Friday, December 3, 2010

Airport Security and Racial Profiling

If found this post in my "drafts" page from when I first started this blog.  I guess I'd forgotten to press "publish".  Anyway, given the recent changes in airport security measures it seems serendipitous that I found it now... 

Wednesday January 27 2010

First of all, I obviously wasn't able to comply with my new schedule. Band rehearsal ended up being post-poned for 30min but I didn't know how long it was going to actually be when I started waiting. One of the guys in the band had to do a media interview so we couldn't start until it was done. Usually they like to do several takes so the duration of interviews is always unknown. While waiting I practised my stick toss. Then I was supposed to do a "meet and greet" after rehearsal, so I got ready and waited...and waited. I guess it got cancelled because the next thing I heard was 20min to curtain. Maybe tomorrow will go according to plan...HA!

Disclaimer: Before anyone gets all huffy about what I'm going to write, I want to say that this is a thought experiment. Policies that I find logically compelling I have trouble with on other grounds. Besides, this is my blog, and if you don't like it, go write your own! (or leave me angry comments...as you please!) One more thing: please feel free to exchange the word "freedom fighter" with "terrorist" depending the side for whom you feel the most political sympathy.
I listened to a really interesting interview on my favourite podcast "Skeptics' Guide to the Universe". It was about racial profiling for airport security. Although I'm not an airport security expert (big shock, I know) I'd always leaned toward racial and geographical profiling in this context. I know, I know, racial profiling is not something one should publicly support unless you are spend your afternoons in rocking chairs swatting flies and telling any newcomers to your neighbourhood "we don' like yur kahd 'round here". However, in the context of airport security there seems to be an intuitively strong case for it. If a given group that is identifiable racially and/or geographically openly declares that they will attempt acts of terrorism against you, does it not make intuitive sense to screen people that fit this profile a little more carefully than others? I know it's not fair for the innocent majority members of this group to be screened more that non-group members but who said security had to be fair?
The interviewee was an international airport security expert and he asserted unequivocally that racial profiling does not work for airport security. Airport security is much more effective when secondary screening is random. Apparently, this isn't even a debate amongst security experts. That's food for thought. The reason racial or geographic profiling doesn't work is because the terrorists already know what security is looking for, so they will send someone who doesn't meet the profile. For example, when Chechnian terrorists blew up two planes both bombers were women. The terrorists knew that Russian security rarely gives secondary screening to female passengers, so the obvious thing to do was to recruit women. Security will never see what it is not looking for and it is not difficult for the freedom fighters to know what airport security is looking for. When secondary security screening is done randomly, there is no way for the freedom fighters to anticipate who will be screen and on what grounds.
Basically, the way airport security operates now, almost by definition, cannot work because they are always fighting the last battle. When terrorists knew that shoes wouldn't be screened, they put da' bomb in da' shoe. When carry-ons were screened for solid explosives, the terrorists used liquids. When security started screening shoes, terrorists put da' bomb in da' underwear, and so on.
According to the security expert the solution is to take the resources (money and people) that are being wasted in "security theatre" and put it it intelligence gathering and the training of all airport employees to be able to do security interviews and learn to recognize suspicious behaviour. When it comes to profiling, behaviour profiling will yield better results so airport employees should be trained to be able to profile based on behaviour.
Why am I writing about airport security? A couple of reasons I guess. First is that I travel a lot so I directly suffer the consequences of onerous airport "security". Second, prior to listening to the interview I thought that racial profiling for airport security made sense from a logical point of view, although I was never quite comfortable with it's unintended side effects. Listening to the interview showed me my logic was flawed and I learned something new. Contrary to many people I actually enjoy when a belief I hold is methodically disproven with evidence. I like the feeling I get when I can discard an erroneous point of view. Third, I wrote this as a practice for when I'm back in school and I have to be able to assimilate and reproduce what a professor has said during a lecture. Finally, I thought that I there may be a chance that I am not the only one in my circle of friends who isn't a security expert and might also benefit from listening to what an expert has to say on the topic.
In the end, however, I agree with the security expert's arguments for measures that would lead to actual airport security but regarding racial profiling I'm not entirely convinced that it needs to be discarded entirely. I agree that racial profiling on it's own is not the best security method but I'm still unsure why his method would have to exclude racial profiling. I still think that from a purely logical point of view, in cases where there is a racially and/or geographically identifiable group that overtly promises to attempt terrorist acts, there is a reasonable case for profiling. Here's an grossly oversimplified example to illustrate the principle: Imagine you lived in a country where everybody had 5 eyes and there was a country of people with 3 eyes. A group of people from your country, because they harbour grievances against the 3-eyed people, openly declare on behalf of all 5-eyed people (but without consulting you) that they will attempt acts of violence against 3-eyed people. Would it not be reasonable for the 3-eyed people to regard any 5-eyed person with suspicion? How can the 3-eyed people reasonably discern who are the terrorists among the 5-eyed people in their airports? I suppose they could conduct lengthy interviews with each passenger, regardless of the number of eyes: it would be effective and fair, but security wait lines would take even longer than now. Or you can just secondarily screen all people with 5 eyes.
It sounds good in the over simplified example and establishes the principle but racial profiling on it's own might not translate too well in the complex real world of US airport security. First of all, there are several examples of home grown, white terrorists (McVeigh, Unibomber), so if you are only secondarily screening Arabs, the home grown crazies can easily slip through. Also, and this may come as a shock to most of you, but Americans are largely ignorant when it comes to distinguishing between Hindus, Arabs, Israelis, Turks, and other darker skinned people or knowing who their political allies are. An Israeli friend of mine that I work with told me a funny/sad story that happened shortly after 9/11. He was talking to a group of girls who were either in the military or who's husbands were in the military. They asked him were he was from because he has an accent. When he told them "Israel" one of them yelled "you better get the fuck away from us before I try to kick you ass"! Yup, somehow the "subtle" political distinction between the Israelis and Arabs escaped her keen military mind....(by the way, for my American friends reading this---as defined by the state department, the Israelis are your friends!) Anyway, you'd probably have better luck training airport security personnel methods of behavioural profiling rather than trying to undo a life time of ignorance :)
So, where am I going with all this unnecessary rambling? I'm not sure. Maybe something like this: scanning bags, shoes, etc.. has benefits, but by itself it's not the best solution. It probably acts more as a deterrent to the average crazy person than to the well organized, highly motivated terrorist. Behavioural profiling and intelligence gathering are the security expert's methods of choice but it's going to be years before current staff are adequately trained and intelligence networks put in place. Racial profiling is far from perfect but if there are members of an identifiable group that overtly announce their violent intentions, it is not unreasonable to check people that fit the profile.
As one of my heros, Mark Crislip says, please post your comments and hate mail below, I'd love to hear from you"!

Stuck/Quine/Epistemology Natualized

Ok, what I'm about to do may turn out to be a very poor decision but at this point I don't have much to lose.  Here's the situation.  I have one term paper left and for some reason I just cannot make any progress.  I'm so confused by now that I can't tell if the problem is that I don't understand the content of if I can't synthesize the information or if I need more information, or if I have too much and it's just disorganized.  I don't know.  I'm embarrassed to admit how much time I have devoted to this stinkin' paper and I have nothing but a few pages of unfinished thoughts that lead nowhere.  So, here's my plan.  I'm going to summarized the issues here in my blog.  I feel like I'm talking to friends, you're not going to grade me, and my writing seems to flow better when it's "for fun".  The risk is that nothing may come of this and I will have wasted more valuable time that maybe should have been spent doing more research or rereading material.  Anyway, speaking of wasting time, enough with the preamble:

  So my paper is about a philosopher named Quine and the problem of knowing whether we can justify our system of scientific knowledge.  I know what your thinking...um, just put on lab coat, and run some experiments and if the evidence confirms a theory, then ta-da! it works.  Why do philosophers need to make everything so complicated?  In a sense this response is correct, and it is in some ways related to what Quine says.  
  The general topic that we are investigating is "how we know what we know".  This area of philosophy is called epistemology.  Traditional epistemology saw one of its primary roles as trying to find a rational justification for our knowledge of the world.  This type of reasoning must be distinguished from experiential reasoning, or as philosophers call is "synthetic" or "a posteriori"(after the fact).  The problem can be framed like this "how do we know what we know?" Answer: "we learn through experience" Question: "how do you know that experience gives you a reliable picture of the world?"  Typical answer, "because further experiences confirm that my beliefs are true".  Did you catch the problem?  If you didn't, here it is.  This is an instance of circular reasoning.  You are setting out to prove the reliability of beliefs derived from experience by appealing to experience.  It's not a good argument if you support what you are trying to prove with the thing you are trying to prove.
  So what does Quine say about all this?  He says, all this talk of trying to prove science (experiential knowledge) from something other than experience is just crazy talk.  We haven't made any progress since Descartes, who formulated the problem in its modern form, and Hume, who pointed out even more problems.  Lets just stop all this madness and allow for a certain amount of circularity.  Why? because unless you want to postulate divine revelation, clairvoyance and other untenable ways of knowing, all we have is appeal to experience.  So, instead of continuing on this fools errand of trying come up with a logical proof for how it is that science (seems) to work, lets look at how it actually does work.  In other words, instead of looking at how we ought to derive beliefs, lets look at how we actually do derive our beliefs (in science).  How do we do this?  We appeal to empirical psychology.  
  Empirical psychology is the study of how we go from sensory stimulation to some sort of behavioural (maybe assertions) output indicating belief.  Now I know what some of you are thinking: Whoa! Nelly! Stop right there.  That sounds like behaviourism and I remember hearing that behaviourism is the devil!  Lets make a distinction here between philosophical behaviourism and methodological behaviourism.  The former, which is probably the one that most people object to, is the position that all there is is behaviour.  There are not mental states, emotions, etc....we are basically all just machines.  Well, you'd be right to object, and in academia that view died a long time ago.  Quine is proposing methodological behaviourism.  This is the idea that, while people may have mental states, emotions, and sensations, the only thing that we can observe as outsiders is behaviour.  Behaviour here should be construed in a broad sense: from the macro scale all the way down neural and sensory receptor stimulations, firing of synapses and so on.  There is no way to observe what someone's feeling of feels like but we can observe their behavioural responses.  That's all this means.
  Moving on.  So, suppose we accept this model and ignore, for the time being accusations of circularity. Are there any important questions that this new paradigm will not be able to answer? A common argument against this model is that it provides no guidance in areas of norms.  Let me explain:  Norms are standards by which we measure success.  Norms are the basic level of competence required for an operation to be considered successful.  For example, the academic norms for grade 3 are that you are able to read and write at a given level and maybe know all your multiplication tables.  If you are below the norms, you cannot successfully carry out these tasks.  
  So, what's an example of a case where we need to know something about epistemic (to do with knowledge) norms?  Lets say 2 people are having an argument about the results of a science experiment. The experiment was, in a beaker (I have to say that word any chance I get!) they pour 1 cup of sugar into one cup of water: the result is that the beaker now shows the meniscus to be a 1 1/4 cups.  Scientist 1 (I'll call her Mary) says this proves that 1+1=1 1/4.  Scientist 2 (Bob) says, no: this does not disprove the laws of arithmetic.  
  So what does this all have to do with using psychology to solve epistemic problems?  Here it is:  psychology can give us an account of how the light reflecting off the beaker and solution stimulated the visual system and, after a long causal chain, produced a behaviour in the scientists that indicated they perceive the meniscus to have risen to indicate 1 1/4 cups.  If we appealed to neuro-psychology we could learn the neuro-pathways that were stimulated when Mary and Bob came to their respective conclusions.  To summarize, the scientific method could tell us how the reasoning occurred but it cannot tell us which method of reasoning is preferable.  This is because science is descriptive, not prescriptive or normative.  
  So, does it end here?  Case closed?  I think there's more to the story.  I think there are ways that science can tell us which ways of reasoning are preferable.  Let me elaborate:  We know that depending what type of problem we are asked to solve and how we frame a problem it will influence how we solve the problem.  For instance, sometimes people voice opinions/beliefs based on emotions, other times they use logic.  Depending on what type of "reasoning" process they are using, neuro-psychology tells us that different parts of the brain are engaged.  It is safe to assume that for most problems, resolution is preferably achieved through using the rational, rather than emotional part of the brain.  So, since we know which part of the brain is preferable and the conditions which bias toward our engaging it, we can set up conditions for us to yield "true" beliefs.  
  Lets take a step back so we can go into more detail.  When I said that I assumed that for most problems it is preferable to engage the rational part of our brain, in a sense I was expressing a norm.  Strictly speaking science cannot say "this method is better because it produces true beliefs".  This presupposes a norm or value, that of producing true belief.  So, there are 2 issues here, that of expressing preference for an outcome and the notion of truth.  
  Scientific theories cannot tell us what is true.  They can only tell us that to which experience, thus far, has conformed.  Science history is littered with examples of theories that where at one point as sure as, well...lets just say they were held as true.  Sometimes there are competing theories to explain the same phenomena.  The point is the the notion of truth in science, as in most places, is not "true come what may".  Again, science simply gives us a best account of the evidence at hand.
  So, how can science distinguish between thought patterns for justification?  The same way that it adopts any theory.  Setting up circumstances that biasing our thinking in ways that use the rational part of our brain will lead to beliefs that will conform better with experience and help us to better navigate our world than if we had chosen a cognitive process that would bias toward other beliefs.
  Anyway, those are two of the central themes.  There's a lot more but I'm struggling with it.  If you made it to the bottom I thank you for taking the time to read this crap.  I'm so screwed.  I have no idea what I'm going to write....

By the way, if there are any philosophers out there reading this, any helpful suggestions will be graciously accepted....