Terrorist leaders are not about Terror

by gwern1 min read3rd May 201129 comments


Personal Blog

From "Academics Doubt Impact of Osama bin Laden’s Death":

"...Fifty-three percent of the terrorist organizations that suffered such a violent leadership loss fell apart — which sounds impressive until you discover that 70 percent of groups who did not deal with an assassination no longer exist.

Further crunching of the numbers revealed that leadership decapitation becomes more counterproductive the older the group is. The difference in collapse rates (between groups that did and did not have a leader assassinated) is fairly small among organizations less than 20 years old but quite large for those more than 20 years in age, and even larger for those that have been around more than 30 years.

Assassination of a leader does seem to negatively impact smaller terrorist groups: The data shows organizations with fewer than 500 members are more likely to collapse if they suffer such a leadership loss. But organizations with more than 500 members are actually more likely to survive after an assassination, making this strategy “highly counterproductive for larger groups,” Jordan writes."

See also Lost Purposes, The Importance of Goodhart's Law, & Faster than Science.

29 comments, sorted by Highlighting new comments since Today at 8:45 PM
New Comment

I think the second quoted academic's conclusions are more relevant in this case. The first researcher assessed people in a leadership role, presumably the terrorist equivalent of a C-suite. The second looked at actual decapitations, where the head of an organization is killed, not just leadership generally, and found it to be an effective strategy.

This makes sense: the heads of organizations serve as focal points for internal negotiations and, simply by occupying the top slot, could help reduce competition amongst the leaders. Losing the head like that could lead to a succession crisis where infighting or uncoordination dominate as the C-suiters jockey for that position.

Big confounding variable is that not all assassination attempts are the same. More costly attempts being more successful will bias it one way. Better-organized terrorist groups might evade attempts more often, biasing it the other way. And of course terrorist groups might change names and structures but still have the same people and funds doing the same attacks. Oh, and being unsuccessful at assassination might be correlated with taking other measures to stop the group.

So although this is evidence one way (and the other study in the article is evidence the other way), it's quite weak, and doesn't overcome my prior based on what I'd do if I were a terrorist and my leader were assassinated.

This is a simple model and as such might make a decent prior if you know nothing else. In the Al Qaeda case we probably know substantially more than (size, age) for reasoning about it.

Do we know how to reason about that other information?

Certainly to an extent we would. If we looked in the data above and noticed a dramatic difference between African and Middle Eastern based terrorist data, we may want to add that variable to our model such that it considers (size, age, location). Data modelling techniques are generally useful. Random Decision Forests and that sort of thing. Humans are pretty good at generating hypothesis from sparse data because they have good 'common sense' understanding of the causality structure of the world.

I wouldn't claim that we'll have a particularly accurate result, but the above strikes me as the kind of conclusion that one might be certain about because of it's mathiness, and yet because reality is nonlinear, any extra considerations beyond two variables might swing the results around wildly.

Well, we have theories about how to work with it. But the study of terrorism has one of the highest words:applications ratios I've ever heard of, and unambiguous successes seem thin on the ground despite a large volume of theory. Of course, it's also possible that the limits of information availability are distorting my picture of the field.

The US Army's FM 3-24 on counterinsurgency operations might be the best summary of the mainstream perspective (whatever that means in this context) that I've read, adjusting for its authorship, age, and goals.

There was a Washington Post Article about the fact that bin Laden apparently had goals of damaging the U.S. economically instead of purely by death-destruction-terror. It goes on to discuss that in that respect, he might have been very successful.

[This comment is no longer endorsed by its author]Reply

There is some highly specious reasoning in that piece. First, the notion that the Afghan war bankrupted the USSR and led to their demise was a symptom of Bin Laden's delusions, not a fact about the world. The war wasn't unimportant, but the collapse of the USSR was much more complicated than that, and reflected the over-all trend of trying to outspend the US DOD with significantly less GDP.

The Federal Reserve, worried about a fear-induced recession, slashed interest rates after the attack on the World Trade Center, and then kept them low to combat skyrocketing oil prices, a byproduct of the war in Iraq. That decade of loose monetary policy may well have contributed to the credit bubble that crashed the economy in 2007 and 2008.

This is even more dubious. Certainly there is some relationship between reserve rates and mortgage markets but most of the excess money supply was "created" by the various financial arrangements that buried default risks, artificially lowering the price and returns on the secondary markets.

This piece reads more to me like a not-very-subtle dig at the author's domestic political grievances.

It's an interesting statement, but back in 2008 I read a compilation of Laden statements 1994-2004, and one of the interesting things (besides how bin Laden waffled on 9/11 and how he admitted it worked out much better than expected) was how his economic reasoning does not show up until the end - suggesting that he had no such thing in mind. And you see on occasion some pretty weird economic thinking, like this '97 bit:

Ladin: Correct! This is the right method to throw America out of the Gulf. The United States is beefing up its forces in the Gulf in a bid to control the oil resources. Since 1973, prices of all other items have increased but oil prices did not rise much. Since 1973, the price of petrol has increased only 8 dollars per barrel while the prices of other items have gone up three times. The oil prices should also have gone up three times but this did not happen. Price of American wheat has increased three-fold but the price of Arab oil has not risen three-fold. The increase was not more than a few dollars over a period of 24 years because the United States is dictating to the Arabs at gunpoint. We are suffering a loss of 115 dollars per barrel every day. Only Saudi Arabia produces 10 million barrels oil per day and thus the loss is one billion dollar per day. Total loss is more than 2 billion dollars. In the past 13 years, the United States has caused US a loss of more 1100 billion dollars. We must get this money back from the United States. The total population of Muslims all over the world is more than 1 billion. If every Muslim family is given 11,000 dollars then the 1100 billion dollars will be repaid. Muslims are starving to death and the United States is stealing their oil. It buys oil from US at a low prices and then makes US buy its tanks and fighter airplanes by projecting Israel as a threat. In this manner the United States takes all its money back.

This is hardly new, didn't he say something similar himself in his videos? The point was to trick the US into radicalizing the populace via increasingly barbaric acts of warfare and secret policing. He wouldn't be the first terrorist to make such a claim and, tempting as the "don't you see you are playing into their hands!" argument is, it always smells like post-hoc justification for a failed military campaign.

Edit: Sorry should have read the article didn't realise it mentioned the mans own arguments.

Also, (non) survival of the group should not be the only goal. Rather, it should be about the terrorist actions performed by the group or individuals potentially belonging to the group. Perhaps assassinating the leader will reduce the groups effectiveness in planning attacks, for example.

It's true that the discussed statistics do not rule out an interpretation like the new leaders add to longevity by steering the organization into tamer less effective strategies, as opposed to my own favored implied interpretation, that 'old leaders stay too long and drag down organizational performance & survival'.

But you could probably cross-check her list of organizations with assassinated heads, and compare against Max Abrahms' list of the 7% successful terrorist organizations and see whether survival correlates with success. (I'd bet it does.)

Parenthetically, I wonder if the SIAI cognoscenti would prefer that bin Laden was buried in liquid nitrogen rather than at sea.

I vote liquid nitrogen. I was pretty disturbed that people actually cheered his death.

[-][anonymous]10y 0

Cryogenically freezing someone who believes in the afterlife might - or so he might think - prevent him from truly dying and therefore from entering the afterlife. We might cryogenically freeze believers in the afterlife specifically to punish them, by keeping them chained to the mortal coil.

You'd think they'd either have their consciousness suddenly jump until they wake up, in which case it doesn't take any more subjective time to get to the afterlife, or he'd stay in the afterlife in the mean time.

In any case, my point was that, as long as he can be kept from harming others, I'd rather have him not die.

Considering he was killed by a gunshot to the head, I doubt it matters much either way.

[-][anonymous]10y 1

It's my understanding that cryogenic freezing does a tremendous amount of currently irreversible damage to the brain. If so, then the future technology that can reverse the damage, or else read the contents of the brain despite the damage, might be able to do the same for a gunshot to the head.

I'm not as enthusiastic about cryo as some here, and neither am I an expert in it. That said, my understanding of the procedure is that in successfully vitrified tissue it kills neurons via cryoprotectant toxicity but preserves neural structure and enough chemistry to make a pretty good guess about their internal state. You'll find neither in a brain scrambled by gunshot: at least in the damaged portions, and barring truly Clarkeian technology, that information is permanently lost to entropy.

[-][anonymous]10y 0

Depends on how scrambled. If you take a porcelain vase and drop in on a hard floor, it will shatter. But you can put the pieces together. The bits of brain may be indistinguishable to the human eye, but unless they are literally creamed, the individual cells separated from each other on an individual level (a possibility - I don't know the specifics), it might be possible to re-assemble the bits.

Cells are not so resilient that a bullet will simply separate them and leave them otherwise intact. It may or may not be possible in principle to tell "which cell goes where" if you were to separate them, but a bullet to the head doesn't even give you something that convenient to work with.

[-][anonymous]10y 2

You're talking about a "creamed" brain - one in which the individual cells are separated from each other on an individual level. I was explicitly specifying a brain that is not creamed, but rather, we might say, shredded. I did admit that I didn't know the specifics, so I'm not 100% sure that a bullet merely shreds the brain. I'm no ballistics expert.

Remember paper shredding? We used to do this to hide secrets. Many people still do it. But it doesn't work any more. The shreds can be reconstructed. Now with scanners and computers the process can be automated. Quoting Wpedia:

After the capture of the United States embassy in Tehran during the Iran hostage crisis, shredded documents were turned over for painstaking manual reconstruction, which revealed to Iran some U.S. operations including spies. Today, scanners and computers can reconstruct shredded documents very quickly.

So, shredding a document was previously considered to effectively lose the information to entropy. No longer are we under any such delusion - with today's technology, shredding is no longer sufficient. Now we need to burn our secrets - hence, the burn bag.

Now apply this point to the brain. If the brain was shredded, that doesn't mean that the chunks of it can't be reconstructed. Not today, obviously, but I wasn't talking about today. One of the methods of revival speculated about is scanning the brain and using the data to create an upload personality. But it's precisely scanning that makes it so easy today to reconstruct shredded documents. So even if scrambled brains seem like a mess to us - seem like a shredded document - that doesn't mean they will provide a serious challenge to the sort of computing power involved in uploading personalities.

Also, a person can often survive massive chunks of his brain being removed. A person can survive with half his brain removed. All I see in that famous photo of Bin Laden is a little hole over one of his eyes. Maybe the whole brain was turned into cream. Or maybe only one half of his brain was creamed. If the latter, then the possibility remained that he could have been revived at a future date, had he been frozen.

Parenthetically, I wonder if the SIAI cognoscenti would prefer that bin Laden was buried in liquid nitrogen rather than at sea.

I don't think I'm an SIAI cognoscentus (I don't know if that's the correct singular), but in general while I'm in favor of general preservation of individuals, I think a utilitarian calculus strongly suggests that preserving this individual would be very likely be a net negative.

This article talks about how likely groups are to fall apart after the death of a leader, but how much do we know about what killing a leader does to group effectiveness? Do groups that have lost the leader that made them into a powerful force run into problems when planning new operations, or do the younger replacements have new ideas?

Killing bin Laden was not about strategy. Killing bin Laden was about revenge.

Also about winning elections.

How confident are you about this? The morale boost seems to be a significant strategic benefit, from what little I see.

Why can't it be about both?

But revenge, in turn, is about counterfactually preventing the deed you want to get revenge for. Although revenge does not "undo" that (past) deed, agents that act as if it did are victims of less wrongdoing (under plausible models of human incentive structures).

Disclaimer: Not trying to voice an opinion about this particular person, his killing, or the impact thereof, just making a general point about the decision-theoretics of revenge.