"...[O]ne thing to watch for will be how the president himself addresses the importance of public opinion in explaining his eventual decision on how to proceed -- and how he explains himself should he decide to defy the public's wishes. It will give us an important glimpse into what this young and largely untested leader is made of and how he views his presidency."

That's from the Plumline's Greg Sargent, who notes CNN polling showing nearly 60% of the public opposes sending more troops to Afghanistan and that a small majority believes that the war is turning into Vietnam.

What is the picture inside of our heads about the war in Afghanistan? The American people seem to want their president to take public opinion into account. But does a president need to? Does it matter, firstly, whether a war can be just unless a majority supports it? Does public backing make it easier to prosecute a war? Does public opposition to a war increase the chances of defeat? How does public opinion track the progress of a war or conflict?

According to political science literature, war and public opinion intermix in several, fairly circumscribed ways. When a conflict begins, particularly as troops are leaving the homeland for the first time, people tend to support the Commander in Chief -- a rally round the flag effect, as John Meuller of the University of Ohio calls it.

For a war of more than a brief duration, several theories have been put forth. One is that the public looks to elites for cues about how to react. The elite, right now, clearly believe that the war in Afghanistan is going poorly; ergo, the public's growing opposition is just a reflection. Another theory holds that the public is capable of assessing risks and benefits and does so fairly independently of the elite. Mueller has written that, as the human costs of war inevitably accumulate, the public will inevitably come to oppose a war whose duration exceeds expectations. There is some evidence that support for a war is linked to -- and the direction here isn't clear -- presidential approval more than it is to casualty levels or an appraisal of whether the war is going well or not. Peter Feaver, a Duke professor who served in the Bush administration, believes that "support for war is a function of two attitudes: the retrospective attitude of whether the war was the right thing in the first place, and the prospective attitude of whether the war will be won."

In his mind, the Obama administration isn't sending the right cues. If Obama doesn't say that he thinks the war will be won -- only that it must be fought -- he is betraying a doubt that public opinion will come to reflect -- and has come to reflect. In other words -- is something in America's interest if Americans don't think it's in their interest?

In Iraq, it took about two and a half years before the public became convinced that the war in Iraq wasn't going well, and it took about another year before a majority started advocating for U.S. troops to withdraw. For President Bush, the war became a proxy for his presidency. It is not clear whether President Obama faces the same calcification challenge as President Bush. That is, even though public opinion right now on Afghanistan closely approximates public opinion on Iraq when opposition to the Bush administration's theory of international relations hardened, there is no reason to simply assume that, a year into office, President Obama will face the same type of pressure. Obama's challenge is different. Because of 9/11 and a concerted PR case for war, the Bush had Americans' hearts and minds on Iraq, and then lost them when the 9/11 rationale didn't check, the WMDs weren't found, and casualties began to mount. So far as hearts and minds go on Afghanistan, Obama has never acquired them, so he has less to lose. And it may well be that the less Americans are directly engaged in a war, the less they care about it -- even as they insist that they do care about it.

Some on the right believe that Americans don't like to lose wars, period, and that they'll punish President Obama for withdrawing too early from a conflict to save face. It may be better to borrow from Dwight D. Eisenhower's example, here. He was rewarded, in the presidential campaign of 1952, for promising to end the sputtering Korean War, which did not necessarily mean that he promised to win it. Indeed, over the objections of his generals and his CIA director, he signed the truce, even though, to that time, America had never actually lost a war. For that he was called -- yes -- an appeaser. Eisenhower had credentials that Obama lacks -- five stars being the main one -- and so the historical analogy doesn't quite hold up here.  But by the time Eisenhower came to office, the war was not popular. He did not try to win it; he ended it, and was rewarded politically. Afghanistan and Korea share several features, too: both are/were attrition wars, guerrilla conflicts, where a nation-state with undefined goals is trying to adjust to an enemy whose principal means of fighting is terrorism.

Arguably, the President is trying to convince Americans that Taliban and Al Qaeda fighters in Afghanistan represent a direct threat to U.S. interests; such a threat was never really assimilated by the public during Korea or Vietnam. Obviously, perceiving the presence of a direct threat makes Americans more attentive to the war. But, as is true in the case of the first Iraq war, it does not have to be present in order for a war to be popular. The first Iraq war succeeded, from a public opinion standpoint, probably because the elites were almost universally supportive and the risks were seen as minimal. President Clinton's first major conflict, Bosnia, was seen as a morass, was confusing to understand, involved a tangle of ethnicities and religions and land claims, and despite humanitarian-based upticks when, for example, mass graves were discovered, it was never popular. It also never really hurt President Clinton.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.