Friday, September 30, 2011

Airport digital strip search machines: to see or not to see...

To see or not to see, that is the question

In the political arena – and airport security examined in the OU/BBC’s Bang Goes the Theory(26/9/’11) is a good example – the desire to reduce risk at almost any cost is notable. Authorities on both sides of the Atlantic regularly advocate the need, in the context of the ‘war on terror’, to introduce measures like airport scanners, sometimes described as digital strip search machines, in order to be seen to be improving public safety.  Often, the more expensive and hi-tech and visible the approach is seen to be the better.

We are asked to give up a little bit of privacy to improve overall security. People with nothing to hide, we are told, have nothing to fear.[1] The systems proposed will help catch the baddies and protect the goodies. I will touch on the validity of these claims a little later but for the moment I have a question. Is it true that people always want the level of risk they are facing in life to be reduced? 

Evolution has effectively turned every single one of us into our own personal risk managers. We make decisions every day that involve taking risks. Some people eat, drink and smoke too much and do not take enough exercise, in spite of the health hazards associated with such a lifestyle. They may not always be consciously aware of it but they are balancing their interpretation of the long-term risks of damage to health against the short-term rewards of enjoying their food, drink and smoking.

Many actively seek out the thrill of risky pursuits or professions like mountain climbing or racing car driving. The motto of the SAS is “who dares wins”. Just helping the kids to safely cross a busy road to get to the local park involves the balancing of risks and rewards. So perhaps a more legitimate pursuit for those professionally involved in risk management at airports would be the balancing of risks against benefits rather than the reduction of risks? We need to weigh the costs against the benefits when making decisions about airport security.

Unfortunately this can be a complicated business.  We need to consider security trade offs - the facts, effectiveness, economics, societal values, technology, health, safety, political, legal and ethical issues.

Fortunately, security expert, Bruce Schneier [2] has developed a user friendly check list that virtually anyone can use to do a pretty decent evaluation of security measures deployed in all kinds of contexts, including airports.  Schneier suggests we ask a series of relatively straightforward questions.

  1. What problem are you trying to solve?
  2. What is your solution?
  3. How well does your solution address the problem?
  4. What other problems does your solution create
  5. How much does it cost?
  6. Is it worth it?
Let’s try to apply Schneier’s questions to airport scanners.

  1. What problem are you trying to solve?
We hope to prevent attackers with malign intent getting onto airliners with dangerous weapons and the adverse consequences arising from such circumstances.

  1. What is your solution?
Full body scanners (digital strip search machines).  In the Bang programme presenter, Liz Bonnin, extolled the virtues of a millimetre wave scanner but it is important to note that there are generally two generic types of scanners deployed in airports (and other contexts) – (a) millimetre wave scanners and (b) X-ray backscattermachines. The latter seems to be the most commonly used in the US. 

3. How well does your solution solve your problem?

Well here is where things get interesting.  If you have a compliant attacker, like a Bang presenter, who tries to smuggle through a potential weapon which the scanner can detect – like Liz’s ceramic knife, secured under her bra strap – then if the security personnel are paying attention they will spot the knife. And indeed the expert, Paul Mason from Redline Aviation Security, did detect the knife on the scanner image.  Queue imaginary impressed music - dah dah – and Liz’s “awesome… excellent stuff” reaction. She had got her ceramic knife through the metal detector without being spotted after all. But would she have been as impressed if the knife had been detected through a pat down search?

The key problem though is not that Liz’s knife could have been spotted through a simpler, cheaper route, although these physical searches create their own difficult issues.[3] The key problem is that a number of researchers have shown that the machines are ineffective at detecting explosives and other weapons. In 2010 the US Government Accountability Office, the independent agency which investigates how the federal government spends tax revenues, concluded [4] that airport scanners might not have found the explosives used by notorious underwear bomber, Umar Farouk Abdulmutallab.

Security expert, Werner Gruber, gave a stark demonstration on German television in early 2010 that the scanners were ineffective. He used bomb making components concealed on his person and undetected by the scanner to start an intense fire in a controlled experiment outside the studio immediately after he was scanned.[5] On the plus side, the machine did detect his mobile phone, penknife and microphone.

Also in Germany the millimetre wave type machines in trials at Hamburg airport last year were shown to have problems with creases in clothes.

So there are significant questions about the effectiveness of these machines.[6]

  1. What other problems does your solution create?
There are significant security, ethical, operational, legal and health and safety problems with the routine deployment of these machines in airports (and other contexts like train stations).

The security problems are not obvious until someone points them out.  More and more complicated and intrusive airport security causes long queues – lots of people gathered together in a small space. The result is that an attacker armed with a bomb doesn’t have to make it onto a plane.  He can harm a large number of people without ever having to go through security. And contrary to the suggestion in Bang there is no evidence that strip search machines make the process of getting through airport security quicker. Bruce Schneier calls it “security theatre” when you create the impression of improved security without actually improving security. The misplaced belief or confidence that the machine will catch the bad guys can undermine our security as we pay less attention to other indicators of nefarious intent.

I occasionally wonder what Winston Churchill would have made of these types of technological “solutions”.

“We will fight them on the beaches, we will fight them with no dignity…”. What kind of values is a society displaying when routine (even if only of the virtual variety) strip searching is considered acceptable?  Note I’m not passing judgment here just asking the question if respect for personal privacy is a value that is important any more.

The operational issues are legion.  In Bang it was suggested that displays are remote from the scanners, viewed by same sex operators who only see the image and never the individual scanned.  Whereas that is the case in some airports the set up is not universal.  How do we know, by the way, that the scanned image is only being transmitted to the remote operator (if that be the case) at the airport? In truth we do not.  The Bang crew suggested full body images are never recorded. Actually many, if not all, of these systems have recording capability [7] and lots of images have been leaked onto the Internet.[8]

There are numerous stories [9] about young women and ethnic minorities being disproportionately selected for scanning and other passengers being pressurized or maltreated by security staff and/or police when deciding to opt out of being scanned.

The other operational issue that was mentioned in the program was the question of how revealing the images produced by the machines were. Liz said the “technology is based on contours – your underwear is so close to your body that it kinda helps conceal the more detailed bits of your anatomy”; or as one of her co-presenters, Dallas Campbell put it “Your bits and bobs, your bits and bobs.” With reference to the images displayed during the course of the program that was arguably the case.  What wasn’t said, however, was that the machines can be fine tuned, increasing the intensity of the beam, to leave little to the imagination.  Tuning is in the hands of the airport authorities and ultimately the operators.

The legal issues with these strip search machines are complicated. It is possible that their use is in breach of Article 8 the European Convention on Human Rights [10] which guarantees a right to privacy. [11] Their operation, in the case of children, is likely to breach UK and other jurisdictions’ child protection legislation regarding the making of indecent images of children.[12] There are also data protection and data retention questions.

Then there are the health questions.  It is probable, as stated in the Bang program, that the health risks are small with the millimetre wave machines that Liz Bonnin reviewed. The chances are you’ll be exposed to more radiation during an average flight that anything fired at you in a millimetre wave machine but the scanners have not been independently tested to rigorous academic standards. So there is no independent verification of how they work or what the specific exposure levels might be.

The X-ray backscatter machines are a different story, however. In April 2010 Professor John Sedat and a number of his highly respected academic colleagues at the University of California - biochemistry & biophysics, cancer, X-ray and imaging experts and members of the US National Academy of Sciences – wrote [13] to Dr John P. Holdren, President Obama’s chief science adviser, [14] explaining their “concerns about the potential serious health risks of the recently adopted whole body back scatter X-ray airport security scanners”. They point out that though the machines nominally use low energy beams the dose is concentrated into the skin and immediately adjacent tissue. Since this is such a small fraction of body weight, “the real dose to the skin is now high”.  They also point out that the use of these machines with specific categories of individuals raises particular concerns. These include over 65s, females with a predisposition to develop breast cancer, pregnant women, HIV or cancer patients and children. They worry about operators increasing the intensity of the beam if images are not clear enough for them and conclude:

“There is good reason to believe these scanners will increase the risk of cancer to children and other vulnerable populations.  We are unanimous in believing that the potential health consequences need to be rigorously studied.”

So in answer to our fourth question – what other problems does your solution create – it seems there are significant ethical, operational, legal and health issues to be considered.

  1. How much does your solution cost?
I understand airport scanning machines cost between £60k and £200k. In addition there are the ethical, security, operational, legal and safety costs mentioned above.

  1. Is it worth it?
Frankly no.

To sum up – they are very expensive, they don’t work very well,[15] can compromise security and they create lots of other problems. There is no need to peer through the clothes of people traversing through airports.  If there are tens of millions of pounds available to install these machines it would be better spent on intelligence gathering and pursuing rigorous investigations

So why are so many of these expensive scanners being deployed in airports and other places?[16] Essentially it is because no politician or public official, in the aftermath of the next major incident or attack, wants to be the one held up to ridicule and abuse for not putting enough prior resources into security.[17] They must be seen to be “doing something” after every attack and they must be able to claim “no one could have foreseen” or “we did as much as humanly possible”.  And when it comes to airport security it seems there are no limits to the resources (and stupidity) [18] that gets poured into it.


[1] See Nothing to Hide: The False Tradeoff between Privacy and Security by Daniel J. Solove. Yale University Press, 2011, New Haven & London
[2] See for example Beyond Fear: thinking about security in an uncertain world by Bruce Schneier Copernicus Books, 2003.  New York
[3] So-called “enhanced” pat down searches introduced in the US have led to all kinds of problems. See for example Rape Survivor Devastated by TSA Enhanced Pat Down at http://pncminnesota.wordpress.com/2010/11/08/rape-survivor-devasted-by-tsa-enhanced-pat-down/ and TSA pat-down leaves traveler covered in urine http://www.msnbc.msn.com/id/40291856/ns/travel-news/t/tsa-pat-down-leaves-traveler-covered-urine/
[4] GAO Statement for the Record To the Committee on Homeland Security, House of Representatives Homeland Security: Better Use of Terrorist Watchlist Information and Improvements in Deployment of Passenger Screening Checkpoint Technologies Could Further Strengthen Security (January, 2010) http://www.gao.gov/products/GAO-10-401T
[5] Video with some English subtitles available on YouTube at http://youtu.be/idICUSiGcqo
[7] The Electronic Privacy Information Center, EPIC, a civil rights group in the US obtained evidence through freedom of information requests in 2010 that the body scanners being used at federal courts house, for example, can store and record images. http://epic.org/privacy/airtravel/backscatter/ EPIC are pursuing a federal legal case to have airport scanners banned on constitutional grounds, so far without success.
[9] Bruce Schneier has a selection of links to some of these on his blog at http://www.schneier.com/blog/archives/2010/11/tsa_backscatter.html
[11] And on the other the side of the pond the 4th amendment to the constitution that guarantees protection from unreasonable search and seizure
[14] Assistant to the President for Science and Technology, Director of the White House Office of Science and Technology Policy, and Co-Chair of the President's Council of Advisors on Science and Technology (PCAST)
[15] See also Leon Kaufman and Joseph W. Carlson (2010) "An evaluation of airport x-ray backscatter units based on image characteristics" Journal of Transporation Security, Volume 4, Number 1, 73-94.  http://www.springerlink.com/content/g6620thk08679160
[16] EG the Z back scanner van http://en.wikipedia.org/wiki/Backscatter_X-ray
[17] Schneier calls this CYA security and I respectively refer the reader to his essay on the same at http://www.schneier.com/blog/archives/2007/02/cya_security_1.html
[18] [sic] ban on liquids etc. Choose your favourite dopey airport security measure.

Update: edited to cure broken links.