Beauty of Gray

Friday, November 15, 2002

One more note on terrorist warnings

Yesterday I criticized what I feel is some sloppy thinking on the part of those generating and analyzing terrorism warning. But there’s on additional note that occurred to me on reflection. Assuming that the analysts who are doing the worrying have thought it through, the fact that they believe there’s a possibility that increased message traffic or “chatter,” and the bin Laden tape being released could be correlated to an upcoming terrorist strike means that, despite some reports to the contrary, al Qaeda still has a fully functioning hierarchy. It has not been reduced to a collection of isolated cells, cut off from each other and from their superiors.

Consider: if al Qaeda now were simply 500 isolated cells, each would only communicate amongst themselves, or perhaps with one or two connected cells. If this were the case, even if an impending attack were signaled by increased message traffic, it would only be an increase on the part of 1/500 of the total group. It’s only when there’s top down planning or coordination that an upcoming strike by one group could produce an increase in message traffic more or less across the board.

Similarly, unless bin Laden is still either directing operations or close enough to those who are to know what’s going on, there’s no way for him to coordinate any release of a personal message with a future strike by al Qaeda.

So, reading between the lines and giving the US intelligence community the benefit of the doubt (which might not be justified, of course), it seems that they believe the operation in Afghanistan, while it may have degraded their operational capabilities, has not really broken up the al Qaeda chain of command

Link posted by Doug Turnbull at 1:29 PM

Thursday, November 14, 2002
Surrender, POW’s, and winning wars

While I criticized Niall Ferguson’s book The Pity of War below for sloppy argumentation, he does make quite a few interesting points along the way. One of the most interesting is his emphasis on the role that surrender plays in war. I think he makes a mistake in raising it to the decisive element—he claims that the main difference between 1918 and 1916 was that the Germans were surrendering in large numbers—but he makes the very good point that an important factor for soldiers making a decision to surrender is their likely treatment afterwards.

We generally think of the various accords about proper treatment of POW’s as being a humanitarian issue, making sure that people aren’t abused or mistreated. Ferguson points out that this is also a policy that advances your self-interest. If you are known to treat prisoners well, then opposing soliders in tight situations will be more likely to give up and surrender. In contrast, if you are brutal and torture or kill captured enemy soldiers, then they will fight to the death rather than surrender, which will make defeating an enemy army much more difficult.

And this is not a purely academic issue for western armies, something that only “savages” do. Ferguson shows that on the Western Front, the casual execution of surrendered soldiers was a relatively common occurrence. And not just in the heat of battle, but also on the way back from the front, or in follow-on, consolidation maneuvers.

Ferguson goes too far when he claims that a change in this situation on the Allies side, partially due to the incoming Americans who tended to treat POW’s better, was a decisive factor in the end of the war. I think his analysis is very limited and ignores the military situation on the ground—the Germans were being driven backwards, a situation that almost always results in increased surrenders—as well as other factors. But it’s nonetheless an interesting perspective, and one which may very well play out in a second war with Iraq. Hundreds of thousands of Iraqi soldiers surrendered to the US forces and were treated fairly and eventually returned to Iraq. Any soldier serving now almost certainly knows one or more veterans of the first Gulf War, and knows that surrender most certainly is an option, one that carries few costs.

Link posted by Doug Turnbull at 1:48 PM

Magical thinking and terrorist attacks

No, I’m not talking about magical thinking on the part of the terrorists; I’m talking about the CIA. This article from today’s Washington Post is a prime example of the questionable assumptions which seem to be driving US intelligence.

There are different kinds of magical thinking, but in this case I’m referring to a combination of the natural human tendency to see patterns, the assumption that correlation implies causation, and the tendency of people to find and remember data that supports a pre-existing position and to ignore contradictory evidence.

Put these three tendencies together, and you have the following:

The volume of threat information waned in the weeks following Tenet's Oct. 17 testimony, only to begin rising again over the last week to 10 days, sources said. The magnitude is again approaching pre-Sept.11 levels, these sources said.
"That threat environment level was high then, and it has not lessened," a senior administration official said yesterday. Factors contributing to the increase could include the threat of war against Iraq or the celebration of the Muslim holy month of Ramadan, which began this week.

Bruce Hoffman, a terrorism expert who heads the Rand Corp.'s office here, said the tape "could mean that something has been set in motion."
"Bin Laden's appearances have always been carefully orchestrated, and unfortunately they've often presaged a major al Qaeda attack or development," Hoffman said.

Absent fairly specific threat information (which does occasionally come in), US intelligence has little idea when a terrorist strike is in the offing. But there’s lots of information, so the human mind constructs these assumed correlations, none of which, it seems to me, are really meaningful. “Chatter” has spiked. Well, yes, but it’s spiked before and nothing has happened. And does it always spike before any attacks? Here is perceived pattern number one, the most plausible of the lot. But I’m skeptical of it’s real value, although I don’t have the clearance or the access to look and compare if “chatter” levels have any predictive values. If they do, I bet it’s low.

Beyond the chatter, which at least has an operational basis to it, you have the proposed correlation of Bin Laden tapes with attacks. Again, I’m skeptical. There have been several tapes released since 9-11, and the only major attack, in Bali, took place during a period of silence. In the years leading up to 9-11, there were a number of tapes and appearances, yet I don’t recall any clear link with follow-on terrorist attacks. You add in the option of “major attacks and developments” and now the supposed correlation is so broad that you’re almost certain to fin something.

And the final canard—that attacks are chosen to correspond to significant dates—has the exact same problem but worse. There are so many significant dates that if you look, you can often find something. Yet many attacks still occur on completely meaningless dates.

Now, the US has avoided any major attacks since 9-11, so the intelligence agencies are apparently doing some things right. But I sincerely hope they’re spending most of their time actually analyzing real data, and are not letting these preconceived “patterns” interfere with their analysis, since they are so likely meaningles

Link posted by Doug Turnbull at 9:18 AM

Wednesday, November 13, 2002
Revisionist history

I recently finished reading an interesting but flawed revisionist history of WWI, The Pity of War by Niall Ferguson. To keep with my policy of blogging everything two days late, here the week after Remembrance Day is a good time to comment on the book, and on revisionist works in general.

On the one hand, assuming you already have a good understanding of the period, a well thought out revisionist history can be tremendously rewarding to read. Either you are convinced and come to a new understanding of a historical situation, or else you are forced to defend your own understanding, which forces you to really master the details of your own position. So it can be a win-win, if the revisionist history is done right.

On the other hand, I go into any revisionist history with several ounces of skepticism. First, because there’s usually a good reason why something has become the conventional wisdom in a field. Unless some new information has come to light, the revisionist is stuck arguing against quite a large collected weight of authority. It doesn’t mean he’s wrong, of course, it makes this more likely. It takes a certain amount of arrogance to believe that hundreds of colleagues are wrong, and only you are right.

And second, because there’s more than a bit of self-interest involved in writing a revisionist work. If you can stir up a controversy, you’ll sell more books and make a name for yourself. Assuming the work isn’t too shoddy, even if you end up wrong, it can help your career and make a name for yourself. There is also a romantic pleasure in being an iconoclast, speaking out against authority and pitting your own thought against the world. And finally, while much weaker and hence less destructive than in art, there is a certain pressure towards novelty in all scholastic fields. Synthesizing, collecting, and narrating accepted facts is serviceable yeoman’s work, but to achieve renown you have to say something really new. And an easy way to do that is to challenge existing orthodoxies. (More valuable and much less common, as it requires some genius, is establishing whole new ways of looking at the world, establishing a new set of orthodoxies.)

Ferguson’s book in particular should set off alarm bells in the reader, since at the outset he puts forth his objective. Not to present one revisionist account, but to take on 10 different large questions about the war, and in each case to show that the conventional wisdom is wrong, and he is right. This is an amazing display of hubris, and right away makes me think that a couple of things might be the case. I wonder if he, either for careerist reasons or for the joy of being an intellectual rebel, didn’t overreach himself. Starting from a few good positions, and then deciding that he must overthrow all orthodoxy in the field?

First, some of the CW positions he’s overthrowing might not actually be the CW—he might be storming through an open door, so to speak. In other cases, his reasoning might be faulty—he might present evidence that does not support his grand claims of overthrowing existing positions. And he might be right on a few issues, or at least right enough to require some refinement of our understanding. (This is not surprising—history is hugely complex, so almost any straightforward explanation, as most CW is, is likely to be inaccurate in many details, even if it is right in the main.)

The Pity of War, like The Skeptical Environmentalist, has all these flaws, as well as the same occasional virtue. In my next post I want to focus on some interesting issues that I think he got right. But these hits were outweighed by the misses.

I don’t want to get into a laundry list of flaws, which would be unrewarding to write as well as read. But probably most disturbing is the shoddy reasoning evident throughout the work. In case after case, he presents evidence that is insufficient to back up his claims. He also seems to hold an odd, scientific view of history, where any explanation is 100% true, so that some isolated data points that don’t fit in are reason to toss the whole edifice out. But of course, history is complicated and contradictory, which is why it’s an art more than a science.

I think these flaws stem from the search for contradiction, the desire to be revisionist. He picks positions he wants to contradict and then filters the evidence to select pieces that support his position, rather than taking the total of the evidence and trying to fit it together in a more or less coherent way. This is one of the most fundamental flaws in human thinking—the tendency to filter evidence to pick out that which support our pre-existing positions, and Ferguson does this over and over again. Ferguson has apparently made a name for himself as a rising young historian in England. But based on this work and the fundamental problems in it, I doubt he produces much, if anything, that stands the test of time.

Link posted by Doug Turnbull at 2:40 PM

A Renaissance blog: Politics, sports, literature, history, and whatever else strikes my fancy.


Sneaking Suspicions
Crooked Timber
Unqualified Offerings
Asymmetrical Information
Cranky Professor
The Poorman
newsrack blog

Powered by Blogger