Profiling, Political Correctness, and Airport Security

Mark of New Jersey

Mark is a Founding Editor of The League of Ordinary Gentlemen, the predecessor of Ordinary Times.

Related Post Roulette

63 Responses

  1. John Howard Griffin says:

    Imagine for a moment that the government was in charge of airport screening in the pre-911 world.

    Wouldn’t E.D. use that as an additional argument about why a private company would be better – after all, private security companies didn’t let 911 happen?Report

  2. gregiank says:

    Good reply Mark. It’s always depressing how quickly many people deride concerns about profiling and basic fairness as “political correctness.” This countries history with race is far to mixed, to put it gently, to not be concerned with treating all groups fairly.

    Regarding the entire, very silly, just privatize it argument. One thing not mentioned it that the people who go into private security business are people who used to be in military or police work. That makes sense, they have experience with providing high level, serious security. Well they have the same background and world view as the people currently in gov. Thinking that some private contractor is going to have some radically new vision is just not likely. I’m thinking that when anybody just throws out the “just privatize” line they should be required to also present a business plan to show how a business makes money by doing such and such. How many private companies have a few th0usand metal detectors and tens of thousands of trained security people just hanging around?Report

  3. Trumwill says:

    One of the biggest arguments against the Israeli system is that it’s extraordinarily expensive. Israel can do it because they have a limited number of ports-of-entry, meanwhile for this to be effective we would have to do it at every airport in the country (lest the go through security in Toledo to hijack or blow up a plane headed from Chicago to NYC). I haven’t heard a great counterargument to this.

    Regarding “they’ll just get Albanians/Africans/etc”, the profiling wouldn’t just be of Arabs. It would be arabs and foreigners more generally. I don’t find the “it wouldn’t be effective!” argument to be very compelling. It would almost certainly be more effective. But, literally and more important figuratively, at what price?Report

    • Mark Thompson in reply to Trumwill says:

      Regarding “they’ll just get Albanians/Africans/etc”, the profiling wouldn’t just be of Arabs. It would be arabs and foreigners more generally. .

      So then they’ll just another Adam Ghadan or Jose Padilla. Point being that given the size (small compared to Israel and smaller in proportion to the total amount of air travel in and out of this country) and nature (relatively broad-based in terms of the pool of potential terrorists) of the threat we face, a profiling algorithm would be extraordinarily easy to beat. Added on top of existing procedures, it may provide a very marginal increase in security, but if combined with any kind of a reduction in the existing procedures (ineffective and inefficient as they are), it would probably be a net decrease in security.Report

      • Trumwill in reply to Mark Thompson says:

        I think that Americans are harder to recruit for suicide bombers than foreigners in general. I think if you strike at who they can recruit, you’ve actually made a good deal of headway. And you don’t need a Ghadan. You need a Ghadan who has been careful not to attract attention. You need a Padilla who is smarter than Padilla was. There is no full-proof method, but I think this would be a pretty effective one.

        But again, we have the issue of costs, literal and figurative. That’s the main reason for my opposition to it. I think it’s a little easy to fall into the trap where “I don’t like it, therefore it is not effective” when we wouldn’t like it even if it were effective.

        Of course, from a political perspective, attacking the efficacy makes sense since I don’t have a huge amount of faith in my countrymen to care about costs (just a few more dollars to the national debt and it’s better they get searched than we and all that). But realistically, I just don’t think it’s true.Report

    • Boonton in reply to Trumwill says:

      Israel IMO may not be the best case here. How much of Israel’s success at its airports is due to its tight security and how much is due to the fact that Israel is a lot easier to hit on the ground either with suicide bombers or rockets from Hammas? How much is due to the fact that Israel has other places that can be hit more easily by Palestinians living more or less within its occupied terrorities whereas the US requires its terrorists to either travel into the US from abroad or recruit the odd domestic terrorist.Report

      • Trumwill in reply to Boonton says:

        You strike on a good, broader point. Since they can’t really hijack planes anymore (reinforced doors, non-compliant passengers), there are worse things they could do to us than blow up a plane. They could bomb a levy. They could do ticky-tack things like blowing up a bus here and a shopping center there. The DC Snipers killed a dozen or so people and had an entire city living in abject fear.

        This is another reason why I think it’s so ridiculous to focus so intently on air travel to the point of striking at civil liberties. Everything is a risk and flying less so. The real 9/11 coup was what they did with the planes. That we can stop.Report

        • Mark Thompson in reply to Trumwill says:

          I agree with this entirely, but just wanted to re-emphasize a point that seems to get lost in the shuffle far too often: they could also just attack an airport security checkpoint. The longer those lines are, the more inviting a target they make.Report

  4. Boonton says:

    OK let’s stop the silliness for a moment here.

    First, profiling already happens. I don’t doubt for a moment that a 20 yr old male travelling on a student visa with stamps in his passport from Yeman or Pakistan isn’t more likely to get selected for more intense screening than, say, your hypothetical 80 yr old grandmother. There are ‘no fly’ lists and ‘fly but scrutinize’ lists as well.

    Second, you clearly haven’t thought about Type I and Type II errors. That’s ok, few of us like to recall college statistics but they provide a useful way of thinking about things here.

    Imagine you’re driving back from Mexico into the US. At the traffic jam near the crossing, a police officer walks with a drug sniffing dog. Every now and then the dog barks and that car is searched for drugs. The dog is almost always right, he never barks at a car that drugs are not found. This is an example of keeping a Type I error low.

    Type I error – You say the guy isn’t ‘normal’ when in fact he is. The ‘perfect’ dog never barks at an innocent car.

    In fact, I recall once reading about a search getting tossed out because the dog in question had barked nearly 5 times out of ten when there were no drugs present. The court ruled that the poor dog was no better than flipping a coin. But maybe the dog was just very good at eliminating the other type of error:

    Type II error – The guy isn’t ‘normal’ but you fail to bark.

    See you’re waiting to cross the border in a line of cars and the dog barks at some of the cars. But you probably know some cars do have drugs but the dog doesn’t bark because they did a good job of wrapping them up so well the dog is unable to smell them.

    So maybe that ‘bad dog’ who was no better than a coin toss was actually a great dog for Type II errors. Imagine out of 100 cars 10 have drugs. The ‘perfect dog’ barks at 5 cars and they always have drugs. So people say he’s 100%. But he isn’t in the Type II sense. He’s letting 1 out of every 2 smugglers go. Say the ‘bad dog’ barks at 20 out of 100 cars and 10 have drugs. He’s catching 100% of the drug smuggling, but the courts dismiss his barks because 1 out of 2 barks is an innocent person.

    Now hopefully you see the problem with ‘profiling’. It quite frankly doesn’t work. We are willing to live with the ‘good dog’ who let’s most of the drug smugglers get by provided he never harms an innocent person. We aren’t willing to play this way with the guy who wants to take out a plane. The argument for ‘more profiling’ is built an a faulty premise, that we need to lower the Type I errors. But it ignores the fact that in this context Type II errors are very, very important.

    How would ‘profiling’ do with Type II errors? Well look at Google. Here is a company that probably has thousands of data points about me, many very intimate. It knows my searches, clicks, even reads my emails. Also Google has some of the world’s smartest people who have built the most sophisticated data mining algorithms ever created all under the intense profit motive of the market to improve their search results. Yet for all that they have yet to serve me a ‘pay per click’ ad for anything I wanted to buy.

    Now with this real fact in mind, kindly explain to me how the gov’t with a lot fewer data points is going to construct a ‘profiling algorithm’ that isn’t going to have a huge Type II error problem?Report

    • MFarmer in reply to Boonton says:

      The silliness is to think you can find something when you aren’t looking for the characteristics of what you seek to find. It’s not the end all, be all, but in a security situtation, personell should be trained to notice certain characteristics of the people who pose a risk. If the pressure is to ignore all forms of profiling, and this is strictly enforced so that you could lose your job, then security officers will purposefully avoid making these judgements. Maybe I’m wrong, but making these common sense observations appears to be vital to good security. I just think that if trained security officers are alert and observant, strange behavior and the appearance of risk will be detected, and that this is good enough. If trained terrorists want to blow something up, they will, but what are the chances of continuous terrorist attacks in America? I’ve always said that if terrorism really succeeds in America, it fails — the reality, for good or bad, is that with our military capabilities, if we ever become truly terrorized, like worried about our very existence in an immediate way, the demand to destroy whatever moves funny in the Middle east will be so great, and our military response will be so devastating, that no terrorist group or country wants to push us to that point. If terrorists wanted to really terrorize us, they could do it at any time, but what would they gain? It’s best for them to disrupt our way of life and keep us on edge, and slowly bleed us of what’s important to our way of life. I refuse to be on edge.Report

      • Boonton in reply to MFarmer says:

        I don’t disagree but advocates of profiling tend to frame it as an either-or problem. As in “granny has to get body scanned because those silly PC liberals won’t let us ‘profile’ Ahmed from Yeman”. I totally agree if a security officer notices that something seems ‘odd’ about someone he should have the training and the perogrative to probe a bit more deeply. I haven’t seen any evidence, though, that this is not already the case.

        Here, though, you see how the Israel model starts to break down when scaled up. Being a small area with a limited amount of diversity in the travelling public Israel is able to get its officials really well versed and highly skilled. The US has a much larger ground to cover and a lot of diversity in its population. That type of skill will be harder to produce on a mass scale which, IMO, increases the danger that terrorists will game the system by simply looking like sterotypical Americans. For example, some of the 9/11 hijackers were reported to have hung out at gyms and even hit strip clubs and drank while never bothering to attend any US mosques for prayer services. Would a ‘profiling system’ be easily gamed by having the terrorist act like a college student cutting loose on spring break while the TSA officer interrogates the devoute Muslim who carries a Koran around?Report

        • Mark Thompson in reply to Boonton says:

          I’m a bit confused – wasn’t this exactly my point above? I might not have specifically referenced Type I and Type II errors, but I thought one of my points was that given the nature of the threat faced by the US as compared to Israel, profiling based solely on data points would be remarkably ineffective. But the key part of the Israeli model, as I understand it at least, is that all passengers get intensely interviewed by a security agent, not just the ones whose data points fit a particular profile. This, of course, is entirely impracticable in the US.Report

          • Boonton in reply to Mark Thompson says:

            That, then, would sound like a system that would minimize Type II errors. I’m not sure how libertarian minded Americans would take to a gov’t employee intensively questioning them about family relationships, their jobs, politics etc. I suspect you’d start getting Type I errors where angry Americans start giving ‘smart ass’ answers which then results in them being flagged as ‘suspects’ that need more intense scrutiny.

            The Type II would still be a problem IMO. I doubt gov’t questioners could get so good that they could more often than not snuff out the terrorist.

            I think the back scanners would start to look a lot better.Report

            • Mark Thompson in reply to Boonton says:

              More Type I errors than the existing system? I’m not so sure about that. But, regardless, my point here is basically the same as yours: profiling wouldn’t work in the US.

              On Type II errors, I don’t disagree (at least as it would be applied in the US), but that’s largely the other underlying point in this debate, which is simply that there is no way to completely – or necessarily even mostly – eliminate Type II errors significantly more than just luggage x-rays and maybe bomb-sniffing dogs would accomplish. You’re looking for a single needle in a very, very large haystack, except that needle is sentient enough to recognize how you’re looking for it and thus has the opportunity to disguise itself pretty well. By comparison, Israel is dealing with a vastly smaller haystack and a higher number of needles with less ability to disguise themselves.

              So, given that, the best that can be achieved is to limit the Type I errors (within the confines of restrictions such as the Constitution, budgets, and economic costs). One could argue – and indeed, many opponents of the backscatter machines do effectively argue – that the backscatter machines combined with the pat-downs turn every innocent party who goes through them into a Type I error.Report

              • Boonton in reply to Mark Thompson says:

                On Type II errors, I don’t disagree (at least as it would be applied in the US), but that’s largely the other underlying point in this debate, which is simply that there is no way to completely – or necessarily even mostly – eliminate Type II errors significantly more than just luggage x-rays and maybe bomb-sniffing dogs would accomplish.

                Agreed, we are basically at a level where Type II errors are nearly zero. It is, IMO, very hard if not impossible to get a bomb or serious weapon aboard a large commercial plane when the current system (backscanners and all) are fully implemented. I’ll say nearly zero because there’s human error, corruption, the possibility of mind bending ‘Oceans 11’ type operations and the possibility that you could produce an undetectible bomb if you had the help of the KGB, CIA, or James Bond’s support system. That aside we are at a nearly zero level Type II error rate but a very high Type I error rate.

                Adding profiling on top of the system, then, doesn’t help much. There’s not much Type II error left to get rid of and we already have more than enough Type I error. If you want to replace the system with profiling then you will lower Type I errors but increase Type II errors, probably dramatically as I don’t buy that any amount of human intelligent, data mining or ‘smart profiling’ can get close to the near zero Type II error rate you get when you basically search everyone.

                Rather than adding Type I errors (having ‘profile empowered TSA agents’ grill you about your religious beliefs and try to determine by ‘hunch’ if you’re a terrorist), I think a more sensible strategy would be to keep the ‘zero Type II error’ model and work on making it less intrusive. Ensure the radiation from the scanners is lower than what you get from flying on a plane to begin with. Use software to analyze the ‘nudie images’ only giving it to TSA agents if something looks amiss. The pat downs for those opting out would probably have to continue. Sorry but we had a guy who smuggled a non-trivial bomb on board a plane in his crotch. I’m not going to bet my life on our ability to ‘data mine’ out just the right list of people for intensive pat downs.

                One could argue – and indeed, many opponents of the backscatter machines do effectively argue – that the backscatter machines combined with the pat-downs turn every innocent party who goes through them into a Type I error.

                They do to an extent, I suppose you could also argue that the drug sniffing dog walking up and down the line does the same. After all, whose to say you couldn’t create an ‘electronic nose’ that would do the same thing the dog does? Whose to say it couldn’t be modified to detect, say, if you had sex in the last 48 hours?

                But here’s the thing, there are times you live with a high Type II error rate and times you don’t. In the case of speeders on a highway, or drug smugglers you live with a high type II rate. The cop understands he will not pull over every speeder and so does the public. The public wants to make sure the cop, when he does pull someone over, is pulling over someone guilty of speeding and not someone innocent. Hence all the hub bub over whether radar guns are ‘calibrated’ and the officer properly trained etc. In the case of bombs on a plane, though, I don’t want to live with a high Type II error and neither do most others.Report

        • MFarmer in reply to Boonton says:

          Well, to a certain extent, political correctness has hampered a common sense approach to profiling — witness some of the responses to E.D.’s post — the word “profile” automatically brings out cries of Muslim-bashing. Yes, it’s more difficult to find the anamolies in America because we are so diverse, but I think what we can can find will be sufficient, and that groping everyone or getting nude photos are unnecessary. I think we have to go back to the problem and why we are doing these things to start with — the response is out of proportion to the problem.Report

          • Mark Thompson in reply to MFarmer says:

            I think it’s a very limited extent, though. I agree that the reflexive response from some quarters that “profiling” is indistinguishable from “discrimination against Muslims” completely misses the boat (of course, people like McCarthy don’t help in that regard), but ultimately I’ve got little doubt that the overwhelming majority of Americans would be perfectly comfortable – at least in concept – with a profiling system; in fact, this Gallup poll put the number at 71% in favor of an explicitly ethnicity-based profiling system – and that was long before the backscatter machines and patdowns began: http://www.usatoday.com/news/nation/2010-01-12-poll-terrorism-obama_N.htm . It’s probably safe to assume that an even-larger majority would support a non-ethnicity-based profiling system. Or, put another way, profiling is almost certainly even more popular than the backscatter machines allegedly are.

            Such a system would of course almost certainly violate the Equal Protection Clause, which is one reason it hasn’t been attempted. But Constitutional limitations didn’t stop Bush, and Obama hasn’t shown that he’s terribly concerned with Constitutional limitations either. That tells me that the primary reasons profiling hasn’t been seriously discussed are something other than political correctness.Report

            • Trumwill in reply to Mark Thompson says:

              It’s probably safe to assume that an even-larger majority would support a non-ethnicity-based profiling system.

              I don’t think so, actually.

              I am a smoker and by extension a serial loiterer, standing around smoking in public places. Periodically, I draw the attention of the cops who want to know what I am doing wherever I am and, even when I am in a perfectly legal place, telling me to move on. I think to them I look like a drug dealer or someone looking to buy drugs or something. They don’t arrest me or anything, but they do ask to see my ID, ask me a few questions, and tell me to move along now.

              When I mention this to my friends, they are often quite incensed. I was, too, the first time or two it happened. But I think underlying this outrage is that it is obvious that people like me should never be questioned in such a manner. Some of the people outraged on my behalf are also people that think that a wide interpretation of the Arizona law, that cops should be able to simply ask a few questions if they think the guy might be here illegally (which means subjecting Hispanics to the same sort of thing I was subjected to) is a-ok.

              My point with all of this is that people have a much lower tolerance for any sort of profiling (such as picking on the white guy in the park who is just standing there and acting like somebody looking to buy drugs) that might infringe on their freedom (to do things like stand in the park and be left alone).

              This is one of my huge problems with the Arizona law and racial profiling. When people know or think that they won’t be affected, they tend to minimize the inconvenience of being profiled. “Come on, he only asked a few questions and to see your ID.” Inconveniencing everybody is actually a way to make sure that nobody is inconvenienced too harshly without backlash.Report

            • Boonton in reply to Mark Thompson says:

              Gallup poll put the number at 71% in favor of an explicitly ethnicity-based profiling system

              To what end though? Right now the system makes a Type II error very, very low. It’s very difficult to get a bomb on board a plane or a weapon serious enough to plausibly hijack it. If you put some ‘ethnicity based’ profiling on top of that what would be accomplished being that you’ve already gotten very tight on Type II errors other than inflating what everyone already knows is a very high Type I error rate (that is bothering people who are totally innocent again).

              I think that poll reveals most Americans have no problem lowering the Type II error rate even more if there’s not cost to them in terms of Type I (white, non-Muslim will likely exclude one from the ‘profile’). But since Type II errors are already very low why buy even less Type II at the cost of yet more Type I? Supposedly the whole ‘backscanner revolt’ is because Type I errors are too high?

              If you replaced the current system with a system where only, say, Arabs got scanned you’d almost certainly get a Type II error as a terrorist who doesn’t fit the profile walks thru with a bomb or weapon.

              Keep in mind the problem with profiling is that you won’t get a profile until you get a successful terrorist attack. In fact, the more successful attacks the better your profile gets. Israel got burned by this when for a long period there was almost never a female suicide bomber. The result was that the ‘profile’ began with ‘male’ and females did not get much scrutiny. At some point, though, female suicide bombers started appearing. At the end Israel really hasn’t gone with ‘profiling’ but has gone the route of ‘screw Type I errors, close off the entire border so we get no more Type IIs’.Report

              • MadRocketScientist in reply to Boonton says:

                The problem is that the previous system worked very well to prevent Type II errors, so adding the extra security takes us past the point of diminishing returns (and falls into the realm of “looking like we are doing something”). I mean, when was the last time a bomb detonated on a plane? That was the security level that was probably most effective.

                The bad guys don’t even have to be successful anymore, they just have to make a freshman effort. Bruce Schneier highlights how easy it is for them to bleed us on the cheap.Report

              • “The bad guys don’t even have to be successful anymore, they just have to make a freshman effort. Bruce Schneier highlights how easy it is for them to bleed us on the cheap.”

                This makes me think of that line from Citizen Kane where Kane’s newspaper is losing money and his advisor is bugging him. What does he say? Something like “yea I’ll loose a million this year, I’ll loose another million next year….at this rate I’ll be broke….in sixty years!”

                Don’t be fooled. The US isn’t going broke and Al Qaeda needs actual victories. To us scanners are a few extra dollars on an airline ticket. To them losing donations means they don’t eat, maybe even have to get real jobs (and many can’t easily just return home, Saudi Arabia and Egypt isn’t going to greet them with open arms). Their promises of a worldwide Islamic revolt against the west and corrupt Arab states hasn’t happened. Even radicals are questioning the morality of harming civilians. And the fumbling attempts indicate their organization has been very badly disrupted by targetted killings and making Afghanistan a place that is no longer so easy to work out of. That a crotch bomb didn’t bring down a plane, that UPS packages didn’t destroy some Jewish centers and that a Tim McViegh style explosion didn’ t happen in Times Square are all serious blows to Al Qaeda. Terrorist attacks are high profile which is why the strategy to minimize Type II errors rather than the more common Type I makes sense.Report

              • DensityDuck in reply to Boonton says:

                “The US isn’t going broke…[t]o us scanners are a few extra dollars on an airline ticket.”

                Exactly. “Oh, we’ve spent hundreds of millions of dollars on these scanners!” Yeah? How many hundreds of BILLIONS did the stock market lose on 9/11?

                Complaining about the cost of security is like complaining about the cost of polio vaccinations.Report

              • The thing is that 9/11, and particularly its economic effects, was a unique event that cannot be repeated since it relied entirely on passive passengers; after 9/11 it’s pretty clear that passengers will use their superior numbers to fight back and prevent the hijacking, even if that means the plane crashes in the process. The threat to aviation now is thus limited to bombings. While that threat should not be trivialized, it’s not a threat that would have the same kind of fallout as 9/11 had. There’s also no evidence that the security measures imposed since 9/11 (beyond perhaps better scanning of checked luggage) actually reduce the threat of bombings – keep in mind that both attempted bombings of aircraft since then failed not because of more intensive screening at the airports, but because of alert passengers. By its very nature, this screening requires personnel to look only for known threats based on a well-publicized list of contraband (and ultimately, there’s no way to effectively screen for contraband that can be hidden in body cavities). The terrorist just needs to figure out one of the infinite number of unknown threats and attempt that. And, again, this says nothing of the question of whether increased screening measures actually causes more deaths than it could even hypothetically prevent given that it creates a new potential “soft” target in the form of extraordinarily long and densely-packed security lines and given that it also encourages people to choose a far more dangerous, on average, form of transportation: driving.Report

              • Boonton in reply to DensityDuck says:

                I don’t think a hijacking is impossible today simply because of passenger resistence. Yes it couldn’t happen in the manner the 9/11 ones did but it could still happen….if you had multiple hijackers who had the ability to get serious weapons on board (like guns) and if they found a way to storm the cockpit first and then barricade themselves against revolting passengers it might still work.

                In terms of bombings, well we do know a few things. First the searching and scanning of checked baggage has more or less eliminated the problem of the bombing a plane remotely. The Lockerbie bombing, for example, did not require a suicide bomber. Today you probably would which makes it tougher (not impossible but tougher) for a terrorist to pull it off. The shoe bomber and underwear bomber both got on board the plane outside the US yet inside the US the TSA has never caught a shoe or underwear bomber. This has two possible explanations:

                1. Shoe/underwear bombers have gotten by the TSA but for whatever reason opted not to set their bombs off.

                2. Bombers have opted not to try to get bombs thru the TSA.

                I don’t think #1 is very likely so #2 comes down to:

                A. Wannabe bombers think the odds of getting a bomb on a plane are too low to try.

                B. There’s a lack of bombers in the US so to date no one has tried so there’s no terrorists for the TSA to catch.

                By its very nature, this screening requires personnel to look only for known threats based on a well-publicized list of contraband (and ultimately, there’s no way to effectively screen for contraband that can be hidden in body cavities). The terrorist just needs to figure out one of the infinite number of unknown threats and attempt that.

                That’s a pretty big use of the word “just”. I’m sure the KGB or James Bond’s support team could get a bomb on a plane thru the TSA (think of that ‘red light greed light’ gum from the first Mission Impossible movie). But given the scans and the limits on fluids the number of plausible bombs that are powerful enough to really do the job & sneak thru in a ‘body cavity’ is a lot smaller than ‘infinite’. Just like the cold, though, you address this with layers of protection. At this point I think we are driving the odds of a successful plane based attack pretty low *but* the fact remains that airlines, for multiple reasons, remain very tempting terrorist targets.Report

              • MadRocketScientist in reply to Boonton says:

                While the cost of a scanner is relatively small, the cost of lost ticket sales because people are unwilling to travel by air unless they have to is considerably greater.Report

              • MadRocketScientist in reply to Boonton says:

                ALso, as this video shows, when you hire untrained amateurs to do your security, you get amateur security.Report

              • MadRocketScientist in reply to MadRocketScientist says:

                Except we’ve been able to successfully keep guns off of planes for decades now, using old fashioned security techniques. I think it’d be an interesting exercise to see if 30 people could smuggle guns through security and all get on the same flight.

                Note that none of the passenger screening touches on the fact that an airport has tons of employees who never go through security screening, and enjoy background checks less in-depth than the one I had when I got my CPL. If I was a terrorist, I’d get one guy inside the airport and have him smuggle the guns in to my 20 guys who will board the plane. Why even bother with getting things through security?

                As for what happens when you have 20-30 armed men on a plane? Well, unless they are all highly trained and schooled in weapons tactics & discipline onboard an aircraft, they are just as likely to shoot each other as they are to kill passengers. They are highly unlikely to get into the cockpit without heavy power tools (although if they have rifle caliber weapons, they might be able to shoot through the cockpit door & kill the pilots).

                So in the age of secured cockpits, armed pilots, air marshals, and actively resisting passengers, the only way a terrorist can use a plane is to blow it up (bomb) or cause it to crash uncontrollably (which is bad, but not as bad as ramming a skyscraper or having hostages to make demands with).

                Even bombing a plane is remarkably difficult unless you can get a significant amount of explosive into the right spot (the PanAm flight was destroyed not because the bomb was big, but because it happened to be put in a good spot (by accident, I believe). There are very few places passengers can get to where a small bomb can do enough damage to down an airliner.Report

              • What if you targetted the cockpit early one when the plane is on the ground? If your goal is ramming then being able to take off with the plane (with or without passengers) would work just fine.

                As for bombs I think it would depend on the size of the plane. Smaller planes have been taken out by oxygen tanks exploding in the cargo hold. On the hops from Florida to NY that I’ve been on, I would be a bit surprised that an explosive as potent as a hand grenade wouldn’t be enough to bring the plane down….or at least put it in serious danger of coming down.

                But I think your post does show how the system has worked in a sense. Taking a plane hostage or blowing it up requires serious firepower. While it may be possible to get weapons and bombs thru the TSA system, to get enough thru you need a lot of things to work just right for you. For example, spreading the explosive out among multiple terrorists raises the chances of at least one getting snagged in the TSA lines which ends the whole plot (it also requires more coordination and more recruiting of trustworthy suicide bombers….that’s hard when you have to smuggle your instructions out on cassette tapes from a hole in northern Pakistan because using cell phones puts you in the sights of the drones).Report

              • Boonton, if I were a terrorist, I am not sure I would really like my odds of being able to take a grounded plane to air and avoid scrutiny as long as required to ram my target. There’s a lot of communication involved in getting the plane taken off without a torrent of scrutiny. The 9/11 terrorists were able to do what they did in large part because they were able to avoid detection until they were in the air. An empty plane stands a chance of being shot down. A full plane stands a chance of the same (though there would be more hesitation) and non-compliant passengers.

                For it to work, you would not only need to be able to get into the cockpit, but you would have to be able to impersonate the pilots and bluff your way through an air control that’s less likely to be caught flat-footed than in the past. Not impossible, but requiring a lot of things to go your way.

                I don’t doubt that they could blow another plane up with or without the new TSA regs. I am skeptical that even with pre-9/11 security measures that they could pull off another 9/11. If I were a terrorist, I would probably be looking into packing a private plane in a private airfield somewhere near the target with explosives.Report

              • Well taking off from Newark NJ you’re almost immediately within reach of the major buildings of NYC as well as NJ. Unless the USAF has fighters in the area immediately you could probably make a dash for, say, the Empire State Building. I’m also assuming the terrorist would just take the plane off, not wait to get clearance from ground control. Perhaps a more plausible attack would come from taking several smaller planes and using them to ram incoming larger planes.

                I’ve always wondered about the huge UPS and Fed Ex planes I see. They seem to me to be good candidates for a redo of 9/11. Don’t get me wrong, I consider a redo of 9/11 to be very low. It would require a real ‘Mission Impossible’ type operation combined with a lot of luck for the terrorists. I think this is why many of the newer terrorists they catch are planning Tim McVeigh type operations instead.Report

              • MadRocketScientist in reply to MadRocketScientist says:

                The problem with storming the cockpit while at the gate (the only time the cockpit is not sealed), is that planes don’t have a reverse gear. This means a terrorist would have to take over the cockpit with no one noticing, and then convince the ground crew to remove the wheel chocks & hook up the tractor for a push back. If he got onto the taxi-way, he’d still have to bluff ground control & air control, or make a run for the runway. If he does “run for it”, he’ll find that is not as easy as it seems, especially at a busy airport (ground vehicles, other planes, etc.). Also, disabling an airliner on the ground is pretty easy (attack the landing gear, or get something big & hard enough into one engine; you need all your engines to take-off).

                Hijacking an airplane these days is about as likely as an aircraft falling out of the sky. Sure, it can happen, but sooooo many things have to go wrong (or right, depending on your view) before the end result is achieved.

                Oxygen bottles actually do represent a healthy amount of explosive, but even oxygen bottles are no guarantee of downing a plane. Again, they have to be in the right spot, and they have to detonate in a specific way, or with enough force, to down an aircraft.

                Cargo planes are even harder to hijack, because they don’t take passengers, so you have to either sneak into the cargo facility, take over a plane, and still get ground crew to work for you. Even if you get hired at the place, you still have the same problems. Sure, it can be done, but again, lots of failure points along the way.

                Ramming an airliner with a small plane is actually a lot more realistic, but you’ll have to do it on, or near, the ground, or the speed & size of the airliner will make it near to impossible to get a small plane close enough for impact.

                The greatest threat to airliners is a bad guy with a couple of man portable surface to air missles parked near the end of the runway. He wouldn’t get too many shots off before the police got to him, but he might get one or two.Report

          • James Vonder Haar in reply to MFarmer says:

            Given the quarters from which the desire for profiling is coming, color me skeptical that an actually implemented profiling program really will be blind to race.Report

            • But, if watchdog groups discovered that race was the only criteria, someone would be in big trouble, so even if they wanted to be racist, they couldn’t get by with it — you know they would be watched like hawks and would have to justify every singling out for questioning, which is a good thing.Report

              • Boonton in reply to MFarmer says:

                Again I’ll use Google as a comparison. A company with thousands of intimate data points about me, the most sophisticated algorithms ever developed and the best minds in the world still can’t serve me a Pay Per Click ad that I’m interested in. What reason is there to think that the gov’t will do better with much less brains, much less experience and a lot less data points?Report

              • MFarmer in reply to Boonton says:

                Once again, I’m totally wrong. Damn, you’d think I’d get something right every now and then.

                Not to be argumentative, but I had good success with Google adwords and pay-per-click,Report

              • My point is that the majority of advocates who are in favor of profiling do not have an issue with racial profiling. If they were ever in a position to implement their preferred policy, it is unlikely that the backlash from the profiling being in fact racist would be enough to dislodge the policy, since most of their proponents see no issue with it.Report

            • Heidegger in reply to James Vonder Haar says:

              Why should a profiling program be blind to race? For that matter, how could it be possibly effective if it was? Hypothetically, if multiple witnesses saw a 20-something African American exit a house with a woman’s head in one hand, should they scour the neighborhood looking for Eskimos, er, I mean, Inuits? (I’m also assuming the encounter between the perpetrator and the woman was not consensual.) I’m thinking of that incident in Germany when a man answered an honest to goodness cannibal ad, where indeed, he was going to be the main course. Sounds scrumptous. In any case, HE was going to be the eaten, not the eatee. And wouldn’t you know it, sure enough, soon after the dinner bell rang, he was quickly dismembered and, well, eaten. Most odd, it turns out, there was actually a WAITING line for diners being the main entree! I mean, people were making reservations to be this delectable feast. Different tastes, I guess.

              Now, why couldn’t I start up an airline with the absolute bare minimum security–close to none at all? All you’d need would be able and willing pilots (also, very, very, well paid) and thrill seeking, intrepid, ice water in their vein passengers. Given the statistics that have been bandied back and forth on this subject, it would seem to be that you’d still be dealing in the very high 90s for a safe, terrorist-free flight. For that matter, why not allow guns on board? I’m thinking, passengers, flight crew, and pilots all could be armed. And a Koran and Bible provided for anyone so inclined. Just a thought….Report

              • Boonton in reply to Heidegger says:

                OK, I have thirty terrorists who all book seats on the same flight, all armed with guns. ‘Regular passengers’ are outnumbered. We take over the plane and now we have 9/11 all over again.Report

              • Jaybird in reply to Boonton says:

                Get sixty and have them be the only ones on the plane.

                As it crashes into the ocean, they can ask “wait, who planned this again?”Report

              • Heidegger in reply to Jaybird says:

                HEH HEH HEH HEH–God that made me laugh, Jaybird!! I’l be laughing all day over that–THANKS! Off to work—“As it crashes into the ocean, they can ask “wait, who planned this again?”
                Damn, laughing CAN hurt!

                ReplyReport

              • Heidegger in reply to Boonton says:

                Boonton—you’ve forgotten–MAD. As in, mutually assured destruction. Usually works. There are other remedies for the more fanatical. Perhaps, have piglets and pig blood on board. The hardest of the hardcore Jihadist will NOT die with pig’s blood on his hands. There are of of course, many other stronger tactics that will compel the Allah Akbar screaming passenger get back to his seat and sit quietly.Report

              • Boonton in reply to Heidegger says:

                How exactly is MAD going to work against a 9/11 style hijacker who has taken control of a plane?Report

              • Heidegger in reply to Boonton says:

                Forgot to mention, short of a suitcase nuke, the cockpit door would be impenetrable. NO ONE will be able to mess with the pilots. As for the MAD disincentive, I would not at all be opposed to an above ground atmospheric test of a thermonuclear weapon, as close to Mecca as politically possible, say, perhaps, a Minuteman III with a W87-1 with a 670 kiloton yield. Just for show, gentlemen. Remember, there are no more ifs in the equation. Only whens. We have always been a gentle country except when being efffed with. And one more devastating attack by the 72-virgin loons will be the last one. They’re going down.Report

    • Koz in reply to Boonton says:

      “I don’t doubt for a moment that a 20 yr old male travelling on a student visa with stamps in his passport from Yeman or Pakistan isn’t more likely to get selected for more intense screening than, say, your hypothetical 80 yr old grandmother.”

      I doubt this very much, which seems to be a significant part of the problem.Report

  5. Jaybird says:

    According to the comments over there, we’re a Randian echo chamber.

    This makes me knit my brow.Report

  6. Rufus F. says:

    I’m in France, so this will have to be quick, but has anyone here actually been hauled in? My wife and I got it one New Year’s at the border because the guard thought she was “giving attitude”. My wife had a cold and was croaking everything she said. Also, I’d never tell my wife this, but it did sound like she was giving him attitude.

    Anyway, we were held for questioning for 3 hours, in order to make a point about the arbitrariness of power and authority no doubt. We saw everyone else who got pulled in for extensive questioning. Let me tell you, I’d be amazed if there isn’t racial profiling as is because we saw every possible variation on “from the Orient” from Arab to Indian. The only other white people we saw were some kids who tried to cross with cocaine on the floor of their car. Anyway, I’m pretty skeptical when people say that some sort of political correctness is stopping them from profiling already.Report

    • Rufus F. in reply to Rufus F. says:

      Okay, Boonton already made my point. Yes, I find it hard to believe that profiling doesn’t already happen. That was my point.Report

    • James Hanley in reply to Rufus F. says:

      My wife and I were forced to go in for questioning at the Detroit-Windsor border once because we accidentally got into the truck lane (we learned later that the direction sign was covered with blown snow, so we couldn’t see it). There was no reason except the border guard was pissed off that we had accidentally gotten into the wrong lane. Once we got out of the building where we were questioned, we accidentally got turned around and headed back across into the U.S., where the U.S. border guard berated me for not having a passport. (Me: “But that law doesn’t take effect until next year.” Him: “Well you should be following it anyway!”)

      Not airport security, to be sure, but pretty much what we can expect from security guards of all stripes. I’ve rarely crossed the border between the U.S. and Canada without some degree of hassle (although on one of those occasions, I admit that my traveling companion might as well have pulled out a sign that said, “I’m a drug smuggler!”), and when I was a bike messenger in San Francisco there was continual hassle from office building security guards (who seemed to be under the impression that the sidewalks in front of their buildings were private property). Anyone who thinks airport security won’t end up engaging in petty tyranny is arguing against Lord Acton. Frankly, given my overview of history, I’ll take his word over theirs.Report

    • Boonton in reply to Rufus F. says:

      I think profiling advocates should do a fantasy versus reality check. In the fantasy, profiling is done by someone like Gregory House who can tell you’ve been indoctrinated into a radical belief system by the way you choose to button your shirt. In reality profiling is done by a $13 hr guy whose expected to screen 300 people per hour.

      Maybe Israel is able to pick out the ‘House’ type screeners for their one or two airports, but are there enough ‘House’ types to fill all the other airports in the world? And if Israel is so good at profiling then why have they veered towards the ‘max Type I error’ style by walling off the occupied territories and making it almost impossible for Palestinians to travel in and out of Israel? Can’t their profilers spot the few with suicide bomb intentions?Report

  7. Rufus F. says:

    Also, most of the airports and public transit buildings I’ve used in France and Italy have had army guys with submachine guns walking through them. I don’t know if they do that anywhere in the US, aside from NYC, where I think they do. But you see it all over the place here. It’s okay for me in France because I can speak the language, but I almost got hauled in in Rome because I thought the plain-clothes cop was lying. But they had them and the guys with machine guns there too. I’ve heard it’s the same in Sweden, for some reason.Report

  8. DensityDuck says:

    One thing to remember is that nobody’s going around pretending that the threat to Israel is nonexistent or that Israel’s level of response is an overreaction to a once-in-a-million accident.

    I flew on September 16th, 2001, to go to my first job on the other side of the country. At the airport, both my checked luggage and my carry-ons were dismantled and thoroughly searched before I even got *near* a metal detector. Same happened to everyone traveling that day. And you know what? I didn’t hear one complaint.Report