TSA Chief: 'We'll Never Eliminate Risk'

John Pistole says he's "much more interested" in screening people, not the items they carry onto planes

Ever since new airport security procedures went into effect in late October, the Transportation Security Administration has been at the center of controversy. The combination of enhanced image screening with invasive patdowns for those who opt out has rankled civil liberties advocates and some of the flying public. John Pistole, the head of the TSA, is a 26-year veteran of the FBI -- an expert in counterterrorism and for six years the bureau's deputy director. As TSA administrator since July, he finds himself having to defend the new measures. James Fallows and Jeffrey Goldberg spoke with Pistole on Monday. An edited version of the interview is below; their posts about the conversation are here and here.
John Pistole.jpg

Getty Images

James Fallows: I'd like to start with a question both Jeff and I have raised, which is the whole question of the balance between security, on the one hand, and liberty and privacy concerns, on the other. Is it TSA's job to set that balance? Or how do you think that balance is set?
John Pistole: The way I view it is for TSA to develop the security protocols that afford the best security, while recognizing that there is a balance. The best security would be something way beyond what we're doing.
Jeffrey Goldberg: The best security would be to just not allow people on planes. That's perfect security.
Pistole: That's "risk elimination." And we're not in the risk-elimination business, we're in risk mitigation, informed by the latest intelligence, informed by our friends [in the intelligence agencies], and informed by the results of our covert testing.
Those things inform judgments and actions and then we take that information -- I take that information -- and then ask the experts how can we address these threats? They come up with different things based on all the information they have, and then they make a recommendation, and then it's up to me to say, OK, does that exceed what I think is appropriate in terms of privacy?
So that's my responsibility. To say, does this give us security, without violating something that would be a Fourth Amendment issue? That's what I did in this situation [the latest controversy] -- being informed by my prior background [with the FBI], and seeing this as different from a law-enforcement Fourth Amendment search, for example on the pat-down, and the privacy of the AIT (the controversial scanner).
How do we go about deploying this privacy screening? Separate officer, separate room, not retaining the images. You've seen these posters, and you realize, it's not as graphic (as has been reported). A lot of the media reporting showed these graphic images, and they rotated through 360 degrees so you get almost a crotch shot - that's not what these are. A lot of informed people describe these, I think in some of your writing, as the "naked" machine, or the porno -
Fallows: -- Jeff has a description ["porn machine"] that I don't use --
Goldberg: -- Fallows is very statesmanlike.
Pistole: I thought of bringing something in that actually was pornographic. I could say to you, then, "This is porno, this is not." I'll recognize it when I see it.
Goldberg: Exactly, the Supreme Court standard.
Do you agree with the civil libertarian view that airports -- not just because of the TSA -- but airports have become these Fourth Amendment-free zones? This is the fundamental concern, that you're giving up your Fourth Amendment rights when you buy a ticket.
Pistole: I don't agree with that. But I do agree with the difference between what most people think of in terms of a reasonable search-and-seizure for purposes of law enforcement, versus a public-safety administrative search. I don't know if people are drawing that distinction, either, from a legal standpoint or a practical application. I think people don't look at the public safety aspect. They look at it strictly from -- as I did for almost 27 years (in the FBI) -- a law enforcement search perspective, in which I need probable cause, and I've got to be reasonable in that search.
Goldberg: It is true that you now have more power as a TSA administrator to search people without probable cause than you ever had as an FBI official.
Pistole: Oh sure, yeah. If people take an affirmative act of engaging in, in this case, aviation -- they want to get on a plane -- they're taking an affirmative act to do that. Then, yes, there is authority to do the administrative search for public safety purposes. As I've said a number of times, I think reasonable people could disagree as to the precise technique used on each person. So for you, it may be patting around the knees or the armpits. You might be sensitive there. For others, it is groins. For women, it is breasts, for which procedure hasn't changed. Our protocols on the breast have been the same for years now.
Fallows: You were quoted in the Atlanta Journal Constitution as saying that you were opposed to the "cookie-cutter" approach to security. I think that the criticism from the public now is that it seems like we're seeing a cookie-cutter approach--disabled people, old people --
Goldberg: -- Nuns --
Pistole: If they are nuns. That's the question. So how do we verify who that person is?
Fallows: So how is what you're doing now not a cookie-cutter approach?
Pistole: So, starting with pilots. Pilots were the biggest group of those I assessed as being low risk to civil aviation. I mean come on, they're in charge of the yoke, they can put the plane down, like the co-pilot in Egypt Air in 990 did. I worked that investigation in October 1999. It was just a horrific recovery effort -- men, women, children, body parts, everything.
No amount of physical screening is going to identify what was going on in that co-pilot's head. So why are we doing physical screening? Instead of physical screening, we do an identity management base system to say, OK, this ID matches up with this airline's current records as of five minutes ago or whatever it is, so if they got fired yesterday... And then the airline has the responsibility -- shared responsibility -- to make sure [of the identity] of anybody that is a licensed pilot with their airline, that they're actually supposed to be in control of the aircraft, pilot or copilot.
For me that's the first step. Then you start looking at who are the other trusted travelers. What about flight attendants? So we made some modifications to the type of physical screening they go through. We haven't advertised that, but we've made some modifications to how they are screened, and we're looking at the possibility of doing something similar to what we did with the pilots.
I've had members of Congress say, "Look, I am a member of Congress. I am not a terrorist; This is absurd. Why do I have to go through a physical screening? It's an insult." And I say, "Well, in the legislation that authorizes TSA every year, they specifically include provisions that members of Congress will go through physical screening." By the way, I go through screening every time, I went through AIT three times last week in L.A., Long Beach, and Las Vegas.
Goldberg: Secretary Napolitano doesn't, though.
Pistole: No she has a security detail and she doesn't. And she usually goes on Coast Guard aviation, so yes, that's a different construct.
Goldberg: Talk about the difference between commercial aviation, and general aviation, or private aviation -- corporate jets, private planes and the like.
Pistole: There's a number of issues here. Much of our approach is risk-based. To say, "OK, what was the last threat from general aviation?" And we did have this guy down in Texas last year fly into the IRS building [in Austin]. Killed himself and unfortunately an IRS agent. But being smaller aircraft, typically, than a full jumbo jet or something similar, with the fuel capacity and the damage that could be done, there are certain risk-assessments that have to be done. We're trying to look at what makes sense from a risk management perspective, while also trying to promote the free flow of people and goods. And what's the balance there, just like what's the right balance on a pat-down, in terms of security and privacy?
Fallows: I wanted to follow-up the question about the TSA's role in setting the balance. It sounded to me as if you were saying that it partly is your job as TSA administrator to set a balance.
Pistole: Yes. That's part of my job, to identify the issues, then discuss it with a broader community -- within the department, and elsewhere -- as to: Does this make sense? What equities will be affected by this decision? How does that work?
Fallows: You've mentioned the new rules for pilots, and the unpublicized new ones for flight attendants, as an adjustment of the balance. Those are relatively recent changes, right? Does that reflect a response to the pushback of recent months?
Pistole: So when I came in [to this job] on July 1, I asked for a review of any number of things. Who do we screen and why? What's on the prohibited item list, and why? Are we harmonized with the EU [European Union], in particular, because of all the traffic in those areas. For example with the EU, they say to us, you don't have 100% staff screening at your airports, talking about "insider risk."
Goldberg: The EU has 100% staff screening?
Pistole: That's what they say. We say, we have a 100% check -- but it's identity based, along with random checks. So we know every worker, with their background checks. If somebody's on a watch list, they don't have access.
Goldberg: But a worker can get access to the tarmac without going through a physical screening?
Pistole: Yes, in many airports.
Goldberg: "Many" airports?
Pistole: Yes, there is a small number where we do 100% staff screening, but to do that for the tens of thousands, hundreds of thousands of airport workers. I mean, people who work in the kiosks, the stores, food vendors, and then the airline employees, the mechanics, and all those people--that's a big lift in terms of additional screening. And other than drug smuggling, we haven't seen terrorists exploit that. They tried to last year, Al Qaeda of the Arabian Peninsula, dealing with some Brits. There's a little bit written about that, but they tried to use a prominent airline insider, an employee, to recruit a couple of other airline employees to do something bad. So we see that as a risk, but it subsides.
Fallows: One of the general critiques of the TSA approach is that it's not scalable. That is, somebody has a shoe bomb, which cost him a thousand dollars. Then it costs us a billion dollars to respond to that. Somebody has an underwear bomb, which is cheap, and it costs us a lot of money to react to. How do you approach this problem?
Pistole: As terrorists become more precise in their design and concealment, they make this "proportionality" question even greater. They did two toner-cartridge bombs--why not do four, six, eight ten, that are just waiting to go.
Goldberg: But you do get the impression over time that the TSA and other agencies are constantly chasing the last plot? On Thanksgiving it was dips and cream-based sauces on the list of banned items. It gets to be a level where they have you running in circles.
Pistole: And that's the question. Do we not address yesterday's threats? I don't think anybody would agree that we shouldn't have hardened cockpit doors, that we had to deal with box cutters. At least until technology gets to the point -- which I'm hopeful is very soon -- that we can carry liquids on, or you can keep your shoes on -- even coming into TSA headquarters! - or you don't have to take your jacket off, or you can keep your laptop in their bags, the current rules will be necessary. And I'm hoping to get to the point where the invasiveness of screening becomes less and less, as it becomes more identity-based. Of course I know your [Goldberg's] experience from Israel, at least what I've read about, so that is a construct that without profiling becomes a challenge.
NEXT: Why Pistole is open to profiling passengers -- and how the TSA already does it

Goldberg: Do you ever see a point after, God forbid, another attack, where we move toward an interrogation-based system as opposed to a material-based system in our airports? Or is it simply not doable on the scale of American aviation?
Pistole: I'm much more interested in the person than the items the person is carrying. I want to know more about that person, and I want to be able to use all available intelligence, and "Secure Flight," which we just completed in terms of the international roll out, is one step in that regard. Just those three basic data fields [required by "Secure Flight"] -- name, date of birth, and gender -- helps knowing about the person.
Every morning I start off my day with an intelligence briefing, and looking seventy-two hours out in terms of who is on any watch list, terrorist watch list, who might be interested in traveling. Who has purchased tickets. So we can look at, out of 27,000 flights that are occurring today in the U.S., where are people traveling that might be of interest? Do we need additional federal air marshal coverage on those flights?
We had a situation a couple months ago where we saw five individuals, four on a (watch list) and one on the no-fly list, who bought tickets from around the country to the same city, flying to another city, on the same day. And we thought, "Hmmmm, OK, that's of concern, potentially." So we pass that information to the joint terrorism task force, our colleagues in that city. They do the little check and say no, there's an iman conference this weekend in that city. So that's intelligence. Now we still did additional screening of the four who did fly, and of course the no-fly didn't fly, fortunately.
Now, if you're saying in an ideal world, if I also had all the information about where a person purchased that ticket, or the co-travelers -- and that's one of the things I do see on the 72-hour look ahead, are there co-travelers, that sort of thing. What does the credit history on that person show? You know, there's a lot of commercially-available data about people that we are not allowed to look at. The Brits refer to this as "rich picture." They talk about knowing everything about a person -- that the "richer" the picture that we have of somebody, the better job we could do of security screening with less hands-on. Less physical screening, more informational screening.
But as a society, as soon as you use the word "profiling," for some that invokes a just terrible image. So I think what we are seeing is a public policy debate over which is better: Is it hands-on pat downs for less than 3% of traveling public or advanced imaging technology, or is it [some form of profiling].
Fallows: But there's crude profiling and there's more sophisticated profiling.
Pistole: Exactly. Good profiling is not saying, I don't like the way you look, you're wearing a red tie, so come on over, we're going to take a look.
Goldberg: Even if you can get over the privacy concerns that the Israeli system raises -- the interrogation about where you're coming from, your business, those things -- do you think that it would be ever be possible to institute that system on our scale?
Pistole: I think it's possible, I think it would require a different construct, both for TSA and the airlines. We would require a different workforce. So I wouldn't want to do it the way we're presently configured or employed. But yes, I'm a big proponent of behavior detection.
We've got several thousand behavioral detection officers [BDOs] who are in the airports, and looking at things. I especially like when we do what we call "plays." We have a canine team walk through a terminal with plainclothes behavior detection officers following, to see how people respond to the canine team. I would have loved to be in Amsterdam airport last Christmas. If there had been a canine team and Abdelmutallab was transferring from his flight, to see how he would respond to a canine team. That's the sort of thing that the BDOs are looking for.
We don't advertise everything we do. We use a lot more intelligence than people realize, but we don't advertise for the same reason I didn't want to go out several months ahead of time to say we're going to be doing these pat-downs in a couple of months. I wanted to inform the American people, but I didn't want to inform the terrorists. Because every time you inform, you give them an opportunity to shift their tactics.
Fallows: Are you saying that you recognize the proportionality question as a challenge, and you think you are able to get ahead of it?
Pistole: Well, we have people from the intelligence community and the law enforcement community and also bomb experts, who are working on what's the next device. Nobody predicted toner cartridges. I wish we had. That's why we need tactical intelligence versus strategic intelligence. I've had members of Congress call and say: I have friends who are concerned about going to Europe over the holidays, we've got this travel advisory, what do you recommend? So strategic intelligence is good, but I'm always looking for the tactical. The proportionality issue comes down to how do we ensure that yesterday's plots don't succeed a second time. Because shame on us if they do. I mean, as the American people. I think we say, OK, burn us once, yeah, we'll learn from that, but you're not going to do it twice.
Goldberg: But it's infinite. I mean literally, talk about the problem of cavity bombs. Do you think it is a possible tactic that someone could use? And do you think that your current technology could stop it?
Pistole: No. It is infinite, so how do we deal with yesterday's threat? How can we be informed by those, and yet try to be predictive about what's the next plot.
Fallows: Everything in life fails at some point. At some point there will be an attack that succeeds. Is that something that you should be talking about? How should public officials be talking about that, and about the importance of resilience in response?
Pistole: I think that's a great point, because I think that the next attack is inevitable. Given my bureau experience, there all these domestic lone wolves. Whether it's the guy at Springfield, Illinois last year, or Houston, or the guy in Portland most recently. A group around Newburgh, NY, the Bronx. People who are dealing with what they believe to be jihadists. And they want to kill Americans. They're not interested in doing some fireworks, they want to kill Americans, here in the US. Fortunately it's not aviation-related.
Given that perspective of the threats, the threats are real, how do we best devise a layered system of defenses that has no single point of failure? That's my concern.
If I'm drawing a map, look where TSA is on the continuum. We've got NSA, CIA, then foreign intel services, law enforcement services overseas, foreign agencies. Their SIGINT [Signals Intelligence) and their human collection and all that. Let's say somebody gets through all these defenses. Nobody's identified putative terrorists. They get through our terrorism task forces, 103 of those around the country, and let's say they don't identify anybody. The 750,000-plus police officers, sheriffs, deputies, they don't identify anybody. The concerned public, nobody sees something and says something. They get through all these layers of security, then it comes down really just to the TSA.
Goldberg: That's the impossibility of your job, in a way.
Pistole: So if they get through all those defenses, they get to Reagan [National Airport] over here, and they've got an underwear bomb, they got a body cavity bomb -- what's reasonable to expect TSA to do? Hopefully our behavior detection people will see somebody sweating, or they're dancing on their shoes or something, or they're fiddling with something. Our explosives specialists, they'll do something - they do hand swabs at random, unpredictably. If that doesn't work then they go through (the enhanced scanner). And these machines give the best opportunity to detect a non-metallic device, but they're not foolproof.
Goldberg: Are these actually better than pat downs, properly done?
Pistole: Well that's the key, because these have a human dependency, until it goes to ATR -- automated target recognition -- there's a technology dependency here, there's a human dependency. So it may show us a glaring something, anomaly, on here, but if the Transportation Security Officer doesn't observe it... And yes, the pat-downs do work, too. Again, I don't like a single point of failure.

Goldberg: But there will be failure.
Pistole: We're not in the risk elimination business. The only way you can eliminate car accidents from happening is by not driving. OK, that's not acceptable. The only way you can eliminate the risk of planes blowing up is nobody flies.
Fallows: Do you think that in your position you can start advancing that argument, or does it have to be an actual politician who admits that security can't be perfect?
Pistole: Perhaps a combination. Say, someone with experience in the law enforcement intelligence community, and being in this job now. I think I can do that. But I think it will [also] take more of a political voice. I think it's Congress, in large part, because the President can get up and say it, and the reaction will be, he's saying that just to protect himself when the next attack happens. I think this is a tough, tough issue.
Fallows: There's a related issue that I have beaten the drum on a number of times, about the "ratchet" effect of security. That once you impose a security measure, no politician can take responsibility for removing it. Am I wrong to identify this as a problem, and are there solutions to it?
Pistole: I think that's an issue that is being put on the radar now given all this controversy. What is an acceptable level of risk? And if Congress, the Administration and everybody agrees, "Now, look. Thorough pat-downs are too invasive for most of the public" -- which we know polls show, half and half, people are generally all right with; and of course AIT, which from two-thirds to eighty percent think they're important for security. I think all this gets into the issue of, what is acceptable? And what could we do without?
For example, the Brits, ... I shouldn't say in detail, but they're addressing some budget issues within aviation security in different ways that perhaps we would be talking about here.
Goldberg: But this is the question. Assume you have the technology that you think will detect some sort of explosive in liquid, or some sort of component of explosive in liquid. And you announce, to the general joy of the American public, "OK you can bring your water bottles back on." And then, God forbid, three months later something happens. You might lose your job. I mean, and it's not necessarily fair for you lose your job, is the point.
Pistole: Every day that goes by that nothing bad happens to one of these planes, that's a good day. That's my perspective. Knowing what the threats are, knowing the intelligence, knowing how determined the bad guys are to do something in aviation... so every day that goes by is a good day.
Goldberg: But the point is, can you ratchet back in a significant way without worrying about the political risk?
Pistole: Yes. I mean, look, I got 26 and three-quarter years as a career person, I got five months on the political side. So there's no personal issues for me on that. All the debate over this for me simply comes down to, what is the most effective security to ensure that people get from here to there safely. I mean, that's the bottom-line equation for me. You can politicize, you can say whatever, but what can we do that balances the risk and security in a meaningful fashion? And right now, so I see us as doing that best possible job right now. I'm hoping, as the technology develops, that we can do less physically invasive.
Goldberg: Do you see a day when you won't be able to opt out for a pat-down, because this machine is going to do the job?
Pistole: Right now we have, well, depending on what Congress does with the budget going into 2012, we'll have a thousand machines out, more or less, by the end of next year. There are 2,200 lanes at security checkpoints around the country, so it's going to take a lot of additional funding to have those at each lane. Some would argue, well, that's a vulnerability, then. So how do you deal with that? And so that's part of this discussion. We can't be all things to all people at all places at all times, we can't eliminate risk, so how do we best try to mitigate risk by using our resources?
Goldberg: But you don't see a time when people won't be allowed to opt out?
Pistole: Not in the near future, anyway. Because we won't have these at all checkpoints, is the point.
Fallows: So one other question. If you want to destroy or terrorize American society, there are lots of other ways than aviation: you can blow up a truck in the Holland Tunnel, you can shoot people in a mall. Do you think that the terrorist obsession with aviation is an enduring thing?
Pistole: Going back to 9/11 and even Pan Am 103 from 1988, there's a fascination, I think, with blowing planes -- especially passenger planes -- out of the air. There is a psychological trauma that the terrorists see. That's their gold standard. So that being said, we've hardened our targets so well that my concern is, why haven't we seen a Madrid? 911 days after 9/11, March of '04, the subway, and then 7/7/05 in London, and of course Moscow, March of this year. Or somebody going into the the Mall of America and shooting that up?
Fallows: Or D.C. snipers.
Pistole: Yes, D.C. snipers. Well, aren't our queues for the airport vulnerable?
Goldberg: I wanted to get to that.
Pistole: Yes, they are vulnerable. Just like people stand in line to get their tickets at the ticket counter. Or stand in line to see a show. Or go shopping. So we'll never eliminate risk.
Goldberg: Yes. On that specific thing, since you brought it up, what are you doing to guarantee that those TSA security checkpoints are safe?
Pistole: Nothing.
Goldberg: Nothing!
Pistole: To guarantee that they're safe
Goldberg: Well, not guarantee.
Pistole: OK.
Goldberg: I'm using the wrong language obviously.
Pistole: So we work closely with airport police. Most major airports have their own police force. We have protection on officers. We have the canines. We have explosive tracers.
Goldberg: And they're out in the terminal, not just after security?
Pistole: Oh yes. They're not all in the sterile side. They're out in the area where people are congregating, and if there's someone who comes in that looks out of sorts or whatever, then the goal is for the air protection officers to see that person and do something. But look, we're doing a whole review, from the curbside to the cockpit. What is the best technology, what does the checkpoint of the future look like? So that's all underway. The TSA of the future, what should that look like? And I'd be open to your opinions. Ten years from now, in the year 2020, how should we be doing it, and what technology should we be using to accomplish that?
I've asked a number of people that, for their ideas, and we've asked internally, what should the transportation security officer, the TSO, the one that most people deal with- what should that position look like? What should the educational requirements, what should the professional requirements be? So there's a whole number of issues that I've undertaken a review of, looking to the future, recognizing that we have to protect as well as we can today. But I want to build a different TSA for the future, frankly. I want to see a different organization that can do a lot of things that I know worked, worked other places. But I need the support of Congress and the American people to do that.
Goldberg: You're talking about moving away from material inspection toward passenger inspection?
Pistole: Yes. Yes. Recognizing that we'll always need some kind of material inspection. What we have right now, frankly, is an interim solution to a long-term problem.
Goldberg: Do you think that the American people will stand for the privacy violation - the verbal privacy violation - of more passenger-based profiling security.
Pistole: Some will. Just like enhanced pat-downs.
It becomes a question of, those people who see their civil liberties or privacy violated, if they're one of those three percent who get an enhanced pat-down, how do they view more security. I think some people, they want to fly anonymously, with no security screening. That's their ideal. OK, we're never going to be that. Others are effusively supportive, saying "thank you for what you're doing. I want to make sure everyone else on that flight has been thoroughly screened." And you can't tell that because this person looks like an Arab male , he's a threat -- we've got the [non-Arab] McVeighs, we've got the Rudolphs, we've got the Kaczynskis, so there you go.
Goldberg: I once flew business on El-Al, and they had steak knives at dinner. I asked someone about it, and they said, "We know who everyone is on the plane."
Pistole: Yeah, what are you going to do with it?
Goldberg: I don't know if you'll ever get to that point, with scalability issues, but it's quite an image.
Pistole: That is. So I want to use the latest intelligence to inform our judgments and actions, and use the best technology when we don't have intelligence. So there's a huge gap there. So here are the threats, here are capabilities, here are gaps. So how do we fill those gaps? And right now we do it with a somewhat blunt approach.