[ExI] Terrorist? Who can tell?

Lee Corbin lcorbin at rawbw.com
Sat Sep 6 15:19:59 UTC 2008

Harvey writes

> On Saturday 30 August 2008 17:19:23 Lee Corbin wrote:
>> On the literal reading of my question above, which was my intention
>> and which has no necessary direct bearing on profiling, does a six
>> member *group* of known members of Al-Qaeda differ significantly
>> in appearance from a random sample of six age-appropriate men
>> chosen at random from London? Sorry, but I still don't understand
>> how you and Damien can persist in saying that the answer is "no".
>> For one thing, aren't the odds of anyone who is Asian or black
>> being in the UK component of Al Qaeda quite small?
> Not at all.  There may be fewer asian Al Qaeda in Bagdad, but they are the 
> majority of Al Qaeda in Indonesia.  (Remember the Bali bombing?)  Likewise, 
> blacks would be the majority of muslims in Africa.  (A judge  ruled last year 
> that the Sudanese government was liable for the U.S.S. Cole suicide bombing.)  

Once again we see what is from my perspective an almost
genetic predisposition to ever concede anything---e.g., why
not say, "Well, yes, in your stupid hypothetical totally useless
thought experiement, OF COURSE in London six Al Qaeda
members would not look like six people picked at random:
too few whites, too few blacks, etc. And I know that you know
that this would, of course, not be 100% reliable, but if you deal
with groups of people picked at random, then NATURALLY
what you are saying is correct, so?"

But wringing such confessions (it's too hard for them to be merely
admissions) is like pulling teeth.

Okay, so why my question?  At the time I was asking,  I was
too ignorant of the demographic makeup of London (which
has been corrected mostly by BillK). So one naturally sets
bounds upon what the interlocutor thinks.  <sigh> why should
I bother? Even thinking of extreme situations is not likely to
result in any kind of admission or boundary on the possibilities.

Or it could be partly what I call the Andy J. phenomenon. My good
friend Andy in high school would always answer like this if
I began to ask a Socratic like string of questions:  "Just where
is this going, Lee?  I refuse to answer any longer until I know
what your eventual conclusion is supposed to be."   Nothing
but sheer intellectual cowardice on his part. Thank goodness
I've never seen (here or anywhere) anything quite the magnitude
of that!

> I tried to perform your experiment with Google, although I admit that this is 
> a very small unscientific sample.  But I Googled "terrorist conviction london 
> 2008" to see what the first six convicted terrorists in london looked like.  
> What I found were:
>   - A 16-year-old boy 
> <http://findarticles.com/p/articles/mi_qn4158/is_20080819/ai_n28020975>.
>   - A woman <http://www.guardian.co.uk/uk/2008/jun/17/uksecurity.ukcrime>.
>   - A Ugandan african <http://www.euro-
> islam.info/spip/article.php3?id_article=2180>.
>   - Five muslim men (declared innocent, their convictions overturned) 
> <http://mek1980.wordpress.com/2008/02/13/students-win-appeal-against-
> terrorism-conviction/>.
>   - A 50-year-old muslim man <http://news.bbc.co.uk/1/hi/uk/7256859.stm>.
>   - Four younger muslim men 
> <http://www.timesonline.co.uk/tol/news/uk/crime/article3437865.ece>.
> So in trying your experiment, I did not get any sterotypical muslim men in the 
> first six.  The only muslim men in the first eight were later declared innocent 
> and had their convictions overturned.  The ninth person I found was a 50-year-
> old muslim man.  I had to go to the tenth through thirteens persons to get 
> what I think you are describing as the stereotypical younger muslim man.
> So by way of a single example, I would submit that your experiment wouldn't 
> always work as you expected.

Blast it!   #!%^!@@$!!.  Double blast it.  I *never* said that it
would *always* work.   I carefully placed probability bounds on
it. Okay, okay, I get it.  You're simply not going to admit any more
than you already I (I'm speaking of a collective you, here). My
experiement would indeed succeed (oh, say, 80% of the time
with six people, 98% of the time with one hundred people, etc.)
but DO NOT WORRY.  I am going nowhere with this line of
questioning.  I wanted to know the limits.  Okay, now I do.
Thanks for the concrete info, though.

>>   > > In short, you are claiming that the relative homogeneity of my
>>   > > group A (or at least I thought so), would be insufficient to allow
>>   > > for a 19 in 20 test.
> I don't understand what good it does to be able to distinguish a homogenous 
> group from a random group.  We could do this test with any ethnic, racial, 
> religious, socio-economic, gender, sexual-orientation, or other grouping with 
> similar results as you predict.  All it proves is that people can distinguish 
> between diverse groups and homogenous groups.  I'm not sure what the point of 
> that is, or how it helps anything.

Yup.  Explained above.  You're right.  It doesn't help in profiling
(in this case).  Profiling does work sometimes, as perhaps (or
perhaps not) you will admit. If the Siamese authorities are looking
for a white man in small almost entirely Thai village miles and 
miles outside Bangkok, then profiling works. 

>> Can you guess what fraction this might be? If it's at all high,
>> then this is the source of my confusion.
> I think this is indeed your source of confusion.  I don't have statistics on 
> al qaeda converts, but here are some statistics on muslims in the united 
> states from <http://pewforum.org/news/display.php?NewsID=15040>:
> "Muslims are the most racially diverse group in America. Approximately one in 
> three Muslims are white, roughly one in four are black, and one in five are 
> Asian."  

Er, we were talking about *terrorists* or *terrorist bombers* or
Al Qaeda. I never spoke of the racial composition of entire
religions, and I was specifically not speaking about anywhere
except London (except in my wild counterexample about Thailand).

The rest of your remarks here are quite well known to me,
whether or not I travel any  :-)


(P.S.   My spell-checker had to change about 15 uncapitalized
uses of muslim to Muslim.)

> I think you are forgetting how old the muslim religion is, and how far it has 
> spread into asia, africa, and elsewhere.  Most muslims in the U.S. are NOT 
> middle-eastern.  Racial profiling would not help, unless you believe that there 
> is a racial component to violence and that the white, black, and asian muslims 
> are less likely to be violent than arab muslims.
> The other thing to remember is that most muslims are non-violent, with only a 
> small percentage of them being so radical.  We also have violence from other 
> religions, as history shows with the Irish Liberation Army in London, 
> Christian abortion clinic bombers in the U.S., militia types in the U.S. 
> (Unabomber, Timothy McVeigh, etc.).  So one religion does not have more 
> violence than another.  If there were a good statistical lead of one race or 
> religion over another in terms of violence, I assure you that security experts 
> would be all over it.  But the numbers just don't work out.  There simply is 
> not statistical advantage for profiling one such group over another.  Like 
> torture, it sounds like it should work in theory, but it really doesn't pan 
> out in practice.

More information about the extropy-chat mailing list