The rocky road to security: travelers have faced 'no-fly' lists, more stringent searches, and increased demands for identification in the last few years. But are we more secure as a result? And at what expense to our civil liberties?

Since September 11, 2001, Americans' notions of security have changed dramatically, and government agencies have acted to put enhanced travel-safety measures into effect.

For example, the Transportation Security Administration (TSA) proposed a passenger screening program called Secure Flight. Under this program, the TSA, rather than the airlines, would compare passenger names against names of suspected terrorists. As the agency conducted preliminary testing for Secure Flight, it collected personal information on passengers from commercial databases as well as from airlines. The Government Accountability Of rice (GAO)said the TSA violated the Privacy Act by failing to fully disclose the scope of its testing and its collection of commercial data, and the program is now on hold until the GAO approves it.

Also on the horizon are national ID cards--a driver's license that contains more information--and passports that include radio frequency identification (RFID) chips, which operate like tollbooth speed passes and building-entry cards.

Some of these measures go too far, and some are ineffective to begin with, according to BARRY STEINHARDT, director of the American Civil Liberties Union's Program on Technology and Liberty. He spoke with TRIAL Associate Editor ALLISON TORRES BURTKA about how these efforts to increase security may leave people vulnerable to identity theft, discrimination, invasion of privacy, and other violations of constitutional rights.

TRIAL: No-fly lists have caught public attention for detaining well-known figures, such as Sen. Edward Kennedy, as suspected terrorists. What privacy concerns do these lists create?

Steinhardt: All Americans ought to be concerned about this. The TSA has now had four years to develop a workable program to screen passengers, and they've been utterly incompetent in doing it. The TSA is still doing nothing more than providing a poorly constructed no-fly list to the airlines, which have implemented it in a variety of ways, many of them badly. It's not their fault; the airlines are not in the business of security screening.

This failure is sort of the airline equivalent of failing to build the levees higher in New Orleans even though we knew that eventually a huge storm would come along and the levees would break. We know that the TSA doesn't trust the airlines with information about the most serious terrorist suspects. It's not even clear whether Osama bin Laden, for example, is on the no-fly list--or at least a no-fly list that's ever been provided to an airline.

The ACLU wouldn't oppose a simple program that would prevent someone we know to be or have good reason to suspect to be a terrorist from flying. But they keep trying to build these jury-rigged programs. It began with a program called the Computer-Assisted Passenger Profiling System (CAPPS), which changed to CAPPS II, then to Secure Flight--none of which, according to GAO studies, prove to be capable of either producing a real security benefit or protecting the civil liberties of passengers. So it's the worst of both worlds.

TRIAL: The Secure Flight program seemed to be a way to shift responsibility for checking passengers' names against terrorist watch lists from the airlines to the TSA. Where did it go wrong?

Steinhardt: With each iteration, Congress told the administration to demonstrate that the program was effective. Was it going to make us safer? Was it worth the money that we're proposing to spend on it? Each time those questions were put to the TSA and Department of Homeland Security, they responded in away that showed the answer to both questions was "no." They say, "We'll fix it, we'll make it better." But they haven't so far.

To the extent that terrorists are foolish enough to use their real names and to carry identity documents (which are easily forged) showing their real names, it makes sense to maintain a list of people who are known threats and to do a relatively simple match. This probably has a marginal security benefit. But that's not what Secure Flight is. They keep trying to build this Rube Goldberg contraption rather than a simple program that might actually work.

Nor are they willing to even consider the possibility that a sceening system may be fundamentally flawed--maybe we simply can't screen something like 12 million airline passengers a day. Perhaps we ought to continue concentrating on improving physical security, like reinforcing cockpit doors.

The TSA seems to miss the fundamental point that before you even reach the question of whether or not there is an adequate trade-off between the security benefit that we're receiving and the civil liberties consequences or harm that's being done, you've got to first demonstrate that there's some security benefit.

TRIAL: In the TSA's testing for Secure Flight, the agency ordered the airlines to provide passenger information and hired a private contractor to collect commercial data. Have problems already surfaced for people whose information was collected?

Steinhandt: We have no idea, because we're still finding out exactly what they looked at. The TSA filed a notice under the Privacy Act, saying they wouldn't use commercial data, and they went ahead and did it anyway. Then they filed an amended notice, swing they would use the data for testing purposes--after they had done it. And these are the people we're trusting with both our privacy and our security.

TRIAL: You've said that one of the most serious threats to privacy is the Real ID Act, passed in May 2005, which requires states to issue a national ID card. How is this card different from the average state-issued driver's license?

Steinhardt: This is a national ID on steroids. Everybody's card, whether you live in Alaska or Alabama, is going to be essentially the same, and one integrated database will contain all the information. It will be available to any government official. The statute requires that the information that's on the card also be available in a "machine readable" format, which might be an RFID chip. It might be a two-dimensional bar code; we don't know the exact physical structure yet.

It's going to be available to the private sector--stores, your landlord, and banks--as well. They can scan it and sell the information to others, who can then add it to their databases.

TRIAL: What type of biometric data is Real ID likely to include?

Steinhardt: We know for sure that it will display a digital photograph, but there's also discussion of fingerprints and iris scans.

TRIAL: How was the program proposed?

Steinhardt: I don't think Congress would have passed the act had they been asked straight on whether or not they wished to have a national ID. They passed Real ID under the guise of a uniform state driver's license.

The program is decentralized. Instead of having a single national ID card, Congress says to the states, "If you want your citizens to be recognized for any federal purpose--from boarding an airplane to opening a bank account--you shall have a driver's license that conforms to our standards; you shall participate in this nationwide database that's going to be widely available; it must have the following components in it, like a machine-readable stripe; and it must have the following information in it." The federal government imposes it, but the states carry it out.

TRIAL: Is the fact that Real ID was tacked on to an appropriations bill part of the reason it passed?

Steinhardt: Yes, it was tacked on to the appropriations bill that, among other things, funded the Iraq war. The bill's supporters knew that with our soldiers dying over there and the urgent need to fund the war effort, they could avoid a genuine debate over Real ID. The Senate actually would not have passed this straight up, but as an amendment to this bill, it became impossible to defeat.

The civil liberties community wasn't alone in opposing it. The National Governors Association, the National Conference of State Legislatures, and the American Association of Motor-Vehicle Administrators (the association of state DMVs) all claimed the law was unworkable and would intrude on state authority. These groups said: If we're going to propose a national ID, let's debate it openly. Let's not sneak it in the back door.

TRIAL: What practical problems does Real ID cause for states?

Steinhardt: To begin with, there's the problem of cost. Many states are going to have to completely remake their driver's licensing systems, which is going to be fantastically expensive. Estimates range from $12 billion to $15 billion for the states collectively. Congress has appropriated $30 million for the pilot project. This is a classic unfunded mandate. So we'll either pay for it through higher driver's license fees or increased state taxes, or the federal government is going to have to find the money to pay for it itself. Thirty million dollars in this case is a drop in the ocean.

Also, this has potential to be an identity theft nightmare. You're creating a honey pot of data that's going to be very attractive to identity thieves and other criminals. Already, thieves and insiders alike attack the systems of motor vehicle departments to steal personal information that can be used to make phony IDs and assume the identity of innocent Americans. The Real ID will be an identity thief's dream. Through fraud, they can create a universally accepted identity card that has your name on it but the identity thief's biometric data--the identity thief's physical characteristics. That is an identity theft from which you may never be able to recover.

We're still waiting for the Department of Homeland Security to issue the regulations. They tell us they're about to have meetings with some of the state officials, but I wouldn't bet the ranch on them making the 2008 deadline. When they meet with the states, they will have even a greater appreciation of just how daunting this task is.

TRIAL: Is it a valid concern that information in the national database might be shared with third parties on purpose?

Steinhardt: It could be. Under the Driver's Privacy Protection Act, states may not sell your personal information. But it will be available in the open market anyway. Your personal information will be embedded in the card and can be harvested every time a store uses it to verify your identity when you buy something. The store can simply scan the data and sell it to someone else--a data aggregator such as ChoicePoint, who will add it to the information they may have already collected about you.

Ultimately, these third parties can compile a database even richer than the government's--it may contain more inaccuracies but have more data in it. The card will be used as an identifier more often than driver's licenses are now. It will be required for all sorts of government and private transactions.

TRIAL: A few airports have adopted face-recognition technology as part of their general surveillance through video cameras, but it doesn't seem to be very accurate. Is anyone using it successfully?

Steinhardt: No. The technology just doesn't exist to accurately do "one-to-many" face recognition, identifying a specific individual by using a large database of photos of suspected terrorists or wanted criminals. The face-recognition industry has begun to concede that they can't do one-to-many matches very well.

TRIAL: Since September 11, the TSA has deemed some information "sensitive security information" that can't be disclosed in civil litigation, even under a protective order.. How does withholding such information complicate airline discrimination cases and other, related issues?

Steinhardt: It is not just in litigation; it is also withheld under the Freedom of Information Act. It makes it very, very difficult to know precisely what they're doing.

To give you an example, I sat at a meeting with representatives of the TSA and the Department of Homeland Security, and we were talking about their then-program doing aggressive pat-down searches, principally on women [after reports of female terrorists carrying bombs in their bras]. I was trying to find some common ground, and I asked them: At what point do they plan to replace the metal detectors, which can only detect metal, with more advanced technology that looks for the chemical signature of dangerous nonmetallic items--for example, plastic explosives? And I was told, with great insistence and a straight face, that they could not tell me whether or not the metal detector could detect plastic, because that was sensitive security information.

If you step back, you realize how absurd that is. First, it's not exactly a state secret that a metal detector detects metal and not plastic. But you can't have a rational conversation with an agency that is so secretive that it won't concede that the metal detector detects only metal.

TRIAL: What happened with the aggressive searches?

Steinhardt: They ultimately were forced to change this policy, which allowed screeners to conduct demeaning pat-downs on female passengers--from 2 to 90 in age--searching for "unusual body form." Of course, no one alive knows what that means.

The policy was adopted after two planes were bombed in Chechnya. Investigators suspected that the bombers were two women who got the explosives onto the planes by carrying them in their bras. Initially, the screeners had to follow explicit rules in doing these searches, but eventually they were given unlimited discretion--and, not surprisingly, some screeners were being very aggressive and abusing it. It's since been scaled back.

TRIAL: The ACLU filed suit against the New York Police Department in August 2005, alleging Fourth Amendment violations for suspicionless and random searches in subways. Do you think these searches were random to avoid the appearance of racial profiling?

Steinhardt: I'm sure that's the rationale. The problem with this effort is that it's cosmetic. It is largely designed to make people feel better. The New York City subway system is immense, and searching it on a systematic basis would require a huge law enforcement presence, which simply isn't practical. So put aside whether the searches are random in the sense that they pick out every sixth passenger--they're random in the sense that the police can't search every station. And they're easily evaded.

TRIAL: You can just get on the subway at a different station, right?

Steinhardt: Yes, and you may even be able to get on at a different entrance to the same station.

The first question ought to be: Does it work? Does it make us any safer? Because if it's a waste or a bad use of police resources, why do we need to even argue about the constitutionality of it? They shouldn't be doing it. They should be doing something that's more effective and has a greater chance of success.

TRIAL: Is there some way to conduct searches that would be more effective?

Steinhardt: If you're looking for the next threat to transportation, it may well be to something other than a subway system. That's the paradigm they look at in New York now because that's what happened in London. But, in the words of a former federal security official, the terrorists "adapt." They switch their targets. They've gone from airplanes to trains to subway systems in London, and there's no reason to believe they won't move to another target of opportunity. It doesn't make a lot of security sense to be doing ineffective searches on subways, when the terrorists are just as likely to have other targets.

We're arguing about civil liberties principles as if there was some tension between civil liberties principles and effective law enforcement. They've got to demonstrate first that there's effective law enforcement.

TRIAL: How do issues of discrimination complicate matters?

Steinhardt: Frequently, the discrimination issue comes to the fore because the suspicion is based on ethnicity or race or, in the case of these aggressive pat-down searches, on gender. These can result in acts of discrimination or due process and privacy violations, ff you look at all the activity around airline passenger profiling and the use of lists to try to pick out people who require additional scrutiny, there's not a lot of rhyme or reason to it, and that's part of the reason it's been so controversial.

TRIAL: People are more willing to surrender their civil liberties in times of war. How does that make it more difficult to protect Americans' privacy?

Steinhardt: The so-called war on terrorism, which has really been a war on our liberties, has greatly accelerated the creation of what amounts to a surveillance society fueled by new technology and a weakening of the laws that protect us.

As lawyers--and as Americans--we need to begin demanding that chains be put on this growing surveillance monster, before it devours our privacy and our liberty.