A majority of Americans believe that President Bush deliberately misled the nation into the war in Iraq. "Bush lied, people died." His defenders have many responses: if Bush lied, so did Democrats; Bush's statements were not lies but rather based on poor intelligence from the CIA; if Bush lied, so what? This is just a sampling. Many of these defenses are based on a very restrictive notion of a lie: "To lie means to say something one knows to be false." Of course, this leaves out lies of omission, which are the main complaint Democrats in Congress have about the information that Bush provided.

Let's leave the specifics of the debate aside, for now, and take a different approach to understanding why people might think that the Bush administration lied about Iraq. What does it mean to tell the truth about something we know? More basically, what does it mean to "know" something?

An obvious answer might be that we know something when we believe that it's true and it actually is true. But this isn't quite right: John might buy a lottery ticket believing that his number will come up tomorrow, and the next day it actually does. (I take this example from Simon Blackburn's excellent little book, Think.) Can John be said to have known that this would happen? Not really. A refinement of the "obvious" answer goes back to Socrates: in addition to John's believing X and X's turning out to be true, John has to have reasonable justification for his belief. (Philosophers aren't completely happy with this revised answer, either, but their arguments are too subtle for me.)

So when Donald Rumsfeld said, "We know where [Saddam's weapons of mass destruction] are. They're in the area around Tikrit and Baghdad and east, west, south and north somewhat," it seems most likely that he was wrong, in that WMDs did not turn up north, south, east, or west of Tikrit and Baghdad. Assuming that Rumsfeld was being honest in his statement, what's interesting about this example is that Rumsfeld was wrong in that what he believed was not true, but there's the strong possibility that he was also wrong because his belief was not justified. To take this a step further, even if WMDs had been found in Iraq, it's not clear (given what we know about Curveball and other Iraqi informants) that the people in authority were at all reasonably justified in believing that WMDs would be found.

This may seem unintuitive: if something turns out to be true, then wasn't it right to believe that it was true? But remember the example of the lottery player. Invading Iraq was an enormous gamble. If it pays off in the long run, I will be very happy. But I doubt I will ever be happy with the arguments for having done it in the first place.

Update: In comments, m raises a very good point that I'd unaccountably overlooked: Doesn't this line of thought condemn everyone, Democrats as well as Republicans, who said, "We know Saddam has WMD?" as liars? Hoist by my own petard. Now I'm not sure that what I've written actually contributes much light to the situation, since everyone already does seem to be focused on justification of belief. I'll give it further thought. For now I'll admit that the implicit link above between lying and making statements that both turn out to be false and lack justification is much too strong. Mistakes produce the same results.