into the Gulf war, showing, to the extent available evidence permits, how calculations concerning public opinion influenced each decision. For evidence I will rely on published accounts of decision-making, which at this point are mostly journalistic, and on my interviews of top staff aides to key decision-makers and to about ten ordinary members of Congress.[5] These aides were, in most cases, the principal foreign policy assistant to a decision-maker. Many attended decision-making meetings in the Congress and, in a few cases, the White House. As importantly, all reported casual conversations in which their bosses planned strategy or reflected on the issue. It is quite possible that these aides can give accounts at least as accurate as those of the principals, and perhaps also less self-serving. The interviews were conducted in the winter of 1992-93, about two years after the events of the Gulf crisis. The four events I examine are as follows: * President Bush's decision to stake his political reputation on the expulsion of Iraqi troops from Kuwait, which he did when he declared that the aggression "will not stand." * The decision of Congressional Democrats to give virtually unanimous support to Bush's decision in August to send 200,000 U.S. troops to Saudi Arabia. * Bush's decision in November to send an additional 200,000 troops to the gulf. In so doing, Bush was, in effect, choosing to forego reliance on economic sanctions and to rely instead on either the threat of force or the application of force to resolve the crisis. * The vote of Congress in January, 1991 to authorize the president to use "all necessary means" to force the Iraqi army to withdraw from Kuwait. A difficulty in trying to understand these decisions is that the public record of them, and to a lesser extent my interviews as well, heavily stress moralistic considerations. I do not doubt the importance of such considerations and will not fail to report them. It is, however, the job of the political analyst to search for the political reasons that make moral arguments compelling, and this is the largest part of what I will be doing in these pages. The reader may then judge which type of explanation is more compelling. FOUR KEY DECISIONS I. "The Line in the Sand" Four days after Iraq's invasion of Kuwait, President Bush declared to reporters in an impromptu news conference, "This will not stand, this aggression against Kuwait." There is little doubt that when the president uttered these words he was staking his political reputation on the removal of Iraqi troops from Kuwait. "A President Puts Himself on the Spot" was the how the New York Times described Bush's action in a headline the following day.[6] "If Mr. Bush is blustering during an international crisis," added columnist William Safire, "he will be the one who 'will not stand.'"[7] What political reasons might a president have had for taking the fateful and politically risky step of pledging himself to force Iraqi troops from Kuwait? One possible explanation for Bush's actions can be confidently dismissed: He was not responding to any discernible pressure from public opinion or the media for a stronger stand against Iraq. Measures of public opinion did not exist, since no polls had yet been taken in what seemed, up to that point, a relatively mild crisis, and media coverage of the President's initial actions -- a strong condemnation of the invasion coupled with winning U.N. support for an economic embargo against Iraq -- was wholly favorable. As the New York Times editorialized: The U.S. has no treaty obligation to come to Kuwait's aid. But the gulf states still look to Washington for leadership and help in organizing action. President Bush has responded with the right lead -- a strong national stand and a strong push for collective diplomacy.[8] Published accounts of Bush's motives for making the pledge stress the president's own beliefs about the right course of action, which were apparently based on the so-called "lessons of Munich."[9] Bush was reportedly determined to stop Saddam before he could, like Hitler after his takeover of Czechoslovakia in 1938, use the fruits of aggression to become even more dangerous. Bush was further concerned that, coming at the beginning of the "New World order," the conquest of Kuwait could invite similar aggression by other countries unless dealt with effectively. A senior executive branch official to whom I spoke also underscored the importance of Bush's own view of the crisis. "This president to an extraordinary degree said, This is right and I'm not going to be influenced by the polls if I'm confident I'm on the right course.'" The official also said repeatedly that the president knew he was taking huge political risks, but was willing to be judged by the final success or failure of his policies. There is, then, no evidence that the pledge to liberate Kuwait was based on political calculation, except perhaps that when the dust settled the president would be vindicated. Notwithstanding the available evidence, however, it is important to consider other political calculations the president might have made, for two reasons. First, even if Bush had taken greater account of politics than the evidence presently shows, one would expect discussions to have been limited to his closest aides, who would be expected to remain loyally silent. Second, if political imperatives are big enough and clear enough, they need not be discussed or even consciously thought about in order to exert influence. It is, as it happens, easy to find domestic political rationales for Bush's stance. First, success against Iraq would obviate criticism that Bush had responded weakly to an aggression that was certain to drive up oil prices, exacerbate the impending U.S. recession and perhaps even bring on a new energy crisis. (The effect of higher oil prices on the weak U.S. economy was discussed in a key decision-making meeting, but there is no indication Bush considered this point important.[10]) In light of the Bush administration's efforts to build up Iraqi power in the 1980s and its weak diplomatic response to Iraq's threats against Kuwait prior to invading it, it was likely that Democratic criticism would have been especially sharp.[11] Second, a successful rollback of the Iraqi invasion could constitute an important part of the president's record in the 1992 election. In his first two years in office, the president had accomplished relatively little he could take credit for.[12] Indeed, his most notable action had been to back down from his famous pledge of "Read my lips, no new taxes" earlier in the summer of 1990. Finally, the crisis afforded the president an opportunity to overcome long-standing criticism that he was a weak and unprincipled leader, a moral "wimp." Gary Trudeau detailed the nature of the allegation during the 1988 campaign. Bush, he maintained, is "not a wimp in the sense of lacking physical bravery." The issue is rather one of moral courage, the willingness to place oneself at risk for one's principles.... The unfailing inclination to hold everybody else's coats during the great conflicts of our times has led to a persona of shimmering translucence -- hence the Nowhere man ... President unDukakis ... the classic cipher.[13] Such complaints were widespread in the summer of 1992 and carried over smoothly into the Gulf crisis. Commenting on Bush's acceptance of a budget agreement with Congress that included big tax increases, columnist David Broder wrote The budget fiasco in the capital has left few politicians untarnished, but the damage to President Bush is particularly serious -- for good reason. The president has revealed to the nation's voters that you can't have the courage of your convictions if you lack any convictions. He -- and we -- will be fortunate if the lesson is not seized upon by Saddam Hussein and other foreign antagonists.[14] Elizabeth Drew (1992, p. 83) reported in the New Yorker, A Senator said to me recently, reluctantly, "We all know instinctively that this is not a strong man. It's greatly disturbing. I try not to think about it. I don't know anyone who's honest with himself who doesn't think this." It is hard to imagine a president who wouldn't be personally disturbed by such criticism and eager to put it to rest. But psychology aside, a president with Bush's reputation for moral indecision who stood by while U.S. interests were damaged, and who had the misfortune of a bad economy at the same time, would be risking severe political damage. There was, of course, also great risk in attempting to liberate Kuwait. But if our question is, what political reasons might the president have had for taking this risk, we don't need to look hard to find several good ones. It is notable that, as soon as Bush made the pledge to liberate Kuwait, the press reached a nearly unanimous verdict that, if redeemed, the promise "could virtually assure reelection."[15] It is also worth noting that Bush's publicly stated goals are not necessarily very different from the motives I have suggested. After all, a president who orchestrates an effective international response to, as Bush claimed, the greatest threat to world peace since 1941, thereby playing Churchill to Saddam's Hitler, would clearly avoid any criticism for "losing Iraq" or being a wimp. Lofty international goals may satisfy mundane political needs, and may be pursued just for this reason. But all this said, it must be reiterated that the public record of Bush's motivations, as it now exists, refers only to statesmanlike considerations. If Bush was thinking about public opinion at all, it was the laudatory public opinion that would exist after, as he hoped, Iraq had been evicted from Kuwait. II. The Congressional rally behind the president A few days after declaring that Iraqi aggression against Kuwait would not stand, Bush announced that he was sending combat troops to Saudi Arabia to protect that country from invasion by Iraq. Congressional support for this decision was immediate and clear. As Senator George Mitchell, the highest ranking Democrat in Washington, told the press: American interests and our long-standing ties with Saudi Arabia make the president's decision to help defend Saudi Arabia the correct one... It is important for the nation to unite behind the president in this time of challenge to American interests.[16] Many members of Congress made similar supporting statements and none publicly opposed the decision to commit U.S. ground forces to the region (CQ Almanac, p. 727). Congressional support for Bush in this phase of the crisis appears to have been quite consequential. If Congress supports the president, the media will be deprived of oppositional sources, which will tend to make them more supportive as well, which in turn enhances public support. With Congress, the media, and the public all supporting the Bush, the president's threat to use force becomes gains credibility abroad, which strengthens his hand in attempting to pull together an international coalition against Iraq (see below). The initial Congressional and public support for his policies may also have encouraged Bush to escalate the level of military confrontation later on (see below). Elite support which generates mass rallies behind presidential policies is common in foreign policy crises (see Brody, this volume). How can it be explained in this case? When I asked one senior aide why Congressional leaders had rallied so strongly behind the president's decision to send troops to Saudi Arabia, the question seemed silly to him. "Why not support the president when he stands up for American interests? You can always withdraw your support later if you want to. In the meantime, go along." This aide also said that he had been shown intelligence reports strongly suggesting that Iraq intended to invade Saudi Arabia. "I've seen lots of this kind of data in other cases, and not all of it showed what the president said it showed," the aide said. "But this time, the evidence was strong." These factors came up repeatedly in interviews. The Iraqi invasion was perceived as a genuine threat to American interests, and there was little political cost to supporting the president, so why not go along, at least for now. It should not be assumed, however, that Congressional leaders would acquiesce in any presidential action in a crisis. Although some legislators doubted that the U.S. interest in Kuwait was strong enough to justify war, few doubted that the U.S. had an important interest in the region. Such judgments are best understood as reflections of reigning conceptions of geopolitical reality, and appeared in my interviews to be the most important determinant of the Congressional rally in August. If leading members of Congress had disagreed with the ideological premises of Bush's action, their response to his decision would probably have been entirely different. That Congressional leaders were toadying to public opinion in their support for Bush's policies, rather than expressing their own feelings, appears unlikely. For one thing, Congressional endorsement of Bush's decision to send troops to the gulf began in the same news cycle in which the policy was announced, before public reaction was known. For another, some Congressional leaders who praised the president in August criticized him in November when he announced a policy they disliked. Although polls had by then shown Bush's gulf policies to be popular, these leaders (though not necessarily all member of Congress) showed little hesitation in lambasting the president. So here it appears that Washington elites took an action -- public endorsement of presidential policy in the gulf -- on the basis of their own convictions. In so doing, they were not responding to public opinion, they were helping to shape it. III. The November troop deployment The August deployment of 200,000 troops to Saudi Arabia was too small and too lightly armed to take offensive action against the 400,000 Iraqi troops then in Kuwait. The purpose was to defend Saudi Arabia while economic sanctions were allowed to have their effect. But on November 8, Bush announced the deployment of enough additional troops to give the United States and its allies the capacity to launch offensive actions. The additional troops profoundly transformed the crisis. Until then, the expectation was that events would play out over months or perhaps years while sanctions ran their course. But the American force was now too large to sit in the desert heat for a year or more while sanctions ran slowly on, so the timetable for resolution of the crisis speeded up. Either Iraq would withdraw within three or four months, or the U.S. would go to war. Congressional Democrats were furious over the decision, claiming that it "boxed us in" to a position in which they would have two bad choices: Acquiesce in a presidential decision to use force, which most Democrats thought was bad policy, or oppose the use of force, which, with an army poised for battle, would seem like undercutting the troops. Some Republicans, however, sympathized with Bush's decision to speed up the crisis. If U.S. troops were kept in the desert for an extended period, said an aide to a Republican Senate leader, "we would start getting letters from families with loved ones overseas, stories about little babies who had never seen their fathers, then the inevitable accident in which lots of GI's are killed ... then pictures of children in Iraq starving from the boycott." Congress, he said, responds to this kind of thing. "I think you would have seen a process of weakening of American resolve" which would have made it politically difficult to start a war if sanctions turned out not to work. "This is a classic case of the 'best' policy [i.e., trying sanctions first] being politically impossible."[17] Another aide to a Republican senator saw the same nightmare of eroding support if sanctions were allowed to run their months-long course. "It would have been a delight to the Democrats," who would be constantly "sniping at Bush" and "eroding support for his policy" with the public. And then, once the public support was gone, the Democrats would want Congress to vote. "This is a very partisan place ... a sick place," he said. In my interview with a National Security official who met regularly with President Bush during the crisis and helped plan the second troop deployment, I asked whether the domestic political effects of the second troop deployment -- boxing in the Democrats, and avoiding a drawn out affair in which public support would erode -- had been intended by the administration, or were merely fortuitous side-benefits. The second troop deployment, he said, was entirely determined by strategic concerns. "We didn't have any choice. Without sending the additional troops, we had no chance of persuading Saddam to leave Kuwait," he said. "Why should he leave if we couldn't make him leave?" I next asked whether the decision to send extra troops had been affected by the Vietnam experience, in which flagging popular support made it politically impossible to send additional forces once the fighting got difficult. Perhaps troops were sent in November because public support for sending them wouldn't exist at the time when they came to be really needed. "That's the same question you just asked," he replied. "We weren't thinking about domestic support when we decided to send the additional troops." I next asked if any thought had been given to how public and Congressional support for the president's policies would have eroded if the embargo had been given more time, and whether it was partly to avoid the nightmare scenario envisioned by Republican Senate aides that the administration decided for an early war against Iraq. "It didn't come to that," he said. "We were more worried that our international coalition would fail. We didn't feel we had a lot of time," especially with the Arab allies. He agreed that domestic support might also wear thin during an embargo, but "we always assumed that we'd have the support to do what needed to be done." At this point I suggested that, from what he was saying, the Bush administration had not been much concerned about public opinion during the crisis. Not at all, the official replied. "In order to make our threats credible to Saddam and to hold our international coalition together, it was tremendously important for us to demonstrate that we had national unity in support of our policies." What practically was done to assure such public support, I asked. "We worked a lot harder than usual on putting out statements, getting the president out there explaining the policy, sending people out to speak to groups." Seeking the approval of the U.N. security council for the use of force was also part of a strategy to bolster domestic public opinion. Although helpful with the international coalition, the official said, the U.N. vote was more important at home. "Domestically, we knew the U.N. vote would make it hard for people to disagree with us." "Public relations was not a strength of this administration," in large part because Bush didn't like having to make speeches, the official said. "But on this we really tried." Taken at face value, this interview suggests that public opinion had no influence on the critical November decision to create an offensive capacity against Iraq. The administration was concerned about public opinion, but this concern manifested itself chiefly in efforts to convince the public to accept the policies that it judged best. I hesitate, however, to accept this conclusion. The second troop deployment, after all, was taken in a context in which mass opinion supported Bush's initiatives so far. For example, 66 percent said in CBS-New York Times survey just before the announcement that it was the "right thing" to send troops to Saudi Arabia. The public was certainly not clamoring for more aggressive policies, but there were clear indications that it would go along with stronger action if so urged. (See Table 1 and associated discussion in the earlier chapter, "Elite Leadership of Mass Opinion.") A plausible presidential reading of public opinion at the time of the second troop announcement might therefore be: A president who waits for the public to demand war against Iraq may wait a long time; but a president who simply starts the country on the road to war may expect the public to follow. So if the Bush administration didn't worry much about public opinion in deliberating over the second troop deployment, it was probably because it didn't have to. The public had supported Bush's initiatives in the past and could be expected to continue to do so. The effect that the generally permissive public opinion of this period had on President Bush is suggested by a conversation in late August between the president and his national security advisor, Brent Scowcroft. In the course of a morning spent trolling for bluefish in the president's speedboat, Bush revealed his long-range thinking to the man who was, by all accounts, his closest confidante in the crisis: Bush said there could be real opportunities for American leadership abroad, after the conflict with Iraq was resolved. Bush outlined his vision... He was thinking... about the meaning of the coalition that had begun to form against Iraq. It was a dramatic example of how diverse nations could come together to address a common problem, he said. At home, he saw a kind of post-Vietnam "maturity" among the public. The mood has manifested itself, Bush told Scowcroft, in the strong support reflected in the recent public opinion polls on American policy in the Gulf. .... A large part of the President's conception had to do with the startling changes in the Soviet Union and Eastern Europe... A new world order was in the making, and the president believed that America would help to shape it. (U.S. News, 1992, p. 141-2) Of course, what the president saw as a "mature" public -- if that is the right word for it -- was by no means an attribute of the public alone. It was predominantly an attribute of other political elites, notably leading members of Congress, who had given Bush virtually total support up to that point in the crisis and who, in so doing, induced the public to support him as well. In taking encouragement from public opinion polls, the president was really taking encouragement from the supportive attitudes of other elites. Was there, then, any influence of public opinion at the time of the second troop deployment? I would say yes, in the sense that, in following the president and other elites so willingly down the road to confrontation with Iraq, the public encouraged Bush to anticipate that it would continue to follow down the same road in the future, including, if necessary, a war. IV. Congress authorizes the use of force against Iraq On January 12, Congress voted on resolutions authorizing the president to use "all necessary means" to drive the Iraqi army from Kuwait. After much debate, the resolutions passed by votes of 52-47 in the Senate and 250-183 in the House. These vote margins are roughly in line with opinion polls, which showed the public divided but leaning toward war.[18] This correspondence suggests that a majority of members of Congress may have been responding to mass opinion on this issue. However, a close analysis of the dynamics of Congressional decision-making reveals a far more complicated picture, one in which responsiveness to current opinion was only one of several determinants of the vote and perhaps not the most important one. For this analysis, we must consider three types of legislator: 1) those with substantial interest or expertise in foreign affairs, especially the heads of relevant committees; 2) the leadership of the Democratic party; and 3) remaining members of Congress. Motivations within the three groups were radically different. Legislators who specialized in defense or foreign policy wanted to influence public opinion and help make policy. The gulf crisis might be the most important issue to fall into their domains of expertise in their legislative careers, and they wanted to be part of the action. Ordinary legislators felt just the opposite: With no ax to grind, they mostly wanted to play it safe. Their dilemma was figuring out what safe consisted of. The Democratic leadership, finally, was cross-pressured between acting on its own preferences, which were to oppose the war resolutions, and trying to uphold the status of Congress as an institution. A. The expert and interested. The House of Representatives is an institution which pays considerable attention to the opinions of members who have specialized in a particular subject. Probably the most knowledgeable member of the House on defense issues at the time of the crisis was Les Aspin, chair of the Armed Services Committee. From the beginning of the crisis, Aspin produced a whirl of public hearings and statements, all manifestly designed to influence opinion either inside of Congress or outside it or both. When I asked members of his staff whether Aspin had been influenced by public opinion, the answer was a resounding no. A junior staffer, speaking in awed tones, said "Les is almost professorial. He studies issues very carefully and then does what he thinks is right." A more experienced staffer put it this way: "Les is a very ambitious man, but the opinions of the folks back home aren't what he's paying attention to." The most important single voice in the Senate was Sam Nunn, a normally reliable hawk from a conservative state who, on this issue, took the role of leading Congressional dove. His post-Thanksgiving hearings in the Senate amounted to the strongest attack on Bush administration policies of the entire crisis (See Entman and Page, this volume). In explanation of Nunn's unusual behavior, two staff aides (not from Nunn's staff) recalled how, shortly before Bush announced the second troop deployment in November, Nunn was called out of a meeting for a phone call and quickly "consulted" by the Bush administration on the decision. Nunn, already skeptical that Bush was too inclined to use force, reportedly returned red-faced, furious, and implacably opposed to Bush policy. Of course, no one knows whether the administration's unwillingness to consult with Nunn in a more dignified way had any effect. But I was told several times that personality and ego matter a great deal on Capital Hill, and that the Bush administration paid less attention to them than it should have. By many accounts, Aspin and Nunn determined many more votes in Congress on this issue than their own, though, of course, their influence ran in opposing directions. Their difference was apparently an important reason the war resolutions passed more easily in the House than the Senate. Two other members of Congress often cited as influential on this issue were House Democrats Ron Dellums and Stephen Solarz. Dellums helped organize a Congressional lawsuit to enjoin the Bush administration from going to war without Congressional approval, while Solarz led an extremely active "whip organization" in Congress to round up votes in support of the use of force resolution. Although both men came from districts that could have been expected to support their actions -- Dellums represents a liberal inner city district in Oakland and Berkeley California, Solarz a heavily pro-Israel district in New York -- both went far beyond the requirements of political responsiveness. Like Aspin and Nunn, these leading legislators were active on this issue because they held strong views and wanted to affect the course of events. B. Members who are non-specialists. In the middle of November, as war was growing more likely, Defense Secretary Dick Cheney gave a briefing to about 100 members of the House. As Bob Woodward described the scene in The Commanders: After an hour, Cheney said, "I assume all of you guys want to vote up or down on the proposition." The room erupted. There were shouts of no and yes (p. 311). Many members of Congress did not initially want to vote on whether the U.S. should go to war against Iraq. Some legislators, most commonly true believers on either side of the issue, were eager to put themselves on the record. But the majority of members who had no strong interest in the issue were often reluctant. They would get little credit no matter how the war turned out, but might become objects of retribution if they either opposed a successful war or supported a disastrous one. Facing a situation in which they couldn't win but could easily lose, these members didn't want to vote until as much information as possible had come in. How Congress came to vote on the war resolution was a leadership decision and will be discussed in the next section. Here I note only that when Congress did eventually vote, it did so in the most solemn manner. Partisan acrimony was put aside, as member after member went before their colleagues and, of course, the television audience to give an account of their votes. Congressional deliberations were described in the press as a sober and sincere affair that marked one of Congress's "finest hours." When I asked staff aides whether they felt the vote on the war resolution had been much affected by public opinion, all but one or two initially said the public's influence had been minimal. The vote was, in a phrase that came up over and over, a "conscience vote," one that many members had anguished over more than any they had ever cast. "When you're voting whether to send soldiers into battle, you don't think about politics," I was told many, many times. A few legislative aides, however, went on to observe that the vote was a conscience vote mainly because current public opinion was split and the most politically relevant aspect of the decision, the outcome of the war, was unknown. [O]ne congressional aide explained to me afterward ... "The vote was a potential career-killer -- either way." The members of Congress knew that they could be caught on the wrong side of history -- but when they voted there was no telling which side that would be (Drew, 1992, p. 86). A staff member with many years of experience on Capitol Hill put it this way, You're talking about a body of politicians about as good as any in the world, and suddenly they're not getting a clear message from the public. Some guys, of course, are all conscience on every vote, but most members wanted to make a political decision and couldn't ... A lot of the anguish you heard so much about was just because members couldn't get a clear message from the public. The notion that many members voted their conscience only because they were forced to do so by circumstance was, as I indicated, volunteered only a few times. But after the idea was suggested, I raised it many more times myself, and in only one case did anyone disagree, saying it was "too cynical"; in the other cases, I got a silent nod, a statement of "well, of course," or some other form of assent. Two other features of the Congressional vote on the war are worth noting. In ambiguous or dangerous circumstances, it is essential for members to cast what is called in Congress "an explainable vote" -- a vote that appears well-reasoned, well-informed, and sincere. And this is what the long, somber debate on the war, which was not really a debate at all but a series of unrelated monologues, permitted members to do. The other notable feature of the vote was its strongly partisan composition. Roughly 97 percent of Republicans and 70 percent of Democrats voted on the same side of the issue as their fellow party members and against the position of the other party. The existence of this partisan difference is by no means surprising. Democratic and Republican legislators come from districts that differ in partisan orientation and tend to reflect those differences in their voting. Moreover, Democrats who came from districts in which Bush had run relatively well on 1988 were significantly more likely to support the president on the war resolution than other Democrats (Jacobson, in press). Whether one counts this as the influence of public opinion, since different publics select different types of members, or the influence of conscience, since members, once elected, cast ideological votes that are akin to conscience votes, is not clear. Undoubtedly, both types of influence contributed to the partisan coloration of the war vote, as did one other type of influence. Several aides noted that, in ambiguous situations, the safest vote for a member to cast is a party vote. This was most clearly true for Republicans, who had an interest in seeing the Republican president do well and who could, in addition, have expected to find the Bush administration less likely to do them favors if they went against it on the war. But it was also true for Democrats, whose leaders control the flow of legislation in the Congress. As a result, Congressional Democrats and Republicans were far more partisan than were party members in the general public. Within the Congress, the two party groups differed by 67 percentage points in level of support for the war resolution, whereas Democrats and Republicans in the general public differed by only 16 percentage points.[19] Given this disjuncture, it is hard to argue that the partisan division of the vote in Congress simply reflected the partisan divisions that existed in the country. My investigation, thus, has turned up two quite different types of explanation for how vote decisions on the war were made. One stressed that, after much soul-searching, members voted their consciences, which turned out in many cases to be their ideology or partisanship. The other type of explanation stressed' political calculation -- that any vote at all was risky, that the only public opinion that would count would be the opinion that existed after the war ended, that the safest vote was a party vote, and that one's vote should be, above all, carefully and convincingly "explained." It is striking that, on either explanation, public opinion as measured in polls at the time of the vote was not a major factor. Public opinion as a determinant of the ideological composition of Congress was very important,[20] though its effect seems to have been exaggerated by the tendency of members to ally themselves with their party leadership. The public opinion that would exist after the war was potentially of great importance, but because it was difficult to predict, its only effect was to motivate a search for safe haven. C. The Democratic Leadership. President Bush's policies set the agenda to which the Democratic leadership had to respond. The essence of Bush's policy was to "threaten war to prevent war," that is, make a credible threat to evict Iraq from Kuwait in the hope that it would then withdraw on its own accord. The president further asserted that his role as commander-in-chief gave him all the authority he needed to make good on this threat. The Democratic leadership of Congress -- in particular, House Speaker Thomas Foley and Senate Leader George Mitchell -- did not object, at least publicly, to threatening Iraq, but the leadership opposed going to war until sanctions had been given a long try, and it objected strenuously to the assertion that the president could start a war without authorization from Congress. "Under the American Constitution, the president has no legal authority -- none whatsoever -- to commit the United States to war," asserted Senator Mitchell. "Only Congress can make that grave decision."[21] From the president's side, disagreement with Congress was a source of concern but no apparent anguish. The president, according to published accounts and my interviews, worried that Congressional opposition might give encouragement to Iraq, but felt confident that he had both the moral right and sufficient political support to go to war without Congress. "Nobody cared about Congress," as a military official bluntly stated. Public opinion was more important to the Bush administration, but here the president was confident. "Low levels of public support for war before the war started were no problem," explained the military official. "We felt the country basically supported the military effort, and that as soon as the fighting started, there would be a surge of increased support." Then, if the war went well, all would be fine. The complement of presidential confidence was Congressional trepidation. The leadership's problem was how to oppose the president without bringing discredit on the party or the Congress. This was no easy task. A fierce war debate could undermine the strategy of threatening war to avoid war, thereby making Democrats liable to the charge that, in opposing the president's policy, they were making war more likely. Why, after all, should Saddam back down peacefully if Congressional Democrats were going to tie Bush's hands? This was a point the Republicans were prepared to drive home in the most cutting fashion. Republican Senator Trent Lott said in November that if Congress allowed a debate to become "a backing away from the [administration's] policy, it could be giving aid and comfort to Saddam." In using language from the Constitutional definition of treason, Lott sent a message that some Democrats were still angry about more than two years later. Bush made the same point by showing a group of legislators a clipping from an Iraqi newspaper purporting to describe Congressional opposition to war. With these incidents in mind, I asked a Republican Senate aide whether Democrats might have faced partisan attacks that verged on questioning their patriotism if they had too vigorously opposed Bush policy. "That's not a possibility," he replied. "That's a fact." The House leadership, I was told, was seriously worried about such attacks and regularly discussed how to make opposition to Bush's policies seem "a position of strength." The leadership, as I was told, wanted to "avoid reinforcing the public perception that the Democratic party was weak on defense, weak on foreign policy." This problem was complicated by the fact that the party included members, sometimes described by others as the crazy left or looney left, who were prepared to go to great lengths to defeat a declaration of war, including perhaps a filibuster in the Senate. Any such endless, unfocused debate could be a major embarrassment to the party and the institution. So when proposals were made for a special session after the election to debate authorizing war, the House leadership turned it down partly from fear it would turn into "a year-end circus," as an aide put it. Finally, it was not clear that Bush would respect a Congressional vote that went against him. This was perhaps the most sensitive point. "There was," an aide said, "a fear that Bush would go ahead no matter what Congress did." The prospect of U.S. troops poised for battle, or perhaps going into combat, fighting and dying, while the courts refereed a dispute between Congress and the president was regarded as "horrendous." It could give the whole government, but especially Congress and the Democratic party, a black- eye in public opinion that would last for decades. And there was no guarantee it would stop the war. In these circumstances, the avoidance of "a constitutional crisis within a foreign policy crisis," as an aide put it, became a top leadership priority. There was, however, one type of Congressional vote the president might feel compelled to honor, namely, one that he himself had requested. Yet the leadership knew that Bush would never request a vote unless he was sure in advance he would win. So, although the House leadership reportedly believed that it had a good chance to win a vote against the war if it were taken in November or December,[22] it refused to schedule one at that time, trying instead to induce the president to request the vote himself. Why, if the leadership knew it was likely to lose a vote that Bush requested, did it care about having one? Because it felt that a presidential request would create a precedent that would strengthen Congress' role in the foreign policy-making process. That, apparently, was the most that the House leadership felt it could get from the crisis. Finally, amid complaints of procrastination and cowardice,[23] Foley and Mitchell scheduled a debate on war to begin January 10.[24] This was one day after the end of the final diplomatic exchange between the U.S. and Iraq, so Congress could no longer be accused of undermining the policy of "threatening war to avoid war." But it was also just five days before the U.N. deadline, at which time many expected fighting to begin. This made voting against the war seem like voting against the troops who would fight it, a perception that, by many accounts, strengthened the pro-war position. It is striking that Foley, in announcing his decision to hold a vote on the eve of war, announced at the same time that he expected the president to prevail. He also later declared that the vote would be a "conscience vote" in which the leadership would make no attempt to line up votes against the president.[25] Mitchell, in announcing the Senate vote, made no prediction about who would win in that chamber, but two other Senators predicted the war resolution would pass.[26] Shortly after the vote was scheduled, President Bush wrote a letter requesting that Congress approve a war resolution. As the New York Times reported at the time, "Bush made the request only after Congressional leaders said in recent days that he was almost certain to receive Congressional endorsement," provided that he worked hard for it.[27] So, in contrast to the Democratic leadership, Bush did not declare a conscience vote. Rather, he ordered an all-out lobbying operation, which complemented the large and well-organized pro-war whip organization that had been operating unopposed in the House since November under the leadership of Representative Solarz. This pattern of evidence, much of which is on the public record,[28] indicates that the Democratic leadership did not go all-out to defeat Bush's war policy, and may have been just as happy to lose. Its highest priority was apparently to avoid either undercutting Bush's policy of "threatening war to avoid war" or creating a constitutional crisis. I pressed hard in my interviews to find out why exactly the House leadership was so concerned about a constitutional crisis. One aide, a political scientist who has since returned to academia, said the leadership was motivated by the belief that conflict with the president would be dangerous for the troops in the field and bad for democratic government generally. He said he was "astounded" by the extent to which the leadership ignored partisanship and tried to do what was right for the country. "In my time on the Hill I saw nothing as apolitical as this vote," he said. "There was a strong feeling of moral righteousness." The other aide, a career staffer, stressed Foley's concern to do what was right for the country and to protect the reputation of Congress. Undermining national unity or provoking a constitutional crisis would accomplish neither of these goals. Was Foley, I asked, not also concerned about what was good for the Democratic party? "This was a case when good policy was good politics," the aide replied. I would draw particular attention to the last remark, except I would restate it as, "This was a case when national unity was good politics." The stress on national unity was never greater than in Foley's speech at the end of the House debate. He spoke strongly against war, but ended with the invocation that "however you vote... let us come together after the vote with the notion that we are Americans here, not Democrats and Republicans, all anxious to do the best for our country ... without anything but the solemn pride that we voted as our conscience and judgment told us we should."[29] These are the words of a man voting against a measure he knew would become extremely popular in a few days when the war began, and who didn't want himself or his party, despite their antiwar beliefs, to be guilty of anything that could be called obstructionism. He was stating his views, but he wasn't getting in the way. Public opinion, though not much emphasized in my account so far, may be seen as the essential determinant of this outcome. I remarked in my introduction that the job of the political analyst is to find the political logic that makes moral arguments seem compelling. If it is fair to argue that President Bush was encouraged by his domestic vulnerabilities to discover lofty international values in opposing Iraq's aggression against Kuwait, it is also fair to argue that the Democratic leadership was encouraged by the party's traditional vulnerability on foreign policy issues, as well as by the standard criticism of Congress as too fractious to be trusted with a role in foreign affairs, to avoid undermining national unity in a crisis situation. To fail in this would be to invite disaster. Why? Because the public didn't want an ugly leadership squabble on the eve of war. It wanted an efficient, confident, and unified national effort, and if it got a partisan mess instead, it would be more inclined to blame the Democrats and Congress than to blame the president. Nobody would win, but Congress as an institution and the Democratic party would be the biggest losers. Thus, when I asked a veteran Democratic foreign policy aide what would have happened if Congress had, on its own initiative, voted in December against authorization of war, he was hardly able to take the question seriously: "We would have been accused of appeasement, of playing into Iraq's hands... We'd never do that, it would be a disaster... It couldn't happen." Clearly, however, not everyone in Congress saw it this way. Some members began demanding a vote on the gulf war as early as November but, since both sides seemed equally confident of winning, one must wonder whether either side gave as much consideration as it should have to what would have happened if it lost. I was especially curious about Republican Leader Robert Dole, a loyal supporter of the president who was among the earliest and most insistent in demanding that Congress go on the record with a vote. When I asked an aide familiar with Dole's thinking why the Senator had taken this position when it might precipitate a constitutional crisis, he replied that the Congressional deliberation on the war had been ... a political exercise between warring parties trying to set each other up. That's the way it works here, and Dole wanted to draw the lines clearly. I encounter essentially similar attitudes on the antiwar side, where the feeling seemed to be that Congress should go all-out and let the chips fall where they may.[30] It is interesting that none of the partisan gladiators on either side of the issue suffered politically from their stands on the war, and some may have marginally increased their national visibility. Indeed, in the end, almost everyone had something to be thankful for in the gulf vote. The gladiators were able to make principled declamations on national television; the cautious were given ample opportunity in a dignified setting to cast explainable votes; Congress enhanced its institutional status because of the somber manner in which it conducted its 'debate,"[31] and because the president did finally request its approval before going to war; and, not least important, the country was spared an ugly constitutional crisis. The leadership had done its job -- not by following public opinion, which simply did not exist on the critical question of when a Congressional vote should be taken, but by successfully anticipating how the public was likely to react to future events. One other point is worth pondering. As is often remarked upon, the military triumph in the Gulf War, although a boon to President Bush's short- term political standing, failed to carry over into the 1992 election. The usual explanation is that the public turned its attention to a more immediately salient matter, the economy. Another factor, however, may have been the way the war vote was handled. Although many Congressional Democrats felt and spoke strongly against the war, their opposition was so carefully modulated and packaged that it hardly penetrated to much of the public. This is apparent in the results of a June 1991 survey by the National Election Studies: Before the war actually started, do you think one political party was more in favor of using military force in the Persian Gulf than the other party was, or do you think they were about equal in their support for using force? (emphasis original) Democrats favored force more 1% Both supported force equally 60 Republicans favored force more 38 Don't know 1 With most of the public unable to remember the Democratic party's position on the war just five months after the vote occurred, it is hardly surprising that Bush had trouble making an issue of it in the election. But what if the Democrats had succeeded in defeating the war resolution and provoking a constitutional crisis? That, surely, would have been memorable, and memorable in a way that would not have been helpful to the electoral prospects of the Democrats. CONCLUDING REMARKS Theories of international politics often assume that nations in conflict with one another behave as "unitary actors." Although the United States was, in its customary way, divided and confused over how to proceed in the crisis, it managed nonetheless to behave with an amazing degree of national unity. It would be implausible to explain this outcome by reference to the Bush's skill in handling Congress, or to the his powers as a communicator, or even to his legal powers as commander-in-chief, since Congress was clearly capable of creating a messy challenge to these powers had it chosen to. My analysis suggests that the basis of the national unity was in elite perceptions of future public opinion -- perceptions that were probably quite well-founded. The Bush administration, judging that the public would follow and ultimately approve of its leadership in the crisis, steered a strong, straight-ahead course, and the leaders of Congress, judging that the public would be repulsed by the effects of a strong challenge to Bush's leadership, declined to go all-out against it. Some readers may wish to ascribe greater importance than I do to the effect of contemporary opinion on Congress at the time of the vote on the war resolution. In view of the correspondence between opinion polls on war with Iraq (roughtly three-to-two in favor of war ) and the outcome of the Congressional vote, it is impossible to rule out an important effect of this type. Yet whatever such direct effect of public opinion on ordinary members of Congress might have existed -- and it is by no means certain that there was much of one, as I argued earlier -- it must not be overlooked that the public opinion that existed at the time of the war vote was in large degree the creation of the president and Congressional elites (see "Elite Leadership of Mass Opinion," this volume). Thus, the Democratic leadership could probably have staged events so as to have brought about a public opinion that was significantly less supportive of war than existed. It could also have scheduled the vote at a time when public opinion and other factors were more propitious for an anti-war vote.[32] The leadership refrained from doing so because of its judgment that, in the long run, such action would be politically self-defeating. My argument in this paper, as with all attempts to explain complex events, is far from proven. But it does have the virtues of explaining, by reference to normal political variables, what seem the most notable features of the domestic politics of the Gulf crisis: Why Bush's domestic leadership, which cannot be called skilled, produced such masterful effects on the homefront during the buildup to war, and why the Democratic leadership, despite its anti-war policy preferences, exhibited so much restraint. One reason the evidence for this argument is weak may be that most of the scholarly literature on public opinion -- of which my paper on elite opinion leadership in this volume is an example -- has concentrated too exclusively on attitudes that can readily be measured in standard surveys, and too little on inchoate or even primordial feelings and instincts. Such feelings and instincts, I suspect, were what the Congressional leadership feared, and what induced it to give the national leader a free hand in the crisis. The primary finding in this study, then, is that, many exaggerated reports of its demise to the contrary, the democratic interplay between leaders and followers was by no means extinguished by the onset of the Gulf crisis. As the president himself expressed it, "I know whose backside is at stake and rightfully so ..."[33] The Democratic leadership of Congress, thinking of its own posterior, felt the same way and acted accordingly. The interplay between leaders and followers did not turn on the mechanical translation of poll results into public policy, but it was a lively interplay nonetheless. References Bennett, W. Lance 1991. "Toward a theory of press-state relations in the U.S." Journal of Communication. Brody, Richard 1991. Assessing Presidential Character: The Media, Elite Opinion, and Public Support. Stanford: Stanford University Press. Cohen, Bernard 1973. The Public's Impact on Foreign Policy. Boston: Little, Brown. Congressional Quarterly 1990 Almanac. Washington, D.C. Gamson, William and Andre Modigliani 1966. "Knowledge and foreign policy opinions: some models for consideration." Public Opinion Quarterly, 30, 187-99. Drew, Elizabeth 1991. "Letter from Washington." New Yorker, February 4, p. 82-90. Jacobson, Gary In press. "Congress: Unusual year, unusual election." To appear in ... Key, V. O., Jr. 1961. Public Opinion and American Democracy, New York: Knopf. Mueller, John 1973. War, Presidents, and Public Opinion. New York: Wiley. Mueller, John Forthcoming. Policy and Opinion in the Gulf War. Chicago: University of Chicago Press. Sigal, Leon 1973. Reporters and Officials. Lexington, KY: D. C. Heath. Smith, Jean Edward 1992. George Bush's War. New York: Nenry Holt U.S. News and World Report, 1992. Triumph Without Victory. New York: Norton. Woodward, Bob 1992. The Commanders.. New York: Pocket Books. Zaller, John 1992. The Nature and Origins of Mass Opinion. New York: Cambridge. ENDNOTES [1] The "indexing hypothesis" was proposed by Bennett (1991), who based his study on earlier work by Cohen (1963) and Sigal (1973), among others. The chapters by Dorman & Livingston and Entman and Page in this volume corroborate this hypothesis. [2] Important statements of this view may be found in Gamson and Modigliani (1966), Mueller (1973), Brody (1991), and Zaller (1992). The chapters by Brody, Iyengar & Simon, and Zaller in this volume corroborate this view. [3] For a general assessment of this argument, see Zaller, 1992, chapters 11-12. [4] This argument is nicely consistent with the argument of Fiorina (1981) that citizens' retrospective evaluations of government actions determine electoral outcomes. [5] The only important legislator about whom I am still lacking insight is George Mitchell; I hope to be able to make up this lack before this paper goes to press. [6] Article by R. W. Apple, Jr., August 6, 1990, p. A7. [7] "Now or Later," August 7, 1990, p. A19. [8] "Iraq's Naked Aggression," August 3, 1990. Editorials in the Washington Post, Los Angeles Times, Chicago Tribune, and San Francisco Chronicle take similar positions; the Wall Street Journal, was a bit more hawkish, but by no means urged military action. [9] See Smith (1992) and U.S. News (1992) for summaries of available accounts. [10] U.S. News, 1992, p. 50. [11] The basis for such criticism was present from the beginning; see Dorman and Livingston, this volume, and David Hoffman, "U.S. Misjudgment of Saddam Seen: Early Evidence of Belicosity, Drive for Dominance Noted," Washington Post, August 8, p. A1. [12] The Soviet empire in eastern Europe was crumbling, but a New York Times headline at the time of the unification of Germany indicates Bush's difficulty in claiming credit for these events: "Bush Declares He Does Not Feel Left Out by Gorbachev and Kohl," July 18, 1990, p. A1. [13] "Still a Wimp," Washington Post, November 4, 1988. [14] "An Absence of Guiding Principle," October 12, 1990, p. A21. [15] Jack Nelson, "Conflict in Gulf is Make or Break Test for President," Los Angeles Times, August 9, 1991, p. AS. Also David Shirbman, "Iraq Crisis Presents Bush with Crucial Opportunity to Define His Presidency, Determine His Future," Wall Street Journal, August 7, 1990, p. A7. [16] Karen Tumulty, "Bush Gets Solid Backing from Congress," Los Angeles Times, August 9, 1990, p. A8. [17] See also "A Dangerous Mirage," Washington Post, November 1, 1990, p. B7, where Henry Kissinger writes that "by the time it is evident that sanctions alone cannot succeed, a credible military option will probably no longer exist." [18] For other analyses, see Mueller, forthcoming; William Schneider, "Public Backs Gulf War as Last Option," National Journal, January 5, 1991, p. 5; Richard Morin, "Two Ways of Reading the Public's Lips on Gulf Policy," Washington Post, January 14, 1991. [19] CBS-New York Times poll, January 11-13, 1991. [20] Although the Congress that voted on the war resolutions was mostly elected in the midst of the gulf crisis, the crisis was not an important issue in the elections. See J. W. Apple, "War Clouds, No Thunder" New York Times, November 6, 1990, p. A1. [21] CQ Almanac, p. 736. [22] It is not clear that this estimate was correct since no whip counts were apparently done. Contemporary journalistic accounts make it seem doubtful that the pro-war position would carry (see R. Jeffrey Smith, "Capitol Hill Hearings Seem to be Setback for Bush's Policy," Washington Post, December 1, p. Axx); but most of those I interviewed, perhaps influenced by intervening events, thought otherwise. By all accounts, however, support for the war was significantly weaker in Congress in November than it was in January when the vote was finally taken. Bush, by published accounts, was among those who feared he would lose on an early vote. [23] See Jeffrey Birnbaum, "New Congress, Full of Sound and Fury Over Iraq, Fuels Bipartisan Outrage by Signifying Nothing," Wall Street Journal, January 4, 1991, AS. [24] Why exactly the leadership scheduled the vote when it did, before Bush actually requested it, is unclear. Restive back-benchers appear to have prevailed upon the leadership to act sooner than it wished. See Walter Pincus, "Loss of Korea Initiative in 1950 Lingers on Hill," Washington Post, January 5, 1991, p. A17. [25] I have mixed reports on whether Mitchell whipped the Senate vote, but will find out. [26] See Adam Clymer, "Votes Backing Use of Force Are Predicted in Congress," New York Times, January 7, 1991, p. A7. [27] "Bush Asks Congress to Back Use of Force," January 9, 1991, p. A6. The estimates of success were apparently not based on whip counts, since later accounts indicated that some Senators were still undecided, but on leaders' estimates of what was likely. [28] See Christopher Madison, "Sideline Players," National Journal, December 15, 1990, p. 3024-3026; Carroll J. Doherty, "Congress Faces Grave Choices as Clock Ticks Toward War," CQ, January 5, 1991, p. 7-10, especially at p. 9; remarks by Rep. Lee Hamilton, a member of the House leadership for purposes of this issue, in "Where Is Congress on the Gulf?" New York Times editorial, January 3, 1991, p. A20. The only element of my account not suggested by the public record is that the leadership was worried that a vote against Bush would precipitate a constitutional crisis; to the contrary, everything the Democrats said implied that Bush would respect Congress' action. [29] Congressional Record, January 12, 1991, p. 441-442. [30] There were, however, also signs of restraint. A staff aide to Nunn said that "no one contemplated there would be a constitutional crisis because Bush would abide by Congress' decision, having asked for it." But what if Congress had voted against the war without having been asked by Bush? "I don't think there was ever any dwelling on that possibility," because few members wanted that. [31] According the CBS-New York Times polls, the percent of the public saying that Congress was doing a fair or good job went from 23 percent in the fall of 1990 to about 48 percent in February, 1991. [32] It is worth noting that, if the leadership had taken actions which altered public opinion and thereby changed the outcome of the war vote, it would not have had any necessary effect on the strength of the relationship between district-level opinion and the roll-call vote in Congress. The strength of relationship between district opinion and legislators' vote seems, at least in this case, a very fallible indicator of popular influence. [33] Woodward, 1991, p. 331.