Schneier on Security
A blog covering security and security technology.
« FTC Privacy Report |
| Sane Comments on Terrorism »
December 7, 2010
Profiling Lone Terrorists
Masters Thesis from the Naval Postgraduate School: "Patterns of Radicalization: Identifying the Markers and Warning Signs of Domestic Lone Wolf Terrorists in Our Midst."
This thesis will scrutinize the histories of our nation's three most prolific domestic lone wolf terrorists: Tim McVeigh, Ted Kaczynski, and Eric Rudolph. It will establish a chronological pattern to their radicalization and reveal that their communal ideological beliefs, psychology, attributes, traits, and training take place along a common chronological timeline. Their pattern of radicalization can be used as an indicator of lone wolf terrorist radicalization development in future cases. This thesis establishes a strikingly similar chronological pattern of radicalization that was present in each terrorist's biography. This pattern can identify future lone wolf terrorist radicalization activity upstream. It can provide a valuable portent to apply in the analysis of potential lone terrorists, potentially enabling law enforcement to prevent tragedies emerging from the identified population through psychological assistance, evaluation, training, or, in the worst case, detention.
Posted on December 7, 2010 at 6:43 AM
• 72 Comments
To receive these entries once a month by e-mail, sign up for the Crypto-Gram Newsletter.
I have neither the time nor the background to actually parse that thesis, which I regret, but the abstract concerns me: Mr. Springer doesn't seem to be considering the possiblity of false positives. I.e. he is making the case (which seems quite plausible to me) that lone wolf terrorists have a set of common pre-existing characteristics, but does he establish that all those with the pre-existing characteristics become lone wolf terrorists?
A psychological review of events and characteristics of domestic terrorists is a nice idea. I take issue with labeling McVeigh an "lone wolf", when he clearly acted with assistance from others.
Given the events in Portland, one might conclude that the FBI has already identified likely radicalized lone wolf offenders ripe for exploitation.
Building a pattern from three data points looks like quite a leap to me from a statistical standpoint.
Drawing far-fetched conclusions from three cases of a very rare event?
I call BS.
I must agree with Lalufu and Peter A. - three cases is not conclusive data.
And thus Minority Report becomes reality.
This dramatic sentence from the conclusion bothers me:
"The next lone wolf domestic terrorist lurks in our midst, and could be following the same chronological pattern [...]"
I did not see evidence in the paper to support that such a "next lone wolf" exists, with certainty or any other probability.
I will therefore assert with absolute certainty - and no evidence - that the conclusion about the "next lone wolf" will be used as partial justification for future policies. *sigh*
I'm having trouble getting that paper but I'd have to agree with the other posters here. Taking 3 cases of an extremely rare occurrence and extrapolating anything is probably going to result in a LOT of false positives.
@cliffsull, Brandioch, Peter A, et al.
Put it in perspective, people. Yes, drawing profile conclusions from 3 loner-type domestic terrorists is difficult, but you have to work with the data you have. This is basic research churned out for a Masters Thesis, not applied doctrine.
I, for one, am glad someone is taking an interest in this kind of data instead of simply not trying. Profiling, in this sense, has been a part of good police work for a long time.
It would be nice to have commentary from someone who has th "time and background" to "parse" this report.
Beyond "load of SHIT!" that is.
The points about insufficient data are well founded, but as with many types of studies, correlation is often a first step to indicating directions for elucidating a causal mechanism.
What is notably missing in this work is that focusing on these news-heavy events omit so many others from bioterrorism-to-influence elections (The Dalles, OR) to numerous "going postal" incidents, all of which are easily accessed, relevant to the discussion and contribute to understanding casual mechanisms.
They should immediately put all wolves on the no-fly list, lone or not. This is a very worrying development and must not be allowed to be buried by the war on wikileaks.
Thought strikes me...
"Historical Profiling". Imagine for a moment someone who fits the "lone wolf" profile being discussed in this paper. As others have said, I would fear that false-positive aspects of this, because I suspect that there are many loners who are not wolves, maybe dogs at worst. So consider that our "lone wolf" realizes his radicalization, decides to do something about it, and realizes that he fits the profile. If he's smart, his next act would be to de-profile himself, join social groups, quit acting like a lone wolf, etc, while he plans his lone-wolf deeds.
So perhaps we need to profile for "profile changes" as well as profiles. But then we start to catch the loner who decides to "turn his life around" as well as the devious lone wolf above. Or what if the devious lone wolf decides that he likes his new life and abandons terror plans? (I would count that as a success.)
I always thought the wolf was a pack animal. I'm off to watch NatGeo.
Oh and any Department of Pre-Crime is insanity in my book (but it's a coloring book, so there you have it).
I don't see the "non-statistical problem" -- there's no claim to statistical validity, this is a case study which can be suggestive. There's some fields that you can do no better.
The problem is that the analysis is atrocious. What a shocking surprise that a "lone wolf terrorist" would also be a non-conformist nerd who couldn't get laid.
What a shocker! What deep insight!
And what does this give anyone? Given that non-conformist nerds that can't get laid are the basis of numerous industries today, and most do nothing worse than being Libertarians or Social Democrats or play with trains in their basements -- what the hell is the point? It's a terribly common pattern in the US, and without these folks our most productive industries would collapse.
Trivial and pointless (and not very well written, as well). The lack of statistics is the least of it's problems.
If it was true that one's own "profile" could be changed at will, so to speak, then the whole idea of criminal profiling would be worthless. Which it isn't.
Even if one was to profile profile changers, then you'd have to profile for profile changer profile changers. And so on.
Well, based on this analysis, there's obviously no longer a need to search women or non-white men at airports, while white male math geeks need extra special attention - especially if they sport beards (sorry Bruce). Victory is at hand!
@noble_serf: Of course the wolf is a pack animal. That is why the "lone wolf" -- an individual that has been driven from the pack and has to hunt alone -- is considered dangerous.
Now, if the Federal Bureau of Sheep Protection started to identify lone wolves, create a bogus pack for them to join, and then trick them into attacking a Peterbilt truck doing 65 ...
Wolves are pack animals. The lone wolf is the exception to the rule, and is often a wolf that leaves the pack due to lower status within a less successful pack. Some do better hunting alone, many more do not - some die, others return to their packs and a few may join other packs.
We use the term 'lone wolf' to refer to people who exhibit personality characteristics that suggest that they are more comfortable acting alone, even if they are not explicitly leaving their social groups. We also use the term in a military sense to describe military operatives that act alone without a central command. There are risks associated with conflating either of these common uses to align with deranged psychopaths who seek to blow people up. These individuals may exhibit traits associated with both uses of the term, but these uses are not the same, and neither is generally related to actual sociopathy.
That said, I'm neither a psychologist nor a linguist. Have at it.
Naval Post graduate studies, really says it all, Naval means mass murder and serial murder at sea and in ports. However our navy has had to create more ground troops to work in landlocked afganistan, apparently the marines were too independent so now we have other types of 'navy land infantry/armor forces. that started with vietnam when zumwalt wanted to get someones skin into the fight, (sent his son among others) Industrial scale mass murder is political, its not science based, and identifying the victims is a new twist on it. In el salvador, it was enough to call someone a communist in public to justify their assasination. The GWOT is just the new justification.
So dont look for science in a paper that only means to demagogue a set of enemies that the author proposes creating. He however should know that our side uses murder, mass or serial, just as indiscriminatly in spite of the propaganda, and the killing is not personal, its political, which in our state of civilization is considered normal.
I agree with @SnallaBolaget that the paper needs examining in it's own right and not be judged based on it's abstract.
If we make a conclusion based on this abstract it's likely to be based on our own predjudices and preconceptions.
Can extremely rare complex events even be modeled? Persinger thinks so in his work.
Other complexities needing consideration,
intellect & education
emotional (did the mother smoke and drink during pregnancy)
re: Wolf as "pack" animal. Wolves change thier behaviors to meet the avaliable food in their territory. Wolves are social but form packs (greater in size than the family band) when cooperative hunting is needed during scarity. A wolf, such as denning mother, or a juvenile seeking their own pack, can be capable solitary hunters and not necessarily a social misfit or reject.
To judge human behavior by comparison to a natural analogy from a term first documented in the early 20th c. is specious. If we were to apply this analogy than ANY unattached human becomes a 'lone wolf'. As humans are un/attached at different points in their lifecycles ... It don't work.
McVey worked with at least 2 other people to build his bombs. How does that make him a lone terrorist?
As an introvert, I feel rather oppressed by this sort of thing. When will "they" be coming after me just because I like to keep to myself.
"Profiling, in this sense, has been a part of good police work for a long time."
No. You have that backwards.
The police start with a crime and then profile the person who would have committed that crime based upon the evidence they find. Part of that evidence is the location where the crime was committed.
What this paper is doing is taking crimes ... and then generalizing the characteristics of the people who committed those crimes ... and then saying that people who fit those characteristics are the kind of people who commit those crimes.
No. Because terrorists are so rare that there would be a million people who fit that profile who would NOT be terrorists.
This reminds me of the scene in WarGames where the feds describe David Lightman: "He does fit the profile perfectly. He's intelligent, but an under-achiever; alienated from his parents; has few friends. Classic case for recruitment by the Soviets."
Several characters in that scene realized that description fit the vast majority of smart teenagers.
Bruce, you consider this worthy as a post to your blog? Really consider this clutching at straws.
Part of me says that it's a good thing they're trying to look at the problem a bit more intelligently rather than basket sweeping (or groping) everyone in sight; but like most other posters here I also realize just how much potential there is for them to get this wrong - or depend on it too much.
The paper is very poorly written. It seems more like a book report than a thesis. We really need a better education system in this county.
Worst thing I have read in a while. Constantly pointing out there loneliness while simultaneously citing interaction with army friends and buddies. They could not meet girls, except for the one who did. Struggled with education except the one who graduated from harvard. Pointing out the military connection except the one who did not join. So what you are left with is:
3. used bombs
Then at the end the author tries to attach significance to the "chronology" but then does not really identify what was similar. I mean, everyone has their unhappy childhood before their unhappy career, right?
I hope Springer did not graduate, the writing is what I would expect from a high school paper.
the most proflific organisation for mass murder trained McVey to think of killing as a business, The US military have never been equaled in raining death on grass roofed villages and raining medals on the chests of those who ordered and organized it.
Three case studies to build a thesis hardly makes for hard science or even professional academic research. Either the author got his inspiration from Minority Report, or it is nothing more than a tool for special interest entities to justify funds for yet another "security" agency tasked with scrutinizing and profiling oddballs. I think we've seen this before in countries such as the former Soviet Union, GDR, Romania, China and the like.
Step 1. Identify all high risk potential lone wolves.
Method: Administer MMP every 5 years to everyone in US.
The government may class Bruce as a "lone wolf".
He has an obsession with terrorism and airport security, its all he blogs about.
I've been reading his entries for a while and thought "What really goes on inside the mind of Bruce Schneier."
Step 1. Identify all high risk potential lone wolves.
Method: Administer MMPI every 5 years to everyone in US.
@Andrew "may class Bruce as a "lone wolf". "
Lone badger, wouldn't you say Clive?
I agree with the commenters who conclude that the author is dealing with far too small a sample size, but I intend to make time to read the paper in its entireity, since the human behaviour unit of the British Secret Intelligence Service (aka MI6) concluded a couple of years ago (search the Guardian archives) that home-grown terrorists *can't* be profiled for.
Of course, there's also the matter that "our home-grown terrorists" and "your home-grown terrorists" have very different ideological motivations...
@ Everyone saying 3 data points is insufficient...
This research is targeted at govt officials. How many data points is sufficient for this type of person?
9-11 was enough to create TSA, DHS, and all that came with them. The Chrismas underoo bomber was enough for nudoe scanners and lawful sexual assault.
If single data points are sufficient to trigger billions in spending, just think what 3 data can justify...and a study to go with them!
"He has an obsession with terrorism and airport security, its all he blogs about."
and squids; this makes him extra freaky.
Despite the pain of this thesis being poorly written, it gets funnier almost by the page. First of all, he's used his thesis advisor as a main source, which is not, I suppose, technically wrong, but it would perhaps have been better to use independent sources?
Second, the man clearly states that there's been almost nothing like this done before, and goes on with this (a couple of pages later): "The authors use the FBI manhunts of lone wolf terrorists Theodore Kaczynski and Eric Rudolph and the case files of eight other lone terrorists to create a profile of the lone wolf terrorist in the United States"
So... what's the use in this thesis again? (Why was it approved in the first place...)
...so please everyone; study the study.
"[...]and specifically the study of lone wolf terrorism, require further study."
I suspect the vision of the dept of humble secrecy is to find 300 million lone wolf terrorists in the U.S.
I wouldn't get too wrapped around the axles about this - the purpose of the paper is not to sway anyone's thinking. It's purpose is merely to get the author a pass.
You should worry more about the organized terrorists a.k.a the US Gov't and their foreign friends. After all they pulled off terrorism in 2001. 9/11 and Israel, here:
@Jon - "It's purpose is merely to get the author a pass."
The problem with such papers is that when someone with an agenda shows up the paper becomes a tool to justify their pet policies. All too frequently papers get cited but no one reads them critically, instead just taking them at face value.
@casey - "Worst thing I have read in a while."
The use of "exceptions that prove the rule" is the essence of the voodoo that is "criminal profiling." This essay has been referenced on this blog before:
@ BF Skinner,
"Lone badger, wouldn't you say Clive?"
Bruce certainly does 'have a bit of Badger about him' 8) but 'lone' there I'm not sure as I have not observed his set (you have to be carefull in the UK about admitting to 'bader watching' as it has a double meaning since a politico used it as an excuse when questioned by the police).
However Bruce needs to be carefull if venturing into the England country side as the Government has by a proposed change in legislation put badgers on the "most wanted" list with farmers issued with licences to 'shoot on sight' :(
(For those the do not know what "has a bit of badger about him" means compare Bruce's full face photo at the top of this page to the full face photo ot the badger in the above linked article).
Don't knock the report; maybe we can spin it so that the government ensures that everyone gets laid.
Searching Bruce at the airport would be pointless. No mere weapon he could carry onboard could compare to his own fists or beard.
"... the human behaviour unit of the British Secret Intelligence Service (aka MI6) concluded a couple of years ago (search the Guardian archives) that home-grown terrorists *can't* be profiled for."
That sort of sweeping negative is always dubious (since it presupposes that every possible combination of observables have been examined), but I'm not sure this is the paper to offer anything original -- the author's ideas about where to look for signs of a 'chronological pattern of radicalization' aren't very surprising, so they're probably already among the well-examined.
@ BF Skinner - no he was an american. Terrorists are foreigners, or at least a different colour
With all the data collection going on, I think the sole purpose is to identify about 300 million lone wolf terrorists in the U.S.
"As an introvert, I feel rather oppressed by this sort of thing."
In many policy matters it's reasonable to ignore minorities like introverts, but for the purposes of profiling this is actually an issue. I've never encountered any research on the subject, but I have to wonder if the 'dysfunctional loner' stereotype isn't really 'dysfunctional isolated extravert', and if there isn't a corresponding 'dysfunctional oversocialized introvert' pattern that hasn't been identified, or at least hasn't been publicized (associated with high density living situations, perhaps, or a forced loss of privacy).
Theodore Kaczynski killed three people, Eric Rudolph killed two.
Army major Nidal Malik Hasan killed twelve at Fort Hood last year.
The only conclusion one can draw from this thesis is that government desperately needs to start a program of providing willing women to fix sex lives of the survivalists.
"Nidal Malik Hasan killed twelve at Fort Hood last year."
less "romantic" (no gothic novel appeal for the public, and no formally released "manifesto"), less elaborately planned, yet more "effective".
iirc, Kaczynski injured some. but i don't know how that "score" compares to other "lone wolf" killers.
i thought I'd read that the 'introvert' loner cliche was debunked. perhaps somebody realized they hadn't accounted for the loners they'd never discovered?
if nobody, perhaps exhumation and waterboarding would be prudent.
"less "romantic" (no gothic novel appeal "
"inspiration from Minority Report"
Even using the term Lone Wolf should be suspect. It's being used as short hand to express a concept that may have no basis in behavioral science. A romantic notion as @- says.
The lone gunman.
Are they trying to build a classification system based on how the actor acts? As we saw with malware and intrusion that's not easy.
For years my basic profile for a hacker was... "a person, btw the ages of 16 and 69, male or female, usually anglo, but frequently black, hispanic or asian, may be working inside or outside the company. Comptuer skills will be low to high."
If all we want is to smite sinners then focusing on the act is probably the way to go.
But if we want to stop the behavior, stop the person from choosing to terrorist acts-going to motivation and short circuiting it there is probably the way to go.
Lone nutcases (to use a psychiatric term) will cleave to causes just as a socially concerned well adjusted person does. Were the Americans who supported the IRA with money and action just or terrorist supporters. How 'bout the ones who supported and supplied the ANC?
Maybe the key is to eliminate all the causes. It reaches something deep in people that they want to kill and die for.
Knowing what's motivating OBL is easy. He's told us any number of times.
Are we ready to go there? Kick over the House of Saud and let the Saudi people breath free, free from tyranny, to determine their own destiny. (that's the usual formula right?)
Are we ready to allow a 3rd world country dictate to global corporations where they can or can't exploit resources? (well, mind you, Venezuala)
Based on the US's history for the last hundred and fifty years? Likely not. The fight against 'terrorists' is the surgical tent of a MASH. It's there to reduce the impact of the war outside; not stop it.
Perhaps 'load ofShit!' was a bit strong.
But lets try and put 'profiling' and 'preempting' terrorists into context for non-American citizens.
For us - we look at all the inncoents who have been detained and Tortured(as shown by wikileaks and admitted to by W Bush - 'waterboarding was legal') and we see that they were 'profiled' in the same manner as suggested by this Thesis.
It is a dangerous idea and will never work properly and if it leads toeven 1 miscarriage of justice or violation of a wrongly profiled person - it is self-defeating.
terrorists are radicalized by some oppression like palestinians, who are characterized as radical or terrorist, but given the fact that europe committed a crime that they are paying for by being victimized by the former victims from europe. or afgans who were raised in a war between the US and USSR or if you like between a comune based economy or a capitalist economy for which millions of southeast asians also supplied the body count. If you dont' represent an oppressed people, you arent a terrorist, your just crazy.
but in the end, it takes a government to really do mass murder on a cambodian scale, 5years of bombing weather and 3 years of khmer rouge. and governments are never held accountable, not the wolfowitzes of this world who wormtongue the war to some callow bush
If we use the no-fly list as an example, you only had to meet one criteria, name, and could no longer fly (probably one million on the list so far). All they are looking for in the lone-wolf category, is that you meet a single criteria, whatever it is (the criteria list is probably very large or vague so anybody can qualify), and you will be classified. Thus, sooner or later, we will all qualify. The purpose is to intimidate and control the population. Expect the no-fly list to be augmented by a no-drive list, soon to be followed by "your papers please" checkpoints, or the equivalent of iPass or zPass chip scanning stations everywhere.
I've read the paper (twice) and still haven't been able to find anything substantial in it to support the thesis.
A collection of generalizations and vaguely pertinent, sometimes patently irrelevant, references which are supposed to be the basis for the conclusion or recommendation.
Master's? B-, maybe, from Degrees-R-Us.com. I pretty must think Jon is right - fill in the good-thesis template, get the credit, get the Master's of the month, and move on.
I'll agree that you can't change everything about your profile, but you can change some aspects of it. In this case, if you look like a loner and want to change that you can start cultivating social outlets. For one, you could start going to a "safe" (as profiled) church well-known for lukewarm mainstream views, and socialize with those people.
I tried to touch briefly on the silliness/problems of profiling for profile changers, etc. I'm struck by something my psychology professor once mentioned. When doing aptitude tests, they understand that some people try to skew the tests to get a desired outcome. They have other cross-checking questions in there to find those people. They found that they make good salesmen.
My god ¿is that a master thesis? Suddenly, if the jury was Larry, Curly and Moe
Why is "Minority Report" brought up whenever someone talks about behavior profiling? It'd be a better reference if the thesis was called "How Three Babbling Psychics Strapped to Chairs Can Generate Punch Cards for Later Investigation May Prevent Crime"
This week, a man was arrested for allegedly trying to bomb a polling place. He definitely doesn't fit the profile. He has tons of friends defending him, many of them women. http://tinyurl.com/2cwa5c2 I wonder what we should make of this.
At least he used more than one data point...
Their background all seems perfectly normal to me (life is chock-full of shit), the only connection being the outlandish way they each reacted to their respective situations.
I feel bad for their parents; they don't deserve to be blamed.
Like most others, I can't believe the thesis adviser approved it.
If we seriously invested in children as a society, these guys might not have been twisted and murdered people.
If we took violence against women, violence against children even by their caregivers, and rhetoric against women's health services providers seriously fewer people would die. We could recognize that men who hate women actually hate people.
Hmmm. Investing in education and women's health to reduce violence? That's crazy talk! We just need to detain weird kids.
One point of Springer's thesis jumped out at me... the inability of the various bombers to form strong social bonds with women.
If you will forgive some gross generalizations, I suggest that 'sexual partnerships' between young men and women represent a 'civilizing' force on young men. Through such partnerships, the looming biological imperative is assuaged and, if children are the result, the father acquires an important societal investment and a reason for *pragmatic* change. In short,women and children 'ground' men. (Did I mention I was engaging in gross generalizations?)
Why this matters...
I predict that the inability of young men to partner with young women is a statistically significant cause of radicalization. Thus, countries and communities which limit the ability of young men to partner with young women will produce an elevated number of radicalized young men. That is, radicalization will emerge from areas in which the supply of 'marriage material' is restricted by, for example, polygamy and selective abortions.
Relying upon The Economist's 2007 Pocket World in Figures, I see that the UAE, Qatar, Kuwait, Oman and Saudi Arabia all have male:female population ratios of 117:100 or greater. Even without polygamy, it is clear men in these countries will find it difficult to marry and produce children Other notable countries include Jordan (108:100); Afgahanistan and Libya (107:100) as well as China and Pakistan (106:100). As a comparison, note Turkey (102:100); Australia and Denmark (98:100); USA (97:100) and Japan and Mexico (96:100).
If my prediction proves correct, the 'war on terror' requires that we work to harmonize birth-rates across the world and to end polygamous hoarding of woman by the haves over the have nots.
*briefly, I have wholly ignored homosexuality while simultaneously treating women as mere chattel. In a full length treatment, I would fully defend homosexual and women's rights. I have not so here in an effort to limit the scope of a long argument.
"If you will forgive some gross generalizations, I suggest that 'sexual partnerships' between young men and women represent a 'civilizing' force on young men. Through such partnerships, the looming biological imperative is assuaged and, if children are the result, the father acquires an important societal investment and a reason for *pragmatic* change. In short,women and children 'ground' men. (Did I mention I was engaging in gross generalizations?)"
It might be a "gross generalization" but it's actually wrong in quite a few cases....
Look at the very recent UK bod that killed himself in Stockholm, married and three children. Many of the other terrorists of recent times from the UK have been married with children.
In their part of the world Hammas actually pays the immediate family of "martyrs" a large chunk of money and provides free education and health care to the partner and children.
As in all things "human" the issue is both complex and varied.
For instance the issue of the "moral compas" in young men when taken out of a strongly patriarchal culture and dropped into one that is not.
They are not used to making moral choices as they have been made for them by "the head of the family" thus when "out from under" the thumb and away in a foreign country at University they respond easily to an authority figure who in effect replaces the "head of the family". It is why until fairly recently Universities where hunting grounds for recruiters. However in the UK a number of Unis and LEO's have taken to keeping a close eye on Uni "societies" and have also brought in local known leaders from adjacent communities.
Now I have no idea if it has worked as a policy but... it is noticable that the UK home grown terrorists are comming not from Uni's but communities. And yes unlike America we do appear to have one or two real live (untill just recently) terrorists.
At the end of the day the way to stop radicals is not by locking them up (that just makes them martyrs to the cause) but by helping them grow up to apreciate a wider view point. And to do that we need a lot more honesty in the way we go about things politicaly and corporately. Exploiting people and resources may make a fast buck but there is always going to be blow back. When you stop taking the short term quick buck viewpoint you start to realise that partnerships are more profitable than exploitation (Commonwealths -v- Empires). And when people have a real stake in their society with a clearly visable way to move forward then they do genuinely have something to lose and this does indeed ground them.
But none of this is new Europe had to go through it's middle ages and political problems wraped up as religious wars. In recent times Northern Ireland has turned away from terrorism. The journy is far from over and the steps may falter from time to time but the general direction is now one that "floats all boats". And the economic prosperity is spreading out into the communities and people have the real hope of a self determined future. The trick will be to sustain it. As has been observed commerce crossess all divides be they political or religious in peace or war.
Schneier.com is a personal website. Opinions expressed are not necessarily those of BT.