Since September 11, two large questions have dominated the discussions of how to prevent terrorist attacks on American soil. First, how do we ensure that the government has the authority it needs to gather, share, and use information about potential terrorist activities? Second, how do we ensure that this authority is not abused? The first question evokes images of “connecting dots” and “eliminating stovepipes.” The second raises issues about democratic accountability and the proper balance between individual liberties and national security.

Each question can be profitably pursued on its own terms. And so they were, separately, in the report of the National Commission on Terrorist Acts Upon the United States (a.k.a. the 9/11 Commission), which for the most part focused on coordination failures within the federal government, while noting the “need for balance as our government responds to the real and ongoing threat of terrorist attacks.” But their separation will ultimately be self-defeating, because there is a deep connection between building an effective domestic-intelligence network and building one that is democratically accountable and balanced. Asserting such a close connection between effectiveness and accountability may sound like empty idealism, but a structural basis for the claim can be found in the building blocks from which the domestic-intelligence network will be constructed. State and local governments must be full partners in any effective strategy for preventing acts of terror: without their participation, the federal government cannot possibly know what “dots” to connect. But the relationships that local police departments have formed with the communities they serve can be threatened when these departments are asked to participate in federal intelligence-gathering efforts. If this conflict can be defused, greater police involvement will bring with it not just better intelligence but a system that will be more accountable to the communities from which the information is gathered.

To make the case for the close connection between effectiveness and accountability in combating terrorism, I will explore the complex tensions between federal and local interests and show why these tensions are not just inevitable but valuable. Their strength is underestimated by those who believe that a nationally coordinated response to threats of terrorism can be effective only if it is dominated by the federal government. Their value is also underestimated by those who fear that any coordinated national response to terrorism will lead to the creation of a police state. The interactions between levels of government that have attracted the most attention since September 11 have been oppositional, with local governments complaining about inadequate federal funding or condemning federal policies on civil-liberties grounds. But the most fruitful contributions that local governments can make to the emerging intelligence network will arise from their participation, not their criticism. Better understood, and embraced, the distinct interests of federal, state, and local governments will help ensure that the network is true to the political values of the nation it protects.

• • •

To understand current relations between the federal government and state and local law-enforcement officials, we need to start in the decades before 9/11, when violent crime, not counterterrorism, lay at the heart of intergovernmental interactions.

Historically, our federal system has assigned the function of controlling episodic violent crime to state and local authorities—and usually local authorities, since very few states have integrated law-enforcement hierarchies. It is therefore the local police, working with local district attorneys, and county sheriffs, working with county attorneys, who have primary responsibility for keeping the streets safe and going after rapists, robbers, and murderers. They are at the other end of the telephone line that victims and eyewitnesses call; they will be judged (fairly or not) by movements in the “uniform crime report” statistics, and they will be held directly or indirectly accountable by local voters.

In the 1960s, however, President Johnson, goaded by Barry Goldwater’s use of crime as a campaign issue, began a tradition of federal funding for state and local criminal enforcement. By the 1990s both the Bush and Clinton administrations had flagship programs that committed investigative and adjudicative resources against street criminals. And both in their dollar amounts and in the discretion they gave to state and local enforcers, federal grant programs took off during the 1990s. A lot of the federal funding, like the money that the Edward Byrne Memorial State and Local Law Enforcement Assistance program committed to narcotics enforcement, was administered by the states. But the election of Bill Clinton—who promised to put 100,000 new police officers on the streets—brought the COPS (Community Oriented Policing Services) program, which put money directly into the hands of local police departments, particularly in big cities.

The merits of these federal grant programs—whether they achieved their stated goal of crime reduction, and how many more cops ended up patrolling the streets—are contested issues: crime reductions in the 1990s may have come more from demographic changes or the performance of the economy than from police presence. However these debates are resolved, what matters here is that the latter part of the 1990s marked the high-water mark of a federal–state–local relationship that nicely served a wide range of political interests. Federal officials did well. Presidential administrations of both parties got to tout their commitment to the Fight against Crime and the War on Drugs, as did legislators, who readily appropriated large sums of money for these endeavors. And because the violent crime targeted by the enforcement and funding programs was essentially local, legislators could also reap the benefits of the money flowing into their own districts. The interests of federal enforcement agencies were also well served by the new violent-crime priorities. The general public was happy to see the “feds” fighting local bad guys—street gangs, armed robbers, and murderers. And the championing of these cases by local legislators could only redound to the benefit of agencies at funding time, and to field offices in their relations with headquarters (since local alliances could be triangulated against centralized control). Even better was the extent of federal enforcement discretion. Since violent crime was still primarily a local responsibility, federal agencies could be quite strategic in their case selections.

Local enforcers did well too. While the Supreme Court might bemoan federal intrusion into a traditionally local domain, local enforcers were less concerned with constitutional prerogatives and more with direct grants, overtime pay, and the aid-in-kind provided by federal enforcement activity. After all, federal agencies can rarely pursue violent-crime cases without the help of the local police, who alone control the informational networks on which federal law-enforcement agencies must rely when pursuing episodic criminal activity. So although judges decried the flood of “state” cases into federal court, and defendants who expected to be prosecuted in state court complained when they were somewhat arbitrarily singled out for harsher federal sentences, state and local law enforcers generally looked on this federal “intrusion” with equanimity, even glee. Federal funds also supported a variety of local programs that allowed the police to be more aggressive in their crime-fighting strategies.

• • •

This decade of direct grants, block grants, and enforcement assistance from the federal government to state and local authorities continued with the election of George W. Bush. In a May 2001 memo to department heads, Attorney General John Ashcroft included as two of his seven goals reducing gun violence and drug trafficking and helping states with anti-crime programs. (Terrorism did not make the list.) The new administration did announce plans to phase out the COPS program—not particularly surprising, given that Republicans had long questioned the efficacy of this Clinton program, which tended to funnel most of its money to urban Democratic strongholds. But the plan envisioned a reconfiguring of federal aid to violent-crime prevention programs, not a transfer away from it.

Then came September 11.

• • •

We are often told that September 11 “changed everything.” An overstatement in some contexts, this claim is quite apt when applied to the relations between the federal government and state and local governments in the area of law enforcement.

The shock to the federal enforcement bureaucracy was extraordinary. A system whose defining luxury had been its general lack of primary enforcement responsibility (and accountability) was suddenly saddled with a politically unavoidable, and all-but-impossible responsibility: preventing another such attack. Given the nature of the perceived terrorist threat—the sleeper cells waiting to strike again—this responsibility required the feds to reach out to state and local agencies for intelligence-gathering assistance and create the institutional linkages that would allow for two-way data sharing. In the United Kingdom, which has lived longer with the threat of domestic terrorism, special branches in each local police force mediate between the national security agencies and local communities in a relationship that a recent British government report celebrated as the “golden thread.” Now the United States would have to do something similar, on the fly. The only available templates were the Joint Terrorist Task Forces, of which there were only 35 in selected cities before 9/11. And while successful in New York, these had only mixed success elsewhere.

The federal government’s need for state and local assistance is not simply a matter of manpower—there are about 700,000 state and local police officers on the job compared with about 90,000 federal law-enforcement officers (more than a quarter of whom are in the Federal Bureau of Prisons)—nor even a matter of the many things that cops learn on street patrol. It also stems from the role of state and local enforcers in bringing the bulk of serious criminal charges in the United States, because the threat of prosecution (even prosecutions having nothing to do with terrorism) is one of best tools around for prying loose closely held information. Local police also play a central role in maintaining order and ensuring public safety, and this gives them a more balanced “portfolio” in dealing with community leaders. The police officer who seeks information from a local Arab-American community leader has probably met and assisted that leader before—protecting his property, ironing out some administrative complexity, or ensuring his safe worship.

Recognizing these local advantages, the Justice Department quickly went beyond vague talk of “information sharing” and asked for local assistance in a large-scale program to interview thousands of people (mostly young Middle Eastern males) who had entered the United States on non-immigrant visas. In the spring of 2002, the department went further and announced its plan to place the names of certain aliens who had violated their visa requirements into the national database of wanted suspects. It asked state and local police to arrest these “absconders,” and noted that, as a legal matter, such assistance was “within the inherent authority of the states.”

The predominant response of local police forces to the new War on Terror was to line up at the recruiting station and complain about delays. The drumbeat from police departments during 2002 and into 2003 was that the feds weren’t sharing information with the locals.

That said, conditions were not altogether propitious for police cooperation. In a number of cities and states officials passed resolutions criticizing the administration’s counterterrorism efforts. (As of press time, four states and 355 local governments have done so; see Elaine Scarry’s “Resolving to Resist,” February/March 2004 Boston Review.) Many of these resolutions were costless. The president of the Denver police union called that city’s stance “as relevant as the nuclear-free zone in Boulder.” Some stands were more than symbolic. In December 2001, despite the opposition of the Oregon attorney general and the local district attorney, the Portland City Attorney announced that a state law barring police from “detecting or apprehending” people who have violated only federal immigration law, and from collecting information on political, social, or religious beliefs unless it pertains to a criminal investigation, prohibited Portland police from asking some of the 33 questions that the Justice Department wanted posed to 23 foreigners in the area.

Some of this scattered resistance may have arisen from partisan politics or liberal conviction. But there was a historical basis as well: the federal efforts to recruit local police into a national intelligence network brought back memories of COINTELPRO, CHAOS, and other abuses of the late 1960s, when the feds had urged the locals to create “intelligence units” to gather and disseminate information on potential “civil disorders.” The Senate Select Committee to Study Governmental Operations with Respect to Intelligence Activities (a.k.a. Church Committee) later recounted the result of these efforts: “Local police intelligence provided a convenient manner for the FBI to acquire information it wanted while avoiding criticism for using covert techniques such as developing campus informants. . . . These Federal policies contributed to the proliferation of local police intelligence activities, often without adequate controls.” All this was not ancient history to local authorities in places like Denver, which in 2002 was still negotiating a settlement with the ACLU relating to abuses by the police intelligence bureau, and New York City, where a consent decree arising out of similar abuses is still in place.

Local concerns have not been merely partisan, philosophical, or historical, but have also reflected the political economy of counterterrorism. When the federal–local interaction revolved around violent crime, federal initiatives brought significant local benefits, including credit for local leaders and maybe even improved local safety. The counterterrorism dynamic has been precisely the reverse (save in some exceptional locations, like New York City). There is no reason to expect that terrorists pose a particular threat to the many places where they or information about them will be found. Thus, the gains from domestic intelligence-gathering are felt primarily, perhaps even exclusively, at the national level, whereas the costs fall on the localities—not just the fiscal costs but the significant intrusions and hostilities that attend any large-scale investigations of immigrant activities in communities with sizeable proportions of immigrants.

Police officials have had their own pragmatic concerns about federal counterterrorism initiatives, particularly those involving the use of federal immigration statutes. As the Web site of the International Association of Chiefs of Police explained in October 2002,

Enforcement of civil immigration laws by local law enforcement would have a chilling effect on both legal and illegal aliens reporting criminal activity or assisting police in criminal investigations. Local police want illegal aliens to come forward when they have been the victims of, or witnesses to, crimes. Police depend on the cooperation of immigrant communities to help them solve all sorts of crimes and to maintain public order. Without the assurances that they will not be deported, many illegal immigrants with critical information would not come forward.

Few non-federal enforcement officials have been disposed to assist the feds in using the immigration laws as a law enforcement tool, even though, as one federal official noted, “the only barriers to executing such arrests are statutes or policies that states or municipalities have imposed upon themselves.” State officials have responded with somewhat more alacrity. In Maryland, for instance, according to a Baltimore Sun article, “Many local police departments, including those in Baltimore and in Anne Arundel and Baltimore counties, generally will not report illegal immigrants unless they have committed crimes.” However, “state police policy is to inform federal authorities about any suspected of being in the country illegally.” As of April 2004, only Florida and Alabama had formally signed on to the federal initiative, and in Alabama it appears that only state troopers are involved. A bill has been introduced in the House—the Clear Law Enforcement for Criminal Alien Removal Act of 2003—that would require state and local authorities to enforce federal immigration laws, on pain of losing federal funds. At an October 2003 hearing, the National League of Cities expressed its opposition for what it characterized as an “unfunded mandate,” and the legislation’s future remains unclear. (A similar bill is pending in the Senate, with hearings held in April 2004.)

Why the difference between the attitudes of local police departments and their statewide counterparts? Part of the answer is political: Republican governors versus more-liberal local officials in cities. But institutional obligations (or the lack thereof) likely play a role as well. For it is at the local level and particularly in big cities that the costs imposed by the federal enforcement initiative on relationships with immigrant communities would hit the hardest.

Even had nothing else changed in the relationship between federal and local enforcers, the federal counterterrorism initiatives would have imposed new intelligence-gathering responsibilities on the police and threatened to make it harder to maintain order in areas with significant immigrant populations, including most big cities. But the pre-9/11 story is also important, for the new federal initiatives seemed also to come at the cost of prior federal commitments of resources to addressing violent crime and narcotics.

The point here is not that the shift in federal resources was a mistake but that the federal commitment to street-crime law enforcement during the 1980s and 1990s set a new baseline for local expectations of federal assistance. And these expectations are no longer being met.

State and local governments have additional fiscal grievances. Post-9/11 antiterrorism programs have required massive expenditures. And, as the 9/11 Commission sympathetically noted, many have complained vigorously that federal monies have been distributed without regard to need or risk. Because of the congressional allocation formula (reinforced by Department of Homeland Security decisions), the most populous states took the biggest hit, putting up the most money and receiving the least. In 2003, Wyoming received $61 a person in federal homeland-security grants and Alaska, $58. But New York got less than $25, and California got $14.

If one source of intergovernmental tension is the level of homeland-security funding, another (perhaps even more disruptive) source of tension is the way federal funds have been distributed. One police chief complained in a spring 2003 congressional hearing that homeland security

resources do not go directly to local police departments. They cannot be used to hire new police. They cannot be used to pay overtime expenses that we incur each and every time Secretary Ridge changes the alert level. They can be used to purchase equipment, but not by me. I have to wait for a statewide plan to be developed and then I have to hope that a fair share of those funds will filter to my department.

As a mayor put it in another hearing around the same time, “After all, a 911 call does not get a state trooper.” The U.S. Conference of Mayors has (unsurprisingly) expressed similar sentiments. And members have suggested that partisan politics help explain the Republican administration’s reluctance to give money directly to urban areas.

Because local governments will eventually get much of the homeland-security funding, a case can be made for funneling the money through the states and thus encouraging statewide coordination. The governors have embraced these arguments. Speaking for the National Governors Association at a spring 2003 congressional hearing, Massachusetts Governor Mitt Romney testified, “We believe it critical that homeland-security funding and resources be applied against comprehensive and integrated statewide plans. . . . The most critical step to maximizing our resources is developing integrated statewide plans and channeling virtually all homeland security funding through these plans. . . . Without statewide coordination, there is no check on gaps in coverage, incompatible equipment and communications systems, and wasteful duplication.”

Because they see cities as bearing the brunt (in both fiscal and political terms) of any nationwide intelligence-gathering and security-patrolling effort, local officials, particularly from bigger cities, would likely have complained in any case about the statewide funding model for homeland security. But their sense of grievance has been intensified by the perception that it is their violent crime money that is now going to the states. Urban officials have made much of the coincidence that the COPS program is being phased out and other crime control grants reduced just as homeland-security funding plans are being made.

• • •

Regardless of the financial and political costs of intensive counterterrorist intelligence programs, some cities can be counted on to support federal efforts without reservation—especially cities that see themselves as high on the list of future targets. Even with limited federal reimbursement, New York City has been quick to deploy its resources to the fullest, in a massive counterterrorist program. As even a casual visitor will notice, the program includes large-scale and widespread security measures intended to protect the city’s most conspicuous targets. But even more significant is the degree to which the city’s police department has extended its intelligence collection internationally. In January 2003, for example, the NYPD sent a senior counterterrorism official to London as part of a intelligence exchange with the Metropolitan Police, and it has also sent liaison officers to Toronto, Tel Aviv, and Lyons.

Yet even in New York City, tensions have been building. On May 26 the Daily News reported, “City cops are in a nasty fight with the feds over manpower and money that boiled over with the NYPD pulling out of a massive federal deportation sweep aimed at criminal immigrants yesterday.” On June 4 The New York Times noted that “long-simmering tensions between the Police Department and the F.B.I. Office in New York increased markedly last week after Police Commissioner Kelly publicly credited one of his detectives with ‘breaking’ a terrorism case that the agencies had investigated together, according to several officials and an internal F.B.I. e-mail message.” So long as jurisdictions overlap and publicity brings personal and institutional rewards, disputes about credit-grabbing will plague joint task forces. But something more serious is going on as well.

The tensions promise to be far worse in other cities. Detroit, for example, is losing federal enforcement assistance against violent crime, does not rank at the top of the list of potential terrorist targets, and has a large Middle Eastern immigrant population that is both a potential source of intelligence information and of political backlash to aggressive federal counterterrorist tactics. (Of the 3,212 interviews with aliens conducted by the Justice Department after the September 11 attacks, 330 were done in the Detroit area.) According to a May 2004 Detroit News story, police officials in the metropolitan Detroit area have said that “directing the police to investigate the status of immigrant residents makes residents suspicious of police—drying up important sources of information.” One official highlighted both enforcement and fiscal concerns relating to immigration enforcement:

“People would not work with us. It would make it very hard for us to do our job,” said Cpl. Daniel Saab, who runs a community policing office in Warren Avenue for the city of Dearborn, home to about 29,000 Arab-Americans and 3,000 Latinos. “It would double the work for us, too. Just stopping somebody and, say he qualifies for an investigation by the immigration department, it would tie us up not only for hours, but for days.”

Things could get much worse, with an increasing rift between local and federal enforcers, as the complex coordination problems posed by any nationwide counterterrorism effort are compounded by the use of tactics perceived in many communities as overly aggressive and by local grievances arising out of dashed hopes for continuing federal support against violent crime. It is of course possible that federal enforcers will, over time, do a better job of reaching out to communities on their own. In a May 2003 report to the House Judiciary Committee, the Justice Department noted that FBI officials had held over 500 meetings with Muslim leaders since 9/11 to “establish a dialogue” and to tell them how to report civil rights violations to the Bureau. The department’s Inspector General’s Office and Civil Rights Division have also made special outreach efforts. Nonetheless, the relative narrowness of federal enforcement concerns and the comparative distance of federal agencies from the lives of people in densely populated urban communities would likely make a federally centered intelligence gathering effort a poor substitute for a more integrated national network.

Consider the TIPS (Terrorist Information and Prevention System) debacle. First mentioned by President Bush in his 2002 State of the Union speech, the program was pitched by the Justice Department as a way to enlist the observatory powers of service providers around the nation in the War on Terror. Before long, under pressure by many in Congress and elsewhere, the program was reconfigured to “involve only truckers, dock workers, bus drivers and others who are in positions to monitor places and events that are obviously public.” Even that was not enough, and the initiative was soon legislated out of existence. Or consider the federal government’s experience during World War One with the American Protective League, a group of around 250,000 “patriots” that developed an “auxiliary” relationship with the Bureau of Investigation and, according to an official FBI account, “abused the civil rights of Americans in its efforts to identify German spies, draft resisters, and other threats.” Just about midway between these two programs, historically, came COINTELPRO.

• • •

While things could thus get much worse, a far more optimistic ending is also possible: a counterterrorism-based federalism model that courts the assistance of state and local governments by giving them a greater voice in how the federal government interacts with citizens, and particularly with immigrant communities.

Secrecy obviously has its place in any counterterrorist program. If every tip and the planning of how it is to be followed up have to be shared with enforcers and agencies across the nation and at every level of government, operational effectiveness will surely be reduced (if not completely precluded). But how can the feds make sure they receive such tips? This is where the informational networks that only state and local enforcers have access to come in. Indeed, as the 9/11 Commission noted, the relationships that the FBI has forged with these enforcers over time have become one of the best arguments against the creation of a new stand-alone federal domestic-intelligence agency along the lines of MI-5, which would either have to develop its own (with less to offer in return for state and local cooperation) or be ineffective.

The ability of the local police to mediate between federal needs and community sensitivities was highlighted when the Justice Department’s call to interview certain Middle Eastern immigrants went out in the wake of 9/11. As the police chief of Ann Arbor, Michigan, later recounted, local leaders of the Arab-American community asked that a local police officer be present during these interviews. Later, in March 2003, the director of the American Arab Chamber of Commerce in Dearborn noted how the local police department had worked to win the Arab community’s trust and how the FBI and other federal law enforcement agencies had not done enough. He also noted that after the war with Iraq had broken out, the Wayne County sheriff had dispatched his forces to guard mosques.

In our optimistic story, state and local agencies—and local police departments in particular—will exact a toll from federal authorities as the price of gathering and supplying information. Because of current fiscal constraints, the federal government will not pay in cash but will give local police more of a voice in federal domestic-intelligence policy. And that voice will transform the new national network, making it more accountable to the citizenry than would otherwise be possible.

In the community-policing literature, those who argue for greater sensitivity to community preferences have to confront questions of who represents the community and whether the community has been adequately organized to convey its preferences. When one scales up the model to the national level, however, this maybe less of an issue, as there already are institutional structures—cities, towns, counties—that can legitimately speak to local concerns and, while doing so, promote the gathering of local information. Precisely how the federal interest in a national intelligence network can be squared with demands for accountability (and consequent sensitivity) at the local level is something that officials in cities like Portland, Oregon, and Boise, Idaho, have struggled to work out. In both cities, federal regulations restricting access to intelligence information clashed with civilian oversight mechanisms. The contours of such local struggles will undoubtedly vary from place to place. But their effect on the sensitivity of federal officials to local politics—heightened when congressional representatives take up the causes of local officials—will be salutary.

The touted benefits of federalism often merely reflect the virtues of managerial decentralization and not necessarily those of a genuinely federalist and “polycentric” system. This is partly true here. Even officials in a hypothetical national police force that had prime responsibility for pursuing all violent and street crime would think twice about using tactics that risked alienating chunks of the population in their respective patrol sectors (at least to the extent they cared about enforcing the law in all neighborhoods). These are crimes that can be and are measured, with police performance judged accordingly. The contribution that local police forces can make to domestic-intelligence policy makingand collection are thus related to the nature of their beat.

But the political contribution that police involvement can make to federal counterterrorism efforts stems not just from their “informational accountability”—their obligation to deal with any community from which they will need crime tips. It also reflects their greater electoral accountability. Police chiefs themselves are not generally elected, but the mayors and county executives to whom they report are, as are the chief prosecutors. And scrutiny of police performance by the local electorate, although hardly constant, regularly occurs.

Fine theory, some may concede, but what about the lessons of history? Isn’t the lesson of COINTELPRO that we should fear anything that looks like an effort by the federal government to dragoon local police departments into a federally dominated, and potentially repressive, national intelligence network? Given the inherently low visibility of domestic-intelligence operations, such concerns about governmental abuses cannot be dismissed out of hand. Yet there is a lot more history to be considered. Policing, particularly in large urban areas, has changed a lot since 1968. Spurred by rising crime rates, tutored by theorists like Herman Goldstein, James Q. Wilson, and George Kelling, and encouraged by a growing stream of federal grants, police officials began to see an essential connection between improved community relations and effective law enforcement. To be sure, a lot of the revolution in policing has been rhetorical, and in some departments “community” or “problem-solving” policing has been more a grants-writing strategy than an operational reality. Still, deployment patterns really have changed, and police commanders now look to community leaders not simply for tips on whom to arrest but increasingly for guidance on how the police can best improve the quality of life within a precinct. Approaches can vary, with departments that have adopted a “community policing” style more ready to share decision-making authority with the community than those that, wary of committing themselves to ambitious social objectives, would restrict themselves to “problem-solving.” But either way, the last two decades have seen enormous and accelerating changes in the readiness of urban police forces to solicit and address the concerns of the people they serve. And solicitude for the concerns of ethnic or racial minority groups—which are often majorities within a given city or precinct—has increasingly become a non-negotiable part of a police chief’s job description.

*  *  *

Decades of political focus on violent crime, coupled with increasing sophistication in the way crime-fighting is assessed, have put police departments under an unprecedented degree of pressure to actually achieve results. And although the precise contours of each department’s approach will vary, each has made vast strides toward recognizing the essential connection between community relations and effective law enforcement. At the same time, changes in big-city politics have also made it far more difficult to write off criminal enforcement in any ethnic, geographic, or racial community.

The claim here is not that all is for the best in this best of all possible worlds of “community policing.” But given the history, we should not expect local police to simply surrender their voices and interests to the federal government. When the feds sought help the help of local police intelligence units in the late 1960s and early 1970s, they had a common interest in the investigation and prevention of local riots and protests—the police because they sought to protect persons and property within their respective jurisdictions, the feds because they sought information about subversive masterminds. Now the police’s interest cuts the other way: the chief whose only desire is to ensure local peace will be prone to give the feds too little help, not too much.

Part of the response to COINTELPRO fears must also lie in necessity: Even at the time of COINTELPRO it was hard to get broad agreement on the proposition that a federally coordinated investigation into the communist roots of antiwar activism was an important national priority. Right now, in contrast, the need for a national domestic-intelligence network focused on threats of terrorist attack ought not be in dispute. In this second- or third-best possible world that we live in, full police participation in this network is at least a risk worth taking when weighed against the alternatives.

To some, the notion of police departments as bulwarks of civil liberties against federal encroachment might sound a bit odd. We are accustomed to the idea that the federal government is responsible for monitoring local abuses—stepping in with civil suits or civil-rights prosecutions whenever the locals are derelict in their attention to such matters. The rejoinder, of course, is that while we all have our individual estimations of the skills and predilections of each enforcement level, none has a monopoly on virtue. The obligation of local police departments to combat violent crime and maintain order can push them toward aggressive and even abusive control tactics. Yet that same obligation causes police departments to be especially attentive to the costs imposed by abusive interactions with the local community.

How, specifically, can we move closer to our happier ending (with happiness in this area being a relative term)? There are no magic solutions here, and perhaps not even any brilliant new ideas. Counterterrorism has become such a burgeoning start-up industry that what we need is not more ideas, but fewer, or at least an understanding of why the gains from some initiatives can be undermined by others. In May the Bush administration launched the National Criminal Intelligence Sharing Plan. Spearheaded by the International Association of Chiefs of Police, this program appears to promise a new era of collaboration among federal, state, and local agencies, and an infrastructure within which the diverse political and operational concerns of local departments can be raised and addressed. Yet this sensitivity to local concerns has not been reflected in federal immigration-enforcement initiatives, whose costs need to be recognized by both the executive and legislative branches. Broadly assigning the police the task of immigration work, as envisioned by the Bush administration and pending legislative proposals, would in all likelihood result in a net intelligence loss, because in the neighborhoods that could be the richest source of terrorist tips, it changes the beat cop from peacekeeper to potential source of personal ruin. As a recent white paper by Police Executive Research Forum noted, there is considerable room for cooperation between the local police and the Bureau of Immigration and Customs Enforcement. But the dimensions of this interaction need to be a matter of negotiation, not federal fiat. And incidents like federal raids that lead to the deportation of someone who (according to the same white paper) “was helping local police identify crime suspects and was influential in building better community relations with law enforcement” should be avoided.

Another matter worthy of recognition is the cost of making state governments the primary interlocutors with the federal government on homeland-security issues. States are of course constitutionally and operationally indispensable, but their views ought not be seen as necessarily representing those of their urban instrumentalities, as their needs and burdens can be quite different. Until these urban views are given far more weight, homeland-security dollars are unlikely to get full value on the prevention side.

When it comes to actually collecting intelligence, the contribution by local police forces to the national counterterrorism effort cannot be limited to that of officers designated to work with federal agents in the Joint Terrorist Task Forces. The JTTFs are now proving good vehicles for operational coordination in raids, undercover stings, and intensive surveillance. A recent series in the Los Angeles Times about Muslims in Las Vegas also shows how the FBI has been able to work within the JTTF framework to reach out to members of the Muslim community to gather intelligence. But a broad intelligence-collection effort will require a range of contacts between officials and community members far more extensive than any federal agency or task can provide. While federal advisory councils and multicultural forums are helpful, they are not likely to capture the diversity of the Muslim community. If the political, religious, and generational conflicts here are anything like those that Jytte Klausen has found in European Muslim communities, reliance on a narrow class of representatives is dangerous, and counterproductive from an intelligence standpoint. Simply put, no federal outreach effort can substitute for the quality and quantity of contacts that local police officers have within the neighborhoods they serve. And the federal government must do far more to reach a modus vivendi with these departments that fully comprehends, and indeed profits from, their order-maintenance and crime-fighting responsibilities. Every felony arrest is an intelligence opportunity and should be recognized as such; this includes those made outside of ethnically concentrated areas, since recent experience has shown that terrorists do not necessarily interact with Muslim communities. But the same channels used to pass local intelligence up to the feds (and federal intelligence back down) should also be used to pass along local concerns (parochial and political) about federal counterterrorism tactics.

The engagement of local interests in the counterterrorism project cannot be left to federal initiative and persuasion, however. Local authorities need to step to the plate as well. If those who find fault with federal policies really want to affect the way the War on Terror gets fought in their area, they will pass fewer empty resolutions of condemnation and do more to structure the way their intelligence units and beat cops contribute to the national effort.

Much remains to be said about the institutional questions that will need to be resolved as we move forward. This year, in Hamdi v. Rumsfeld and Rasul v. Bush, the Supreme Court began to map out the role that courts will play in reviewing executive decision-making in the War on Terror. But so many of the basic issues relating to domestic counterterrorism are primarily matters of executive discretion and only distantly subject to judicial intervention or even statutory control. Indeed, even if the provisions of the Patriot Act that have been so reviled in local resolutions were to be repealed, the concerns of these opponents and skeptics would not be assuaged. For example, eliminating the provision that allows the FBI to seek library records through administrative subpoena would still leave the FBI free to get the same materials through grand-jury subpoenas—which prosecutors can (and do) freely issue without ever going through a grand jury or showing need. The extent to which public meetings are monitored, immigrants interviewed, Middle Eastern–looking people questioned—these, like most official decisions that will determine upon whom the costs of counterterrorism efforts will fall and with what weight are simply not amenable to effective judicial or legislative constraint or oversight. The primary sources of accountability and due moderation will have to come from institutional structure. In the end, the need to accommodate state and, particularly, local agencies within the developing domestic-intelligence system may thus offer the best promise of appropriately tempered zeal as we move into the post-9/11 era.