The Monkey Cage

This is exactly why a friend of mine uses Pape's work in a first-year research design seminar: to illustrate how not to do social science.

I'm not a research design specialist and I haven't read the paper yet that you link to, but off the cuff I have to say I'm a little wary of this sort of critique. The argument, as presented here, of this contrarian view seems to be based on a simple logical fallacy, specifically if X then Y therefore if Y then X.

That is, Pape says that "If there was a suicidist attack then the suicidist thought his target was an occupier." The paper here tries to reverse this argument to "If a possible target is thought to be an occupier by a possible suicidist then there will be a suicide attack...and if not then Pape's theory is disproved."

Of course conditional arguments aren't reversible in this way even in purely logical forms. But more importantly it denies the variability of human choice beyond simple probability.

But like I said, I haven't read this paper yet so maybe I assume to much from this very brief summary of it.

Thanks for this. We require a 200-level course on writing and research for our undergraduate political science major, and we spent a good deal of time talking about case selection and bias on the dependent variable. I just printed out the first page of the paper and will take it into class today. What's embarrassing is that I have read this piece and still missed this rather obvious flaw (but then, so did APSR).

Most sciences don't use correlation as a substitute for explanation - rather it is used as one tool among many to contribute to explanations. Much social science also refuses to conflate these things as well.

Pape isn't interested in the question of what causes suicide terrorism in this narrow sense. And why should he be? We can combat the rise in suicide terrorism without knowing these propensities.

If this sort of logic is correct, than evolutionary biology isn't science.

Lots of oxygen and tinder make a fire likely, but knowing the exact propensity of fire in that situation is irrelevant, as is knowing if lots of places have high levels of oxygen and tinder and don't light up. The lesson is the same - don't want fire - don't keep tinder in your basement. And don't try putting out a fire with more tinder or oxygen!

Another, simpler, criticism I've heard bandied about is that the majority of Pape's data come from a single conflict (Israel-Palestine). As such, the external validity of the conclusion may also be somewhat in doubt.

While I'm unqualified to comment on the quality of Pape's research, his conclusions do make some sense. The media, US govt, etc. have always downplayed the political element of terrorism while making a big deal of the religious nature of them. Even though the 9/11 hijackers did not target any buildings of a religious nature...

I'm not sure how you can get the majority of the data coming from the
Israel-Palestine conflict. Depending on how you count it, it's very roughly, a third of the number of campaigns or attacks, and about six percent of the casualties. (See Table 1 of the article.)

As Pape points out, the largest number of attacks is associated with one group, LTTE, which is secular, and not Middle Eastern. This group is responsible for a larger number of casualties than those involved in the Israel-Palestine conflict. The largest number of casualties is associated with Al Qaeda (although this is a product of one, very tragic outlier.)

What's more, the patterns he describes hold even if you exclude Israel-Palestine, or anything involving the Middle East.

Maybe I'm missing something - does the count come out differently if you use the Appendix?

Its more likely that my recollection of the criticism is faulty, David.

For those too busy to read the paper this quote might help clear things up:

Pape deserves praise for the substantial work it took to collect these data, data that
make a real contribution towards undermining stereotypes about suicide terrorists. But we think it is important to be clear about what they actually tell us: if a terrorist group chooses suicide tactics, then it is highly probable that a democratic country is occupying
its homeland. This fact is certainly consistent with the existence of a close association between foreign military occupations and the growth of suicide terrorist movements." In the remainder of this paper, however, we show that it is also consistent with either a weak or non-association, as well as with a strong negative association.

Patrick,

Lots of CO2 also makes the presence of a fire likely, and the absence of CO2 makes the presence of a fire extremely unlikely, but dumping CO2 on a fire IS a good thing to do if you want to put out the fire (see CO2 fire extinguishers). The difference between CO2 and the tinder and oxygen is the difference between mere correlation and causation, which is why you DO need to work out correlation and causation, whther you are combating fire or terrorism.