[IvanDenisovich]

Pardon my bayseanity, but the article discusses P(Engineer|Terrorist), not P(Terrorist|Engineer). I'm saying that a similar claim holds for P(Muslim|Terrorist), while your point is regarding P(Terrorist|Muslim).

But this is way off topic and my argument has nothing to do with Islam. I'm just pointing out how easy it is for us to play couch sociologists when weighing the merits and problems of "Engineering" culture, vs how difficult it is to hold a similar discussion about the issues of "standard" religions or cultures.

[BWStearns]

Neither direction is significant. Given reasonable definitions of terrorist (and the ones I think are best) terrorists are not significantly more likely to be muslim. In order to make that the case you need to extend "terrorist" status to a lot of groups that are really governments or insurgencies and not give "terrorist" status to a lot of Christian and political anti-government groups.

P(Muslim | Terrorist the US currently gives a shit about) is interesting, especially given the propensity to regard certain terrorist groups as criminal elements for political reasons.

Also interesting, other countries sometimes negotiate disliked groups onto our lists to legitimize their crackdowns (ironically the crackdowns can push the group to actually become terrorists. Hi Turkey!) and make their international fundraising efforts effectively illegal.

It is important to refute you because the implicit points when you talk about "how difficult it is to hold a similar discussion about the issues of "standard" religions or cultures". It isn't difficult for scholarship to have that discussion. It has been had. It disagrees with what you suggest to be true.