Everything about the war with Iraq was billed as big, from the shocking and awesome dimensions of the proposed initial bombardment to the Hitlerian-sized menace posed by Saddam Hussein.
"What will follow will not be a repeat of any other conflict," said Defense Secretary Donald Rumsfeld. "It will be of a force and scope and scale . . . beyond what has been seen before."
But on March 23, after several days of unchallenged advances through southern Iraq, American forces took their first combat casualties -- 20 soldiers were missing or killed and another 50 were wounded. Television viewers at home saw the stunned faces of American POWs and learned that four other POWs may have been executed.
The bad news from the front lines seemed to have a sudden and dramatic impact on the confidence of the American public. The stock market plunged more than 300 points Monday after reporting some of the biggest gains in 20 years the week before.
A Pew Research Center survey released on Tuesday showed that while support for the war remained high, the percentage of Americans who believe it is "going well" plummeted from 71 percent on Saturday to 38 percent on Monday.
Everything was big about the war, it seemed, except our tolerance for casualties.
Did we really believe that no one would be killed in this war? Are we, as pundits and politicians have said for years, so squeamish about casualties that any bloodshed makes us want to want to abandon the fight? Are we more timid now than 60 years ago when more than 1-million men were killed or wounded fighting the Axis powers?
No matter the war, no matter the cause, the American public's stomach for battlefield casualties has proven remarkably elastic. From the Civil War right through to the imminent Battle for Baghdad, once a war has begun two factors -- one practical and one philosophical -- determine the level casualties Americans will accept.
"Contrary to the expectations of political leaders and the media, the American public is willing to tolerate some casualties if they believe the operation is going to be successful, if we're going to win," said Christopher Gelpi, associate professor of political science at Duke University. "The level of casualties the public is willing to accept also depends on how important they think the mission is."
While conducting research for a book on the perceived animosity between the civilians and the military, Gelpi and his coauthor Peter Feaver surveyed Americans in 2001 about the number of casualties they would find acceptable in a variety of conflict scenarios, including one to destroy weapons of mass destruction.
"A majority of respondents would be willing to tolerate around 500 casualties," Gelpi said.
This may sound like a small number when compared to the 6,639 Americans who were killed on average during each month of World War II, but Gelpi says that it is a significant figure because it bucks the conventional wisdom that Americans insist on an arm's-length, bloodless approach to warfare.
"There is a reservoir of support for taking casualties that any president can tap into," he said.
Gregory Urwin, professor of military history at Temple University, sees examples of American's fluctuating tolerance for battlefield casualties as far back as the Civil War.
In 1864, a presidential election year, Gen. Ulysses S. Grant was marching the Union army toward Richmond. The public expected a continuation of the previous year's successes at Gettysburg and Vicksburg, but on the way to Richmond, Grant suffered 60,000 casualties, approximately half the men with whom he had begun the campaign.
"There was tremendous war weariness. There was tremendous doubt in Grant and in the man who had picked him to lead the Union," Urwin said. "Lincoln thought he was going to lose the election and he had made contingency plans for what he would do about the war between the election and the inauguration."
Victories elsewhere saved Lincoln's presidency, Urwin said, and the war continued until Richmond fell the following April.
Gelpi and Feaver went back and looked at polling data from 1949, a year before the Korean War began, through 1994, during the crisis in Kosovo, and compared public sentiment to reports of casualties from the various conflicts during that period.
"What we found was that public support for the war did not drop simply because there were casualties suffered or simply because the war dragged on over time," Gelpi said.
In Korea, about 5,000 troops had been killed within the first three months, but support for the cause was a robust 66 percent, Gelpi said. But when the Chinese entered the war in December of that year and pushed American forces back down the Korean peninsula, public confidence sank to 39 percent. Why? Because defeat seemed more likely than victory.
In Vietnam, the turning point was the Tet Offensive in early 1968. The Vietcong suffered 37,000 casualties in city fighting, far more than the Americans.
Still, "The Vietnamese had won the war of wills," Urwin said. "Psychologically they whipped us."
A month before Tet, public support for the war had been just under 50 percent, but after Tet that support eroded steadily to less than a third, Gelpi said, as the number of dead mounted in what was perceived as a losing cause. On this date in 1968, for example, 57 Americans died in combat.
"The drop in support comes when casualties come in conjunction with defeat," he said.
Even World War II commanders were not immune from criticism at home when they were losing men by the thousand during island assaults in the Pacific, Urwin said.
Defeat is sometimes in the eye of the person holding the remote control, however. Battlefield losses can stiffen resolve or break the nation's will depending on how they are presented by political and military leaders.
The decision to withdraw from Somalia in 1993 after an armed mob downed two American Black Hawk helicopters and killed 18 highly trained U.S. Army Rangers is often cited as proof of the United States' new reluctance to endure casualties. But Gelpi points out that immediately after that debacle, two-thirds of the American public were still in favor of staying in Somalia to finish the mission.
Had the government made a case that capturing warlord Mohamed Aidid was not only important, but achievable, "there was a reservoir of support the Clinton administration could have tapped into," Gelpi said. "They didn't."
Contrast that decision to the way the Bush administration has dealt with the disturbing images of the American POWs and reports that troops were being ambushed by Iraqis pretending to surrender.
Some disagreement emerged late in the week between the White House and military commanders on the subject of whether the war has slowed because of unanticipated resistance from the Iraqis. But the administration has adopted a mantra of optimism, repeating at every opportunity "The outcome is not in doubt."
"I can assure the American people we're making good progress," Bush said a week ago.
"It isn't a matter of timetable," Bush said Thursday. "It's a matter of victory."
But assurances of ultimate victory will not prop up support at home if the evidence provided by embedded reporters at the front indicates otherwise.
Right now, President Bush is regarded well for his perseverance, but his approval bump was not as great as his father's 12 years ago during the Persian Gulf War, said James McCann, associate professor of political science at Purdue who is a guest scholar at the Brookings Institution.
Resentment at home could build if casualties, both military and civilian, mount without evident progress. At that point, McCann said, the public may come to regard Bush as hardheaded and unwilling to accept a changing reality. "A personal virtue with only a little bit of new information can be turned into a vice," he said.
Public opinion could be especially volatile given the ambivalence Americans felt about the war going in. This is "our first optional war," McCann said, completely different from one in which we respond to another nation's first strike.
A Zogby poll taken before the war began showed American's opposition to the war climbing as the number of potential casualties rose. Half of all respondents were opposed if there were thousands of American casualties. The numbers were almost identical if it were thousands of Iraqi casualties.
"Support is not robust. A lot of people are saying that they are behind it because they want to support the president. That creates a precarious situation," said Steven Kull, director of the program on international policy attitudes at the University of Maryland, in a Washington Post story.
"If there were continuing casualties, and the military operation was perceived as being bogged down, and criticism continued," he said, "then the underlying doubts about the operation would start surfacing. But I don't think we're near that right now."