Secular Right | Reality & Reason

Dec/09

30

Terrorism and health care reform

Share on FacebookTweet about this on TwitterShare on RedditShare on StumbleUponEmail this to someoneShare on TumblrShare on Google+

In 2000, commercial jets carried 1.09 billion people on 18 million flights, according to a no-longer-linkable Boeing document.   Assuming that the number of flyers has not increased since then, that makes for one would-be underwear bomber out of about 10 billion travelers over the last decade.  Does that record represent success or failure?  Are we jacking up physical security measures on planes and in airports because we think that the risk of another underwear bomber has risen since Dec. 25, or because we think that our record of prevention over the last decade was inadequate?   The notion that we should be able to protect against every terrorist incident is understandable, and announcing that we are not going to try to stop every such incident is unthinkable, though former DHS Secretary Chertoff did make tentative noises in that direction regarding cargo screening.  But it’s still intriguing to me why dying in a terrorist-induced airplane crash has a greater hold on the public imagination than driving on the highway, where there are about 40,000 fatalities in the U.S. a year, much higher on a per-mile basis than the number of deaths from non-terror-induced airline crashes, of which there are many more than terror incidents.  We do not have a federal agency checking everyone who gets on a highway for driving safety.  Terror attacks are intentional, not accidental, so the public policy imperative of sending a tough-on-terror message is arguably far greater than for highway crashes.  But that fact doesn’t affect the individual perception of risk, which seems to be influenced by issues of agency, control, possibly even altitude. 

Maybe the proper denominator in assessing risk is rather the number of would-be underwear bombers.  If the number of would-be underwear bombers is small–let’s say, one–then our security system deserves a huge black eye, but our perception of a huge phalanx of ready recruits would need readjusting.  Admittedly, the size of the recruit pool is probably affected by the perception of our security system, so security overkill may be responsible in part for its own disproportionality. 

(Contra my impatience with what appears to me to be security excess at airports, one could argue that we approach all activities with similar levels of risk intolerance.  The difference is that security measures designed to eliminate even very small risks are constantly being engineered into the design of products–such as cars and planes–making them invisible.  That may be.  But then people drive while texting and talking on the cell phone [see below], introducing massively higher levels of risk into their own and others’ lives.)

Perhaps Abdulmutallab’s intentions really should have been as obvious before as after the fact, resulting in the revocation of his visa and his placement on the no-fly list.  But hindsight wisdom can make the chaos of intelligence signals look clearer in retrospect than they are in real-time.  There may be many other people with comparable levels of security suspiciousness who never end up becoming suicide bombers.  It has not been officially decided yet whether Abdulmutallab slipped through the cracks because the threshold for categorizing an imminent terror risk was too high or because officials failed to implement the standards that already exist.  The former problem is easier to fix than the latter, and the dichotomy is overly simplistic in any case, since it is impossible to create a bright-line set of rules for such judgments that eliminates discretion.  But if we want to put a lower threshold on terrorist categorization, thereby increasing the size of the no-fly list and the heavy surveillance of suspects, we’re going to have to do a better job of standing up to the ACLU and the privacy advocates, who complain vociferously at physical screening measures, and at the gathering of intelligence, which ideally renders physical screening less imperative.  It is amusing to see the ACLU’s Jay Stanley argue that whole-body scanners should be used only on people who are subject to extra scrutiny, since it’s not clear that the ACLU has ever seen a security list that it finds acceptable and it certainly would fight any expansion of that list.  I couldn’t give a damn about being X-rayed at an airport, but you’d get a heck of a lot more safety bang for the buck by ending all non-emergency use of cell phones and text-messaging while driving than by adding billions more collective hours spent in airport security lines. 

Holman Jenkins rightly points out that the real story should be how few terrorist attempts there have been in this country.  If Obama were really the force for change and new thinking that he purported to be, he would undertake a study of whether the terrorist threat is such as to justify the massive bureaucracy of DHS.  Of course, there is no chance that DHS will ever be dismantled in any case, it will merely evolve new functions. 

(I have also not seen a full-throated assertion that the Nigerian’s planned bomb could have brought down the plane.  Perhaps that is so obvious that it doesn’t need stating, but I would like to see it stated.)

If the increase in post-12/25 security measures roughly reflects the popular will, and is not simply a bureaucratic reflex unmoored to popular sentiment, the chance of driving down health care spending through more rational use of precautionary testing becomes more unrealistic, it seems to me.  Perhaps it’s not just a ravenous tort bar or doctors’ pecuniary self-interest that drives the aggressive ordering of medical tests, as we have been led to believe, but patient attitudes towards risk.  The revised mammogram guidelines debacle would suggest that to be the case.  Making patients pay for every diagnostic test may increase their tolerance of risk, but such a cost arrangement is never going to happen.

·

6 comments

  • mnuez · December 30, 2009 at 5:14 pm

    Hear hear!

    Three points –

    Cellphones while driving: Easy for you to say Heather, but the laws in NY, NJ and DC have pissed me the hell off. Of course I’m an ignoramous about the most important things in life and thus don’t know how much additional risk I personally am or am not incurring while speaking on the phone while driving (though I do know all about Liz Lemon, House and Jack Bauer! Yeahy for our culture!) but I don’t think that my risk is too greatly increased. Of course on the AGGREGATE though there DOES appear to be quite the increased risk and I am not yet dictator enough to be able to exempt myself (heck, Chuck Schumer even got a spoken-to by an epsilon… and a ratting out by some kind of hall monitor or something).

    Point number two: I forgot. Happy New Year! Who hoo!!

    I see… stupid people…: A bunch of idiots took umbrage with your previous post on the subject. What in the name of Christ is wrong with people if even here, at secular right, with you, Razib and Derb, such idiots are still in abundance! Also, are all the authority-trusting fraidy cats on the secular right of the female variety?

    Point number four: I wish everyone a very very happy and fun drunken time with no attendant physical discomforts (or Driving… for Fuck’s sake!) But entirely videotaped so you can remember with pleasure the freedom you once had (and also have a better grasp of why you lost your job, marriage and left knee, but that’s beside the point). Happy random drunk day!

    mnuez, your friendly neighborhood opinionated fun guy in need of moolah now

  • mnuez · December 30, 2009 at 5:27 pm

    Mnuez is truly an idiot, of the three Jihadyphodibs one calls itself Carl, another Mike (who, to be fair , may or may not have been making a point about prevention, as his subsequent posts either clear up or don’t) and a third Deklane, gender uncertain.

    I really wish that mnuez guy would think before he writes.

    Mnuez

  • Jim · December 30, 2009 at 5:49 pm

    Well said. However, we do know some things about Abdulmutallab. We know he came from an Al Queida red alert state. We know he paid cash for his ticket. We know he is Muslim (sorry but that’s important in the profile), and we know his father ‘outed’ him.

    Somewhere our country’s computer security system could easily know all those things at the same time, without spending any more money than the billions it is already spending.

    So the real question is why Abdulmutallab was not pulled aside and searched by security already in place and paid for, before he got on any airplane coming close to the USA?

    Either the system doesn’t see all those things listed above to meet the bar of personal search and questioning, and I would ask “Why not?” Or the system isn’t in place as I just described. I doubt it is in place, even while we search little white girls in pig tails. The question is ‘why isn’t it in place?’ Because it is easy to implement. And we are certainly paying for it.

    So the comparison to texting is partly incorrect. Unlike the risk of texting, the citizens are angry because they are already paying for security that they are not getting, and now are being inconvenienced further based on a silly system that no rational security company would ever build.

  • Mike H · December 30, 2009 at 8:32 pm

    In spite of the lack of civility and insulting tone on part of a previous commenter I wish to address Ms. McDonald’s post.

    I am by no means a fan of knee-jerk security responses which seem to lack any specific purpose except to signify to the public that something is being done. It appears obvious to me that the fight against terrorism ideally occurs before a would-be terrorist walks through airport doors, into a hotel lobby or onto a subway train namely by aggressively combating those involved in those hostile causes through surveillance, investigation and elimination. That is where the Nigerian fellow should have been stopped, his previous involvement with Islamic causes and the warnings from close family members should have made him a high-risk suspect that should have never boarded a plane to the U.S. That has obviously little to do with whether you can read a magazine 45 minutes before landing or not.

    And of course there are natural limits to transport security measures simply because of cost and practicability concerns hence it is much easier to board a city bus or commuter train than to board a commercial airliner. I have no illusions with regards to that reality and no objection to the point.

    But that doesn’t mean it’s wrong to think about ways of enhancing airport security within the framework of practicability and really my problem is with the implication given here that airport security is unimportant because terrorism is “overrated”. In fact, I took the contrast with car-related fatalities to merely be a rhetorical tool to cast doubt upon the relevance of terrorism (it is usually applied that way in this context), mostly because there is no evidence that outlawing cellphone use whilst driving and combating terrorism are in any way mutually exclusive. I strongly favor both. And airport and airplane security are parts of the overall effort against terrorism and there is good reason to look for enhancements – and I would hope we can all agree on that no matter what we think of specific requirements that have been added recently.

    My main argumentative thrust has thus been the relevance of the Islamic terrorist threat rather than any argument about individual perception of risk. The individual perception of risk will always depend on many factors, the publicity of an issue being a major one, it hardly is rocket science to figure out why people pay attention to the unusual and high profile rather than the routine and mundane. Now I do not think it is wise to suggest we ought to talk less about terrorism as some critics will usually claim because the fact it is talked about so much is directly related to its relevance on a societal level.

    Unlike personal issues like the individual death of someone in a fairly routine fashion, death in a terrorist attack has by default political and thus social implications. Terrorism is an issue of national security and affects the interests of a nation overall so whether I myself feel there is a high risk of myself dying in a terrorist attack or not is in a way irrelevant as my concern and fear might just be about the country (“us”) rather than myself specifically. Maybe we feel the costs are justified because we perceive terrorism to be a particularly vicious form of political warfare aimed at our country and thus ourselves and we wish to see our country win the confrontation and defeat our enemies – not only but also to discourage others from using terror to make political gains. As I commented elsewhere I see a nation standing up to terrorism as morally and socially vital just like I think a community standing up to crime is vital. Terrorism however occurs on a larger stage and with a much broader focus and that’s what makes terrorism an issue of national and historical magnitude (20th century history offers one story after another about terrorism and the effect it can have on the fate of entire nations) and both national public attention and federal resources devoted to it make sense as far as I am concerned. (I don’t mean this to be a constitutional argument but I am pretty sure national security is a federal responsibility by even the strictest originalist viewpoint whereas the creation and enforcement of traffic laws appears to be a local and state matter by any standard).

    That doesn’t mean I agree with how the money is spent, I don’t like, never liked, the DHS because I always figured it’d just be another bureaucracy no better than what was already in place. And the moment something is considered by suits in Washington, it is also going to involve politicking, bureaucratic lack of direction, waste of money etc. No doubt those things were observed in the various administrative agencies during the Civil War and World Wars as well, yet there was no alternative to the use of them and the only thing that could be done were strong efforts to increase their usefulness.

    I also obviously recognize the problem posed by the ACLU and the “civil rights of everyone including suspects and perpetrators always have highest priority” attitude with regards to fighting terrorism and crime of any sort. The current Justice Department is full of advocates of such positions and thus it’s unlikely that much progress will be made on this front under President Obama. However that isn’t exactly a contradiction or refutation of my point but rather something that should motivate everyone to work harder to reduce the influence of those groups and ideas. It would not just positively affect the conflict with Islamist terror but also the troubles of urban crime and illegal immigration which I care about as much as Heather does.

  • John · December 30, 2009 at 10:14 pm

    I don’t agree with the notion that terrorism isn’t a big deal since not as many people die from it as car accidents (although think of how much the governent regulates driving!). If someone dies in a car accident, that is simply an accident. When someone dies in a terrorist attack, it is because a group of people are trying to achive a specific political goal. If someone accidentally hits me, I’m going to react very differently than if he hits me on purpose. That isn’t irrational. One is a moral transgression, the other isn’t. Doing nothing in the the first case won’t have any negative consequences. Doing nothing in the second case invites more aggression. I don’t think we should wait until 40,000 people a year die from terrorism to consider it a serious problem.

    Yes, I find airport security annoying, too. The obvious solution…

    Mrs. 60 year-old Japanese grandmother. Please just walk through the metal detectors. Enjoy your flight.

    Mr. 20 year old single Muslim man who wrote an article called “Infidels Must Die!”. Please step into line B. Thanks.

    This would save both lives and time. Of course it will never happen.

  • Clark · January 1, 2010 at 4:44 pm

    The problem is that terrorists aren’t stupid. Drug dealers long ago. Learned the value of simply getting mules who don’t fit the profile. Don’t get me wrong – I’m not against profiling. Afterall it’s often harder to find people not in the profile willing to work. Which is why so many mules still fit the profile – but of course far too many of the others are simply successful at smuggling…

<<

>>

Theme Design by devolux.nh2.me