TESTIMONY OF

PAUL ROSENZWEIG

SENIOR LEGAL RESEARCH FELLOW

CENTER FOR LEGAL AND JUDICIAL STUDIES

THE HERITAGE FOUNDATION*

214 MASSACHUSETTS AVENUE, NE

WASHINGTON, DC 20002

BEFORE THE UNITED STATES COMMISSION ON CIVIL RIGHTS

REGARDING

ANTI-TERRORISM EFFORTS, CIVIL LIBERTY

AND CIVIL RIGHTS

19 MARCH 2004

Good morning Madam Chairperson and Members of the Subcommittee. Thank you for the opportunity to testify before you today on the challenge of maintaining the balance between security and constitutionally protected freedoms inherent in responding to the threat of terror, in the particular context of civil rights laws and obligations.

For the record, I am a Senior Legal Research Fellow in the Center for Legal and Judicial Studies at The Heritage Foundation, a nonpartisan research and educational organization. I am also an Adjunct Professor of Law at George Mason University where I teach Criminal Procedure and an advanced seminar on White Collar and Corporate Crime. I am a graduate of the University of Chicago Law School and a former law clerk to Judge Anderson of the U.S. Court of Appeals for the Eleventh Circuit. For much of the past 15 years I have served as a prosecutor in the Department of Justice and elsewhere, prosecuting white-collar offenses. During the two years immediately prior to joining The Heritage Foundation, I was in private practice representing principally white-collar criminal defendants. I have been a Senior Fellow at The Heritage Foundation since April 2002. 

My perspective on this matter is that of a lawyer and a prosecutor with a law enforcement background, not that of technologist or an intelligence officer/analyst. I should hasten to add that some of my testimony today is based upon a series of papers I have written on various aspects of this topic and testimony I have given before other bodies in Congress, all of which are available at The Heritage Foundation website (www.heritage.org). Other portions of my testimony today are derived from a forthcoming law review article entitled "Civil Liberty and the Response to Terrorism" which will be published in the Duquesne Law Review Spring 2004 issue. For any who might have read portions of my earlier work, I apologize for the familiarity that will attend this testimony. Repeating myself does have the virtue of maintaining consistency -- I can only hope that any familiarity with my earlier work on the subject does not breed contempt.

* * * * *

The subject for today is, of course, an exceedingly broad one. If we discussed all aspects of anti-terrorism, civil liberty and civil rights we might reasonably spend the next several days. Potential issues for discussion range along the road from traditional civil rights questions, to immigration, to racial profiling, to employment discrimination. And the scope of government action is equally broad - including programs under the USA PATRIOT Act, Executive Orders, and actions taken under the President's traditional war powers. From this wide range of options each speaker must choose some more narrow focus. I have therefore chosen to focus my testimony on two aspects of the question - the USA Patriot Act and racial profiling. I would, of course, be happy to try to answer questions in other areas of concern if you wish.

Patriot Act -- I believe that a governing rule for assessing our response to terror can be readily summarized from the writings of Chief Justice Rehnquist. He wrote: "In any civilized society the most important task is achieving a proper balance between freedom and order. In wartime, reason and history both suggest that this balance shifts in favor of order - in favor of the government's ability to deal with conditions that threaten the national well-being."[1]

Everyone does not share Chief Justice Rehnquist's vision of the balance between liberty and order. The past several months have seen the growth of a new movement - call it the "anti-anti-terrorism" movement, if you will. The thesis of the movement, which has some of the appearances of a political campaign, is that steps being taken domestically to combat the potential for terrorist attacks are too intrusive and a threat to cherished civil liberties. 

The principal focus of the campaign is the USA PATRIOT Act,[2] a law passed with overwhelming support in Congress immediately following the September 11 terrorist attacks. Taking many forms, the campaign argues that various provisions of the Patriot Act, and related laws and practices, have greatly infringed upon American liberties, while failing to deal effectively with the threat of terror. Criticism of the anti-terrorist campaign is not, however, limited to the Patriot Act - many other aspects of the Bush Administration's domestic response to terrorism have come under fire. To some degree, the Patriot Act as conceived by the public is broader than its actual provisions. Its very name has come to serve as a symbol for all of the domestic anti-terrorist law enforcement actions. It has become, if you will, a convenient short hand formulation for all questions about the alteration in the balance between civil liberty and national security that have occurred since September 11.

There are two over-arching themes that animate criticism of the Patriot Act (using the phrase now in the broad, symbolic sense already noted): First, critics of the Patriot Act frequently decry the expansion of executive authority in its own right. They, generically, equate the potential for abuse of Executive Branch authority with the existence of actual abuse. They argue, either implicitly or explicitly, that the growth in executive power is a threat, whether or not the power has, in fact, been misused in the days since the anti-terrorism campaign began. In essence, these critics come from a long tradition of limited government that fears any expansion of executive authority, notwithstanding the potential for benign and beneficial results, because they judge the potential for the abuse of power to outweigh the benefits gained.

The second theme of many criticisms of the Patriot Act and other government responses is one we might call a fear of technology. In service of our efforts to combat terrorism the government has begun to explore ways of taking advantage of America's superior capacity to manage data through new information technologies. The Transportation Security Administration's proposal for a new computer-assisted passenger-screening program (CAPPS II) is one such program.

These new technologies offer two advantages over current investigative practices - they have the potential to both expand the ambit of the information available to federal law enforcement and intelligence agencies and to enhance the efficiency with which those agencies are able to examine and correlate information already in their possession. And both possibilities raise corresponding fears among critics of the programs. Expanded access to information increases executive power. And with great efficiency comes more effective use of power. Thus, the hesitancy to use new technology, though sometimes born of technological apprehension, also resonates with the principal theme of critics, a reluctance to expand the capacity of the government to examine the lives of individuals.[3]

Criticism of the Patriot Act, however, sometimes misapprehends important distinctions: First, the criticism often blurs potential and actuality. To be sure, many aspects of the Patriot Act (and other governmental responses) do expand the power of the government to act. And Americans should be cautious about any expansion of government power, for assuredly such expansion admits of the potential for abuse. But by and large, the potential for abuse of new Executive powers has proven to be far less than critics of the Patriot Act have presumed it would be. 

Second, much of the belief in the potential for abuse stems from a misunderstanding to the true nature of the new powers that government has deployed to combat those threats. To a surprising degree, opposition to the executive response to terror is premised on a mistaken, and sometimes overly apocalyptic, depiction of the powers that have accrued to the government.

More fundamentally, those who fear the expansion of executive power in the war on terrorism offer a mistaken solution - prohibition. While we could afford that solution in the face of traditional criminal conduct we cannot afford that answer in combating the threat of terror. In the context of current circumstances, vigilance and oversight, enforced through legal, organizational and technical means, are the answer to potential abuse - not prohibition. We must keep a watchful eye to control for the risk of excessive encroachment, but if we do the likelihood of erosion of civil liberties can be substantially reduced. 

Thus far, I believe we have succeeded in meeting that goal. With respect to the Patriot Act (now using those words in the narrower and technical sense of a particular law), the record is, in fact, one of success. The Inspector General for the Department of Justice has reported that there have been no instances in which the Patriot Act has been invoked to infringe on civil rights or civil liberties.[4]  This is consistent with the conclusions of others. For example, at a Senate Judiciary Committee Hearing on the Patriot Act Senator Joseph Biden (D-DE) said, "some measure of the criticism [of the Patriot Act] is both misinformed and overblown." His colleague, Senator Dianne Feinstein (D-CA) said: "I have never had a single abuse of the Patriot Act reported to me. My staff . . . asked [the ACLU] for instances of actual abuses. They . . . said they had none." Even the lone Senator to vote against the Patriot Act, Russ Feingold (D-WI) said that he "supported 90 percent of the Patriot Act" and that there is "too much confusion and misinformation" about the Act.[5] These views -- from Senators outside the Administration and an internal watchdog -- are at odds with the fears often expressed by the public.

The Report of the Inspector General is particularly instructive in this regard. According to the IG, the Patriot Act identifies certain specific ethnic groups that would be vulnerable to potential abuse from backlash due to the terrorist attaches of September 11. These include Muslims, Arabs, Sikhs and South Asians. Between June 2003 and December 2003 (the most recent reporting period), the IG received 1,266 complaints suggesting potential civil rights or civil liberties violations (including many that were not within the IG's or DOJ's jurisdiction). Of these 720 were deemed "unrelated" - that is they either cited no improper act by any DOJ employee/contractor or identified no discernible nexus between the alleged conduct and any civil rights/civil liberty violations. Many of these complaints appear to have been frivolous (e.g. allegations that the government was broadcasting harmful electronic signals at an individual). 

Another 384 were complaints outside of the IG's jurisdiction because, for example, they alleged acts by local law enforcement or private businesses. Some of these were significant allegations (e.g. of excessive use of force by local police) and others were less serious (e.g. that INS or TSA inspectors were rude). In any event, all of these allegations, whether true or not, relate to traditional law enforcement issues and are unrelated in any way to the Patriot Act.

Thus, at bottom only 162 complaints were within the scope of the Department's activity (and thus, within the jurisdiction of this Commission's purview). But these 162 were examined simply because they made a prima facie claim of a violations. Examples included alleged excessive force used by BOP officers and alleged fabrication of evidence by FBI agents. In then end only 17 of these investigations were deemed to warrant the opening of an investigation and substantial review.

It's important to stop for a moment and remark on that figure. Since September 11, DOJ agents in myriad capacities have encountered common citizens in literally hundreds of thousands of different situations, perhaps millions: from simple interviews, to full-scale searches, and everything in between. And from those many, many interactions, only 17 warranted close review. That is a remarkably low rate of error - even assuming that all 17 investigations in fact identified actual violations (an issue as to which the IG's report is silent). 

[I should note that IG reports closing several investigations, some of which found improper conduct in, for example, the verbal abuse of prisoners. To be sure, such activity by individual officers of the Bureau of Prisons, or INS is to be condemned. But as with the IG's earlier reports on activity at the Metropolitan Detention Center, there is little suggestion of any systematic or deliberate malfeasance. Rather, the misconduct is (as always) the product of individuals who are inadequately trained and/or overseen. In this it is notable that the IG's report commends DOJ for its response to the earlier report on problems at the Metropolitan Detention Center. "The DOJ has taken significant and responsible steps to implement the OIG's recommendations."]

Finally, even more remarkable is the conclusion of the IG with regard to the Patriot Act itself: "None of the 162 matters [within the IG's jurisdiction] involved complaints alleging misconduct by DOJ employees related to their use of a substantive provision of the Patriot Act." Far from there being any actual violations, there were not even any colorable allegations of a violation of civil rights or civil liberties under the Act. And, notably, this conclusion is from an IG who has not been reticent to criticize the Department where appropriate. 

Racial or National Origin Profiling - Racial profiling poses a deeply difficult and intractable problem. As a society we reject general reliance on immutable characteristics such as race or gender. On the other hand the problems of terrorism pose new and greater dangers. Whenever I teach this aspect of police conduct to my law students, I always emphasize that the proper way to define the "reasonableness" of law enforcement activity is to assess three separate values - the degree of intrusion occasioned by the activity; the harm being averted; and the "closeness of the fit" between the scope of activity in question and the harm being averted. 

Looked at through this prism of analysis, it is easy to see why most racial profiling is wisely rejected. Typically, the harm being averted is a general common law crime and the "fit" is poor, at best and often non-existent. Profiling African-Americans for driving on certain roadways fits in this category. If it exists, it is unjustifiable and unconstitutional

But this also suggests that, in very limited circumstances, the balance might change when the object of our activity is to prevent terrorism, and the use of national origin data and characteristics is much more narrowly applied. Let me begin with some theoretical points that are broadly applicable:

The danger to America posed by terrorists arises from the new and unique nature of potential acts of war. Virtually every terrorism expert in and out of government believes there is a significant risk of another attack. Unlike during the Cold War, the threat of such an attack is asymmetric. In the Cold War era, U.S. analysts assessed Soviet capabilities, thinking that their limitations bounded the nature of the threat the Soviets posed. Because of the terrorists' skillful use of low-tech capabilities (e.g. box cutters) their capacity for harm is essentially limitless. The United States therefore faces the far more difficult task of discerning their intentions. Where the Soviets created "things" that could be observed, the terrorists create only transactions that can be sifted from the noise of everyday activity only with great difficulty. It is a problem of unprecedented scope, and one whose solution is imperative if American lives are to be saved.

As should be clear from the outline of the scope of the problem, the suppression of terrorism will not be accomplished by military means alone. Rather, effective law enforcement and/or intelligence gathering activity are the key to avoiding new terrorist acts. Recent history supports this conclusion.[6] In fact, police have arrested more terrorists than military operations have captured or killed. Police in more than 100 countries have arrested more than 3000 Al Qaeda linked suspects,[7] while the military captured some 650 enemy combatants.[8]  Equally important, it is policing of a different form - preventative rather than reactive, since there is less value in punishing terrorists after the fact when, in some instances, they are willing to perish in the attack.

The foregoing understanding of the nature of the threat from terrorism helps to explain why the traditional law enforcement paradigm needs to be modified (or, in some instances, discarded) in the context of terrorism investigations. The traditional law enforcement model is highly protective of civil liberty in preference to physical security. All lawyers have heard one or another form of the maxim that "it is better that 10 guilty go free than that 1 innocent be mistakenly punished."[9] This embodies a fundamentally moral judgment that when it comes to enforcing criminal law American society, in effect, prefers to have many more Type II errors (false negatives) than it does Type I errors (false positives).[10] That preference arises from two interrelated grounds: one is the historical distrust of government that, as already noted, animates many critics of the Patriot Act. But the other is, at least implicitly, a comparative valuation of the social costs attending the two types of error. We value liberty sufficiently highly that we see a great cost in any Type I error. And, though we realize that Type II errors free the guilty to return to the general population, thereby imposing additional social costs on society, we have a common sense understanding that those costs, while significant, are not so substantial that they threaten large numbers of citizens or core structural aspects of the American polity.

The post-September 11 world changes this calculus in a fundamental way. Most obviously, it changes is the cost of the Type II errors. Whatever the cost of freeing John Gotti or John Muhammed might be, they are substantially less then the potentially horrific costs of failing to stop the next al-Quaeda assault. Thus, the theoretical rights-protective construct under which our law enforcement system operates must, of necessity, be modified to meet the new reality. We simply cannot afford a rule that "better 10 terrorists go free than that 1 innocent be mistakenly screened."[11] Put another way, it may be better that 1 million visitors be screened than that 1 million Americans should die.

Second, and less obviously, it may change our approach to the nature of the Type I errors that must be considered. In the traditional law enforcement paradigm the liberty interests at stake is personal liberty - that is, freedom from the unjustified application of governmental force. We have as a model, the concept of an arrest, the seizure of physical evidence, or the search of a tangible place. And we reject the concept that any of those impositions on freedom should arise from the application of invidious, immutable characteristics.

What these examples demonstrate is not so much that our conception of liberty is based upon absolute privacy expectations,but rather that government impingement on our liberty will occur only with good cause. In the context of a criminal or terror investigation, we expect that the spotlight of scrutiny will not turn upon us individually without some very good reason.

This conception of the liberty interest at stake (the interest that will be lost when Type I errors occur) also emphasizes one other point about privacy - in many ways the implementation of new laws and systems to combat terror are not an unalloyed diminution of privacy. Rather the laws and practices can substitute one privacy intrusion (for example, a search of electronic data about an individual) for another privacy intrusion (the physical intrusiveness of body searches at airports). But this means that legal analysts cannot make broad value judgments - each person weighs the utility of their own privacy by a different metric. For many Americans, the price of a little less electronic privacy might not be too great if it resulted in a little more physical privacy - for others the opposite result might hold. This suggests little in resolving the tension, save that it cautions against allowing the tension to be resolved by unrepresentative institutions like the courts and in favor of allowing more representative institutions, like the legislature, to do their best at evaluating the multi-variable privacy preferences of the population.

What all this means for racial profiling is fairly clear, and in my judgment the new Department of Justice policy strikes the right balance. It recognizes that for "traditional law enforcement activities" involving routine law enforcement actions such as traffic stops, federal law enforcement officers should never rely on race or ethnicity, except to the understandable extent that they have a suspect description that is race or ethnicity specific. The policy also makes clear that in connection with a specific investigation, law enforcement may only consider race or ethnicity if there is a "close fit" to an identified criminal incident either because of geography or a temporal connection. These are good rules - they make clear that, in general, in the case of investigations for common law crimes race and ethnicity are not relevant.

The Department's policy also makes clear, however, that in the case of terrorism - that is in cases where national security or some other catastrophic threat is involved, the general Constitutional framework applies. Thus, in very rare circumstances where substantial predication exists, the Department contemplates a regime permitting investigation that includes a component of ethnicity or national origin. This, too, recognizes the fundamental importance of national security, while honoring the rule of law. We have a right to expect that such instances will be rare - but we must also recognize that on a few occasions they may be necessary.

Finally, we should turn our attention to the inevitable problems that will arise because of a potential invidious use of racial or ethnic classification for reasons that are only masquerading as legitimate. One thinks, immediately, of the security officer who targets for additional screening and inspection men of a particular ethnicity (a case, by the way, that the DOJ guidelines expressly declare improper and prohibit). How can we fight that all too real prospect?

One answer may lie in technology. Not all solutions necessarily trade off Type I and Type II errors, and certainly not in equal measure. Some novel approaches to combating terrorism might, through technology, actually reduce the incidence of both types of error. Consider, for example, various proposals for enhanced electronic data screening procedures, such as the CAPPS II proposal now being developed by the Transportation Security Administration. This is neither the time nor the place to discuss the details of that proposal (and those who are interested in my own views can readily find them in my recent testimony before the House Subcommittee on Aviation). But in considering such proposals we should recognize that systems like CAPPS II (and other similar proposals) do not result in a one-way diminution of privacy. Rather they require trade-offs in different types of privacy: substituting one privacy intrusion (into electronic data) for another privacy intrusion (the physical intrusiveness of body searches at airports). 

In adopting such a trade-off we may actually achieve a collateral gain of importance to the values that underlie this Commission's inquiry. Rules-driven risk assessment systems such as CAPPS II substitute hard data and a priori rules, for instinct and racial stereotypes. Thus, they will undoubtedly have the salutary effect of reducing the need for random searches and eliminate the temptation for screeners to use objectionable characteristics of race, religion, or national origin as a proxy for threat indicators.[12] For many Americans, the price of a little less electronic privacy might not be too great if it resulted in a little more physical privacy, fewer random searches, and a reduction in invidious racial profiling. 

In closing, I would simply note that we face a difficult challenge of enhancing security while protecting liberty. In doing so, we must recognize that liberty is not an absolute value: it depends on security (both personal and national) for its exercise. As Thomas Powers has written: "In a liberal republic, liberty presupposes security; the point of security is liberty."[13]  The growth in danger from Type II errors necessitates altering our tolerance for Type I errors. More fundamentally, our goal should be to minimize both sorts of errors.

* * * * *
Madame Chairperson, thank you for the opportunity to testify before the Subcommittee. I look forward to answering any questions you might have.