Thursday, March 31, 2005

Stupid Intel Tricks

That title is a bit harsher than what I actually have to say, but it will have to do. Besides, I've always liked the "Stupid Tricks" motif.

Several intelligence reports have come out in recent months. There was the final report of the 9/11 commission, back before the election. And now, we've got a report on the Iraq intelligence that was used in 2003 as a casus belli. I'm not linking any of them. They're easy enough to find if you go in for that sort of thing. But there are a few general points I'd like to make. They're all things I've been chewing over for months, and this seemed as good a time as any to go over them.

Point the first: "Being wrong" isn't the same thing as "lying".

This is something I have to remind my daughter of from time to time. Although they are alike in that both involve your having said something that was at variance with reality, there is nevertheless a vital difference. The people who said Columbus was crazy to sail east to find a new path to the Orient because the world was flat weren't lying. They were saying what they truly believed to be true. They were incorrect, but sincere. In his turn, when Columbus said that he'd discovered a new path to the Orient, he wasn't lying, he was just flat-out wrong. He'd discovered America, not China. But again, he was sincere. Sincerely wrong, but sincere.

What's all that in aid of? Well, I think that all the people crying, "Bush lied!" aren't really right. They're sincere in their detestation and hatred of the President, but I think they're wrong in saying that he lied. Granted, the intelligence he used as a justification for going to war was wrong. Vastly incorrect. But he, and everyone he worked with, thought it was correct. Just about everybody thought that Saddam had some sort of WMD program going on. After all, he wouldn't go to such absurd lengths of hide-and-seek with the UN inspectors to hide nothing, would he?

Actually, yes, he would. We found that out the hard and expensive way. Not as hard or as expensive for us as for Saddam, but hard and expensive enough for any ordinary purpose.

But, that's known as "being wrong", not "lying". He thought that to be the case, and acted on the best information he had at the time. Knowing what I knew then, I supported that decision. And in the end, it may still work out OK. But it was a decision made on faulty premises. That is something to bear in mind.

Point the second: It's not always what you don't know that will bite you, it's what you KNOW that ain't so.

Most of us have enough native common sense to keep our hands in our pockets when faced with something we know nothing about. No, what really lands our essentials in the grinder are the things we know with iron-clad certainty that are actually total bat puckey. That's what makes Intelligence so interesting, and so dangerous. We are rarely certain of what we know. What do we know, how well do we know it, and how do we know it to be true anyway? Some kinds of intelligence we do quite well indeed. We're probably the world's best when it comes to image intelligence, or signals, or things of that sort. The techno-geek stuff, we've got in spades. Lately, though, we have tended to come up short on good old-fashioned human intelligence. Agents in place, feet on the ground in places that we need eyes-on-target information about. The right ears listening to the right conversations. The sorts of things that we can use to corroborate bits and pieces of information we get elsewhere. Without that, it's easy to get into sticky situations where you end up acting on information that you're sure of, but that isn't factually correct.

Such as the cruise missile strikes on Afghanistan and Sudan in 1998. There's evidence to suggest that OBL's crew hoaxed up those cell-phone calls, knowing Uncle Sam would be listening in. That's only one recent example.

This is something we have GOT to start doing better.

Point the third: Given that good, reliable information is so rare, a doctrine of pre-emption can be a very dangerous thing.

Intelligence data can't be thought of as nice, clean points on a graph. The points are there, all right, but they have big, fat error bars attached. Now, the premise undergirding any doctrine of pre-emption is that we know what the other guy is up to. The problem is, that's not always the case. A phrase we've heard a lot of lately is "actionable intelligence". It's a very hard thing to get. Intelligence that's truly good enough for you to act upon with a high degree of success is a very rare thing indeed. We had it in 1942, when we were reading the Japanese Navy codes almost as though they were broadcasting in the clear. We gave them a right splendid mauling at Midway that summer, and they never suspected that their codes were compromised. That was proper intel work, that was. But that's a very unusual situation. More often, we have situations like the Song Tay prison camp raid during the Vietnam war. Our special forces were able to infiltrate undetected, only to find that the prisoners had been moved out the week before. Tactically brilliant, but the undergirding intel was flawed.

Jumping the gun on bad intel can sometimes have disastrous consequences. Take the situation in 1984, for instance, when Soviet intelligence was reading a routine series of NATO exercises as a prelude to war. What if they'd taken that as actionable? The consequences would have been monstrously ugly. So much so, that their political leadership opted to take one more look at the situation, and see how things developed ... with the result that we're all still here.

And that's probably the first, last, and only time that I'll ever have anything good to say about that lot of miserable bastards.

Anyhow, I think it's pretty plain that in order to be usable for a doctrine of pre-emption, your intelligence has to be absolutely pristine. Not just pretty good. Not just indicative. It's got to be spot-on, unquestionably reliable. And even then, you're liable to make mistakes.

On the other hand, what if the consequence of not taking action means eating an inbound bucket of sunshine? Or, as Secretary Rice put it a couple of years ago, if the "smoking gun" takes the shape of a mushroom cloud?

Aye, there's the rub.

If the stakes are sufficiently high, you may just have to throw down anyway.

Which brings us 'round to:

Point the fourth: Don't play nuclear Dodge-em with a superpower that's got a case of whoop-ass under each arm and a thirsty gleam in his eye.

If, by some miraculous chance, someone from Iran or North Korea is reading this, take this lesson to heart. Nothing good can possibly come of provoking an atomic scrum with us. You might hurt us. You just might. But it will be the end of your history. We'll hate ourselves afterwards. There'll be absolute torrents of self-loathing, and recriminations galore. You, on the other hand, will be too busy being dead to have time for regrets.

No comments: