I have been thinking about the "Bush Lied Lite" claim that the administration "cherry-picked" the evidence of Saddam's involvement in terrorism or his acquisition of WMD.

No one, I hope, is so blind to evidence to believe that Saddam had forsaken his dream of wielding a menacing military power. He'd have taken a nuclear weapon any way he could get it, and the best you can say on his behalf was that he was biding his time in the hope and expectation that he could undermine the sanctions regime and split the alliance against him, with the aid of useful idiots abroad, and resume his race to build up a WMD arsenal.

That observation, pertinent or not to the topic at hand, never should be left out of any discussion of the WMD issue.

I'm no expert in espionage, except what I've read about it in broader histories of other topics. But it seems to me to be essentially the business of discovering what the other fellow is up to. There are fundamentally two types of mistake you can make at it: to fail to perceive a threat, or to perceive one that does not exist.

Of the two, the former is more catastrophic -- think Pearl Harbor, or Sept. 11 -- and so if the two errors represent the Scylla and Charybdis of the system, the conscientious espionage worker will strive to sail between them, but tack slightly closer toward the error of over-assumption.

The intelligence trade has two components: collection and analysis -- call them hands and head. Agents in the field will gather a mass of data and information: tips from credible sources, rumors, the gleanings of wiretaps and intercepts, newspaper reports. Among them will be some true facts, and some wrong ones and some good guesses, and some bad guesses, and some deliberate deceptions planted by the other side.

It's the analysts' job to try to weed through them and find the best bits of information and use them to construct a coherent picture of what the other fellow is up to. Of course it's "cherry-picking." That's the whole nature of this part of the business. And, again, there likely will be a bias toward seeing something rather than not seeing it.

Americans never have been good at this, frankly. Especially the analysis part. It's part of why we still rely so much on foreign intelligence services to pass along information to us (something you can discover by reading between the lines of the various domestic inquiries into the misleading U.S. intelligence on Iraq).

In World War II, the British had the best-balanced espionage. Americans had decent intelligence-gathering (especially as a result of code-cracking), but iffy analysis. The French, before they got knocked out, had superb espionage in Nazi Germany, but they lacked the political will to act on it. The Soviets also had a vast network of spies -- in the capitals of their nominal allies Britain and America. They didn't know anything about Germany and didn't seem inclined to do anything but accept the tips dutifully passed along to them by the governments in London and Washington. Their analysis was worse than the Americans'.

An intelligence agency can build up a picture that vastly over-rates an enemy's strength and capabilities without any outside political interference. In fact, it seems to be a systemic tendency. Alan Pinkerton, McClellan's spymaster in the Civil War, consistently over-estimated the strength of Confederate forces, which accounts in part for McClellan's cautious ineffectiveness.

An even more remarkable example of American intelligence error turns up at the end of World War II. The U.S. convinced itself Hitler was preparing a vast, impenetrable "national redoubt" in the Alps, a self-sufficient Alpenfestung where he and his most rabid followers planned to hold out for years.

The situation that led to this mistaken belief was structurally similar to the run-up to the Iraq War intelligence fiasco:

• A previous failure to perceive an enemy threat. In World War II, it was the Nazi concentration and attack in the Ardennes in late 1944 that led to the Battle of the Bulge. In Iraq, it was the realization, after the 1991 war, that Saddam was much more dangerously armed and capable than the West had suspected.
• An image consistent with the known character of the enemy. The Allies knew Hitler had sworn there never would be "another 1918" in Germany; he would fight to the last, and he had legions of fanatical followers. The mountains of Bavaria were the birthplace of Nazism and exerted a romantic, almost mystical pull on the party leaders. Furthermore, the Allies knew the Nazis had, in fact, moved some armaments factories underground in other places to escape Allied bombing. In Iraq 60 years later, Saddam did covet WMD and had gone to tremendous lengths to acquire them. His secretiveness and evasions in the face of inspections were legendary. The sanctions against him were known to be porous, and he continued to probe for a hole in them, as when he approached Niger about uranium.
• Deliberate deception. Saddam gave every indication that he had potent weapons, even while he officially denied it. He blustered like a man with a hidden arsenal -- he would have been a fool not to, given his neighborhood. In World War II, Goebbels, when he learned of the American hysteria over the Alpenfestung, set up an entire propaganda office to encourage them to think it was true.

The notion of a Nazi Alpenfestung arose in Switzerland, which had in fact used its natural topography to turn itself into a fortress state. The Swiss, observing across the border into the Tyrol, imagined they saw the Germans doing the same thing. In fact, the Germans had begun to scout defensive locations along the old World War I frontier of Italy as their armies in the peninsula slowly withdrew to the north. Many leading Western intelligence agents were based in Switzerland, as well as newspaper correspondents. The stories began to percolate.

The more the Americans thought about it, the more alarmed they became. A look at the map revealed the Alps as a natural focus of retreat for the hundreds of thousands of German troops still undefeated in northern Italy, Hungary, and Bohemia. There were known to be old salt mines and tunnels deep into the mountains. The prevailing weather would neutralize the Allies' great advantage in air power. It all began to make sense. Soon the stories had grown to tell of monstrous fortifications, underground factories, secret airfields, armies of slave laborers, stockpiles of supplies.

Hitler and the top Nazis and surviving generals would retreat there as the pincers closed around Berlin, and from the Alps they would lead a resistance that could hold out for years. They would hope for a split among the Allies, or gain time to develop a dreaded secret weapon. To storm the Alpenfestung would cost the Allies more casualties than Normandy, would drain vital resources needed in the Pacific, and still likely would leave behind fanatical Nazi remnants who would keep alive for generations the myth of unvanquished National Socialism.

Never mind the obvious evidence that the Nazis at this stage of the war lacked material capabilities to construct such an empire along the hundreds of miles of mountains from Lake Constance to Carinthia.

Allen Dulles, Office of Strategic Services representative in Bern, initially was skeptical of the Alpenfestung reports, but he dutifully passed them up to Washington anyhow. But by February 1945 he was writing as though the redoubt was a reality and something to be deeply concerned about, even though he admitted "it is impossible to put your finger on the particular area where the foodstuffs are being collected, or where these underground factories are being prepared."

Maj. Gen. Kenneth Strong, head of intell at SHAEF, by March 1945 also was treating the Alpenfestung as a reality, despite the obvious paucity of real evidence. "[R]eports of deep dugouts, secret hiding-places, underground factories, and bombproof headquarters were confusing and unconvincing. No single piece of information could be confirmed." But, he added, "After the Ardennes, I was taking no more chances."

Once the idea became fixed in the minds of the U.S. military and espionage leaders, every new observation was fitted into it. Aerial observation of long trains headed south in Germany were seen as confirmation of the Alpenfestung. In fact, they were full of looted art treasures being sent south for safekeeping.

Perhaps nothing illustrates this better than the huge map that hung in Ike's headquarters, labeled "Reported National Redoubt." Every new tip and observation within the suspect area was marked with a red dot. Naturally, red dots proliferated within the space, and didn't exist outside it, and you had to get much closer to the map to see that most of them were marked "unconfirmed."

In fact, the idea was never more than a lark to the Germans. Hans Gontard, the SS Sturmbannführer in the border town of Bregenz had begun feeding the hints into Switzerland. When he saw reports out of Washington taking them seriously, he laughed at the gullible Americans. He described the deception to the gauleiter of Tyrol, who realized that this actually was a good idea, and he sent a memo about it to Martin Bormann in Berlin.

But Hitler may never have heard of it. Bormann wisely pocketed the memo, realizing that the Führer at that point still was intent on a military victory, not a defeatist retreat. Only Goebbels used the idea, and only for propaganda to terrify the Allies.

The idea of a Nazi Götterdämmerung high in the aeries of the Alps seemed to fit Hitler's self-dreams so much better than his eventual fate, to die like a rat amid the drab canals of Berlin. "Allied military officials were thinking more like the Nazis than the Nazis themselves," writes Stephen G. Fritz, who opens his "Endkampf" with an excellent brief account of the redoubt hysteria.

The Americans convinced themselves that the redoubt was real -- or that the likelihood of its being real was too great, and the consequences too serious, to be ignored.

The decisions had consequences. Eisenhower, by wheeling his armies southward, deprived Churchill and Montgomery of their dream of marching to Berlin. This was more a British fantasy than a realistic strategy by early 1945. But the Alpenfestung hysteria did cause Eisenhower to hold back Patton from marching on to Prague, and if he had done so Patton might have kept the Russians out and spared the one Eastern European nation that hadn't already been lost to the pleasures of a Soviet utopia.

The depletion of Western Allied forces in northern Germany also opened the door for the Russians to make a rush toward Denmark (and the rest of Scandinavia) after the fall of Berlin, which they in fact tried and might have accomplished had not Montgomery deliberately blocked them on the Lüneburg Heath. The Cold War might have been fought along a different Iron Curtain line, with different results.