Men and Machines

Confirming what I said in a recent post about the difference in cultures::
Dana Milbank talks about Israeli security using people versus US security using machines:  their version costs about 8 times per passenger what ours does.  And the NYTimes runs a piece on the many robots being developed for our armed forces.

Simple Solutions to the TSA Scanner Policy

I've two simple suggestions (alternatives) to offer for TSA to adopt.  They should either require passengers to go through their scanners, or through the pat-down process, or do one of the following:

• allow up to 5 percent of the passengers on a flight to go without scanning or pat-downs, provided they wear a bright yellow vest with a big question mark on it from the boarding gate, through the time they're on the plane, until they get off.  That way their fellow passengers can keep an eagle eye on them, ready to jump them if they make a suspicious move.
• allow a person to board a flight unscanned if they buy flight insurance indemnifying the airline against loss of plane and passengers against loss of life in case of a terrorist attack.  We've pretty much guarded against planes being taken over in flight to be used as weapons, so the big danger now is simply the downing of a plane. 

The Right Run Into the "War on Terror"

Dan Drezner, with whom I often agree, and Megan McArdle, with whom I sometimes agree, unite in opposing the new scanning/pat down procedures at some airports.  Kevin Drum passes on an apropo observation--this is the professional class being subjected to "government[al] humiliation".

Bottom line--people arrested for good cause, or not so good cause (i.e., driving while black) get subjected to such patdowns and none of us professional types have much problem with that.

I'm assuming the use of these scanners increasing the likelihood of detecting people with explosives in their underwear.  I'm also assuming we believe it's a good thing to keep people with explosives off airliners. People who object presumably have persuaded themselves there's no increased detection ability, or possibly they would never be so unlucky as to be on a plane with a fellow passenger who has explosives.  Or maybe they're just reacting with their emotions, and not their logic.

[Updated: Dave Weigel in Slate says the right has always resisted big government's intrusions based on protecting society; it's just 9/11 and the Bush administration which temporarily changed their tune.]

Props to GW Again

Seems every 6 months or so something comes up where I have to recognize our former President and his accomplishments.  This time it's his approach to Islam after 9/11.  See this Politico piece by Ben Smith and Maggie Haberman on the way the GOP is abandoning his stand and, as titled, GOP takes harsher stand on Islam.

Resolving the Mosque Issue

Seems to me at bottom the controversy over the community center/mosque planned for lower Manhattan is a NIMBY (not in my backyard) issue.  Everyone agrees the group has the legal right to build 2 blocks from the World Trade Center site; it's just some like the ADL and some conservatives don't think it's a good place. In this light there are a couple ways to resolve it, methods which apply whenever NIMBYism raises its head. Of course, since I'm a liberal, they involve using governmental authority:

• use zoning laws to specify that no religious building shall be built within x miles of the WTC site, grandfathering in the existing churches. (It wouldn't be legal to specify no mosques.)
• use eminent doman to buy all the property within x miles of the WTC site, so the land become government owned, just like the Flight 93 memorial in Shanksville.  Of course, the cost would be high.  If the public is willing to pay the price, and not spend the money on other uses, then they can have a buffer zone.

[Updated:  You could also do a "legislative taking", which is what Congress did for some Manassas battleground land back in the day.

Finally, I like Dan Drezner's comments.]

Terrorists and Miranda

Back in the Bush administration there was a flap related to the Army's publishing of its manual on interrogation techniques.  Some argued it was wrong to described permitted and prohibited methods in detail, because it would enable terrorist organizations to train their members to resist interrogation.  That seems to make sense: we can't exaggerate how wily and tricky these terrorist networks are.

But if that makes sense, then surely there's no need to modify the Miranda warning and law with regard to U.S.  citizens and residents.  Any smart terrorist organization understands that these people have rights under the Constitution, rights which aren't dependent on the Miranda warning.  Any libertarian will tell you there's no obligation to say anything to a law enforcement officer.  So smart organizations will train their US citizen recruits in their constitutional rights, and modifying the warning will do nothing.

So there's a choice: believe in smart terrorist organizations and don't change Miranda; or, believe terrorist organizations are less than smart, that Murphy's Law operates there as well as elsewhere,  and the terrorist threat becomes too small to warrant any changes.

The Layers and Layers of Duplicity in the New Yorker

Malcolm Gladwell has a piece in the New Yorker elaborating on the layers of duplicity in intelligence, and counter-intelligence, and counter-counter-intelligence, and....   Matt Yglesias links to it.

Meanwhile, via Best Defense, Steve Coll comments on the possible attitude of Pakistani terrorists to the NY car bomber.

Praise for Bureaucratic Reorganization

Stuart Baker at Volokh Conspiracy posts about TSA's new approach in screening international airline passengers.  And he sneaks in a compliment to the reorganization which created DHS:

"This new system is also a sign that the creation of DHS is in fact paying dividends.  If we still had three separate cabinet departments doing customs, border immigration, and air security, there is no possibility that TSA would be borrowing from and cooperating with border agencies to use their techniques and perhaps their IT systems to screen passengers."

He prefers the approach using a person's individual data to estimate probabilities rather than using more general characteristics, like nationality.

9/11 Problems--Communications

Those who read the report of the 9/11 commission know one of the problems in NYC was the first responders had different communications technology.  It's been a while since then, and one would have thought we'd be well on the way to fixing such problems.

One would have thought, but one would have been wrong.  See this Government Executive discussion.

Terrorist Tipoffs and an Evolutionary Arms Race

Josh Marshall at TPM wonders something which I wonder too.  Namely, when we identify things as tipoffs, as suspicious circumstances warranting more investigation, what stops the terrorist organizations from adapting?  For example, John Doe is committed to blowing up an airliner.  So he boards an airliner paying cash for a one-way ticket with no luggage and using a false name.  All sorts of sirens should be going off, right?  But assuming some resources, what's to stop him, knowing what we consider as red flags, from using a credit card, his correct name, and a full set of luggage on a round trip ticket?

And Ann Althouse points to the Newark incident:

"The fact that these two individuals kissed and walked hand-in-hand does not and should not wash away suspicion. If it did, terrorists would know how to stage a security breach. Have male and female confederates. The woman passes through security and then lets in the man, who has whatever weapons/bombs on him that may be desired. The two act like lovers, and the TSA workers sit back and think ain't love grand. A few hours later, hundreds of human beings are blown to pieces."

 Biologists point to arms races in evolution where prey and predators, the eaten and the eater evolve defenses and counters. Maybe that's what we have here.  We come up with a profile of the likely terrorist, the terrorist organization figures out what it is (not a hard job) and how to counter the profile (carry luggage, travel with a female companion, whatever).  A successful attack, or at least one close to success, tells us we need to change the profile and the process continues.

The Next War and Security Works

Two thoughts coming out of the airline incident:

• I'm not clear on who handles security in Amsterdam, whether it's TSA for US flag airlines or the Dutch.  Assuming it's TSA, I think we should credit their rules as successful.  Apparently terrorists believe the screening process is good enough that they went to some lengths to evade the process--sewing the explosive into underwear.  And the attempt to use it to bring down the plane failed, presumably because of the difficulties of turning the hidden explosive into an effective bomb.  I'd compare it to a football defense like Carolina showed Sunday--effective enough to force the Giants into low-return and/or risky pass plays. If the terrorist had been able to get an effective weapon on board, it would be a different matter.  And it's a separate issue of whether procedures should have identified and prevented the guy from boarding in the first place.
  
• I noted Sen. Lieberman now wants to attack Al Qaeda in Yemen.  That's the downside of Obama's policy in Afghanistan.  He may be right that his new strategy can work and avoid the problems of using lesser force.  But, as long as the strategy is sold as attacking terrorism and preventing a safe haven for terrorism, it's doomed to failure.  The number of failed or impaired states in the world far exceeds the capability of DOD. 
  

The Problem With Lists

Blogs like Ann Althouse and Powerline are suggesting problems in the Obama administration's handling of terrorism as a result of the Detroit incident.  Maybe, maybe not. 

I do want to comment on one aspect: according to the Post today Abdulmutallab's father grew worried about his radicalism and notified authorities a month or two ago.  However, he may have applied for his tourist visa to the US in 2008.  All of the following is tentative, based on assumptions: assume the various lists, the "no-fly list" and the broader "Terrorist Identities" list really are "lists", that is static databases which are updated with adds and deletes periodically. So the person processing the visa request checks the appropriate lists, gets no hits, and goes ahead and approves the request. 

Ideally, of course, one would like two-way communication, if not in real-time, at least daily, between the various lists/  If visa requests are checked by matching against the terrorist list, any changes to the terrorist list should be matched back to the approved visa list.  So when Abdulmutallab is added to the list of possible terrorists, a process is triggered that results in putting his previously approved visa into question.

Achieving that sort of two-way communication is probably about 10 times more bureaucratically difficult than the one-way communication.

Public Enemies and Terrorists

Saw the movie Public Enemies (wife's favorite Johnny Depp as lead) about John Dillinger yesterday. It's good, not as good in my opinion as Collateral, also directed by Michael Mann. One thing Dillinger says, in explaining why he can rob banks: "they (the authorities) have to watch every bank, I only have to hit one" or something close.

I thought of that in watching the ABC news last night, which led with a Senate committee hearing on a GAO report on how easy it was to smuggle into Federal buildings the materials necessary to make a small bomb.

Generally speaking, I think worries about security are overdone. Security details for mayors and governors? How many have been assassinated? Security for USDA buildings? I remember one incident in the South Building of USDA; if I recall correctly it was domestic violence, certainly not terrorism directed at the hard working bureaucrats of Agdom.

We used to have typewriters walk out of the buildings. But I wonder how many typewriters are left in the department. We used to have to account for calculators. But Moore's law took care of that. I suppose they still have to account for PC's, and security could deter their theft. But Moore's law will take care of that as well, as everyone will have his or her own personal electronics implanted at birth.

I wonder whether an economist has studied the evolution of loot for robbers/burglars: does minaturization increase the availability of high-profit loot (no one could steal a mainframe computer) or does Moore's law wipe out such loot? Do the two forces counterbalance?

But, having security makes people feel better, and that's important.

Bypassing Bureaucratic Rules--NYPD

The Post's Book World carries a review of a book on the NY Police Department. In an example of entrepreneurship (yes, bureaucrats can be entrepreneurs just as capitalists can), it's set up a counter-intelligence shop:

Freed from the bureaucratic restraints of Washington, Cohen [ex-CIA man heading the shop] set about building his 600-person unit with astonishing speed and efficiency, infuriating former federal colleagues along the way. In no time, he had twice as many fluent Arabic speakers on his staff as in the entire Federal Bureau of Investigation. His agents speak some 50 languages and dialects in all, which matches the reported linguistic capabilities of the Central Intelligence Agency.

The book is: SECURING THE CITY Inside America's Best Counterterror Force -- the NYPD By Christopher Dickey.

But there's also this:

"Dickey might have dug a little deeper in addressing the persistent but vague allegations in Washington that the NYPD counterterrorism unit cuts legal corners and that some of its methods are unconstitutional. "They do stuff that would get us arrested," says one three-letter guy."

A Good Book

Nicholas Kristof highlights the Greg Mortenson book in today's Times. I should have blogged about it when I read it, if I didn't. Mortenson turned a failure at mountain climbing into a success at building schools in the mountains of Pakistan/Afghanistan. Inspiring and down to earth. The book runs the danger of being saccharine, but it's not.

Identity Checks and Government Blogging

Here's an article on changes being made by DHS in handling their no-fly list and here's the DHS blog's post on it . (If I understand, Ted Kennedy gets stopped all the time, because there's a suspected terrorist (or at least someone on the no-fly list) with a similar name, so they have to establish Ted isn't the same person. Now, under the proposal, if Ted allows his date of birth to be added to the airlines data, he can go right through.)


The proposal makes sense to me, but not to the first four comments on the blog. Maybe they aren't into genealogy, where you have to distinguish among multiple John Rippeys or even worse, William Smiths. Much less try to reconcile the data between ASCS and SCS to determine whether each agency was dealing with the same people. But then, I'm just a retired bureaucrat who tends to trust bureaucracies, at least in some instances.

Kevin Drum and Steven Levitt Agree on Anti-Terrorism Logic

The Levitt/Dubner Freakonomics blog has just moved to the NYTimes server. Levitt's first post discussed ways possible terrorists could attack the U.S. cheaply. (First one, have 20 teams of snipers emulate the DC two, but truly at random.) He caught a lot of heat (surprise!). In his
response today he outlines two alternative interpretations, this is the second:

The alternative interpretation is that the terror risk just isn’t that high and we are greatly overspending on fighting it, or at least appearing to fight it. For most government officials, there is much more pressure to look like you are trying to stop terrorism than there is to actually stop it. The head of the TSA can’t be blamed if a plane gets shot down by a shoulder-launched missile, but he is in serious trouble if a tube of explosive toothpaste takes down a plane. Consequently, we put much more effort into the toothpaste even though it is probably a much less important threat.

Kevin Drum says the same, in the context of the Democrats and the bill on FISA:

Note the way the incentives work here. If you pass the bill, the results are ambiguous. Sure, a lot of people will be angry, but they'll probably get over it eventually (or so the thinking goes). But if you stall the bill and a terrorist strikes, you are firmly and completely screwed. Goodbye political career. So which choice do you think a risk-averse politicians is likely to make?

This same dynamic was at work before the war, too. If you favored the war and things went south, the resulting mess would be long-term and ambiguous. There would almost certainly be a way to weasel your way out of any trouble and stay in office. But if you opposed the war and then, after the invasion went ahead over your objections, the Army discovered a serious nuclear arms program or an advanced bioweapons lab — both considered distinct possibilities at the time — you'd be out of office at the next midterm. For risk-averse politicians, the choice was obvious.

Nobody wants to risk being proved wrong in a way that's so crystal clear there's simply no chance of talking your way out of it. It's this fear that gives national security hawks the upper hand in any terror-related debate. Still.

I have to agree--as far as I can see, the "terrorist cells" that have been captured weren't very scary. If you assume that our security can catch 90 percent of the threats, that means the threats generally aren't potent. If you assume a lower level of averted threats, where's the attacks?

Two Views on Terrorism

Mitt Romney wants to redo the DHS to focus on intelligence and attack prevention, rather than recovery from attacks.

This is from Princeton's blurb for a new book:

Many popular ideas about terrorists and why they seek to harm us are fueled by falsehoods and misinformation. Leading politicians and scholars have argued that poverty and lack of education breed terrorism, despite the wealth of evidence showing that most terrorists come from middle-class, and often college-educated, backgrounds. In What Makes a Terrorist, Alan Krueger argues that if we are to correctly assess the root causes of terrorism and successfully address the threat, we must think more like economists do.

Krueger is an influential economist who has applied rigorous statistical analysis to a range of tough issues, from the minimum wage and education to the occurrence of hate crimes. In this book, he explains why our tactics in the fight against terrorism must be based on more than anecdote and speculation. Krueger closely examines the factors that motivate individuals to participate in terrorism, drawing inferences from terrorists' own backgrounds and the economic, social, and political conditions in the societies from which they come. He describes which countries are the most likely breeding grounds for terrorists, and which ones are most likely to be their targets. Krueger addresses the economic and psychological consequences of terrorism. He puts the terrorist threat squarely into perspective, revealing how our nation's sizeable economy is diverse and resilient enough to withstand the comparatively limited effects of most terrorist strikes. And he calls on the media to be more responsible in reporting on terrorism.

The egghead seems to me to have much the better argument. The US may be attacked by terrorists once for every 10 attacks on EU nations and 1 in 10,000 attacks in Iraq. While some attacks may be scary, and some damaging, we have much more to fear from mother nature. Our general policy should be to do intelligence and defense reasonably well, but respond to disaster very well.

Why Most Terrorists Are Incompetent

That's the title of a piece in Slate by Tim Harford. The focus is really on the background of terrorists, which turns out, according to one study, to be well educated, particularly in comparison to the milieu from which they come. The British doctors involved in the last attempts are an example. All of which accentuates the question--why is someone who is smart enough to get an MD incompetent as a terrorist?

My answer is --first, you never get it right the first time ("Harshaw's law"). Second, the brits seem to have been operating without close guidance from more experienced terrorists. (The Palestinian suicide bombers had a whole infrastructure in place. Even though the bomber was a first (and last) timer, the organization was not. Third, terrorism is probably harder than it seems from the outside.

Sensitive But Unclassified--Bureaucratic Boundary Setting

Elizabeth Williamson in the Post had an article on Sensitive But Unclassified (SBU) markings (things like "For Official Use Only", etc.). These are stamps that government agencies use when they can't justify a "Confidential, Secret, or Top Secret" classification. It seems that they pose a big threat to the information sharing deemed essential to combat terrorism, because each different marking carries its own rules for dissemination and there are 108 different ones. So if the FBI sends info to the state police who relay it down to county sheriffs things can get confused. There's a committee working on simplifying this (to improve the "information-sharing environment").

Why so many SBU's? It's a combination of reasons.

• The official classification system is limited and rigid--only three markings so they have been amplified by modifications.
  
• Bureaucrats are scared--suppose this paper leaks to the Post, that would be embarassing. Or even if it reaches the local gossip. (The Plame affair revealed that even deputy Secretaries of State can love their gossip.)
• There's the high school clique reaction: we know something you don't, ha ha ha.
• Most of all, bureaucrats love to set boundaries and SBU's are a way of marking them.
  

Is it all bad? No. I'm reading William Easterly's "The White Man's Burden". He makes the point that a bureaucracy (foreign aid/foreign development type agency) that tries to do everything (and that has multiple "principals" to report to) is prone to failure. So a bureaucracy that is focused on doing one thing is more apt to be successful.

The problem we have in homeland security is that our bureaucracies have each had their own objective(s). When the global war on terrorism came along, we superimposed new objectives on the old and we still haven't straightened things out yet.