This is a horror. In a column written June 28, I asserted that more Iraqis (civilians) had now been killed in this war than had been killed by Saddam Hussein over his 24-year rule. WRONG. Really, really wrong.

The only problem is figuring out by how large a factor I was wrong. I had been keeping an eye on civilian deaths in Iraq for a couple of months, waiting for the most conservative estimates to creep over 20,000, which I had fixed in my mind as the number of Iraqi civilians Saddam had killed.

The high-end estimate of Iraqi civilian deaths in this war is 100,000, according to a Johns Hopkins University study published in the British medical journal The Lancet last October, but I was sticking to the low-end, most conservative estimates because I didn't want to be accused of exaggeration.

Ha! I could hardly have been more wrong, no matter how you count Saddam's killing of civilians. According to Human Rights Watch, Hussein killed several hundred thousand of his fellow citizens. The massacre of the Kurdish Barzani tribe in 1983 killed at least 8,000; the infamous gas attack on the Kurdish village of Halabja killed 5,000 in 1988; and seized documents from Iraqi security organizations show 182,000 were murdered during the Anfal ethnic cleansing campaign against Kurds, also in 1988.

In 1991, following the first Gulf War, both the Kurds and the Shiites rebelled. The allied forces did not intervene, and Saddam brutally suppressed both uprisings and drained the southern marshes that had been home to a local population for more than 5,000 years.

Saddam's regime left 271 mass graves, with more still being discovered. That figure alone was the source for my original mistaken estimate of 20,000. Saddam's widespread use of systematic torture, including rape, has been verified by the U.N. Committee on Human Rights and other human rights groups over the years.

There are wildly varying estimates of the number of civilians, especially babies and young children, who died as a result of the sanctions that followed the Gulf War. While it is true that the ill-advised sanctions were put in place by the United Nations, I do not see that that lessens Hussein's moral culpability, whatever blame attaches to the sanctions themselves -- particularly since Saddam promptly corrupted the Oil for Food Program put in place to mitigate the effects of the sanctions, and used the proceeds to build more palaces, etc.

There have been estimates as high as 1 million civilians killed by Saddam, though most agree on the 300,000 to 400,000 range, making my comparison to 20,000 civilian dead in this war pathetically wrong.

I was certainly under no illusions regarding Saddam Hussein, whom I have opposed through human rights work for decades. My sincere apologies. It is unforgivable of me not have checked. I am so sorry.

President Bush has given some good speeches lately, including his talk June 29 at Fort Bragg, N.C., in which he stressed some of the reasons for going into Iraq, and his address this past Monday at the FBI Academy at Quantico, Va., in which he talked about the role of intelligence in defeating terrorists and stressed that "the heart of our strategy is this: Free societies are peaceful societies."
But there's another speech Mr. Bush still needs to give. That would be the one in which he says: I told you so--there was a connection between Saddam Hussein and al Qaeda.

In some quarters, that would of course provoke the usual outrage. Since the U.S.-led coalition went outside the corrupt United Nations to topple the Baathist regime in Baghdad more than two years ago, it has become an article of faith that there was no such connection. Typical of the tenor in both the media and western politics is an article that ran last month in The Economist, describing Iraq as Mr. Bush's "most visible disaster" and opining that "even Mr. Bush's supporters admit that he exaggerated Saddam's ties to Al Qaeda."

If anything, Mr. Bush in recent times has not stressed Saddam's ties to al Qaeda nearly enough. More than ever, as we now discuss the bombings in London, or, to name a few others, Madrid, Casablanca, Bali, Nairobi and Dar es Salaam, or the many bombings in Israel--as well as the attacks on the World Trade Center in both 1993 and 2001--it is important to understand that terrorist connections can be real, and lethal, and portend yet more murder, even when they are shadowy, shifting and complex. And it is vital to send the message to regimes in such places as Syria, Saudi Arabia and Iran that in matters of terrorist ties, the Free World is not interested in epistemological debates over what constitutes a connection. We are not engaged in a court case, or a classroom debate. We are fighting a war.
But in the debates over Iraq, that part of the communication has become far too muddied. Documents found in Iraq are doubted; confessions by detainees are received as universally suspect; reports of meetings between officials of the former Iraqi regime and al Qaeda operatives are discounted as having been nothing more than empty formalities, with such characters shuttling between places like Iraq, Afghanistan and Sudan, perhaps to share tea and cookies. Any conclusions or even inferences about contacts between Saddam's regime and al Qaeda are subjected these days to the kind of metaphysical test in which existence itself becomes a highly dubious philosophical problem, mired in the difficulty of ever really being certain about anything at all.

Certainty is then imposed in the form of assurances that there was no connection. This notion that there was no Saddam-al Qaeda connection is invoked as an argument against the decision to go to war in Iraq, and enjoined as part of the case that we were safer with Saddam in power, and that, even now, the U.S. and its allies should simply cut and run.

Actually, there were many connections, as Stephen Hayes, writing in the current issue of the Weekly Standard, spells out under the headline "The Mother of All Connections." Since the fall of Saddam, the U.S. has had extraordinary access to documents of the former Baathist regime, and is still sifting through millions of them. Mr. Hayes takes some of what is already available, combined with other reports, documentation and details, some from before the overthrow of Saddam, some after. For page after page, he lists connections--with names, dates and details such as the longstanding relationship between Osama bin Laden's top deputy, Ayman al-Zawahiri, and Saddam's regime.

Mr. Hayes raises, with good reason, the question of why Saddam gave haven to Abdul Rahman Yasin, one of the men who in 1993 helped make the bomb that ripped through the parking garage of the World Trade Center. He details a contact between Iraqi intelligence and several of the Sept. 11 hijackers in Malaysia, the year before al Qaeda destroyed the twin towers. He recounts the intersection of Iraqi and al Qaeda business interests in Sudan, via, among other things, an Oil for Food contract negotiated by Saddam's regime with the al-Shifa facility that President Clinton targeted for a missile attack following the African embassy bombings because of its apparent connection to al Qaeda. And there is plenty more.

The difficulty lies in piecing together the picture, which is indeed murky (that being part of the aim in covert dealings between tyrants and terrorist groups)--but rich enough in depth and documented detail so that the basic shape is clear. By the time Mr. Hayes is done tabulating the cross-connections, meetings, Iraqi Intelligence memos unearthed after the fall of Saddam, and information obtained from detained terrorist suspects, you have to believe there was significant collaboration between Iraq and al Qaeda. Or you have to inhabit a universe in which there will never be a demonstrable connection between any of the terrorist attacks the world has suffered over the past dozen years, or any tyrant and any aspiring terrorist. In that fantasyland, all such phenomena are independent events.
Mr. Bush, in calling attention to the Iraq-al Qaeda connection in the first place, did the right thing. For the U.S. president to confirm that clearly and directly at this stage, with some of the abundant supporting evidence now available, might seem highly controversial. But reviving that controversy would help settle it more squarely in line with the truth.