Failures of intelligence

THREE WEEKS before the London bombings of July 7, Britain’s Joint Terrorist Analysis Center advised policymakers that ‘‘at present there is not a group with both the current intent and the capability to attack the UK.’’ That reassuring message from the country’s top intelligence and law enforcement officials, The New York Times reported last week, prompted the British government to lower its terror alert. Less than a month later, 52 people were murdered and 700 wounded when three subway trains and a bus were blown up in the worst act of terrorism the United Kingdom has experienced since the bombing of Pan Am Flight 103 in 1988.

Obviously this was a serious intelligence failure. Undoubtedly there will be investigations into the cause of the blunder. Perhaps heads will roll for failing to ‘‘connect the dots’’ in time to prevent the 7/7 atrocities. (Or perhaps not: CIA Director George Tenet not only retained his job long after Sept. 11, 2001, he was even awarded the Presidential Medal of Freedom.) Whatever is ultimately learned, we can safely assume, will promptly become political fodder for British partisans of every stripe.

But the botched terror assessment raises a question for us, too: Which kind of intelligence failure is better — the kind that badly understates a threat, such as the one in London, or the kind that overstates a threat, such as the insistent warnings before the invasion of Iraq that Saddam Hussein was armed with weapons of mass destruction?

Of course no intelligence failure is desirable. But even in the best intelligence services, they are sometimes inevitable. Foresight will never be as sharp as hindsight. Only after the fact — after the Underground blows up, after 9/11, after the stockpiles of biological and chemical weapons are nowhere to be found — is it clear what the picture looks like once the ‘‘dots’’ are ‘‘connected.’’ Before the fact, it isn’t always clear that there are even any dots to search for, let alone what shape they might take or how reliable they might be.

So what kind of culture do we want intelligence agencies to foster among their operatives and analysts: one that tends to be overly focused on possible threats, or one that is more likely to downplay them? In general, would we rather take action to eliminate a danger that turns out to have been overstated — or take no action, and then be stunned when the enemy strikes?

Surely the question answers itself. When the enemy is an international terrorist organization or a violent and dictatorial regime, preemption must trump reaction. Ousting the most brutal and homicidal tyrant in the Arab world, even if we then discover that he didn’t pose the WMD threat we had envisioned, beats watching Osama bin Laden’s acolytes steer jetliners into the World Trade Center. Bombing the Iraqi nuclear plant at Osirak, as Israel did in 1981, beats waiting until Iraq launches its first nuclear strike. International law has always recognized that states have a right of self-defense, including anticipatory self-defense. So did US presidents long before George W. Bush entered the White House.

‘‘We no longer live in a world where only the actual firing of weapons represents a sufficient challenge to a nation’s security to constitute maximum peril,’’ John F. Kennedy said during the Cuban missile crisis. ‘‘Nuclear weapons are so destructive and ballistic missiles are so swift, that any substantially increased possibility of their use or any sudden change in their deployment may well be regarded as a definite threat to peace.’’ If that was true in 1962, it is even truer today.

To those hyperventilating in the ‘‘Bush-lied-people-died’’ fever swamps, of course, all this is irrelevant. In their view, the administration’s prewar message that Saddam was armed with WMD had nothing to do with 12 years of Iraqi intransigence — the flouting of Security Council resolutions, the expulsion of weapons inspectors, the failure to account for toxic agents. Nor, they seem to believe, was Bush restating a warning about Saddam that had been stated just as insistently by many Democrats (‘‘And mark my words, he will develop weapons of mass destruction. He will deploy them, and he will use them’’ — President Clinton, Dec. 16, 1998).

Above all, they scoff at the notion that the war in Iraq was based in part on mistaken intelligence about Saddam’s WMD programs — those were no mistakes, they say, those were lies. It was all a ‘‘fraud,’’ in Ted Kennedy’s words, ‘‘made up in Texas.’’

But most Americans understand that intelligence failures are not the same thing as lies. And the intelligence failures about Saddam Hussein did not begin under the incumbent President Bush. Back when his father was president, before the first Iraq war, the CIA badly underestimated the extent of Saddam’s quest for chemical, biological, and nuclear weapons.

Fortunately, he was stupid enough to invade Kuwait. That triggered the Gulf War, in the course of which the world discovered just how deadly a WMD menace the Iraqi butcher had become. It is chilling to imagine what the planet would be like today if Saddam had managed to complete and deploy those weapons before the United States even realized he had them.

If intelligence failures are inevitable — and in a world of human fallibility, they are — we are better off worrying too much about our enemies and taking steps to defeat them than worrying too little and being caught, unready, when they attack. Worrying too much led the United States and Britain to topple a brutal tyrant. Worrying too little led to 9/11 and 7/7.

Jeff Jacoby’s e-mail address is jacoby@globe.com.