How much liberty should we be willing to give up in order to make ourselves safe from terrorist attack? Few deny that some trade-off is necessary. The terrorists who attacked New York and Washington on September 11, 2001, took advantage of the liberties we enjoy and some of those liberties need to be curtailed if other terrorists, following their example, are not also to take advantage of them with similar or even more terrible results.

But trading off liberty against security has a treacherous logic. It beckons us in with easy cases—the trivial amount of freedom restricted when we are made to take our shoes off at the security checkpoint before we board an airplane is the price of an assurance that we will not be blown up by any imitators of Richard Reid. But it is also a logic that has been used to justify spying without a warrant, mass detentions, incarceration without trial, and abusive interrogation. In each case, we are told, some safeguards and rights that were formerly regarded as civil liberties have to be given up in the interests of security. But after a while we start to wonder what security can possibly mean, when so much of what people have struggled to secure in this country—the Constitution, basic human rights, and the rule of law—seems to be going out the window.

When logic betrays us, we have to retrace our steps—sometimes in a fussy and pedantic sort of way. So let’s think carefully about the trade-off between liberty and security. One crucial distinction is between trade-offs involving only our own personal situations and interpersonal trade-offs involving others.

A simple case of a personal trade-off is this. I accept the burden of a legal requirement to wear a seatbelt, restricting my freedom to sit in my car as I like, because I am convinced that this will make me safer, less liable to injury or death in the event of an accident. If we all do this, then each of us is safer though each of us is a little less free. We can think of it as a straightforward trade-off, once we understand what happens to human bodies in automobile collisions. It’s something like buying more potatoes (safety) and less meat (liberty) when we find that meat is more expensive than we thought.

Another similar case, slightly less straightforward, is when we all accept a restriction on liberty not because our own actions pose a threat to our own safety, but because it is possible that some of us may pose a threat to the others and we don’t know who. This is the logic of the airport security system; and it too seems to make innocuous sense. We all accept certain restrictions in the expectation that we will all enjoy greater security. Again, the trade-offs affect our own well-being: each of us bears the cost and each of us reaps the benefits.

Quite different, however, is the interpersonal case, in which we sacrifice not our own liberty but the liberty of a few people in our midst in order that the rest of us may be (or at least feel) more safe. A passenger notices some Muslim men praying before boarding an aircraft. She makes a fuss and the Muslims are removed from the flight.1 Some liberty is lost, and perhaps some security is gained. But the person who gains the security is not the person who lost the liberty. This is utterly unlike the trade-off by which one wears a seat belt in order to gain more security. It’s a different game: a game of majorities and minorities.

It is different but of course it’s not unusual. For there are winners and losers all the time in politics: a new highway benefits some restaurant owners at the expense of others whose establishments languish boarded up along the route of the old road. But the stakes are much higher in the trade-off between liberty and security. For what is traded off in that case is not just economic interests or mundane freedoms, like the freedom to drive without a seatbelt. Often what is traded off is something that was previously regarded as a right, and the loss of that right may simply be imposed on the people affected. Members of a minority are detained without trial, or spied upon, or beaten or humiliated during an interrogation, and all to make the rest of us more secure. This is troubling because rights are supposed to be guarantees given to individuals and minorities about the outer limits of the sacrifices that might reasonably be required of them. Rights are supposed to restrict trade-offs, not be traded off themselves.


David Cole is a professor of law at Georgetown University, a legal affairs correspondent for The Nation, and a frequent commentator on constitutional and civil liberties issues on public radio and television. Very early on in the discussion of these matters he distinguished himself by refusing to accept the bromide that we were all giving up some of our freedom in order to make all of us more safe. At the end of 2002, Cole published a short piece in the Boston Review entitled “Their Liberties, Our Security,” which traced not just the unequal impact but the discriminatory intention of many of the liberty-affecting measures imposed after September 11.2


Cole focused particularly on the Justice Department’s preventive detention of thousands of young men of Arab and Muslim descent living in the United States in 2001 and 2002, and its enactment of a new regulation extending the time they could be held and interrogated before they had to be charged or released. Cole reminded his readers that the Constitution does not permit detention for purely investigatory purposes. But we gave up this protection for men whose religion, ethnicity, or appearance reminded us of the September 11 hijackers, imposing on an entire class of residents who were not citizens deprivations of rights that we would not tolerate if they were imposed on ourselves. “As a way of striking the difficult balance between liberty and security,” said Cole in his article, “sacrificing foreign citizens’ liberties is undoubtedly tempting…. We can avoid the difficult trade-offs, and have our security and our liberty, too.”

The title of a new book that he has written with Jules Lobel, a professor of law at the University of Pittsburgh, suggests that Professor Cole now wants to take this analysis a step or two further. Less Safe, Less Free argues that the trade-off is not just unprincipled and unequal, but a fraud. We sell our freedom (or someone’s freedom) to make ourselves safer, but it turns out that we are worse off in regard to safety than we would have been without the trade. It is as though we gave up the meat (or someone’s meat) without getting any more potatoes.

Even here, it is necessary to proceed carefully with this treacherous talk of trade-offs. Earlier we asked: “Whose freedom is being sacrificed?” Now, as we assess the argument of Less Safe, Less Free, we have to ask, “And who, exactly, is turning out to be less safe?”

For it is not only the liberty of a minority that is affected by measures taken to combat terrorism. Sometimes their personal safety is affected as well. On July 22, 2005, a Brazilian electrician, Jean Charles de Menezes, was shot dead by police officers in Stockwell tube station on the London Underground. Those who shot him said they thought he looked and acted like one of the terrorist suspects they were watching in anticipation of a possible repetition of the attacks in London earlier that month. But Mr. Menezes was not a terrorist; he was an entirely innocent and legal resident of the United Kingdom, and he was at the time doing nothing that justified firing on him. He was made radically less safe as a result of Britain’s anti-terrorist measures, which included instructions to police to use deadly force to prevent perceived terrorist activity.

The same can be said about those who have been beaten and tortured—some beaten and tortured to death3—by American security forces. The infliction of pain during interrogation renders a person not just less free—though he has to be made unfree (held down) in order to be tortured—but less safe, less secure in a very straightforward sense. The security that we all crave is security against violent attack, but that is exactly what many people lose when they are imprisoned in Guantánamo Bay or in “black” US prisons in Eastern Europe, or when they are “rendered” by US agents to countries like Syria for torture by foreign authorities. Their security is sacrificed in order to make the rest of us more safe.

Cole and Lobel begin their book with the story of Maher Arar, a Canadian engineer, innocent of any involvement in terrorism.4 In September 2002, Arar was apprehended at JFK Airport while changing planes for Montreal. He was held for two weeks in Brooklyn while US officials investigated charges that he had ties to al-Qaeda. Then, while his lawyer was lied to about his whereabouts, Arar was flown on a US government–chartered jet to Jordan and driven to Syria, imprisoned there for a year in a grave-like cell, and tortured by Syrian authorities (for several days beaten for hours with a frayed electrical cable).

The former US attorney general Alberto Gonzales has said that there were assurances from Syria that Arar would not be tortured. But this is very odd. We don’t trust the Syrians’ word on anything else and had the administration wanted to ensure that Arar’s deportation would not result in his torture, they could have sent him to Canada, whose passport he held and where he resided as a citizen. As Cole and Lobel point out, the only conceivable reason for sending Arar to Syria was so that he would be tortured, and tortured in circumstances where no legal recourse was possible.


People like Maher Arar and hundreds of others who have been abused by our interrogators are not more secure from terror as a result of the Bush administration’s security strategy. On the contrary, they are terrorized by us: their terror has been instrumentalized by officials of our government supposedly for our benefit. We may be more secure as a result, but it is a shameful thing to know that our safety has been purchased on the back of a waterboard or at the end of a frayed electrical cable.


But Cole and Lobel go further than this in questioning the security side of the liberty–security trade-off. Apart from the people who have suffered pain and death at the hands of the American authorities, is it even true that the rest of us are safer as a result of these measures?

The Justice Department maintains a Web site—www.lifeandliberty.gov/ subs/a_terr.htm—where it boasts of the Bush administration’s achievements in combating terrorism: nearly two thirds of al-Qaeda’s senior leadership captured or killed, over 150 terrorist threats identified and disrupted, more than three thousand terrorist operatives incapacitated, terror cells broken up in Buffalo, Portland, and elsewhere, 401 people charged in terrorism-related offenses, more than 515 individuals deported, and so on. The most effective chapter in Less Safe, Less Free is the one in which Cole and Lobel go through this laundry list item by item to show how we are being misled.

For example, the authors argue that the 515 individuals deported must refer to immigrants who had expired visas or who were here illegally, but who have been cleared of terrorist involvement, since the FBI does not allow the deportation of those who are suspected of terrorism (except, as in the case of Maher Arar, to render them up to other countries for torture). Where the Web site refers to 401 people charged with terrorism-related offenses, the phrase “terrorism-related” seems to refer to any charge that emerges as a result of the investigations begun after September 11; many of them are for ordinary immigration offenses or for making false statements on a federal form, such as using a false Social Security number. As for the 150 terrorist threats identified and disrupted, a careful reading of the Web site’s warning suggests that these include plots foiled by our allies abroad as well as threats against the United States.5

Cole and Lobel show that almost all of the plots against America that are supposed to have been thwarted have involved little more than idle threats or “preliminary ideas about potential attacks, not terrorist operations that were about to be carried out.”6 (The closest we have come to frustrating a real plot on the point of being put into effect is the apprehension, trial, and imprisonment of Richard Reid; and even there, as Cole and Lobel observe, the plot was foiled not by the Department of Homeland Security, but by a flight attendant who noticed a strange-looking man trying to set fire to his shoe.)

You may ask: “Well, why shouldn’t the administration receive credit for frustrating plots that are at an earlier rather than a later stage of planning? Surely that’s the safest strategy.” Cole and Lobel acknowledge the point. But no one knows whether these are plots that would have resulted in action. What we have are conversations overheard or recorded by agents provocateurs. Compare this to societies that are actually under attack. There is nothing in the record of the Bush administration corresponding, for example, to the numbers that the Israeli security forces report of people shot or intercepted with bombs strapped to their bodies.7 It is not as though the United States is facing frequent terrorist attacks and has managed to fend off 150 of them.

Worse still, action to intercept “plots” at the earliest, inchoate stage may well be counterproductive. One gets the impression that the possibility of acquiring more information by trailing the conspirators for a while has been sacrificed to the need to publicize arrests. (Something has to be posted on the Web site.) And rounding up anyone who says anything vaguely dangerous is probably a way of alienating the very population whose cooperation we need. Cole and Lobel write:

“Better safe than sorry” is the administration’s mantra, and it makes sense where there is solid evidence of concrete plans. But when that impulse leads to the arrests, renditions, military detentions, and pretextual prosecutions of thousands who pose no actual concrete threat, and about whom officials merely have vague, factually unsupported suspicions, it is far from clear that we are better off.

Resources are wasted and “corrosive distrust” is generated in immigrant communities. In the long run, say Cole and Lobel, “the resentment provoked by these measures is the greatest threat to our national security, and the most likely source of the next attack.”

This brings us to the other main argument that Cole and Lobel make. The Bush administration’s strategy for fighting terrorism—preventive detention and abusive interrogation of large numbers of people who may have terrorist connections—almost certainly undermines the very security it is trying to protect.

If the United States is to deal with the threat of terrorism effectively, it needs to secure the cooperation of other countries, the support of minority communities within the United States, and the willingness of those who might otherwise be tempted to engage in terrorism to act peacefully. Cole and Lobel argue that on all three fronts, the administration’s preventive strategy has been a disaster. Extravagant dragnet campaigns have alienated Arab-American and other Muslim communities at home. The administration’s use of abusive interrogation, extraordinary rendition, and indefinite detention without trial has led to a stunning decline in America’s reputation abroad and the willingness of others to work with us. (The effect of the Maher Arar case on future cooperation with the Canadian authorities is a case in point.) A heading in the section of the new US Army/Marine Corps Counterinsurgency Field Manual devoted to detention and interrogation says simply: “Lose Moral Legitimacy, Lose the War.”8 The military knows a thing or two that our civilian leaders have never learned. Legitimacy is an enormous asset in a struggle of this kind, and squandering it is one of the easiest and most dangerous things that a government can do.

The administration’s “Life and Liberty” Web site claims that many of al-Qaeda’s operatives have been incapacitated. That is surely a significant achievement, but the contribution it makes to our security depends on how easily the foot soldiers are replaced. It is a fallacy to think that each time a terrorist is detained or killed, there is one less terrorist in the world. If the methods by which he was incapacitated stir up enough rage, it may mean that there are two or three or ten more terrorists. One analysis cited by Cole and Lobel asks us to think of two concentric circles: a core circle of hard-line fanatics already committed to terrorism as a strategy, and a second circle, larger than the first, of disaffected and vulnerable young men who might be persuaded to join the inner circle if their anger and desperation become great enough.9

Obviously movement toward the core is affected by (among other things) what these young men know about (and how they brood upon) what Americans are doing to their fellow countrymen and their coreligionists in Iraq, Afghanistan, and elsewhere in the “war on terror.” And here the signs are ominous. Arab children carry torture images from Abu Ghraib on their cell phones and al-Qaeda recruitment videos calmly use what is known of US detention and interrogation policies to convince viewers that Americans have no respect for the dignity of Muslims.

The Bush administration thinks that if it is known in the terrorist community that detainees will not get the benefit of ordinary rules that prohibit torture, arbitrary detention, and disappearance, then that will serve as a brutal deterrent. But Cole and Lobel argue that the truth is exactly the opposite:

When we sweep away the rules in the name of preventing the next attack, we foster the conditions that make the next attack that much more likely.

For their part, they recommend “noncoercive preventive safeguards” (like enhanced port security) and multilateral approaches—including institutions like the UN Security Council and even the International Criminal Court—to address this transnational problem in a way that does not alienate the countries that matter in this struggle. The rule of law, they say in a final chapter, needs to be treated as an asset in the fight against terrorism, so that what is known of our practices of detention, surveillance, and interrogation does not make the problem worse.

Whether these options are actually available to the United States right now in a form that holds out any hope of redeeming the legitimacy that we have squandered over the past six years is of course another question. The authors’ cheerful talk at the end of the book about “the key to success” and “the best way forward” does not sit so well with the bleakness of the rest of their analysis—at least, not until one notices how often the phrase “future administrations” is used in the chapters presenting their suggestions.


Meanwhile, how safe are we, in the light of all this? The one fact we do know is that (at the time of writing) there have been no successful terrorist attacks at all on the United States homeland since September 11 (apart from Mr. Reid’s fizzling shoe-bomb and the anthrax scare of late September 2001). That is something for everyone in this debate to reckon with.

The Bush administration, not unreasonably, adduces the lack of any attacks as evidence that its preventive strategy has worked. Surely “no further attacks” is the best bottom line one could look for on the security side (as long as it lasts). On the other hand, some have argued that the threat may have been overblown all along.10 Maybe for all its bluster al-Qaeda has no particular desire to attack the American homeland again, having achieved that one remarkable success. In which case, the administration has sacrificed liberty and the rule of law and the security of a minority for nothing.

For Cole and Lobel, on the other hand, the fact of no further attacks since September 11 is a sort of double embarrassment. On the one hand, it creates an impression that the Bush administration’s strategy has worked. To dispel this, they argue that there is scant evidence that any attempted attack has been thwarted. But they can’t go too far in that direction. For their position is that the administration’s tactics are making us less safe and more likely to be attacked. That position is also imperiled by the hard fact of our having suffered no homeland attack at all during the time in which, by the authors’ own lights, the Bush administration’s policy has been making such an attack more and more likely. It’s a sorry state of affairs when one gets the impression that both sides would be boosted by another attack: the administration would have proof that the threat remained real and its opponents would have evidence that the administration has gambled recklessly with our security—and lost.

Should we ever have felt insecure? There’s a sort of game that academic commentators play which involves comparing the prospect of any of us being caught up in a terrorist attack to various other risks we face, like being struck by lightning or accidentally drowning in a bathtub. In all three cases, the probability of death for any given American is microscopic and our reaction to September 11 can be made to seem hysterical.11

But the game is a cheat. For one thing, the perils are cumulative, not alternatives: we now have to worry about lightning, bathtubs, and terrorists. More important, the risk assessment assumes a strange sort of atomism among the American people, as though each person’s security were utterly independent of anyone else’s. On September 11, 2001, I was one of the 99.999 percent of the US population who were not killed: and had there been three such attacks instead of one, I would still almost certainly have been one of the 99.997 percent who were not killed. So why should I worry about whether that one attack would be repeated? It would affect someone but it almost certainly wouldn’t affect me.

If the right way of reckoning the extent of my safety uses the whole population of the country as a denominator, then the odds are always going to look derisorily low. And they would continue to look derisorily low even if the number of attacks rose to a level that would plainly be socially and politically intolerable. Moreover, individual odds calculated in this way are so microscopic that margins of error are likely to make them appear indifferent to both the consequences of good counterterrorism strategies and the consequences of bad counterterrorism strategies. Clearly something is wrong with this way of proceeding. No amount of lecturing the American people on lightning, bathtubs, and the finer points of probability theory is going to make them shrug off a repetition or two of September 11.

Cole and Lobel flirt a little with this sort of analysis. They cite a poll conducted a few weeks after September 11, which found that the average person believed he or she faced a 20 percent chance of being injured in a terrorist attack within the next year. Their response is scathing:

In fact, statistically speaking, Americans would not have faced that degree of risk even if a terrorist attack of the same magnitude as the 9/11 attacks took place every day for an entire year.

Silly us.

In my view, what all this shows is that security is not just an individual good, enjoyed by each of us as a matter of statistical probability. Although, as we have seen, some people can be made radically insecure in an atmosphere of general safety, in the end security is a collective good that makes possible an environment in which we can live and deal with one another and pursue the various ways of life that people are accustomed to in this country.12 We are none of us secure just because we are physically unscathed; and a whole population is not secure just because its members can all emerge unhurt after weeks of cowering in a sealed room. The security for which we might be tempted to trade off some of our freedom has got to be more than that.

This leads us to one last point about security and liberty. When we trade off meat and potatoes, we assume they are independent commodities: potatoes are not made of meat and meat is not made out of potatoes. We assume something similar too when we talk of a trade-off between liberty and security. But that may be a fallacy.

Once we abandon the crude statistical view of security—my one-in-a-million lifetime chance of being a terrorist victim—and think more collectively about the security of our shared way of life, then the sources of insecurity multiply. The way of life we value includes movies and baseball and church and the supermarket. But it also includes civil liberties and the rule of law; it includes our reputation as an honorable people, respectful of human rights; it includes a deep and resilient expectation that, as Winston Churchill put it, if there is a knock on the door at five o’clock in the morning, it’s probably just the milkman. Our security, on this account, took a terrible hit when the September 11 terrorists attacked; but it also took another hit—a repeated and ongoing hit—when the Bush administration officials abandoned the rule of law and the principles of the Constitution, and began abusing and detaining people whose appearance and ethnicity reminded them of terrorists.

There has been no attack since September 11, but we cannot rule out the possibility that there will be one next year or the year after. It would not be a bad idea to worry in advance, not just about the damage that the terrorists might inflict, but also about the damage we may do once again to ourselves, to our honor and to our traditions, in the panic and anger that would undoubtedly beset us. I fear that this part of the damage could be very bad indeed, worse than anything endured in the last six years.

Very few terrorists are content with the murders they commit and the physical havoc that they wreak. They reckon to undermine our security in other ways too, by eliciting a panic-stricken overreaction which will further erode our power, undermine our legitimacy, and contribute to our general demoralization (in the literal sense of that word). They take aim at our way of life and they know that civil liberties and the rule of law are part of that target, even if our leaders have forgotten. If they do attack again, I hope we will have the likes of David Cole and Jules Lobel to help us think through our response. But the better prospect, surely, is to include in our thoughts about security, right now, ways of protecting our law and traditions against such an overreaction.

—September 25, 2007

This Issue

October 25, 2007