Our Glorious TSA

Once you eliminate the implausible corner solutions—the TSA undertakes body cavity searches of every passenger, or security screenings are abolished—a number of consequences follow.

1.  The TSA must randomize (play a “mixed strategy,” in game-theoretic parlance).  Otherwise, terrorists can predict some of its precautions and evade them.  The same principle explains why police vary patrol routes and road blocks.  A NYT article today makes clear that the TSA is self-consciously randomizing to keep terrorists off guard.

2.  At the social optimum, the number of successful terrorist attacks will be greater than zero.  It might be argued that we have had too few successful terrorist attacks over the last few years rather than too many.  The question is whether the implicit statistical valuation of life in TSA programs is too high.  I suspect that the answer is yes, as is generally the case with airline safety.

3.  Profiling is an effective strategy when, as here, terrorists come from a small group of (relatively) easily identifiable people.  One suspects that this explains Israel’s success.  But profiling places a large portion of the cost of deterrence on a small group, which makes some people morally uneasy.

4.  Once the implausible corner-solutions are ruled out, any security policy or threshold will seem arbitrary because you have to draw the line somewhere, which means that it will be easy to point to some permitted activity that is only slightly different from what is forbidden (for example, carrying on 100 ml of liquid rather than 101 ml).

5.  As for the “security theater” claim–

a.  If ordinary people are fooled into thinking that the TSA is doing more than it is really doing, then at least some potential terrorists will be fooled as well, and so will be deterred from engaging in airplane-terrorism.

b.  Ordinary people will also fly more often, which means that one of the goals of terrorists—to terrorize people so that they will pressure their government to make concessions to terrorists—will have failed.

Categories: Counter-Terrorism Policy    

    79 Comments

    1. george weiss says:

      At the social optimum, the number of successful terrorist attacks will be greater than zero. It might be argued that we have had too few successful terrorist attacks over the last few years rather than too many. The question is whether the implicit statistical valuation of life in TSA programs is too high. I suspect that the answer is yes, as is generally the case with airline safety.

      i love how intellectual this sounds despite how depressing the real meaning of this is.

    2. krs says:

      In response to 5a, I question whether anyone’s genuinely “fooled” into thinking that surly people confiscating toothpaste are doing anything to make flights safer.

      And in response to 5b, speaking only from personal experience I’ve never decided against flying because of fear of terrorism, but I have decided against flying (either by cancelling a trip altogether or going out of my way to drive or train) because of how annoying TSA is.

    3. Dan says:

      Interesting framework in 1-4, but 5 is off base because 1)the problem is TSA is obviously doing more for the sake of doing more, so you have added inconvenience without even a perception of efficacy and 2) making flying more inconvenient (therefore more costly) means ordinary people will fly less. I personally take the train from New York to DC because by the time I am finished with airport security the travel time is a wash.

      I also don’t think there is any evidence that the goal of the terrorists is “to terrorize people so that they will pressure their government to make concessions to terrorists.” Evidence actually suggests that they want to provoke a massive military response from the West because they believe such a war will lead to God smashing the West and reestablishing the “righteous” Caliphate. You make the mistake of imposing rationality on irrational actors.

    4. vic says:

      Profiling is an effective strategy

      yes it is, but once you make it very obvious who or what is being profiled, the terrorists will adapt. So it has be “Gamed”.

    5. Blue says:

      In assessing the cost of a successful terrorist attack (point 2) you cannot simply add up the costs of human lives destroyed. You must also include the costs of the response to that attack. The cost, for example, of 9/11 represented by the destruction of four planes, the WTC, and damage to the Pentagon has been utterly dwarfed by the costs of the response. Imagine the costs of a response to an attack that, say, destroyed the Capitol and killed many of our Representatives and Senators.

      Vic, you make it seem as if that adaption is cost-free. It isn’t…indeed, the continuing absence of Amish Islamic terrorists post 9/11 strongly suggests that the barriers to substitution are very large.

    6. krs says:

      Point 4 seems to be knocking down a strawman.

      vic, perhaps, but it can be difficult to find someone (a) willing to kill him/herself to further al Qaeda’s objectives, (b) competent enough to go through the training and carry out the attack, and (c) able to avoid the “profiling.”

      If (c) adds sufficient difficulty to the planning process, then profiling may be effective. Also, profiling doesn’t mean zero security measures for people who aren’t within the profile.

    7. Hadur says:

      Please delete this post. When I’m President, some of the stuff he says here will get in the way of my nomination of Eric Posner as Legal Adviser of the Department of State.

    8. Adam J says:

      Eric- “The question is whether the implicit statistical valuation of life in TSA programs is too high. I suspect that the answer is yes, as is generally the case with airline safety.” I think you’re viewing this too legally, as to whether they would be liable under the old Burden of Prevention < Probability x Magnitude of Loss formula. However I suspect the industry and TSA are being rational here, they are also considering losses beyond the direct losses- if a terrorist incident occurs the airlines end up suffering much greater future profit losses from future passengers lost faith in the safety of air travel (and the TSA is probably also concerned about the political costs that would occur as a result of a successful strike).

    9. Malvolio says:

      The TSA must randomize (play a “mixed strategy,” in game-theoretic parlance). Otherwise, terrorists can predict some of its precautions and evade them.

      They will evade them anyway.

      Take the most recent nonsense: immobilizing passengers for the last hour of the flight. Assume (counterfactually) that would in fact prevent a terrorist from detonating a bomb.

      If this ridiculous rule is imposed uniformly, the terrorist will merely blow up the plane 61 minutes before landing, less satisfying perhaps but still effective.

      If a random strategy is tried, some percentage of Undi-bombers will be allowed to blow up their flights; the rest will be forced to either choose the 61-minute strategy or just land and catch the next flight and try again.

      Remind me again why Napolitano hasn’t been fired.

      Blue: Imagine the costs of a response to an attack that, say, destroyed the Capitol and killed many of our Representatives and Senators.

      Just the bill for champagne, confetti, and pointy hats would be huge.

    10. Bill Altreuter says:

      I have to take exception to 5(b). Flying is already such a hassle that lots of people (myself included) consider other options when possible. I regularly fly from Buffalo to NYC for court appearances, for example. There is an additional hour tacked on either end of the round trip to get back and forth from JFK, and 20 minutes from my house to Buffalo-Niagara International. That’s about five hours total travel time, door to door, allowing for time to pick up a paper when I get to New York. Add an hour at each end so that the TSA can check to be sure I don’t have any shampoo or sunscreen in my briefcase and you have about the same travel time it would take me to drive.

      It’s one thing if I’m going to London– I’m not going to pilot my personal submarine to get there. It’s another if I’m thinking about a day trip to somewhere on the Eastern Seaboard. People who fly infrequently enough to be confounded by taking off their shoes when they are on the line in front of me may not mind the inconvenience, and may buy into the security theater stuff. Most travelers are not those people.

    11. Oren says:

      Hadur, I’d rather put him in DOJ.

      I love how intellectual this sounds despite how depressing the real meaning of this is.

      As to point (5b), I think this is widely underestimated. The 9/11 hijackers probably caused at least 3000 extra road casualties from those that chose to drive instead of fly in the past decade (driving of course, have a average fatality rate per mile of many times that of flying). Whether they chose to do so because of fear of flying or aversion to the new security requirements (or because the former two reduced economies of scale past the point where flying made sense over driving) is up for grabs, of course.

    12. Mark Field says:

      Profiling is an effective strategy when, as here, terrorists come from a small group of (relatively) easily identifiable people.

      Perhaps you could describe what you consider to be the “small group of (relatively) easily identifiable people”, since this statement, on its face, is so utterly absurd as to destroy your credibility entirely.

      Then you can explain, using data rather than logic, why and how profiling would be an effective strategy.

    13. Fen says:

      You missed one:

      6) Static defenses are permeable. You cannot be strong in all places at all times. Attacks will still get through. Therefore, you must play offense and act pre-emptively.

      If I’ve reduced you to *reacting* to me, I’ve won.

      We probrably wont learn this until after we lose a city.

    14. PeteP says:

      “2 The question is whether the implicit statistical valuation of life in TSA programs is too high. I suspect that the answer is yes”

      Then you disagree with long standing FAA policy ? As re : aircraft safety, that is. Why is flying so safe, statistically speaking ? After all, it’s a high-risk concept – hundreds of people in a little metal box that relies on perfection of performance. Look what happens after a ( non-terror-related ) crash – the FAA dispatches highly trained, well equipped teams to do a complete ad hoc rectal exam of the event, and they don’t hesitate to ground entire fleets for suspected design errors, etc. They don’t care about ‘cost effectivenesss’ of requiring a new rudder design retrofit, or other systems. Redundant and triple-redundant controls and safety systems are routine, from the largest to the smallest. No item is too large or small, or too expensive, for them to say ‘There is a safer option, it SHALL BE implemented NOW, before the next plane leaves the ground, or even goes up to the loading gate’.

      “3 – But profiling places a large portion of the cost of deterrence on a small group”

      NOnsense. No ‘group’ or individual is EVER charged an extra dime for being the subject of hightened scrutiny. The cost are spread across the entire system, every ticket, every passenger, regardless. No one group is EVER affect financially because of their group identity.

      “5. As for the “security theater” claim–”

      Here’s a big one – the National Guard teams patrolling airports in full camo, with M-16′s at the ready. Unless someone anticipates a frontal ground assault on the terminal, this is purest theater.

      The true ‘theater’ is the intentional disregard of the real high-risk points, in the name of ‘political correctness’.

      Given the following passenger manifest :

      Boy scout troop going to a ‘Annual Jamboree’
      Girl scout troop going to the ‘Cookie salesgirl Awards’ ceremony
      Old folks home residents being wheeled to the National Bingo Championships
      5 muslim men, obvious in their garb and a facial appearance, bertween the ages of 17 and 29, huddling together and looking around furtively while holding Koran’s and speaking Arabic, with travel histories that include Yemen and Pakistan, that bought one-way tickets with cash and have no luggage, etc …..

      Our mock ‘security’, our ‘theater’, our blind PC ‘morals’, say that all 300 of these people are equally suspicious, and should be subject to equal scrutiny, without regard to any of the aforementioned characteristics.

      This is lunacy.

    15. Fen says:

      Perhaps you could describe what you consider to be the “small group of (relatively) easily identifiable people”, since this statement, on its face, is so utterly absurd as to destroy your credibility entirely.

      http://principallypolitical.files.wordpress.com/2009/02/9_11-hijackers.jpg

      Idiot.

    16. Ugh says:

      Perhaps you could describe what you consider to be the “small group of (relatively) easily identifiable people”,

      He means, Them. Those People. Not people like him.

    17. David Chesler says:

      Just another voice that 5b may be counter-productive. I love being aloft, but I avoid commercial flying. Then again, if I were one of the many sheep I wouldn’t be here.

      As for 5a, casual passengers are more likely to be fooled than terrorists who are still a threat and are actually giving this some thought. They are a lot more of a danger than the guy who thinks “I’ll just hide it in my shoes. Oops, they check those, nevermind.”

    18. Mark Field says:

      No, moron. He means the profile that includes 95% of the terrorists who have attacked America.

      All those pictured were male. That’s not a “small” group. What else ya got?

    19. JK says:

      “He means, Them. Those People. Not people like him.”
      No, moron. He means the profile that includes 95% of the terrorists who have attacked America.
      And you know what you can do with your PCBS and faux moral righteousness.

      What other terrorist attacks on the US besides 9/11 were done by people who meet your definition (you refuse to articulate it so I assume you mean Arab males ages 18-35 holding middle eastern passports)? This recent bomber wouldn’t fit that profile, Richard Reid wouldn’t, Timothy McVeigh and Terry Nicholes wouldn’t, nor would Ted Kazinski, the guy who shot up the Holocaust museum, or any of the plethora of anti-abortion terrorists.

      I’m not even entirely against profiling, but the idea that it’s some sort of silver bullet that we refuse to use for PC reasons is pure nonsense.

    20. as says:

      The airport security battle that we see at the terminal is futile. Any wingnut with an imaginary motive for a moment of fame has a large mathematically enormous positive chance of succeeding in doing something obscene. While the semantics of whether or not it is theatre are debated, the facts and statistical probabilities make the exercise irrelevant. The likelihood for success for an aspiring jerk with current visible measures in place refute the idea that the measures are really effective. The lack of substantial numbers of successful incidents would suggest a number of hypotheses:

      1. If there are a lot of bad guys out there, they all have an extremely high propensity for complicated mental disabilities that are fortuitously vulnerable to the kinds of measures taken by TSA;

      2. There are not very many bad guys out there at all who also have a strange attraction to airplanes, and the rate of incidents reflects this;

      3. The TSA is employing some other less visible measures that are actually effective and we never hear about the people they intercept;

      4. Terrorists have little to gain from attacking airplanes themselves, and recruit genuinely vulnerable people who will inevitable screw up or be intercepted by fellow passengers.

      I think #1 is ludicrous. #3- who knows. I doubt it. #4,possible, but very uncertain. #2- I think more likely. Even one success is disastrous, however. It’s enormously asymmetric.

      That people want to travel so badly that they want to be strip searched is amazing. That this adds very little if any security to a flight when pitted against a real bad guy should be obvious. It’s cognitive dissonance. I imagine most people really do know this, but mutter religiously about duty to comply, it’s the least we can do, etc., rather than face up to the fact that if the risk is really that high and it bothers them, they should not fly. They should also not drive, cross streets, or use Q-tips in their ears.

      Complainers like me should realize the TSA’s nervous rosary beads are not worth fighting- human psychology is far stronger. I choose not to fly the airlines about 90% of the time because the annoyances truly do bother me enough to own an airplane and fly myself wherever I need to go inside the US, generally for around twice the cost of coach on commercial transport given my current travel desires, the airplane overhead, and the direct costs of flying. This is out of reach for the majority of folks, but nonetheless I am shocked at the lack in development of alternative transport trends. It is a window on the human (and American) mind that privacy is so important but forgettable in reality.

    21. Oren says:

      They don’t care about ‘cost effectivenesss’ of requiring a new rudder design retrofit, or other systems. Redundant and triple-redundant controls and safety systems are routine, from the largest to the smallest. No item is too large or small, or too expensive, for them to say ‘There is a safer option, it SHALL BE implemented NOW, before the next plane leaves the ground, or even goes up to the loading gate’.

      Sure they do, they just assign a very high value towards preventing future accidents. The ETOPS rule, for instance, would never have been adopted since there are plenty of existing quad jets that can service those routes.

    22. Curt Fischer says:

      2. At the social optimum, the number of successful terrorist attacks will be greater than zero. It might be argued that we have had too few successful terrorist attacks over the last few years rather than too many. The question is whether the implicit statistical valuation of life in TSA programs is too high. I suspect that the answer is yes, as is generally the case with airline safety.

      At a social optimum unconstrained by the realities of our legal and political structures, perhaps. But you should recognize that we have a large, lumbering, CYA-prone bureaucratic federal security apparatus, and an ill-informed and somewhat paranoid public. These two factors constrain the “optimal” level of terrorist attacks: an increased level of terrorist attacks could provoke a public will to impose highly onerous, invasive travel restrictions, must worse than what we have now. The public will not realize that the optimum number of terrorist attacks is greater than zero. Add in the CYA-prone DHS bureaucracy. Moving towards your optimum, with greater numbers of terrorist attacks, would inevitably lead to widespread public outrage, which would prod security officials to float oppressive security measures as the only viable response. We would move towards extremes, where maybe even having secret police search American homes at random starts to seem politically plausible. Since that outcome is so obviously undesireable, we must constrain our social optimization of the number of allowable terrorist attacks. Applying these constraints of bureaucracy and public ignorance would lead to a social optimum for terrorist attacks that is much lower than the unconstrained social optimum you consider in your post.

    23. Oren says:

      Complainers like me should realize the TSA’s nervous rosary beads are not worth fighting– human psychology is far stronger. I choose not to fly the airlines about 90% of the time because the annoyances truly do bother me enough to own an airplane and fly myself wherever I need to go inside the US, generally for around twice the cost of coach on commercial transport given my current travel desires, the airplane overhead, and the direct costs of flying.

      For the rest of us that don’t travel as often (especially us academics who have somewhat flexible schedules and can therefor fly for $130 RT in the US and $500 RT to Europe :-P ) the decision is much simpler.

      Plus, while I don’t think airport security is particularly effective or anything, it’s hardly annoying. Maybe slightly vexing but that’s it. I mean, is it really that hard to put your coat and shoes in a box?

    24. Wikitorix says:

      JK: What other terrorist attacks on the US besides 9/11 were done by people who meet your definition (you refuse to articulate it so I assume you mean Arab males ages 18–35 holding middle eastern passports)? This recent bomber wouldn’t fit that profile, Richard Reid wouldn’t, Timothy McVeigh and Terry Nicholes wouldn’t, nor would Ted Kazinski, the guy who shot up the Holocaust museum, or any of the plethora of anti-abortion terrorists.

      I’m not even entirely against profiling, but the idea that it’s some sort of silver bullet that we refuse to use for PC reasons is pure nonsense.

      Er… When did Timothy McVeigh or Terry Nichols or Ted Kaczynski pass through airport security with the intent to hijack or destroy a plane full of people? Maybe you should look at your data again and see if the terrorists who attack planes fit any sort of pattern.

    25. pot meet kettle says:

      All those pictured were male. That’s not a “small” group. What else ya got?

      now, now, mark. don’t make fen’s brain hurt by making him, shudder, think. where’s your holiday spirit?

    26. Tweets that mention The Volokh Conspiracy » Blog Archive » Our Glorious TSA -- Topsy.com says:

      [...] This post was mentioned on Twitter by Jonathan Abolins, Wallace Forman. Wallace Forman said: "The optimal number of terror attacks is greater than zero" http://volokh.com/2009/12/29/our-glorious-tsa/ [...]

    27. JK says:

      Er… When did Timothy McVeigh or Terry Nichols or Ted Kaczynski pass through airport security with the intent to hijack or destroy a plane full of people? Maybe you should look at your data again and see if the terrorists who attack planes fit any sort of pattern.

      Right, Right, before it was “terrorists who have attacked America,” now it is terrorists who “pass through airport security with the intent to hijack or destroy a plane full of people.” Exactly what people does this new set consist of? I’m still not seeing any pattern that is easily identifiable visually or from a passport.

      As far as I can tell the argument still boils down to: 9/11 was committed by Arab males ages 18-35 holding middle eastern passports, therefore it is safe to assume that all, or at least the vast majority of, future terrorist attacks against airplanes will be committed by this same group. And it still seems like some extremely spotty logic to me.

    28. Connecticut Lawyer says:

      The issue is whether to look for bombers or bombs. The latter requires huge amounts of inconvenience to most travellers, and will inevitably end up severely intruding on the privacy of all (once whole body scanners become ubiquitous, which seems to be where we are headed). In any event, no current technology (aside from CT scans or MRIs) would reveal explosives that are swallowed or carried in a body cavity.

      Looking for potential bombers, as Israel does, is IMHO a better approach, and at the very least should be done to supplement very basic bomb screening. Pulling young, foreign muslim men aside for close questioning seems like an obvious strategy. Asking a few questions of all other travelers (where are you going? do you have a boyfriend? Where is he? What was Babe Ruth’s birthplace?) could, in the hands of trained interrogators, lead to other suspicious travelers who would be more closely questioned or searched. The only problem is that the TSA surely does not have personnel qualified to ask close questions and understand the answers.

    29. Blue says:

      JK, it takes a person really committed to self-inflicted blindness not to see that the problem is with young Muslim males from overseas or with overseas connections.

    30. Smooth, like a Rhapsody says:

      So…pot, mark, jk:
      If all Arab Males between 18-35 suddenly went incorrigibly catatonic, the rate of terror attacks (on planes, trains, night clubs, what-have-you), could be expected to ….
      go up?; down?; can’t tell?….???

    31. Mark Field says:

      JK, it takes a person really committed to self-inflicted blindness not to see that the problem is with young Muslim males from overseas or with overseas connections.

      Even if I were to grant your absurd premise that only some terrorists count for purposes of profiling, you’re still talking about a group that includes maybe 250 million people. What good is “profiling” on that level?

    32. Mark Field says:

      So…pot, mark, jk:
      If all Arab Males between 18–35 suddenly went incorrigibly catatonic, the rate of terror attacks (on planes, trains, night clubs, what-have-you), could be expected to ….
      go up?; down?; can’t tell?….???

      I wouldn’t claim to know, but let’s assume it would go down. The problem is, as I pointed out, that this is not a “small” category (and I’m letting you get away with the implicit assumption that TSA has some way to know a person’s religious affiliation). It’s therefore useless for profiling.

    33. orca says:

      Blue: JK, it takes a person really committed to self-inflicted blindness not to see that the problem is with young Muslim males from overseas or with overseas connections.

      Most captured terrorists seem to have a history of mental illness.

      Americans ignore this common thread, though.

      There’s no in dispatching armies of mental health workers to third world countries.

    34. Blue says:

      Mark Field: Even if I were to grant your absurd premise that only some terrorists count for purposes of profiling, you’re still talking about a group that includes maybe 250 million people. What good is “profiling” on that level?

      They may be a large number in the world, but in the US and in the US transportation system they are a very small number. We don’t have to profile the dirt farmer from Somalia. The engineering student from Cairo, on the other hand….

    35. Mr L says:

      All those pictured were male. That’s not a “small” group. What else ya got?

      I see what you mean. Other than gender they had nothing in common – not age, origin, behavior, nothing.

      Obviously the only option is to deploy thousands of agents at the cost of billions and strip search every grandma, 6 year old child, and blind quadriplegic who boards a plane unlike every other country on earth.

    36. Smooth, like a Rhapsody says:

      Mark
      Neither Arab nor male is a religious category.
      You say it is a large group. How large a proportion are they of all US (to and from) airline passengers?…5%?…less?…
      If they are 1/20th of the group how about spending 10x as much time on that group and leave the rest of us alone. That would, on balance, save time, and be more effective.

    37. Chris Travers says:

      Ok. There are a couple of problems here.

      The first, in point 1, is that shifting security measures around, after announcing you are doing so, offers no real benefit. An individual can watch the security line, get a basic idea of what is going on, and then decide whether or not to go forward. The terrorist can do this BEFORE checking in so there would be no difference between a terrorist noshow and a normal customer noshow. Furthermore, if you don’t tell people that they could be in for a 7 hour delay at security, all you are going to do is cause customer service problems for airlines. In essence, you harrass the passengers, quite possibly into taking other transportation.

      And if the security measure was effective, then not placing it at other airports essentially allows terrorists to use those airports.

      You seem to be thinking that the goal of security is to catch terrorists. This is fundamentally incorrect. The goal of airport security is to allow law enforcement to catch the terrorists by requiring plots of sufficient difficulty and complexity that they can be intercepted before they take place. Once a determined and well-prepared terrorist is at the security gate, it is really too late.

      Points 2-4 seem pretty correct afaics. But I would note that profiling really should be based on body language and apparent state of mind. Once you start limiting profiling to small groups of reasonably identifiable people, you encourage movements away from that. It is better to assume that the terrorists are recruiting 70-year-old white women for suicide operations than to assume that they are not. This being said there are a number of impressive flags which were missed in this case. Alleged refugees really should be subject to additional checks including any visas attached to their tickets, etc.

      I would ask you this:

      If I were granted a monopoly on offering a specific service, and then I pretended to offer it far more than actually offering it, would that be honest services fraud? Certainly it does seem to be a scheme to falsely deny us the intangible right to honest services regarding airport security more directly than almost any application of that law, such as that of Jeff Skilling. Is the TSA above the law? (Assume for the moment that the Supreme Court does NOT strike the law down.)

      Secondly, there seems to be a general consensus among even lay people I know that this is just security theater. This includes folks on the right and left, and it includes both foreigners and citizens. I don’t really think they are fooling anybody.

    38. JK says:

      JK, it takes a person really committed to self-inflicted blindness not to see that the problem is with young Muslim males from overseas or with overseas connections.

      You really can’t use “Muslim.” You need characteristics that either identifiable visually on on a passport. I suppose it’s fair to assume that a guy named Abdul Muhammad from Saudi Arabia is Muslim, but “Richard Reid” from The UK? How do we know if someone has “overseas connections”? I have an Uncle in the UK, is that an “overseas connection”?

      So…pot, mark, jk:
      If all Arab Males between 18–35 suddenly went incorrigibly catatonic, the rate of terror attacks (on planes, trains, night clubs, what-have-you), could be expected to ….
      go up?; down?; can’t tell?….???

      Well since 9/12/2001 it would have remained completely static, right? If we go with “Arab” rather than “Muslim” (which I reject as an usable profiling category), we’re really talking about one terrorist attack (9/11). Yes, it was “the big one” but n=1 doesn’t make a pattern no matter how big that example is.

    39. Blue says:

      “It is better to assume that the terrorists are recruiting 70-year-old white women for suicide operations than to assume that they are not.”

      This is a common thought that seems utterly disconnected from reality. Radical, militant Islam has proven very unpalatable to most natives in the West.

    40. Bruce Hayden says:

      PeteP: Here’s a big one — the National Guard teams patrolling airports in full camo, with M-16’s at the ready. Unless someone anticipates a frontal ground assault on the terminal, this is purest theater.

      Worse, they were doing so, at least in the SLC airport after 9/11, without visible ammunition. No magazines in their M-16s, and no magazines anywhere you could see them. No bayonets either. Plus, no radios.

      As you said, purest theater.

    41. Preferred Customer says:

      Blue:

      Blue:
      They may be a large number in the world, but in the US and in the US transportation system they are a very small number.We don’t have to profile the dirt farmer from Somalia.The engineering student from Cairo, on the other hand….

      Even granting that we should profile a certain segment of the populace (which I don’t think would be productive), what percentage of people on a flight that originated in Lagos are likely to be Nigerian? What percentage of those people are Muslim? What percentage of those people are male?

      I have no idea about the answers to any of those questions, but logic suggests that it is more than a “very small number,” at least in percentage terms.

      Moreover, Hamas and Hezbollah have shown that it is not at all impossible to get women to be suicide bombers. And the recruitment of US citizens to fight for AQ in Somalia has been much in the news of late. Any effort to profile, to the extent it was in any way acknowledged publicly, would be self-defeating; the enemy could simply take advantage of that profiling by sending through someone who did not fit the profile.

    42. Blue says:

      the enemy could simply take advantage of that profiling by sending through someone who did not fit the profile.

      Again, that is a far more difficult job from AQ or whoever than you make it out to be.

    43. Ken Arromdee says:

      what percentage of people on a flight that originated in Lagos are likely to be Nigerian?

      I would suggest that the percentage is large, but the percentage–and the number–going to a US destination would be smaller.

    44. Chris Travers says:

      Blue: This is a common thought that seems utterly disconnected from reality. Radical, militant Islam has proven very unpalatable to most natives in the West.

      Most doesn’t cut it. What about Richard Reid?

    45. Bruce Hayden says:

      JK: Well since 9/12/2001 it would have remained completely static, right? If we go with “Arab” rather than “Muslim” (which I reject as an usable profiling category), we’re really talking about one terrorist attack (9/11). Yes, it was “the big one” but n=1 doesn’t make a pattern no matter how big that example is.

      So, why not use obviously Muslim? Sure, Richard Reid traveling under his orignal name might pass. But most of the rest, including this guy, would have been noticeable (and thus profileable) as a Moslem based on his name. Add in coming from Yemen (and originally Nigeria).

      Remember, with profiling, what we would be doing is applying heightened scrutiny to those who are more likely terrorists than those who aren’t. So, there would be nothing wrong with giving someone profiling points for being from an Arab country, for having a Moslem name, being male, between 18 and 35 or so, etc. And subtracting points for being from a small town in the U.S., older than 65, female, etc. Then, multiplying this by some random number function to determine who should get increased screening.

    46. Blue says:

      Chris Travers: Most doesn’t cut it. What about Richard Reid?

      A convert Muslim, you mean? Western converts actually should be one of the profile categories, actually.

    47. Smooth, like a Rhapsody says:

      Chris Travers:
      In a world with finite resources, time and patience, I would argue that “most” most certainly does “cut it”.

    48. JK says:

      So, why not use obviously Muslim? Sure, Richard Reid traveling under his orignal name might pass. But most of the rest, including this guy, would have been noticeable (and thus profileable) as a Moslem based on his name.

      To be clear by “most of the rest” you mean the 9/11 guys and this current bomber Umar Farouk Abdulmutallab? So you managed to get to n=2 from the original claim of “95% of the terrorists who have attacked America.” Additionally, now that we’ve eliminated the mystical Muslim identifier it’s apparent that a forged passport would be an almost certain evasion to the whole system.

      I’ll also note again, as I did in my first post, that I’m not really against profiling per se. It’s just not nearly as simple or effective as advocates seem to suggest. It’s not simple because it’s harder to objectively define the categories then is suggested, and it’s not terribly effective because it can be evaded like any other known security system.

    49. Chris Travers says:

      Blue: A convert Muslim, you mean? Western converts actually should be one of the profile categories, actually.

      Cool, so put Farrakhan on the no-fly-list?

    50. Chris Travers says:

      Bruce Hayden: So, why not use obviously Muslim? Sure, Richard Reid traveling under his orignal name might pass.

      So now Western Muslim converts will be encouraged not to use assumed names. On the other hand, if Mohammed Ali decides to travel, he had better show up at the airport twelve hours in advance….

    51. Malvolio says:

      JK: Well since 9/12/2001 it would have remained completely static, right? If we go with “Arab” rather than “Muslim” (which I reject as an usable profiling category), we’re really talking about one terrorist attack (9/11).

      Clueless much? You really think there’s only been one terrorist attack in the last nine years?

      And yes, all the of the attackers in all of the attacks would have exactly fit the most naive, racist profile you could have made from the 9/11 attackers. Yes, there are a hundred million or so innocent men who would also fit the profile but on any given flight, there will only be one or two, so yes, why not detail some guy to have a friend chat, see what’s up?

      Especially if his dad has already mentioned that, yes, he’s a murderous lunatic.

    52. Preferred Customer says:

      Blue:
      Again, that is a far more difficult job from AQ or whoever than you make it out to be.

      Is it? I can think of three high profile AQ operatives who would not, on casual inspection, fit the ‘profile’ that people here seem to be pushing (male, young and Arab): Jose Padilla, Richard Reid and Umar Farouk Abdulmutallab. Even if we include “apparently Muslim” or “passported in a state with a large Muslim population” into the profile, we only grab one of the preceding people.

      Now, that isn’t to say that I think all profiling is counter-productive. Young men (or women) of any citizenship traveling alone with little or no luggage on flights to the United States might be a good category to take a harder look at, and certainly those people who’ve been identified to the USG as potential terrorists should be subject to higher scrutiny. But those are very different categories than the kind of broad-brush ethno-religious groups that have been discussed here, which are both under- and over-inclusive enough to make them unusable.

    53. yankee says:

      Blue: A convert Muslim, you mean? Western converts actually should be one of the profile categories, actually.

      How do you propose to have the TSA do this? Do the REAL ID’s we were all supposed to need starting Friday also include your religion and conversion history in addition to full legal name, signature, date of birth, gender, photograph, and address of principal residence?

    54. Chris Travers says:

      Smooth, like a Rhapsody: In a world with finite resources, time and patience, I would argue that “most” most certainly does “cut it”.

      Assuming we only look at plots against airliners bound for the US, you have one out of three not fitting your profile though. That isn’t high enough for my standards.

    55. JK says:

      Clueless much? You really think there’s only been one terrorist attack in the last nine years?

      Oh, so now we’re talking about Terrorism worldwide! And we’re no longer limited to Airline attacks! So on top of getting Timothy McVeigh, Terry Nichols, Ted Kazinski, the guy who shot up the Holocaust museum, and anti-abortion terrorists back in the “terrorist” set, we can also include Basque, Chechnyan, FARC (and other south american militant groups), IRA, etc, etc, etc terrorism? So by expanding the set you’ve added three attacks that fit your profile, but added hundreds (Thousands?) of attacks that don’t. That doesn’t seem to help your case.

    56. Guy says:

      JK:
      Oh, so now we’re talking about Terrorism worldwide! And we’re no longer limited to Airline attacks! So on top of getting Timothy McVeigh, Terry Nichols, Ted Kazinski, the guy who shot up the Holocaust museum, and anti-abortion terrorists back in the “terrorist” set, we can also include Basque, Chechnyan, FARC (and other south american militant groups), IRA, etc, etc, etc terrorism?So by expanding the set you’ve added three attacks that fit your profile, but added hundreds (Thousands?) of attacks that don’t.That doesn’t seem to help your case.

      I think it’s time to ask Eugene to institute a system of commenting estoppel.

    57. Mark Field says:

      They may be a large number in the world, but in the US and in the US transportation system they are a very small number.

      Do you have some numbers on this? I have no idea how you could, since “Muslim” isn’t an easily identifiable category. At a guess, I’d say tens of thousands per day still fit. Plus, as others have noted, there are easy ways to avoid the profile.

      All in all, profiling is a waste of time and money. It’s both overinclusive — the 20-odd “recent” hijackers represent about 1/200,000th of a percent of male Muslims in the world, which means we’d have to anger tens of thousands daily in order to catch a single hijacker (and probably create another hijacker in so doing) — and underinclusive, since a known profile will simply be avoided on the next hijack.

    58. pot meet kettle says:

      If all Arab Males between 18–35 suddenly went incorrigibly catatonic, the rate of terror attacks (on planes, trains, night clubs, what-have-you), could be expected to …

      actually, post 9/11, the rate of terror attacks on planes, at least, would be the same. i am not aware of any terror attacks on trains and night clubs in the u.s, so i will say the rate of terror attacks on those would be the same too.

    59. yankee says:

      Malvolio: Clueless much? You really think there’s only been one terrorist attack in the last nine years?

      I assumed JK was referring to terrorist attacks in the U.S., so the Madrid, London, and Mumbai attacks you reference wouldn’t be included. Of course, there have also been plenty of terrorist attacks in the U.S. since 9/12/01, including:

      The 2001 anthrax attacks, committed by American Roman Catholic Bruce Ivins.

      The 2002 Beltway sniper attacks, committed by American-born Muslim convert John Allen Muhammad and Jamaican Muslim convert(?) Lee Boyd Malvo.

      The 2006 UNC SUV attack, committed by Iranian-born American Mohammed Reza Taheri-azar, who was Muslim but seems not to have been particularly devout.

      The 2007 Virginia Tech massacre, committed by South Korean national Seung-Hui Cho, who moved to the U.S. at 8. Cho was raised as a nondenominational Christian but personally seems not to have had any coherent religious beliefs.

      The May 2009 assasination of George Tiller, committed by American Scott Roeder, an evangelical Christian.

      Last month’s Ford Hood attack, committed by American-born Muslim Nidal Malik Hasan.

    60. Chris Travers says:

      Mark Field: Do you have some numbers on this? I have no idea how you could, since “Muslim” isn’t an easily identifiable category. At a guess, I’d say tens of thousands per day still fit. Plus, as others have noted, there are easy ways to avoid the profile.

      I would also be willing to bet that Israel sees “Islam” as fitting a somewhat more narrow definition than any US response would. For example, iirc, under Israeli categories, Druze are not considered Muslims and there seems to be an open questions whether Sabbateans/Donmeh are Jews or Muslims under Israeli law.

    61. silvermine says:

      Yea, 5b is wrong. These rules are so strict I could not follow them safely at all. I have to choose not to fly.

      Look, I’m pregnant. I need to eat and drink and go to the bathroom. I’m also currently diabetic. I have to eat and drink and test my blood. I have 2 kids — can you make a 3 year old pee on command and then hold it for an hour? Yeah, me neither. After the baby is born — how do you keep an infant from needing to be held, change,d or fed for an hour? Short of drugging it into unconsciousness?

      People have to move around on a plane, or get “deep vein thrombosis” (oh, also more likely if you’re pregnant or diabetic, I’ll add).

      These requirements by TSA are not just annoying — they are *unhealthy* for a large portion of the population! So, it’s “just an hour”. Sorry, that doesn’t cut it. It’s theater of the absurd.

    62. Penguins Fan says:

      Connecticut Lawyer: Babe Ruth’s birthplace? Seriously? Isn’t that a little too much inside baseball?

      I’m a white native-born American female of Western European descent, coming up on 30. (Both sides of the family have been here since the 18th century.) I’ve been to a couple MLB games. However, my family are not (as commenter Hoosier once put it) Baseball-Americans. I’m lucky to be able to name the team that Babe Ruth played for in the majors. And, just for the record, I lack an interest in committing acts of terror.

      Also: asking a woman if she has a boyfriend and wanting to know where he is raises all kinds of hackles, particularly if the questioner is male. If personal and intrusive questions must be asked, something a little less loaded like questions about “What do you do?” and “Tell me about your family” should suffice.

      The boyfriend thing, BTW, is not just about not offending people’s delicate feelings. Men who ask out of the blue if you have a boyfriend are usually the kind of men who are about to hit on you in a clumsy way, generally in an inappropriate time or place. (In my experience, anyway.) Even if I managed to choke down a reflexive “That’s really none of your business” my demeanor would instantly shift to something that would probably be perceived as non-cooperative. i.e., face assuming a “What the hell is your problem” look, tone of voice sounding cold, body language becoming unfriendly. And tack on a “Where is he?” and I’m torn between wondering if the person asking thinks women shouldn’t travel alone and the activation of the “this person might be a physical danger to me” concerns.

      In case the previous paragraph was too long: let’s avoid false positives by not asking stupid questions.

    63. Malvolio says:

      yankee: I assumed JK was referring to terrorist attacks in the U.S., so the Madrid, London, and Mumbai attacks you reference wouldn’t be included. Of course, there have also been plenty of terrorist attacks in the U.S. since 9/12/01,

      I don’t know if Cho is really a terrorist, so much as a nutbag in the Dylan/Eric mold, and if you have any proof that Ivins is guilty, the FBI would like to talk to you.

      Even including them, that list shows that half of the US terrorist attacks are the work of the profiled group (young male Muslims), who represent 1% of the population. As Ann Coulter said, at some point it stops being a “profile” and starts being a “description of the suspect”.

      I’m not sure exactly how that helps us, especially considering the difficulty of eyeballing Boyd, for example, as a young male Muslim (and not just some black kid), but the first step is always admitting that you have a problem.

    64. Elfwreck says:

      yankee:
      Of course, there have also been plenty of terrorist attacks in the U.S. since 9/12/01, including:
      … committed by American Roman Catholic Bruce Ivins.
      … committed by American-born Muslim convert John Allen Muhammad and Jamaican Muslim convert(?) Lee Boyd Malvo.
      … committed by Iranian-born American Mohammed Reza Taheri-azar, who was Muslim …
      … committed by South Korean national Seung-Hui Cho, who moved to the U.S. at 8.Cho was raised as a nondenominational Christian….
      …committed by American Scott Roeder, an evangelical Christian.
      … committed by American-born Muslim Nidal Malik Hasan.

      Obviously, the problem is with monotheists. Screen the monotheists, and leave the atheists, pantheists & polytheists alone.

    65. jakecollins says:

      “It might be argued that we have had too few successful terrorist attacks over the last few years rather than too many.”

      Is Mr Posner suggesting that the lack of successful terrorism in America is a problem?
      Conservatives really do hate America for its freedoms.

    66. Guy says:

      Malvolio:
      I don’t know if Cho is really a terrorist, so much as a nutbag in the Dylan/Eric mold, and if you have any proof that Ivins is guilty, the FBI would like to talk to you.Even including them, that list shows that half of the US terrorist attacks are the work of the profiled group (young male Muslims), who represent 1% of the population.As Ann Coulter said, at some point it stops being a “profile” and starts being a “description of the suspect”.I’m not sure exactly how that helps us, especially considering the difficulty of eyeballing Boyd, for example, as a young male Muslim (and not just some black kid), but the first step is always admitting that you have a problem.

      Does it matter who “counts” as a terrorist or not? We’re trying to prevent deaths here, right? The only reason we’re trying to prevent terrorism is towards the goal of prevention of deaths.

    67. Preferred Customer says:

      Guy: y reason we’

      Well, now you are being entirely too rational. If we were really just concerned about preventing deaths, we probably could have prevented a lot more if we’d ignored terrorism altogether and spent all of the resources in the War on Terror on things like medical research or eradicating diseases like malaria.

      Which points to another problem with treating a tactic like “terrorism” as an enemy that must be defeated–we have to spend a lot of time arguing about whether certain actions/actors are really terrorism/terrorists in order to understand whether our war is being won or lost.

    68. Malvolio says:

      Guy: Does it matter who “counts” as a terrorist or not? We’re trying to prevent deaths here, right? The only reason we’re trying to prevent terrorism is towards the goal of prevention of deaths.

      What? No!

      The number of deaths that has been, or even could be, caused by terrorism is negligible. Breast cancer has killed more American men in the last 10 years than terrorism and ask yourself how scared the average man is of breast cancer.

      Terrorism resembles war: it is an act of violence intended to bend the victim towards the terrorist’s will. We are trying to prevent terrorism so that the US can remain a free and independent country, not so that everybody gets to live forever.

    69. yankee says:

      Malvolio: Even including them, that list shows that half of the US terrorist attacks are the work of the profiled group (young male Muslims), who represent 1% of the population.

      Well, “half” isn’t quite right since I wasn’t trying to provide an exhaustive list, and you are right that it’s not exactly easy to distinguish between “terrorists” and “ordinary” mass murderers. I won’t dispute that however you define it terrorism in the U.S. has been disproportionately committed by young Muslim men (and even more disproportionately by male monotheists, who committed 100% of the attacks in question).

      I’m still confused, though, as to how the TSA is supposed to profile Muslims, short of requiring all Muslims to wear little yellow crescents.

      Malvolio: As Ann Coulter said, at some point it stops being a “profile” and starts being a “description of the suspect”.

      Um, no. You can’t have a suspect until a crime has been committed.

    70. The Honest Services of Jeff Skilling and the TSA « cara ellison says:

      [...] Honest Services of Jeff Skilling and the TSA Jump to Comments Our Glorious TSA, a post on Volokh Conspiracy, is a great read, and there was a truly intriguing comment: If I were [...]

    71. SG says:

      I’m still confused, though, as to how the TSA is supposed to profile Muslims, short of requiring all Muslims to wear little yellow crescents.

      I say you offer everyone a (complementary) pork rind on their way through security and give heightened screening to anyone who refuses to eat it. Sure, you’ll get some false positives on vegetarians, but I don’t trust them much either…

    72. Chris Travers says:

      SG: I say you offer everyone a (complementary) pork rind on their way through security and give heightened screening to anyone who refuses to eat it. Sure, you’ll get some false positives on vegetarians, but I don’t trust them much either…

      Just another conspiracy to isolate and harrass the JEWS, right? ;-)

    73. Chris Travers says:

      Malvolio: Terrorism resembles war: it is an act of violence intended to bend the victim towards the terrorist’s will. We are trying to prevent terrorism so that the US can remain a free and independent country, not so that everybody gets to live forever.

      It would be a supreme irony if we throw out our privacy and hence liberty in order to remain a free country.

    74. Tatil says:

      Profiling is an effective strategy when, as here, terrorists come from a small group of (relatively) easily identifiable people.

      Effective? Would that catch the shoebomber Reid or any US citizen who converted to Islam? I doubt it. There is also the cost. Tourism is a big earner in many places. Ruling out more than a billion potential visitors is not going to help them any. Besides, if we decide not to grant visas to any Muslims, how are we going to replace the engineers in many tech companies? Given our declining manufacturing base, services and high tech is pretty much what we have left. Wouldn’t we rather these peaceful people work for and contribute US economy instead of work for competing foreign ones? Then of course there is the possibility of Muslim countries retaliating. More “bigoted” than “effective” is how I would describe it.

    75. Chris Travers says:

      Fen: 6) Static defenses are permeable. You cannot be strong in all places at all times. Attacks will still get through. Therefore, you must play offense and act pre-emptively.

      Depending on how I read this, you are exactly right or exactly wrong. Perhaps this misunderstands the purpose of airport security. Once the terrorist is properly prepared and at the security check point it is too late to avert an attack. What you CAN hope to do is require enough preparation that law enforcement and or other means can pick up and disrupt plots before they get to that phase.

      Though perhaps this is what you mean by playing offence and acting pre-emptively.

    76. Bob from Ohio says:

      Besides, if we decide not to grant visas to any Muslims, how are we going to replace the engineers in many tech companies?

      Are there actually large numbers of such people? All the articles I have found say Chinese, Indians and Europeans form this group. While an Indian can be a Muslim, most are not.

      As for tourism, the percentage of those billion Muslims who come to the US is only a small fraction.

      I don’t think economics is much of an argument against travel bans and/or profiling.

    77. pot meet kettle says:

      Effective? Would that catch the shoebomber Reid

      of course, just add jamaican origin people to the list, mon.

    78. David Sanger says:

      Apparently the TSA has become so concerned about publication of Security Directive SD 1544-09-06 that it has served a personal subpoena on a well-known travel blogger, Christopher Elliot, demanding to how he obtained a copy.

      Full text of my subpoena from the Department of Homeland Security

      At least one other travel blogger, Steven Frischling, posted the text of the original SD plus a later revision, and says he received it directly from the TSA.

      “If this document was intended to be secret why was a copy sent to me? ”

    79. Mike Licht says:

      TSA moved with commendable speed after the Christmas incident. For the new airport security rules, look here:

      http://notionscapital.wordpress.com/2009/12/30/flying-home/