What policy makers want most from their intelligence services is support for whatever it is they want to believe. So President Clinton was delighted to hear there was no foreign terrorist involvement in the Oklahoma City bombing or the downing of TWA 800, and President Bush was pleased to hear that Saddam Hussein's possession of weapons of mass destruction was a "slam dunk."

		Jack Kelly is national security writer for the Post-Gazette and The Blade of Toledo, Ohio (jkelly@post-gazette.com, 412-263-1476).

What policy makers need most from their intelligence services is the truth. But the truth usually is very hard to find out. And even when they learn the truth, intelligence services sometimes fail to recognize it, or are reluctant to tell it.

The director of National Intelligence recently declassified the "key judgments" portion of the most recent National Intelligence Estimate for Iraq.

The NIE is a projection of trends that comprise the thinking of the 16 U.S. intelligence agencies. "Its purpose is to provide the president with an understanding of what the future is likely to be -- and to provide this understanding soon enough, and clearly enough, that if the president doesn't like what lies ahead he can take steps to change the future before it happens," said Herbert Meyer, who used to work on NIEs for legendary CIA Director Bill Casey.

The future prophesied in the latest NIE is stark:

"Iraqi society's growing polarization, the persistent weakness of the security forces and the state in general, and all sides' ready recourse to violence are collectively driving an increase in communal and insurgent violence and political extremism," the NIE said in the foremost of its key judgments. "Unless measures to reverse these conditions show measurable progress during the term of the estimate, the coming 12 to 18 months, we assess that the overall security situation will continue to deteriorate at rates comparable to the latter parts of 2006."

The initial draft of the NIE was completed in December, and, I suspect, had much to do with the change in strategy in Iraq President Bush has made. The entire NIE now is circulating among policy makers, including members of Congress.

Eli Lake reported in The New York Sun Monday that four of the agencies involved in drafting the NIE have filed a formal "dissent" to a major conclusion.

Army and Marine intelligence, the National Geospatial-Intelligence Agency and the Treasury Department's Office of Intelligence and Analysis think that al-Qaida is running the Sunni insurgency in Iraq. The other 12 agencies think the Sunni portion of the violence is dominated by former supporters of Saddam Hussein. Al-Qaida, the majority thinks, is playing only a relatively small role.

The dispute is fraught with political implications. If the majority is right that the insurgency is dominated by ex-Baathists, that bolsters the view of those who think the conflict in Iraq is largely a civil war. But if the dissenters are right that al-Qaida is running the show, this bolsters President Bush's contention that Iraq is the central front in the war on terror.

Major dissents in NIEs have, in recent years, been rare. Typically, disputes in NIEs are "resolved" by reducing conclusions to the lowest common denominator, with lots of "on the one hand this, on the other hand that" qualifying phraseology. This covers wonderfully the posteriors of all involved in drafting the report, but pabulum of this sort provides policy makers with little useful guidance.

Policy makers often prefer pabulum to sharp disagreements within the intelligence community, because then they don't have to choose one point of view over the other. If a president has to choose, he could choose wrongly, and the agencies whose views he overrode could be annoyed.

It makes a very big difference whether the majority or the dissenters are right about who is calling the shots in the Sunni insurgency in Iraq. But who is right?

The majority who say the Baathists are still in charge is large. But the dissenters -- especially Army and Marine intelligence -- represent the agencies in the best position to know what's happening on the ground in Iraq.

The position of the dissenters is close to that expressed in a report last August by Marine Col. Peter Devlin, chief of intelligence for Anbar province. And the CIA didn't exactly cover itself with glory in its prewar predictions. Have the CIA's sources improved? Are they better than the military's?

I don't know the answers to these questions. But I do know that finding them is critically important. Intelligence is the radar of policy. If our radar is busted -- either because we lack the means to find out what is going on, the wit to understand it or the guts to face it, our policy will be broken, too.