Knoxville, Tennessee

David Thacher
Harvard University
John F. Kennedy School of Government
Program in Criminal Justice Policy and Management

Case Study Prepared for the Urban Institute


For much of the 1980s, The Knoxville Police Department (KPD) was something of an organization in limbo. The department had kicked off the decade with a burst of innovation, developing capacities for problem-solving and community partnerships that put it well ahead many of its contemporaries in the police world. But after the city hosted the World’s Fair in 1982—an event that had been the catalyst for many of the initiatives in the first place—, many of these innovations either disappeared or downshifted to a lower level of intensity. The result was a department that remained perfectly competent at performing its basic tasks, but one that had lost some of its cutting-edge character. In the words of Knoxville Mayor Victor Ashe, who took office in 1988 partly on a promise to improve law enforcement in the city, the department became “kind of a go-along, get-along operation. It wasn’t bad, but it wasn’t terribly great.”

This section describes the KPD during this period in more detail, focusing on its operational units, administrative systems, and management. First, however, consider the relationship between the KPD and the outside world, since some observers believe that the difficulties the department faced were a result of its position in city government.

1. Relationship with the Environment

The Authorizing Environment

In the early- to mid-1980s, direct oversight of the KPD was exercised through a mayoral cabinet position called the Safety Director, which bore responsibility for the city’s police and fire departments. The Safety Director held considerable power over day-to-day police decisionmaking, counting both the Chief of Police and the department’s planning and budgeting unit among his direct reports. Thus despite fairly strong civil service protections, which extended all the way up to the Chief, policing was hardly depoliticized in Knoxville during this era.

By all accounts, the KPD’s string of Safety Directors were largely ineffective managers. The position required relatively few qualifications, and many observers believe that it became a political favor rather than a considered choice. For example, Mayor Ashe, who effectively abolished the position when he took office in 1988, maintains: “[It] was just totally political, usually incompetent people that didn’t know anything but needed the job.” Many inside the KPD blame the Safety Directors for some of the lost progress the department experienced during the 1980s. Phil Keith, who worked on the staff of three KPD Chiefs before taking the position himself in 1988, recalls that one Safety Director was “completely laid back.”

Others in the KPD echo these sentiments, explaining that the department’s loss of initiative began at the top.

Community Input in the KPD

Indeed, some of the department’s backpedaling may have had its source even higher, in the previous mayoral administration, which simply did not value some of the initiatives the KPD had started to pursue. Ellen Adcock, a recent president of Knoxville’s Council of Involved Neighborhoods, maintains that before Mayor Ashe, “you had a mayor in here that viewed neighborhoods as his enemies.” She herself attributes this attitude to a conscious political strategy, explaining: “The past administrations had [felt], ‘The more neighborhood groups you have, the more issues you are going to have to deal with;’” and from Adcock’s perspective, the conclusion seemed to be that “keeping the neighborhoods in their place” was the best policy. As a result, neighborhoods had little voice in city politics, and therefore little incentive to form in the first place (Adcock herself estimates that only about three or four neighborhood organizations had much visibility in Knoxville in the years preceding Ashe’s election).

Against this background, the KPD’s World’s Fair-era efforts were perhaps destined to be something of an anomaly, since they went against the dominant philosophy in city government. Crime prevention, for example, took a heavily neighborhood-oriented twist in the KPD: In 1979, the department drew on federal grant money to create a dedicated crime prevention unit that would offer programs and activities to the public, focusing particularly on schools, businesses, and neighborhood organizations (mostly Neighborhood Watch groups, which the unit also helped to organize). To be sure, there were limits to the department’s neighborhood focus, as the unit all but monopolized community relations in the KPD: For example, crime prevention officers rather than the patrol force at large actually gave most presentations and showed up at most Neighborhood Watch meetings. Nevertheless, Deputy Chief Robert Coker maintains that the department as a whole tried to build bridges to the community during this period:

But in the end, these efforts did not apparently take very deep roots in the KPD, as many officers from this period remember little interaction with the community outside of their calls for service. One recalls: “We did not have the partnerships with the community as strongly. There were some out there, but nothing like today, where we are constantly attending meetings with different neighborhood watch groups, neighborhood associations, or our community advisory board.” Another maintains that most officers during this period thought that “the idea of becoming involved with community was a waste of time.”

Coker himself explains the problem as one of leadership, maintaining that the some of the safety directors and chiefs of this era “went back to doing traditional-type policing,” and that “they were not very good at engaging the community—[they had] more of a fortress-type mentality.” Though a cadre of sergeants and lieutenants tried to keep the new initiatives alive, they did not get much support from higher levels. “It wasn’t taboo,” Coker explains. “[But] you didn’t see some of the resistant chiefs out there on the front lines going to the community meetings.” In any case, many officers recall that efforts to engage the community withered during this period, and even the well-staffed crime prevention unit began to dwindle as federal funding dried up: By 1988, only 9 of the original two dozen positions remained.

The Task Environment

All of this is to say that as a matter of citywide philosophy, and with brief exception as a matter departmental philosophy as well, Knoxville did not look much to its neighborhoods for guidance about governmental priorities. In this sense, the KPD did not look outwards to find authorization for its actions.

Knoxville Police were not entirely insular in this period, to be sure, as they sometimes worked with outside agencies in fairly nontraditional ways. For example, officers and community members alike point to a long history of collaboration between the KPD and the city’s code enforcement department, in which the two agencies would team up to deal with neighborhood crime problems. In the criminal justice world, the KPD had long participated in joint initiatives with the local DA and the Knox County Sheriff’s Department. For example, Knoxville Police have spearheaded a number of joint grant applications for local criminal justice agencies, including a career criminal grant with the DA’s office in the early 1980s. And for some time in the 1980s, the KPD and the County Sheriff collaborated on a Metropolitan Organized Crime Unit that carried out both agencies’ narcotics investigations, the idea being that the drug market did not respect arbitrary political jurisdictions. More broadly, in the KPD’s 1988 management survey, most department managers described the relationship between the department and other agencies as positive.

On the other hand, with few exceptions, members of both the KPD and outside agencies remember many missed opportunities for collaboration. Fred DeBruhl, Executive Director of the Knoxville Community Development Corporation (which manages the city’s public housing stock), recalls that his agency and the KPD simply did not share information—police might visit a KCDC apartment several times on emergency calls before the housing agency discovered that it had a problem. Even on the criminal justice side, most interagency ties were weaker in the 1980s than they are today. For example, Rhonda Garren, who at the time worked in Knox County for the state’s Department of Children’s Services, recalls: “We were all doing these jobs and dealing with the same people, but none of us knew one another. Back then, I didn’t know anybody in the police department. So dealing with these serious kids on probation, if we needed [the KPD], we didn’t know who to call.”

A good example of the lack of interagency collaboration, and of the consequences it had, concerns the difficulties Parole Officers faced in serving arrest warrants on their parolees. As a matter of policy, the Tennessee Board of Paroles does not allow its officers to serve warrants themselves, so the agency depends on local law enforcement for this service. But in Knoxville, that arrangement did not work out very well. The crux of the problem seems to have been a bureaucratic shuffling of responsibility between the KPD and the Knox County Sheriff, which had traditionally maintained and served all arrest warrants in the county (even though by state law, any law enforcement officer could serve a warrant). As Parole Board regional director Jim Cosby explains:

As the department began undergoing extensive organizational changes in the late 1980s, the KPD and the Parole Board were able to work this problem out. But in the meantime, parole officers and police had trouble coordinating on some very basic issues.

2. Operations

During this period, most of the KPD’s operational personnel were assigned to either the Patrol Division or the Criminal Investigations Division, and this section will focus on those two areas.1 During the late 1970s and early 1980s, overall department staffing fluctuated widely, but the basic pattern was decline: Although the department employed 350 sworn officers in 1976, it was down to 278 by 1988, despite the fact that call volume had actually increased somewhat.


Throughout much of the 1980s, Knoxville’s patrol force was assigned to five shifts (or “detachments”) that patrolled the city’s three sectors. Time rather than geography was the basic organizing principle for the patrol force. To be sure, officers in some detachments had relatively permanent assignments to their beats, which they could request themselves subject to approval by the administration. But other officers—particularly rookies—often switched beats daily; and in any case, the beats themselves were more of an administrative convenience than a meaningful map of the city, as they split many locally-recognized neighborhoods in two. Moreover, supervision of the patrol force did not follow a geographic structure at all: All officers in a detachment ultimately reported to a single patrol captain, regardless of which sector they patrolled.

Officers in Knoxville had the same responsibilities as most of their contemporaries in other police agencies, and today most of them remember emergency response and random patrol as their main duties. In taking care of this basic work, officers apparently acted more as “free agents” than they do today. First of all, many officers remember that departmental systems did not offer much support to patrol work. Information systems, for example, were poorly-developed: It often took several months for incident information to get logged in the system, so the available information was often outdated. More specialized information, like the addresses of parolees living in the city, existed but was essentially inaccessible; and in any case there was little training in how to use the information the department did maintain. The result was that officers had to rely on their own resourcefulness in patrolling their beats and handling calls. One remembers:

Beyond support for patrol work, there was apparently less guidance as well, for the department seemed to take less interest in what patrol officers did than it does today. For example, the KPD’s pursuit policy was fairly vague, leaving important choices up to officers and their sergeants; and patrol time was reportedly less likely to be programmed to focus on particular problems than it is today.

On the other hand, the department did structure patrol work in some ways. Most simply, the KPD apparently provided strong incentives for officers to make arrests and citations through its personnel evaluations. But more in line with the initiatives to come, the department often tried to push and facilitate some types of problem-solving work through its system of “directed patrols” (or “DPs”). Another World’s Fair-era innovation, the directed patrol represented a limited form of problem-solving that focused the attention of an officer (or occasionally more than one officer) on a particular troublespot in the city. Typically, a supervisor would designate a DP in an area where he or she had noticed repeated calls for service: For example, an intersection that generated numerous complaints or accidents might be targeted for a directed patrol that would focus on traffic enforcement (in fact, the great majority of all DPs seem to have involved traffic problems, as they do today). One or more officers would be directed to focus their attention on that area during whatever “free time” they had; or occasionally, the officers might be taken out of the 911 loop so that they could spend all of their time on a problem area. In many cases, “focusing on” a particular area simply meant maintaining high visibility in the area, as the department considered targeted patrol to be an effective preventative measure—thus the system was, in fact, the means of programming patrol time that its name implied. In some cases, like the traffic DPs, officers received a more specific directive to step up enforcement in the area. Less commonly, officers would turn to non-traditional responses like code enforcement and community involvement to deal with neighborhood problems.

Although the directed patrol system did allow the department to nominate problems for patrol officer attention, it suffered some clear limitations. First of all, DPs faced the same deficiencies of organizational support as did patrol work in general: Officers specifically remember a lack of information, training, and planning capabilities to support their problem-solving work. Second, according to many officers, the department did not give them much protected time to do DPs; instead they were expected to find the time between calls, and as a result the activity seemed less important to them. Some officers believe that they simply had less “free” time, as staffing dwindled throughout this period. In any case, the DP strategy reportedly faded away somewhat as the 1980s progressed, returning officers to their own devices during their time between calls.


During much of this period, the KPD assigned responsibility for investigations to its Criminal Investigation Division (CID), which was broken down into an investigative section, an organized crime section, and an investigative support section (which oversaw services like polygraph tests, criminalistics, and warrant services, and which housed the crime prevention unit). The Division was a relatively large one; for example, in 1987 it employed 51 of the department’s 313 sworn officers, and spent $2.3 million of the department’s $14.7 million annual budget.

CID had a largely positive reputation in the department and in the local criminal justice system. As in most police departments, patrol handed off most cases to investigators through the reports they took at crime scenes,2 which were usually parceled out to separate units focusing on person crimes and property crimes; in the mid 1980s, the KPD also added a repeat offender unit that focused on cases in which the primary suspect was a habitual offender. The logic behind the decision to actually investigate these reports evolved considerably during the 1980s, as many detectives remember a growing concern with case management during this period. For example, one detective maintains that early in the 1970s, CID investigated almost every case that came through the department, but that as time wore on, the growing workload forced the division to develop criteria for “screening cases.” In 1980, as part of its new “managing criminal investigations” initiative, the division developed formal criteria for deciding which cases to pursue, using solvability and seriousness as the major factors. In any event, most cases that the division did pursue were investigated using traditional tactics—namely, by interviewing victims and witnesses and (in more serious cases) analyzing physical evidence.

The major exception to this picture was organized crime, whose cases were mostly driven by informant tips, and whose investigations often relied on covert tactics like undercover buys, fencing stings, and surveillance. The unit was outward-looking in the sense that it worked closely with other law enforcement agencies (including the federal Drug Enforcement Administration, the FBI, and the Tennessee Bureau of Investigations). But it seems to have suffered something of an image problem within the department, as a 1988 management survey revealed some suspicion about what went on inside it, and it also singled the unit out as having particularly strained relationships with the rest of the department—problems some KPD members attribute to the simple fact that the unit was located in its own facility away from police headquarters. Moreover, some people within the unit felt that it was too reactive. For example, asked how his unit prioritized cases ten years ago, one organized crime officer responds: “I don’t think there was any priority.” Another maintains: “Basically you went into the direction your informants led you and that was it. You didn’t guide their actions. They took you where they wanted to take you.”

3. Administrative Systems

In the minds of many KPD members, the department’s administrative systems did not really meet their needs in the years before community policing-related reforms. The department had developed formal systems to routinize many important organizational tasks—like collecting information, making assignments, and developing and enforcing the police budget. But most of these systems were fairly primitive, to the point that some did not really do the things they advertised.

With training, for example, the KPD made use of what was available, but many today believe that the results were inadequate.3 Some of these concerns simply reflect changing norms about what police should be taught: For example, some of Knoxville’s police today feel that the recruit academy did not spend enough time covering police-community interaction (a decade ago, the total number of hours spent on this topic amounted to at most 9% of the academy’s time). But in other areas, the problem was that the department simply lacked any significant capacity for in-house training beyond the basic recruit academy: For example, new sergeants and investigators got their training from outside schools, so that the KPD had to make do with what was available. Many department members were not happy with the results: They criticized many programs as too generic, and in some areas—as with training for new lieutenants, or field training for new recruits—, outside schools did not or could not exist, so that the KPD had to go without formal training altogether.

Other administrative functions offer similar stories: The old systems for patrol deployment accomplished the basic job of assigning officers to beats, but they ignored many important factors that affected an area’s workload. Budgeting systems did program each unit’s annual expenses, but the KPD lacked an automated system for keeping track of how each account was doing with respect to that prediction—thus it was difficult for a unit manager to actually use the budget to manage expenses. And as suggested above, information systems did record relevant information, but the system for accessing and disseminating that information was fairly primitive (with the result that in the 1988 management survey, only 15% of respondents felt that existing information systems met the department’s needs).

4. Management

Some of these deficiencies in administrative systems point to the problems of management in the KPD: For example, given the inadequacies of budget data, it was difficult for department managers to use their budgets to guide and develop their respective units. But according to many interviewees, there was a larger problem with department management: The KPD simply did not have much in the way of a direction, and it lacked the ability to follow through on many of the initiatives it did develop. The slippage that occurred after the World’s Fair is a case in point, as the KPD dropped many of its innovations because the department’s leadership failed to push them. Indeed, many grant-funded initiatives apparently suffered this fate, as the 1988 management survey concluded that “several beneficial programs have been implemented in the department only to be discontinued.” In other words, the department was able to identify new opportunities—typically with prodding from funding agencies—, but it did less well at building the widespread support for them that would be necessary to make them permanent.

The major problem, according to many department members today, was a lack of internal accountability: The department did not have an effective way of holding its personnel to the tasks it laid out for them. Indeed, some KPD members argue that there was nothing to be held accountable for, as management did not develop overriding goals and objectives that would guide each unit’s work; and in any case, its structure blurred responsibility (most notably in patrol, where command was based on shifts rather than areas, so that it was never clear who was to blame for a neighborhood’s chronic safety problems). But more basically, many officers simply report that previous administrations did not expect very much from employees: Many positions were considered sinecures for which little work was expected, and initiative was apparently rarely expected from lower levels (for example, officers remember directed patrols as supervisor-initiative jobs; they were rarely encouraged to identify problems on their own). As a result, it is not surprising that energy-intensive innovations such as the community partnerships that blossomed around 1980 were not sustainable in the KPD. The agency could fulfill all of its basic responsibilities, but after the World’s Fair, it lacked the dynamic quality that innovation demanded.


1. A New Direction for the Knoxville PD

As the 1980s wore on, the Knoxville Police Department was stagnating: Innovation had ground to a halt, and even hiring and promotions had been essentially frozen for five years. Against this background, it is not surprising that police emerged as a major issue in the city’s 1987 mayoral election, as all three leading candidates expressed their intention to improve law enforcement in the city. Winning candidate Victor Ashe helped set the tone for this debate by calling for “improved working conditions” for KPD officers, as well as a “fair wage” for police that would enable Knoxville to compete with other Tennessee cities for talent.

Mayor Ashe and the KPD

Before taking office, Ashe assembled a transition team that looked into the police department and made recommendations for change. Although he did not have a detailed agenda for Knoxville’s police, the new Mayor did want to be sure that the department got out of its rut of complacence, calling, for example, for KPD management to evaluate its strengths and weaknesses and to develop a plan for change. Ashe initially tried to work on these issues with the police administration he had inherited, but he reportedly ran into some trouble in doing so. One member of then-Chief Bob Marshall’s staff explains:

In any case, Ashe concluded that a change in the command staff was needed. The city’s civil service laws did not allow him simply to replace the Chief the way he could replace other department heads. But Marshall had been talking about retiring for years, and in early in 1988 he discussed that possibility with the Mayor.

Marshall proposed leaving the department in December of that year, but Ashe wanted to get started on reforming the KPD immediately. The solution he devised was to appoint a police commissioner above Marshall who would have final authority over departmental decisions. The Chief agreed to this arrangement, and in April of 1988, Ashe gave the commissioner’s job to Phil Keith—the man who Marshall himself had reportedly recommended to succeed him.

Ashe did not know Keith personally at the time, but according to the Mayor, the young lieutenant had “really a sterling reputation.”

Moreover, Keith clearly had his own ideas about how to run a police department, as he had been serving as a consultant to a number of local agencies. Ashe apparently saw this as evidence that Keith had the drive and vision he needed: “He was carrying on an active consulting business with other police departments where they were implementing a lot of the things that he talked about,” Ashe remembers. “But he’d never been able to do it in the Knoxville Police Department, because obviously he wasn’t in charge. . . . But as I told him when we interviewed, ‘You’ve done a lot of plans, you’ve just never seen them implemented. You’ll now have the opportunity to implement them.’”

Ashe initially hoped that the police commissioner job would serve as Keith’s permanent position, offering a way around what he viewed as a cumbersome civil service system.4 But after discussions with his law director, the civil service board, and Keith himself, it soon became clear to the mayor that with the exception of the safety director—a position Ashe wanted to do away with—, only the Chief’s job enjoyed complete legal authority over the department, and as a result he posted for the Chief of Police opening. As it turned out, few people were even eligible for the job: Of 17 captains, only two had the leadership training that civil service required for the job; and even after opening it up to lieutenants (including Keith) as well, not many more than a dozen were qualified. In any case, a handful of applicants went through the assessment center process, and Keith came out on top. Ashe remembers being relieved with that result: “Once they sent me the list, which was three names, and he was at the top of the list, I think I made a decision literally within an hour. It didn’t sit on my desk very long.”

Chief Keith and the Push for Organizational Focus

The new Chief took office with a relatively broad mandate for change, one focused on basic themes like professional development, accountability, and neighborhood focus. Indeed, Ashe purposefully kept his directions general, allowing Keith to flesh out the department’s plan of action himself: “I basically left it up to him,” Ashe remembers, “other than [saying] ‘I want you to make this as good and as effective a department as possible, and make it safe for neighborhoods throughout our city.’”

Keith brought to his new job a managerial vision that put a high value on building competence, accountability, and direction—a philosophy that in retrospect he describes as a push for “organizational focus.”

This latter problem, more than anything, was an ailment Keith wanted to cure. “I’m a firm believer in . . . ‘Let’s don’t talk about this unless we’re absolutely serious about doing it,’” he explains. Indeed, seriousness of purpose and the ability for follow-through are perhaps the cornerstones of his philosophy, underlying many of his more specific views about police reform.

Keith’s vision of a focused police department was something of a contrast to the field’s emerging ideas of community policing, in that it put organizational development before street-level reforms like problem-solving and building community partnerships.5 Keith was well aware of those fashionable ideas—Knoxville had, after all, been an early innovator, participating in things like the national ICAP initiative in the late 1970s—, and he thought that his program for reform would ultimately encompass the core elements of community policing. Nevertheless, he did not want simply to pick up prepackaged ideas and bring them in to Knoxville. “What I tried to do is stay away from cliches and catchy titles, and said, ‘We’ll call it what we want, but we’re going to have a business plan that talks to the issue of how we’re going to police in this community.’”

A Business Plan for the KPD

Keith saw the development of a business plan as perhaps the central aspect of his program of action, since it was the one concrete initiative he could pursue that would catalyze other important developments. “I knew that I had to change the work ethic, and I knew that I had to create some organizational values,” he explains. “Developing a business plan was the closest thing I could see to make us operate that way.” Equally important, a plan would serve to fill in the details of his vision of what the KPD should be doing—a vision that was deliberately indefinite, as it called for a focused organization without prescribing what the department should be focused on. To be sure, the Chief held a strong belief that strengthening accountability and improving the quality of police-citizen interactions should be cornerstones of reform, maintaining that “accountability, respect, [and] interaction with the community” are “the basics” against which the department should always be measured. Nevertheless, Keith felt strongly that others needed to contribute to setting a new organizational direction: “I spent full days with staff talking about vision,” Keith explains. “Getting them to understand that they too had vision, [that] it just wasn’t my vision. If we didn’t have a shared vision, then we couldn’t get there.”

But as the department began taking its first steps in this direction, some stumbling blocks became apparent. Judie Martin, the current head of Planning and Budget for the KPD and one of the participants in the first strategic planning sessions, recalls: “None of us had ever really done anything like that before, and we all were asked to come up with goals and objectives. Well quite frankly, most of them didn’t have a clue what a goal and an objective was.” Indeed, many department members began approaching Keith with their concerns; as he recalls: “There were some people in my department that came to me and said, ‘I recognize what you want done, but I also recognize I can have very little impact on that in helping you because my skills are limited.’ And I really appreciated [that] honesty.” Others were even more blunt: “Some came forth and said, ‘Where you’re going, I would only be a burden. I need to make a decision about my career’.” Keith attributed these problems not to individuals but to the organizational environment: “We had people who had advanced in a system that existed [that] was not producing what we wanted,” Keith explains. “It didn’t have what we needed to go into the future.” A clear need to build competence in the KPD quickly became obvious.

Keith describes his response as a series of “leadership development” initiatives—“everything from accountability, to vision, to strategic planning.” Planning, in particular, was a skill that most managers in the department simply had not been exposed to. “That scared the hell out of lot of people,” Keith recalls.

Beyond basic “skills” like how to do a strategic plan, Keith also felt that KPD management needed to understand the local community. In response, he arranged a series of seminars on Knoxville’s demographics and history.

Keith admits that these activities were not uniformly well-received. “I’m not going to tell you that they all sat there with their ears on,” he concedes. “Just like in any other organization, I had some that could care less: They were there to draw a paycheck, unfortunately, but that comes with the turf.” But on balance, the seminars seemed to him to be worthwhile.

In any case, these leadership development activities fed into the ongoing strategic planning, which amounted to several months’ work of introspection for the department. The process started out with management team meetings that reviewed the department’s strengths and weaknesses, outlined a general direction for the future, and laid out a rough plan for getting there. That, in turn, fed into a number of additional planning groups that focused more narrowly on issues like the mission of the KPD, managing calls-for-service, and patrol deployment. One in particular sought to institutionalize the planning process itself by investigating ways to develop employee participation in change and to improve employee skills and knowledge for doing so. The whole process was guided by Keith and a number of outside consultants, and it drew on extensive surveys of both the community and employees (including a management self-assessment survey undertaken before Keith took office and at the request of Mayor Ashe). Keith describes the effort as “a huge, huge undertaking to go through” but remembers it as one that helped set expectations and a direction for the department for years to come.

In the early planning sessions and those that followed (the management team has since met annually to revisit and refine its plans), a number of distinct themes have emerged as recurrent preoccupations for the department. First, problem-solving has become a central tenet of the KPD’s operating philosophy, as the department sees the ability to truly analyze community problems and develop an appropriate response to them—rather than instinctively resorting to arrest—as central to its mission. Community partnerships have been another major theme, with the department placing a special emphasis on accountability and building confidence in the police. Finally, the department has made offender apprehension a centerpiece of its community policing efforts, developing several innovative programs to target recidivists. As Chief Keith explains:

To support all three of these operational efforts, the department has focused on a fourth theme of “infrastructure development,” which highlights the need for appropriate support systems—particularly education and training to impart needed skills and ethics.

2. Building Support for Change

In the end, many elements of the KPD’s vision for reform had much in common with the fashionable ideas about which Keith expressed his skepticism. But Keith had never been opposed to the ideas per se so much as the common practice of importing them unreflectively—a practice that he felt would make it difficult for the department to make the ideas its own, and thereby undermine needed support for change.

First of all, he felt that announcing a new “community policing program” would be counterproductive because it would simply antagonize the rank-and-file. “If you put a title on something, it becomes a project,” Keith explains. “And it really feeds the organizational resistance issue because it’s viewed as, ‘This too will pass.’” As a result, the department spoke more generally of business plans and a change effort rather than narrowly about a community policing program; and it insisted that what was afoot was not a specific new project, but a long-term, department-wide reform. Many in the KPD believe that this holistic, incremental approach made it much easier for department members to adapt to change.

Second, Keith felt strongly that whatever vision the KPD did pursue, it needed to come from within in order to generate the necessary support. Indeed, the Chief maintains that he put the department’s very emphasis on change on the line when he first met with the management team to discuss where it wanted to go:

This basic intuition—the idea that letting the crew help set the course would generate commitment to the voyage—was to become important throughout the KPD as management looked for ways to build support for change. Directed Patrols, for example, had begun to fall by the wayside as a strategy for solving problems, and one common theory located the problem in the current management style: Supervisors rather than officers typically nominated DPs, and the result was that officers were not invested in them. In response, many KPD managers tried to inspire commitment to DPs by giving their officers power to choose new projects, as with this manager who explains how he has dealt with officer resistance to problem-solving:

One Lieutenant goes so far as to suggest that dealing with resistance may mean that not everyone gets involved in all types of police work: “You’ve got people that are good at certain things and you let them do the things that they would like to do.”

There were limits, however, to that accommodating stance: Once the KPD committed to a vision of organizational focus, the department expected its members to live up to that ideal, whatever their personal predilections. For his part, Chief Keith focused his attention on management: In innumerable meetings and seminars like those that paved the way for the first strategic planning effort, Keith has tried to help KPD managers understand what change demanded of them and to hold them accountable for results; the managers, in turn, were expected to do the same for the officers who worked under them. One sergeant explains how the Chief’s strong work ethic and demands for internal accountability—a central element of his ideal of organizational focus—have made their way down to the troops:

That the pressure was real is attested to by the fact—mentioned by several veteran KPD members and alluded to by Keith above—that many people simply left the department as the pressure to perform increased. “A lot of guys,” the sergeant continues, “did not enjoy someone looking over their shoulder making sure they worked the full eight-and-a-half, ten hours, 12 hours, however long they were here. And they just decided to go ahead and retire.”

Between these two extremes—on the one hand getting buy-in through accommodation and voice, and on the other hand ratcheting-up internal accountability—, the department spent enormous effort getting out the new messages through training. The details of many of these efforts are described below, as they touch on the administrative changes the KPD undertook to support community policing and related efforts: “Training” was not simply a matter of putting together one-shot sessions on community policing, but rather redesigning existing training systems (like the recruit academy) and developing entirely new ones that would become a permanent part of the department (like a leadership development course for newly-promoted lieutenants). Here it is enough to note that top management considered training as perhaps the single most important contribution they could make to changing the way the department thought about its work, and it spent enormous effort and resources developing new training programs.

3. Building Partnerships for Change

The need to build support for change inside the department was obvious to KPD management. But because of the nature of the plans it had been developing, the department also needed to build support in the outside world. First of all, the theme of “community partnerships” made it essential to strengthen relationships with Knoxville residents and businesses. Second, some of the department’s more substantive efforts—particularly its focus on chronic offenders, but also its more general emphasis on problem-solving—turned out to demand extensive collaboration with outside agencies and elected officials.

Reaching Out to Knoxville’s Community

Over the years since the first strategic planning sessions, the KPD has reached out to the Knoxville community through many special programs, including a community newsletter, an extensive volunteer effort, and a citizens police academy. But the core of its community policing efforts have involved attempts to develop partnerships with organized groups that aim to solve specific community problems.

As Chief Keith explains it, the idea of “community partnerships” has a very definite meaning in Knoxville:

In practice, the KPD pursued these two goals of accountability and confidence by trying to listen to community priorities—particularly those concerning which neighborhood problems deserved police attention, but also priorities about larger policy issues. On the other hand, the KPD has discouraged more direct community involvement in initiatives like citizen foot patrols.

Given this philosophy, the obvious question became how the department could learn about community priorities on a regular basis. In the beginning, Keith took on this task personally, trying to take the community’s pulse in a series of meetings with community leaders. As he explains:

But as time progressed, the department began developing more extensive and more formal links to community sentiment. Two major efforts in the city helped this project along: First, a groundswell of neighborhood organizing that began in the neighborhoods themselves, but which was facilitated and encouraged by the Mayor’s office and the KPD; and second, the department’s own organizing drive to develop a Community Advisory Committee and a citizen-driven crime control plan.

Building Community Capacity

In the years before Ashe and Keith took office, Knoxville’s neighborhoods had not played a very strong role in local government. But as the 1980s drew to a close, the city saw a torrent of neighborhood organizing that would in a few short years transform political life in Knoxville.

Ellen Adcock, a community activist who has recently taken a job as Knoxville’s Director of Administration, was on the leading edge of this wave when she helped start the West View neighborhood organization in 1988. Adcock and her neighbors had long been concerned about crime and code enforcement problems in their part of town, but they had never been able to make much headway on those issues because of a feeling that city government did not want to cooperate. Adcock saw Ashe’s inauguration as a signal that it was time to try again, and this time her efforts began to pay off:

Adcock was initially wary of the idea: “It took me about two weeks to get up the courage to call this guy I did not know but [who] lived within spitting distance from my house,” she remembers. But eventually the two met and organized a neighborhood meeting in the local elementary school, making phone calls and passing out flyers the city printed up for them to advertise the event.

Adcock remembers the first meeting as a success. “We had about 20 people that showed up that were just as frustrated as we were,” she remembers. The group started out by following up on the Mayor’s invitation to contact Community Development, and with that agency’s help and some advice from the few other organized neighborhoods that existed at the time, West View organized a neighborhood cleanup. Adcock remembers the effort as a huge success—not just in its immediate aim of cleaning up the neighborhood, but also in building some credibility for the new organization—, and she credits the city for helping to make it possible:

In any case, after the initial clean-up West View wrote up by-laws and launched a formal organization. The group has taken on various tasks over the years—Adcock maintains that they have met “like clockwork” on every second Thursday from the outset—, but its main focus has remained neighborhood cleanup and crime, which it has dealt with by meeting regularly with KPD officers and sharing information with them, and by working with code enforcement to put pressure on problem properties.

Adcock believes that many neighborhoods around the city were undergoing similar experiences at the time. “All of a sudden we found a voice,” she remembers, speaking for Knoxville’s neighborhoods in general. “[We said], ‘Hey, we have got somebody that is not only going to listen, they are going to give us some resources to do something.’ We were kind of like a bunch of infants, and all of a sudden we got some reinforcement and we grew up.” For its part, the city welcomed all of this organizing activity, but at the same time it started to feel overwhelmed. Adcock herself heard about the problem through Laurens Tullock, the Community Development director who had developed a relationship with the West View Group: “He kept seeing his staff bombarded with community groups that were trying to get started,” she remembers, “and he knew that really for it to have a long range effect that he needed to create something outside of his department.”

The solution that Tullock and Ashe developed was a new city agency called the Center for Neighborhood Development (CND), which was charged with helping neighborhoods to organize and even stirring up activity in areas where no groups were active. CND performed many services for groups like West View, including simple tasks like helping to print and design flyers and newsletters, and more complicated ones like its “Transforming Neighborhoods Together” grants, which offered start-up money to new neighborhood groups for various activities. Soon after it opened up shop, CND also funded a staff position for Knoxville’s new Council of Involved Neighborhoods (COIN), a recently-formed umbrella group for the city’s community organizations. COIN had started out small, focusing on the basic tasks of encouraging residents to form organizations and trying to help them work together, but the staff help from CND gave it the wherewithal to grow. The group quickly became an important force in city and even state politics, pushing for legislation and spending on issues that its member organizations found important.

In this way, the newfound synergy between city hall and the neighborhoods led to a spiraling growth in community organization. As Adcock summarizes that history:

Indeed, the Center for Neighborhood Development and the Metropolitan Planning Commission estimate that they have contact with over 200 community organizations in the city. Although comparable figures do not exist for the late 1980s, CND staff agree with Adcock that the number was far smaller then.

The driving force behind this enormous growth came from the neighborhoods themselves and from city hall, but the KPD played an important supporting role. For example, Adcock explains that when the West View group tried to form, the KPD’s crime prevention unit helped out by providing basic organizing help like printing up fliers; indeed, the KPD was probably the only agency in the city that had much experience organizing neighborhoods in this way. Equally important, the KPD, like other city agencies, helped encourage West View simply by responding to its requests—particularly by sending officers to its meetings. Adcock, at least, feels that patrol presence at the group’s meetings (which was arranged through the crime prevention unit) helped the two sides share information and to give the neighborhood some assurance that its concerns weren’t being ignored:

Indeed, Adcock describes information exchange as the most important part of West View’s joint efforts with the police. Some neighborhood residents wanted to go further by starting up efforts like foot patrols, but the KPD, in keeping with the philosophy outlined by Keith above, did not particularly encourage that type of involvement. “KPD was like, ‘Let’s see if we can handle it first,’” Adcock remembers. “The Chief is not real high on that kind of stuff . . . . He certainly does not encourage you to put yourself in any kind of jeopardy, . . . and probably that is a wise [position].”

In any case, the KPD clearly benefited from Knoxville’s explosion of community organizations, as the new groups provided a window into neighborhood concerns. The KPD had, of course, helped organize neighborhood watch groups for a long time, but even the crime prevention unit felt that those single-purpose crime watches were less sustainable than the multipurpose organizations that were beginning to emerge. In any case, the KPD and CND began working closely together, as each would notify the other about new groups that had formed and about the issues that were being raised. The result was a partnership that helped both agencies cement their neighborhood ties.

Community Organizing and the KPD: The Knoxville Crime Control Plan

Throughout this period, the KPD also made many internal changes to strengthen its community partnerships. The most notable shift was perhaps the department’s geographic reorganization into three sectors, each commanded by a captain with significant autonomy from downtown. Captains were charged with identifying and building ties with the various communities within their sectors (something they did by working with existing groups and holding open “roundtable” sessions to work on specific community problems); and officers were given a new geographic focus as well, as more of them were given permanent beat assignments and all of them were encouraged to take ownership of their beats. For example, the KPD now maintains a running list of neighborhood organizations and their meeting schedules so that a beat officer will always be in attendance—something that did not necessarily happen in previous years.

But the department’s most visible attempt to forge community partnerships came to fruition late in 1994, when the KPD undertook its first community-wide planning session to produce a “crime control plan” for the city and create a Community Advisory Committee in the process. The initiative apparently emerged out of a larger KPD effort at community policing, known locally as the “Service Excellence process,” which began around 1993. At that time, the KPD was awarded one of five Bureau of Justice Assistance grants (which was later renewed through the federal COPS office) intended to help develop model community-oriented policing programs. Knoxville’s application outlined a multifaceted approach that centered on the development of a community crime control plan, in which police and residents would jointly identify several public safety goals for the city and carry out specific initiatives to accomplish them.6 For the KPD, the effort was an attempt to strengthen the police-community relationship and thereby realize one of the main goals of the department’s own evolving strategic plan.

The Crime Control Plan was kicked off in 1994 with an informational breakfast hosted by Mayor Ashe, who had consulted with the KPD about the project. Ashe and Keith used the occasion to explain the effort’s purpose to those in attendance, focusing on its potential to help the department better understand community concerns by providing a forum where any Knoxville resident could be heard. As Keith explained it at the time: “No community speaks with one voice, and people who present themselves as community leaders are not always representatives of the neighborhoods served by police.”7 In any case, the assembled group concluded the breakfast by scheduling the first crime control planning session.

The first planning session itself was fairly well-attended, with 125 people in attendance. The group spent an entire day focused on one objective: Developing a set of goals that it would spend the next year trying to achieve. KPD staff and grant-funded consultants were in attendance to help facilitate the work, providing basic guidance about the planning process and breaking it down into several stages. But despite this help, the first planning session ran up against some of the same skills issues that the department itself had faced six years earlier, as community members (like their KPD counterparts in 1988) simply did not have much experience with planning. As Judie Martin, director of the KPD’s Planning and Budget unit explains:

Nevertheless, by the end of the day the group came to agreement on eight workable goals that would serve as the framework for its activities during the coming year—including imperatives like “improve police-community communication,” “expand community education regarding prevention,” and “strengthen housing in neighborhoods.” Residents then volunteered to work towards one of the goals by serving on an “implementation team” charged with realizing it.8

The “police-community partnerships” team can serve as an example of the work the crime control plan has led to and how that work gets done. Team chairperson Elizabeth Harrison, who got involved in the Crime Control Plan through her participation in Neighborhood Watch, explains the team’s purpose in broad terms as an attempt “to literally improve the relationship between the community and the police department.” Team members developed several strategies for realizing that goal, including an anonymous drop box intended to allow residents to voice their concerns and share information with the police, an “adopt-a-church” program that pairs individual officers and churches together on a volunteer basis, and a citizen’s police academy that aims to give citizens a better understanding of the police role. The team itself has primary responsibility for carrying out its projects, and Harrison explains that she and others try to maintain work equity within the half-officer, half-citizen group as well. “We have worked very hard to see to it that both sides of that committee stay involved—that we do not end up having police officers do all of the work, because that is just one more thing for them to do.” For example, team members designed the drop boxes themselves and arranged for the necessary hardware to be donated: “All the police department did was print up the cards for us,” Harrison maintains.

In order to keep track of what each of these teams is doing and to facilitate information-sharing among them, the department created a Community Advisory Committee (CAC) made up of the chairs and co-chairs of each implementation team, as well as any other community members who wish to attend. The advisory committee serves several purposes. First, at each its bimonthly meetings, team chairs report on their progress towards the goal they have been assigned as a way of keeping other teams and KPD management up-to-date on their activities. Second, as its name implies, the Advisory Committee also serves as a forum for community opinion that helps the department understand community priorities: Keith regularly briefs the committee on current departmental concerns, asks for advice on some policy decisions, and solicits any additional concerns committee members want to raise. Finally, the committee frequently invites outside speakers to discuss current concerns; for example, state Senator Bud Gilbert recently spoke to the group about the process of passing a law and about how committee members could most effectively make their concerns known to the legislature. CAC attendance was fairly low at the outset, but it has grown steadily, and about 30-50 people are reportedly in regular attendance today.
Impacts of the Crime Control Plan and Directions for the Future

Beyond the specific “implementation” activities it has catalyzed, the Crime Control Plan was always seen as a broader effort to forge police-community ties and improve communication between the two sides. Many participants, at least, feel that it has succeeded in doing so, particularly as the number who take part has grown.

Harrison, for example, argues that participants in the crime control plan become spokespeople for the department—or “voices out in the neighborhood,” in her words—who help the community understand how the police department works. As an example, she points to a recent police shooting of a black man named Juan Lorenzo Daniels who had threatened to kill himself—an incident that became a cause célèbre in the city, turning out some 350 residents to a city council meeting to protest the police’s actions. According to the KPD, the two hostage negotiators who responded to the call spoke with the man for over an hour but opened fire when he lunged at them with a hunting knife. Particularly controversial was the fact that officers reportedly fired four rounds each at the man’s chest, a barrage that some community members found excessive. In any case, the incident raised many questions about the KPD’s use of force policy, and Harrison, who lives in the middle of the area that is in an uproar about the incident, argues that CAC members like herself played an important role in clearing up some misunderstandings:

Harrison does not, of course, maintain that the CAC alone could quell the type of uproar that came in response to the Daniels shooting, and indeed, the incident has perhaps demonstrated some limits to that institution. One important problem in this regard is that while most of the discontent about the shooting came from the black community, the CAC has had difficulty expanding its small base of participation among blacks, as CAC chairperson Rhonda Garren explains:

The problem of uneven participation is hardly unique to the black community—indeed, Garren and Harrison both maintain that the most underrepresented group on the CAC is Knoxville’s business community; and by Garren’s estimate, four to five of the CAC’s thirty to fifty regular attendees are African-American, a respectable if not perfect ratio in a city where blacks make up about 16% of the total population. Nevertheless, the CAC’s difficulty reaching out to the black community may seem particularly problematic both because of the frightening reasons behind it and because of the fact that it is precisely in the black community that relations with the KPD are most strained. In any case, Garren and others in the CAC have attended tenant association meetings in the city’s predominantly-black public housing developments to try to encourage people to get involved.

In the wake of the Daniels shooting, some KPD critics—notably Dewey Roberts, president of the local chapter of the NAACP, and Knox County Commissioner Diane Jordan—, have called for a different model of community involvement, namely, a civilian review board that would oversee internal affairs investigations and have input into department policy. Keith argued against the proposal partly on the grounds that it seemed to him unlikely to achieve the purposes that motivated it.9 But at the same time, Keith himself has entertained hopes that the CAC would play a more active role reviewing department policy—something of a halfway point between the review board model and the CAC’s current focus on solving discrete safety problems—; and in the wake of the Daniels incident, he discussed ways of getting more citizen input of this sort with department management.

To be sure, Keith and others insist that the CAC does provide some direct input into department policy choices, as described above. Nevertheless, the Chief agrees that community input on policy issues “isn’t systematic like we’d like [it] to be,” speculating that the problem reflects a general difficulty eliciting interest outside of times of crisis. “If you don’t have a real major problem, it’s hard to get people interested,” Keith maintains.10 “When you have a declining crime rate, and you look at [community] surveys and [people] feel safe, the question is, ‘So what’s your point?’ . . . And that’s the challenge that we have: How do we get them to focus beyond the crime problem?” As an example of the difficulty he has faced in broadening community focus, Keith describes how an early CAC initiative that reviewed internal affairs complaints fizzled out due to lack of interest:

Similarly, one CAC member remembers a KPD presentation about its crowd control strategies at football games—presumably aimed to solicit community views on the matter—and reports that those in attendance did not have much interest in the issue. In any case, Keith himself hopes to be able to develop new ways of involving the community that will improve police-community dialogue, reflecting, “We’ve got to look at our strategy on how we’re bringing the neighborhoods in. I think the message is wrong from my perspective . . . You know, we organize around problems—we say we want community input, but community input is around problems, not about how we develop for the future.”

The other difficulty facing community input—particularly in the strong, community control format proposed by Roberts and Jordan—is resistance by some officers to the idea.11 As Deputy Chief Robert Coker explains, “Some officers thought right off the bat [that] citizens should not have too big a role in telling officers what they want [them] to do. . . . They were real hesitant and some of them are still hesitant about how much input the community [has] . . . on the direction of the department.” Nevertheless, Coker—who believes that this sentiment derives from what he calls “that well-intentioned lie that the police have been relying on for so long, ‘If you’ve got a problem, you call me, I’ll fix it’”—believes that many officers have come around as they have discovered that “most of the community are exactly like us.”

In any case, given the twin difficulties of officer wariness and community apathy, the KPD has had to be resourceful to realize its aim of being responsive to community priorities. In large part, the hope seems to be that by cultivating links with individual neighborhoods, and by sponsoring community-wide forums like the Crime Control Planning sessions, the department will develop a keen sense of what the community wants—a sense that KPD management will be able to rely on when they make important choices about policies and operations. As Coker explains it, the department may not run every policy decision past a body like the CAC,

So driven by this desire not to be left with a lot of chicken on its hands, the KPD has tried in many ways to discover what, in fact, the community it serves does want.

Building Support in Government: The Repeat Offender Programs

The KPD’s growing dialogue with the community was apparently its main strategy for building outside support for its new efforts. To be sure, Keith and others in the department clearly understood that the KPD would need support from elected officials for change as well, and to help meet that need the department has invited city council members to its management retreats from the outset. Keith sees these meetings as opportunities to show elected officials that police are serious about their work: “Mayors have their own agenda, schedules, and things like that, as do council members. But when they come in and see a group of police personnel there with sleeves rolled up, . . . they say, ‘They do put a lot of thought into this.’ That’s fifty percent right there.”

But beyond these overarching efforts to bring the community and elected officials along, the KPD has built a political strategy into its planning process for individual initiatives, as managers are trained to ask whose support will be needed to put any new reform into place. In fact, most of the KPD’s work with elected officials and other government agencies has revolved around individual projects, like a recent pawn shop ordinance, repeated CPTED initiatives, and larger neighborhood problem-solving projects. The department’s many interrelated efforts to target repeat offenders can serve to illustrate what this way of working has entailed.

The KPD’s repeat offender initiatives got their start in 1986, when then-Lieutenant Keith and others in Knoxville successfully applied for a federal grant that would make their city one of the nation’s first demonstration sites for the Serious Habitual Offenders Comprehensive Action Program (SHOCAP), which had been created by the federal Office of Juvenile Justice and Delinquency Prevention (OJJDP). The KPD played a central role in coordinating the efforts that followed—particularly through its newly-minted Repeat Offender unit, which helped to focus police attention on the job—, and it quickly helped to create an interagency task force that included agencies like the school system, the juvenile court, and the district attorney’s office. After a week-long training arranged by OJJDP, the task force developed criteria that would allow the juvenile court to identify habitual offenders (specifically, it created a point system that scored individuals based on the number and types of crimes they had committed). The group then set out to develop a strategy for dealing with the offenders, settling on a system that attached strict rules and penalties to anyone designated a “SHOCAPper.”—a system that was backed-up by an extensive regime of information-sharing that effectively put these serious offenders in a fishbowl.

The SHOCAP program itself was widely considered a success in the region, but what is perhaps most significant about it is the way it blossomed into other efforts. First of all, over time the evolving task force applied the point system model to offender groups other than juveniles, creating a Gang Offender Comprehensive Action Program (GOCAP) in 1996 and recently embarking on programs to target adult repeat offenders and at-risk youth. In each case the basic logic has been the same: As Captain Randy Lockmiller, who spearheaded the recent GOCAP initiative, explains, “We sat down and said, ‘All right. If you have an individual’—let’s take gangs, for example—‘if you have an individual that you think is a gang member, who would have contact with this person throughout the systems that are out there?’” The answer to this question then becomes a roster of agencies whose support must be enlisted—eventually through a formal agreement signed by each agency’s head.12 The KPD and other task force leaders have used various strategies to get agency agreement, from appeals to self-interest to an incremental approach that builds on small successes. In any case, the efforts seem to have paid off, as the initial core of about a dozen participating agencies has more than doubled over the years.

Perhaps the clearest way to appreciate the interagency dynamics involved in these efforts is to examine a few examples of the individual agencies that they have brought the KPD in contact with. Besides clarifying the process of partnership-building through greater detail, such close-up pictures reveal how the task forces have spawned far-ranging partnerships that that go well beyond their formal programs. Consider first the regional parole board.
The East Tennessee Board of Paroles

The East Tennessee Board of Paroles was a founding member of efforts like GOCAP and the adult SHOCAP interagency task force, but its relationship with the KPD blossomed outside of those formal bodies. The first overture came from the KPD’s new Repeat Offender Program unit, which at the time was fumbling around for a purpose. As former ROP officer Robert Hubbs remembers:

In the early stages, Hubbs and the other ROP officers simply tried to meet informally with Parole officers to learn the identities of these “career criminals,” who then became a focus for the unit’s attention. For example, if a rash of robberies occurred in a known parolee’s neighborhood, the ROP unit might at least try to rule the parolee out as a first step in its investigation. In the process, the unit’s officers developed strong relationships with Parole Board personnel, helping them deal with problems like their difficulties in getting arrest warrants served.

But by all accounts, the Police-Parole partnership truly took off some six years after SHOCAP and the ROP unit got started around an ID card system for new parolees. The KPD got the idea from SHOCAP itself, which had required repeat youth offenders to carry cards identifying themselves as such, and which thereby made police and others aware of the fact when they were dealing with serious juvenile offenders who had specific conditions attached to their parole or probation. Police thought that the same idea might be applied to parolees, since they too tended to be repeat offenders, and they called a meeting to propose the idea to the parole board.

Jim Cosby, director of the East Tennessee Board of Paroles, remembers his initial reaction to the idea as skeptical, for he felt that the ID cards might unfairly label parolees as incorrigible. “I thought, ‘Well, we’ll put a brand on their head, a big ‘P’ on their head and brand them for life’,” Cosby recalls. But over a series of meetings with the police department, the Parole Board director began to change his mind. First of all, some reflection on his changing caseload convinced him that the parole population was in fact becoming more and more dominated by serious offenders for whom concerns about rehabilitation might unfortunately have to take a back seat to concerns about control. In Cosby’s words, “I think one of the main things that convinced me was simply taking a good, hard look at our clientele and the type of offenders that were getting out the door at that time . . . Because they’re high-profile offenders, and we don’t have any choir boys and girls anymore.” But equally important to Cosby was his perception that the KPD was not being overzealous in its proposal to target parolees. “I was convinced that they were not trying to run a trail ’em, nail ’em, and jail ’em program, and that they were not out to persecute parolees”

In this way, Cosby began to see the KPD proposal not as an attempt to “persecute parolees” and thereby undermine his mission (which in part included the goal of helping parolees to reintegrate into society), but instead a legitimate attempt to keep better tabs on an increasingly hardened population of parolees. “Our mission statements are very, very similar as far as public safety and protecting the public is concerned,” Cosby explains. “At those particular points in time we were just going about it in an independent fashion—the left hand didn’t know what the right hand was doing.”

Having mutually agreed to the system in principle, the two sides spent about six months ironing out the details. The program that emerged was relatively straightforward: When a parolee first meets with his parole officer, he receives a color-coded ID card that identifies the type of crime that landed him in prison and outlines the conditions of his parole—conditions like curfews he must observe, places he cannot frequent, and people with whom he cannot associate. When an officer stops a parolee on the street, the parolee must identify himself by showing his card, and failure to do so can be grounds for revoking his parole or attaching new conditions to it.

The emerging ID card system was quickly joined and even improved by better information-sharing between the two agencies. State law had long required the parole board to notify local law enforcement agencies about new parolees, but the information had always been sent out as individual paper notices that a particular prisoner was being released. In that form, the information was practically useless for a fairly large city like Knoxville, which at best kept the notices in a file cabinet that did not facilitate any analysis. But as the Parole Board became aware of this problem through its growing dialogue with the KPD, it began to offer the information to the KPD by computer—at first on diskettes, and later through a remote server. The KPD’s crime analysis unit quickly developed a sophisticated search engine for analyzing the data, making it possible to print out maps of parolee addresses for beat officers and for investigators to identify potential suspects in unsolved crimes by analyzing the database’s location and criminal history information. Over time, the two agencies added more and more fields to each parolee’s record, so that what began as a simple file listing demographic and physical information about each parolee evolved into an extensive database that included information about each offender’s vehicle, parole officer, employment address, and so on.

The KPD returned these hi-tech favors in two ways. First, it began sending information back to the Parole Board—most notably information about beat officers’ contacts with parolees, which the Parole Board saw as a valuable way to help keep track of its clients. (This particular bit of information had been made possible by the ID card system itself, which had made it easy for the KPD to flag all the Field Investigation cards it filled out for parolees.) Second, as the crime analysis unit perfected its search engine for the parolee data, the KPD decided to offer that software to the Parole Board for free. The offer was primarily a “thank-you” gesture for having received the information in the first place, but crime analysis Sergeant Robert Hubbs admits that getting the Parole Board to use the software may ultimately benefit the KPD as well: “It helps everybody: They’re really going to get involved now in making sure this information is accurate, because they’re going to be using it.”13

In any case, out of these early efforts grew a number of collaborations between the Parole Board and the KPD—including a ride-along program designed to forge relationships between beat officers and parole officers; several ad hoc sting operations intended to help the Parole Board arrest elusive parole violators; and, more recently, a “team supervision” initiative that may enlist service providers and police officers to help parole officers monitor and manage their parolees. Indeed, collaboration between the two agencies has begun to snowball in a way that suggests that some sort of critical mass has been reached.

As an emblem of this evolving relationship, consider the mundane fact that the Parole Board now assigns its cases according to the same geographic system that the KPD uses. In previous years, parole had assigned cases almost at random: One parole officer might supervise offenders in opposite poles of the city. But as the two agencies began developing the ID card system and the ride-along program, Cosby decided to organize caseloads by police districts:

Thus as the two agencies’ relationships became closer, differences between them faded away to the point that they began to share an important part of their organizational structures. Cosby’s initial sense that the police and parole board missions were congruent had evolved into an even clearer artifact of the things that parole and police work had in common.

The Knox County School District and the SHOCAP Transition School

Of the many agencies the repeat offender initiatives have brought the KPD into contact with, the Parole Board has turned out to be perhaps its closest partner. Even there, some important philosophical differences needed to be addressed at the outset, but the challenges were less intractable than they turned out to be with other agencies. A good example of these challenges and the task force’s strategy for dealing with them comes from the local school system.

Knox County schools participated in the SHOCAP task force from its inception, but their concerns about the repeat offender programs have an equally long history. From the task force’s perspective, the school system was one of the most important institutions young offenders came into contact with, and it potentially offered valuable information about SHOCAPpers’ behavior. But from the schools’ perspective, the task force was making unreasonable demands for information (as described above), and it was not doing enough to prevent serious offenders from disrupting school life.

In SHOCAP’s early years, the latter problem was apparently the most serious one. Rhonda Garren, who at the time worked for the Department for Children Services in general probation, recalls that the school system initially took a hard line on dealing with serious young offenders:

At first, the disagreements over how to respond to even minor violations became so severe that the school system actually pulled out of the task force and stopped sending a representative at all. But within a few months, the schools assigned a new representative to the SHOCAP task force, and two strategies for dealing with the schools’ concerns emerged.

The first was simply educating the educators about the constraints various justice agencies operated under. “We had to explain to them that on the violation of probation or after-care you have to have several charges before you can recommit them,” Garren explains. “In the State of Tennessee, we had so few secure institutional beds that we could not lock the students up in institutional beds for missing one day of school. . . . And we had to do a lot of training with the principals that that wasn’t the whole basis for the SHOCAP program, to keep them locked up. It was to keep them in the community, law abiding and [under] watch.”

To a point, these explanations seemed to work, as the superintendent and his new representative to the task force warmed up to the SHOCAP program and agreed to participate in principle. But individual principals still did not always fall into line, and state law gave them significant autonomy to make decisions about their schools; in Garren’s words, “They have the superintendent of schools, but he doesn’t actually tell the principals what they can do in their schools.” For example, one principal found an obscure code in the Tennessee Code Annotated that he interpreted as giving him the power to expel any student with a criminal record. Garren remembers the problems these sorts of actions created as severe: “We had principals that were telling these kids, ‘You can’t go to school here,’ [while] part of their probation was that they had to go to school.”

Garren sympathized with the underlying concerns: “I can understand the principals’ philosophy,” she maintains. “They do not want those students in school.”

In response, some members of the task force made an ambitious proposal to develop a “transition school” for the most hardened young offenders, raising the idea in a meeting between DCS, the school system, the juvenile court, and the KPD. When the local players responded enthusiastically to the idea, the group brought the proposal to the state board of education and received permission to start an entirely separate institution, together with a $200,000 grant to help finance it. The state’s cooperation put the final pieces of the puzzle in place, and the agencies that had attended the original meeting set out to carry their plans forward; for its part, the KPD, assigned a training and education specialist out of its crime prevention unit to help design pieces like the lesson plans that KPD officers would deliver.

When completed, the school gave strong emphasis to discipline and order. Security, for example, was tight: The twenty-student school was initially patrolled by two full-time security guards, and a police Lieutenant was always present at the building as well. Weapons were screened out by multiple searches, including one by the security guard who picked up every student at home, and gang colors were strictly forbidden. Indeed, the surrounding community would settle for no less, as it had initially protested against locating an institution for serious offenders in the neighborhood. As Garren remembers it, “[We] explained to the residents that they were much safer being located near that school than any other high school, because students in transition schools don’t go the bathroom without a security guard with them.” At the same time, the schools’ designers argued that strict discipline was partly for the students’ benefit, as it ensured that they would have a structured environment for learning.

In any case, the discipline-based system was backed-up by the juvenile court, which was as central to the transition school as it was to the rest of the SHOCAP effort. Students were court-ordered into the school, and they were told that there would be no suspension and no truancy from it; if anyone failed to attend, a police officer would be dispatched to pick him up and take him to school. In the extreme, repeated discipline problems would land students back in prison. At first, Garren explains, some students balked at the threat: “They didn’t care if they were sent back to this institution near us. It’s like ‘big deal’: They’ve got a nice place to sleep, good food and everything—better then home.” But the juvenile court judge was able to get the attention of recalcitrant students by threatening to send them not to the local juvenile institution but to a far-away facility called Taft, which had a reputation as a tough place where the state’s worst offenders from its largest cities wound up. “The judge told them if you’re recommitted for violating over at the transition school, you don’t go to this one up here; you go to Taft,” Garren explains. “And the first two kids that we had in court [said], ‘Well, let me think about that then. I don’t want to go to Taft.’” In any case, the point is that the ability to maintain order in the new school depended on the existence of credible threats against misbehavior—which in turn depended on the cooperation and resourcefulness of the juvenile court judge. “Without [the judge’s] support, you couldn’t do much of what we did,” Garren maintains, speaking not only of the transition school but of the SHOCAP effort in general.14

Those involved in developing the transition school were more than pleased with the results, but some observers—particularly the students’ own defense attorneys—criticized the effort on philosophical grounds, arguing that it was an overly punitive institution that only further separated young offenders from the mainstream community. Transition school supporters did not agree with that assessment: They described the school as a “stepping stone” back to regular schools for young people who might otherwise not attend school at all; moreover, they pointed out that the school’s 10-to-1 student-teacher ratio meant that its students got much closer attention than their counterparts in the rest of the public school system. Defense attorneys, in particular, were reportedly not convinced by those arguments: For them, the transition school was just one more misguided attempt to label their clients as incorrigible—an attempt that underlay the entire SHOCAP effort. But those objections effectively ended at the juvenile courtroom door; as Garren remembers it:

Thus the new school did not please everyone, but it had the support it needed to function, and it was able to provide the outlet the schools wanted for those youth they considered disruptive.

4. Implementation

All of these broad efforts to build support for change both inside and outside the department went hand-in-hand with more concrete administrative reforms. To be sure, Keith insists that the department’s changes cannot be reduced to any particular new policy or system: “It wasn’t an engineering effort,” he maintains. “It was almost like a biological change—I mean, we just kind of shifted.” Nevertheless, the department did pursue innumerable reforms of its organizational systems, and together, those reforms add up to a substantial portion of the change it went through. For the sake of example, consider first the department’s changes in its personnel systems—one of the more intensive efforts the KPD undertook, and one that underlay many of its other initiatives. (As Keith puts it, personnel policies are central in labor-intensive organizations like police departments because that is “where the majority of our resources are vested”).

Architectural Responsibility: Personnel Reform in the KPD

From the moment he first took over as police commissioner, Phil Keith tried to shape the personnel he had inherited through training and reassignments. But he had apparently always seen these actions as short-term measures: For him, the KPD’s “competency” problem in areas like leadership skills and technology stemmed from the historical system that department members had advanced in—thus getting to the root of that problem meant reshaping the systems that had generated it. Consequently, in addition to a short-term challenge of leadership, Keith believed that he faced a long-term challenge of organizational design. With respect to hiring systems, for example, he describes the changes made during his administration in the language of an artisan: “When I first became Chief, I obviously didn’t hire any of those people. But today out of the nearly four hundred people we have employed, I’ve hired three hundred and thirty or three hundred and forty of them. So I can’t run from whatever is out there. I have to take architectural responsibility for it.”

These architectural efforts to shape the department’s personnel took many forms, most of them aimed to increase competency, create a focused workforce, and encourage the more specific skills that ideals like community policing demanded. There were, of course, other changes to personnel systems as well, notably a massive rationalization of the department’s patrol deployment system (which had traditionally made almost no allowances for variations in call load) and a redeployment of many desk-bound officers to the street. But reforms of the systems designed to shape the character and motivations of people who work various departmental jobs—those systems focused on training, evaluations, hiring and firing, and assignments—can serve to exemplify the way the KPD used reform of administrative systems to help create a focused organization.

Leadership Development

Training was a central element of administrative reform in the KPD, especially in the area of leadership development, where the department developed entirely new programs for its sergeants and lieutenants. Keith maintains that the need for this training was severe: “Leadership development—we didn’t have any,” he explains.

The solution was to develop separate in-house courses for sergeants and lieutenants, designed by a small team that included Keith, Deputy Chief Robert Coker (who oversees the patrol force), and Deputy Chief Jerry Day (who oversees investigations). The team started out by learning what they could about the few existing programs that were offered around the country, blending those ideas with their own in order to develop a tailor-made curriculum for their evolving department. The first-line supervisors training, for example, was able to pay considerable attention to the ways in which sergeants could facilitate problem-solving—something that was becoming more important to the KPD but which was not covered effectively in the state training program the department had traditionally sent its sergeants to.15 As Coker explains:

In any case, such specific topics were blended with a more general supervisory curriculum that covered a broad spectrum of material—everything from basic tools like patrol deployment to more general principles like the importance of values and integrity in leadership. The result was an intensive four-week course that all new sergeants went through upon promotion, and a five-week course for new lieutenants.

Training the KPD Patrol Force

At the same time, the department also sought to revamp its training for patrol officers at all levels, revisiting the recruit academy, developing a new Field Training Officer program, and greatly expanding its in-service offerings. As with many KPD initiatives, most of these reforms were guided by a special committee on the topic formed after the initial strategic planning sessions. This committee did not receive direct instructions from Keith or the command staff; instead it was given fairly wide latitude to develop training that would support the department’s emerging mission.

Its basic strategy for doing so was to let the organization and the community tell it what training needs were most pressing. The committee’s very design reflects this strategy, as it regularly brings together employees from all parts of the department (both sworn and nonsworn) to brainstorm about current training needs. But the committee also reviews extensive survey information relevant to training—including the city’s annual budget surveys (which solicit community opinion about the KPD and other Knoxville agencies) and a custom-designed training survey developed to ask KPD employees what they felt their training weakness were. All of this information is used to help the committee prioritize training needs each year, so that curricula can constantly adjust to changing perceptions about what the department needs (for example, the committee can reprogram in-service training each year in response to weaknesses identified through the surveys, the only constraints being state-mandated topics like the eight hours of firearms training that all officers must undergo). In the words of Sergeant Don Jones, who coordinates training for the department: “Basically we’re just like big sponges over there: We absorb all that we can, and we try to disperse a product that will solve all of the needs that everybody has come up with.”16

The approach is an incremental and additive one, in the sense that the committee typically identifies new needs that it should try to fill each year, and the result has been constant growth in how much training department members receive. The recruit academy, for example, has shot up from 18 weeks of training in 1986 to 25 weeks in 1997 (Tennessee mandates only 10 weeks) in order to offer new or expanded instruction in areas that committee members considered important; for example, the academy has more than doubled the amount of time it spends on interacting with the community. Many of the new courses are decidedly non-traditional fare, focusing on topics like cultural awareness, active listening, and public speaking. But Jones maintains that few have criticized the new curriculum as “soft” because of the way in which the changes were introduced—namely, at a gradual pace and by supplementing (rather than replacing) the existing courses: “Going from that eighteen week academy . . . to the twenty-five week academy has been a slow process,” he explains.

The cost of this additive approach to reform has been a literal one: Both the academy and the department’s in-service offerings have become longer and more extensive, and their pricetags have grown proportionally. The department has drawn on different sources to pay for this growth, including the city’s general fund. But Keith maintains that grant funding has been central to the effort—particularly for the many in-service training sessions developed to educate long-time KPD employees in the new ideas: “Had we not had some of the resources that were made available to us by grants, we may not—not even we may not: We would not have achieved a level of change that we have. Somewhere down the road we might have gotten there. But it wouldn’t be 1997.”

The most recent training grant the department received came from the federal COPS office to fund a regional training facility for community policing, one of 33 around the country. The KPD had recently “gone regional” with its recruit academy, receiving the state accreditation necessary to train new officers and deputies from nearby jurisdictions. But the regional training institute offered a further opportunity to create a community policing curriculum for police employees and community members throughout the eight-state area around Tennessee. In any case, when the department received an RFP for the grant program, it seized on the opportunity, seeing it as an occasion to share its growing expertise in community policing and to increase its visibility with neighboring agencies.

As required by the grant, the KPD proposed a joint program with the University of Tennessee and the Metropolitan Drug Commission—an agency that the KPD had long used to help build ties to the community, and which it hoped would help bring community members into the design of the program and into the training itself. The details of the training programs were left somewhat open, as the three agencies will soon conduct a needs assessment to set the curriculum’s framework, but the basic focus will be the KPD’s self-defined specialty of “infrastructure development” for community policing (including topics like crime analysis, managing calls for service, working effectively with neighborhoods, and leadership roles). In any case, the three agencies have recently hired a director for the effort whose mandate includes both getting the effort off the ground and finding funding to make it sustainable (the one-year grant is renewable for up to three years, but beyond that period it must fund itself).

The regional training facility clearly will not take over all of the KPD’s new in-house training programs. But it is interesting as an indicator of how far the department’s training efforts have come: When Phil Keith took office as police commissioner, he viewed existing statewide training as inadequate to the KPD’s changing needs, and the department was forced to develop its own in-house training capabilities. With the regional training institute, the KPD hopes to bring those maturing capabilities to the region as a whole—reshaping the regional context whose shortcomings had pushed training reform to begin with.

Other Personnel Reforms

Training may have been the single largest investment of time and energy among the KPD’s administrative reforms, but the department revamped many other personnel systems as well. Transfers, for example, have been tightened up, in the sense that the KPD rigidly holds applicants to the selection criteria for every advertised position. In the past, an applicant might have been accepted for a position if he or she met three of the job’s four selection criteria. But today no such exceptions are made. If no qualified employees apply for an open position, the personnel division will actively recruit new applicants until the position is filled. The result, the department hopes, is a system where each job is filled by someone who is truly competent to perform it.

Personnel evaluations have also received an overhaul, notably by incorporating problem-solving into their criteria. This reform followed much the same path as those in training: It was initially proposed and agreed to by an employee committee, which determined that problem-solving should be an important dimension for measuring performance. But implementation decisions eventually reached the level of city government, which needed to sign off on the changes to performance pay that were associated with changing performance measures. The reform took some time to settle in, suffering, it appears, from the unintended and unwelcome effect of turning problem-solving into a numbers game for a while. But ultimately, many feel, it became the common understanding that officers would not be rated solely on their quantity of problem-solving, but also on its quality.

Finally, the department has used personnel systems to influence what type of people work for the agency to begin with. First, as described above, turnover definitely fed into the new administration’s change efforts. Deputy Chief Coker explains: “We had decided that after a while, you can only leave the door open for change for so long and you’re either for it or against. And if you’re against it, it’s probably time to go on and look at a different opportunity.” History conspired with the administration to make turnover a feasible strategy for change: Knoxville had annexed an enormous amount of territory around 1963, and the many new hires the department took on at that time were reaching the 25-year retirement mark when Keith took over the department.

But retirements were only half of the equation between turnover and change: The other half concerned how the department would fill its newly-vacant positions (and, for that matter, the many new positions the department added in the years after Keith took over the KPD). To this end, the department revamped its hiring policies and practices in order to bring in new officers who would contribute to the agency’s changing mission. For example, minority hiring became an important area of concern, and the KPD stepped-up its recruitment efforts by enlisting the help of community leaders (such as area pastors or representatives from the Urban League) and by contacting minority colleges throughout the nation (the department had traditionally recruited primarily from the state of Tennessee). As a result, the number of minorities in the KPD’s ranks has doubled since 1989, though the agency recognizes that it still has more work to do, as its makeup does not completely reflect the community it polices.


The KPD retooled many other administrative systems after Keith took office—from a massive effort to update its information systems, to a long-term initiative to develop its planning capabilities, to a gradual change in the way it carried out budgeting. Indeed, administrative reform has apparently dominated the KPD’s change efforts, largely because the vision that underlay it had been framed in the broad terms of “competence” and “organizational focus.” By contrast, Knoxville has apparently been skeptical of new street level programs and units: Keith, for example, argues that “you can have real successful special unit programs, but your overall mission is not reached’.” In a similar vein, many department members are quick to emphasize that for them, “community policing” is not a special program along the lines of the storefronts and bicycle patrols that are becoming so common around the country but instead a department-wide way of doing business.

That is not, of course, to say that new operational innovations were absent in the KPD, and Knoxville has gone on to develop many of the street-level innovations associated with community policing (including, in fact, those very storefronts and bicycle patrols, together with other efforts like new crime prevention programs, a domestic violence unit, and decentralized investigation of property crimes). But to remain true to its overarching vision, the KPD has tried to approach these efforts in a holistic way that ties them into its larger efforts. As an example of the way it does so, consider the department’s efforts to expand its capabilities for problem-solving.

Problem-Solving in the KPD

Knoxville determined early on that it would not develop a dedicated problem-solving unit the way many police departments had begun to do: Instead, it would make problem-solving a part of every officer’s duties. Its first strategy for doing so was to revive the city’s directed patrol program, which had started up during preparations for the World’s Fair but had gradually been neglected over the course of the 1980s. That development was among the first that the new administration questioned: “It worked incredibly effectively for about five years,” Keith recalls, referring to the directed patrol strategy. “[But] we kind of got away from it, [so] I said, ‘If it was effective, why did we stop doing it?’” In response, the KPD tried to bring the program back to life, which at first essentially meant getting supervisors to identify problem areas and to direct their officers to patrol those areas between calls.

The second strategy for bringing problem-solving to the patrol force was a broad- based effort to develop problem-solving competence. In part, this job included a 40-hour training program that the department designed and required for all officers. But more ambitiously, it included the in-house development of a new framework that went beyond DPs. Indeed, Keith considered problem-solving as an idea that transcended the patrol force altogether: “We looked at the whole mechanism,” the chief remembers. “It could include bigger organizational problem-solving, like ‘how do we solve employee grievances?’” But the center of this effort seems to have been the problem-solving kits the department developed for patrol.

KPD command staff were aware at the time of widely-used frameworks like the SARA model (named after its four steps of scanning, analysis, response, and assessment). But although they agreed with the basic SARA framework, the administration felt that the system suffered some weaknesses. as Deputy Chief Robert Coker explains,

Coker and others particularly felt that SARA didn’t provide enough structure on the analysis phase, in part because it did not explicitly require officers to look at multiple possibilities and choose the best among them.

Given these perceived weaknesses in the most common problem-solving framework, the KPD set out to develop its own by building on a device known as “the crime triangle.” In a nutshell, the triangle is an argument that that all crimes require three things to be committed: An offender, a victim, and an opportunity. Traditional police strategies focus on removing offenders from the community, but the crime triangle points out other modes of intervention: First, one can reduce the number of potential victims by helping the public take preventative measures; and second, one can reduce the number of opportunities by inserting a “guardian” between the offender and the victim (as in the cases of directed patrols and community surveillance). The problem solving kits, according to Coker, were designed to help officers consider all three legs of the triangle:

Specifically, the new kits incorporated these ideas into the “analysis” phase of the five-stage process they outlined, requiring officers to list anything known about the likely offenders, targets, and opportunities that might be responsible for the problem.

With one exception, the remaining steps followed this same pattern, building on the existing SARA framework by fleshing it out in more detail (the “response” stage, for example, asks officers to list individual tasks, the people who will accomplish them, the resources they require, and the dates when they will start and finish). The exception came between analysis and response, where the KPD inserted an intermediate step called “developing a strategy.” In that step officers were required to lay out at least three potential strategies and rate each one according to specific criteria (potential effectiveness, practicality, community support, and resource availability). The strategy with the highest combined score then became the one that the officer would carry forward.

The administration was more than pleased with the kit itself and the way officers began using it, feeling that it eliminated SARA’s core weaknesses—particularly its failure to operationalize key terms like “analysis.” Nevertheless, some department members have reacted somewhat negatively to the kits, and while these reactions hardly mark the new tool as a failure, they do suggest that the pitfalls of problem-solving are less an error to be corrected than a dilemma to be managed. As one officer puts it:

Other officers echo these sentiments: One, for example, maintains that the kits “intimidate a lot of officers,” and the result seems to be that department members lean towards the simpler directed patrol strategy wherever possible. “It’s human nature,” one officer explains. “If you have a blank DP form and they can fill out one piece of paper and work on something they want to work on, and then you hand them a problem solving kit, it’s obvious which one they’re going to gravitate towards.”

This reaction seems to suggest a fundamental tension in problem-solving: On the one hand, simple frameworks encourage problem-solving by making it easy to do, but the efforts that result will likely be superficial; on the other hand, complex frameworks force officers to take problem-solving seriously, but at the same they may discourage them from taking the initiative in the first place. Indeed, the KPD’s two levels of problem-solving—the directed patrols and the problem-solving kits—may be seen as one answer to this dilemma, as they give officers a ready outlet for simple problems while preserving a carefully-controlled structure for more complex efforts.

Quality Control

In any case, problem-solving proceeded apace in the years following the revival of the DP and the development of the “kits.” The main strategies to institutionalize this relatively new form of police work were apparently training and leadership of the sort already described above. But in 1995, the KPD added a new tool that sought to develop it further.

At that time, the department posted for a new position called the “operations analyst” that would serve as technical support for the deputy chief in charge of patrol—particularly in the area of deployment, since the KPD still lacked a scientific way of distributing manpower across shifts and areas. But over time, the position developed an additional function of coordinating and controlling the quality of problem-solving efforts, and of helping the organization to learn from its experience with them.

These new duties did not emerge by plan but rather out of the initiative of two department members.17 The first was Robert Coker, who took over the patrol force as its deputy chief a short while after the analyst’s job was created. Coker apparently wanted to go beyond the position’s limited mandate, particularly by using it as a way to do some basic quality control on the various problem-solving tools. Officer Terry Moyers, who has held the operations analyst position since its inception, explains the basic idea in terms of the way Coker talks with him about patrol work:

Before Coker took the reins in patrol, the deputy chief apparently left such matters to the district commands, for he did not ask Moyers to collect information about the kits or the DPs. But this arrangement turned out to be problematic, for the newly-decentralized patrol force undermined coordination across districts. At best, decentralization meant that each district did not have any opportunity to learn from the experiences of the others. At worst, it led to redundancy, as neighboring districts and other KPD units targeted the same problems on the same nights, entirely unaware of each others’ plans. In any case, Moyers recalls that Coker asked him to start reviewing problem-solving projects, saying: “I want you to start collecting all the data on problem solving kits and DPs. This is what I want you to look at, and anything else you think I need to know.”

With this open-ended direction, Moyers himself became the second personality to influence the job duties of the operations analyst, as he determined to try to evaluate the various efforts informally. As he explains it, “If I was going to waste time to look at [the problem-solving projects] and see if they were doing the same thing . . ., I could take an extra five minutes to find out why they weren’t working.” Moyers describes this job as a series of questions he asks about each initiative: “[First,] we want to look and see if the tactic was really carried out: Did they really get to do it? Or did it get pulled off for this reason here? Did it fail because of a lack of backing by the department? Did it fail because of a lack of funding? Did it fail because of the community not getting involved in enough of it?” In any case, Moyers felt that over months of reviewing problem-solving projects he had developed a good sense of which ones seemed to lead to trouble and which ones seemed to work, and he was determined to share that information with Coker and the districts.18 Moreover, as his overview of departmental problem-solving made him into something of local expert on the topic, Moyers began receiving requests for assistance. “They’ll come to me [and ask], ‘This is not working for us. What’s wrong with it?’” he explains. “Then I’m able to look at it, and a lot of times I’m not able to tell them. [But] sometimes I am: It just depends on what is wrong with it [and] if it is fixable.”

In any case, Moyers gradually became a clearinghouse of problem-solving knowledge in the department. Each month, he regularly reviews all active problem-solving kits (which on average consist of something over a dozen major projects) and a sample of the departments six-hundred or so directed patrols, scanning the lot to make sure the efforts are what they purport to be, that they do not duplicate other efforts, and that they will not repeat old mistakes. The information is then sent to Chief Coker for use in his monthly meetings with the district commanders, where the group can coordinate and troubleshoot whichever problem-solving projects need the attention. In that way, the department hopes, problem-solving will become integrated into patrol work rather than running the uncertain course of an isolated program.

5. Grant’s in Knoxville’s Past and Present

Throughout these changes, as Keith maintained, grant funding has played a very visible role in the KPD. That fact is not surprising, for grants have been important to departmental innovations for at least the past two decades.

As remembered by Judie Martin, who today oversees grants from the KPD’s planning and budget office, the department’s ties to the grant world emerged somewhat adventitiously. “Both the Chief and I, it’s just been a piece of what we have done,” she explains, referring to then-Lieutenant Phil Keith. “I don’t know that anybody ever said ‘this is what you will do,’ [but] that’s primarily what I [started doing].”

The federal Law Enforcement Assistance Administration (LEAA) programs of the 1970s were a particularly important catalyst for the KPD, as Martin effectively became the county grantsperson under the LEAA coordinating council system. That arrangement meant that she did planning and helped spearhead grant applications not only for the police department but also for other county criminal justice agencies (such as the DA’s office and the Sheriff’s department, with whom the KPD pursued several joint applications). By the time LEAA money had dried up, Martin was something of a local expert on grant-getting, and she and Keith continued to put together applications for whatever new programs came up. For much of the 1980s those opportunities were slim, and the department had to compete for the limited discretionary funding that the federal government was offering (for example, the KPD received grants for managing criminal investigations, crime prevention, and the SHOCAP program during this period). But later on in the decade, Byrne grant money became widely available through the state, followed closely by the Police Hiring Supplement Grants and the various COPS programs.

The KPD vigorously pursued all of this funding: As Martin describes it, the department’s strategy has been to pursue whatever money is available so long as it is compatible with the department’s objectives. Indeed, grant-getting has become institutionalized in the KPD to the degree that grant-money is an expected part of each year’s budget: “We certainly planned for it in our budgets,” Martin explains. “We put in funding for the match moneys and that type of thing even though we may not have known exactly what’s coming down the pipe.”

In any case, Martin describes two ways in which the department typically goes after grant funding:

In either case, the department has felt a decided preference for equipment grants and new programs. “We look at grants traditionally as seed programs,” Martin explains. “We tend to go for getting [programs] started with the equipment, the supplies, and the travel. Kind of the peripheral things rather than the personnel, because we know we obviously have to pick up personnel, and we don’t want to lay anybody off.”19 In this way, grant funding has long been tied to innovation in the KPD (and indeed, it is interesting to note that its periods of change correlate almost perfectly with fluctuations in the federal revenue streams).

The Title I COPS Grants in Knoxville

Given this philosophy of grant-getting—and particularly its aversion to hiring programs—, the personnel-based COPS grants came onto the scene as something of an anomaly for Knoxville. The department was not about to ignore the money, since it always sought to make use of whatever outside funding was available; but the new federal program nevertheless presented it with a dilemma. In part, the KPD’s response was simply to focus on the minority of COPS programs that did not entail new hiring. But in part, Knoxville tried to adjust to the personnel grants by trying to soften the impact that salary phase-outs would have.

Strategic Hiring

Mayor Victor Ashe shared the KPD’s discomfort with the three-year COPS hiring grants, explaining that “from a budgetary standpoint, I did not want my budget to [skyrocket] all of a sudden. I mean, it’s going to jump up, but [I didn’t want it to] jump up astronomically to where a tax increase or laying off officers were your only two options. Either one is difficult to do.” The result, according to Ashe, was that the city used the program somewhat conservatively: “We have not utilized it to its full extent,” Ashe explains, referring to the COPS hiring programs, “because of that three-year outlay.”

Even so, the department did apply for a substantial fraction of the personnel it might have been eligible for, hiring a total of 57 officers, three civilian crime analysts, and (through the YWCA) a domestic violence counselor with COPS money (the Police Hiring Supplement grant funded 18 more officers). In order to do so and still stay clear of the city’s concerns about phase-outs, the department tried to think strategically about the timing of new hires, particularly in laying out its projected hiring schedule for its first Universal Hiring application (as the COPS office required). “The chief knew where he wanted to be,” Martin explains, “and that was a logical progression of looking at increases in the city [population], the annexation issues, and the calls for service.” But the magic number, 416, was more than 50 officers above the current level (and almost as far above authorized strength), so Keith determined to go slowly:

In any case, in the first two years of the COPS and PHS grants, the department received relatively small grants that together added up to 28 new hires; and in 1996, it was awarded funding for 47 new officers staggered over two years. By 1997, the KPD had enough hires in the pipeline to bring it up to its new authorized strength, so it chose not to apply for more officers (in fact, it is eligible to hire 7 more officers with federal funding but has decided to hold off for now).20

The new hires helped expand the department’s numbers dramatically. “We would have still had some increases if it had to be entirely locally funded,” Ashe maintains. “But probably not at the same level.” Normal attrition meant that sworn strength never quite reached 416. But it came close, and in any case the growth was marked and noticeable (for example, from 1993 to 1997, the number of officers per 1,000 residents rose almost 25%, from 1.82 to 2.25). Officers on the street are aware of the increase and the fact that much of it was funded by grants—as well as the fact that the department must pick up the costs after a few years—, and although many still claim to feel pressed for time when topics like problem-solving come up, most seem to admit that growth has helped release their workload pressures.

The new officers have had other impacts as well: For example, the Knoxville Community Development Corporation (KCDC), which manages the city’s public housing, had long contracted with the KPD to have officers patrol its developments at overtimes rates. But as its funding from HUD began to fall in recent years, KCDC asked the KPD if it could create a more formal arrangement for its developments—namely, to assign a team of regular-duty officers to patrol public housing, thereby saving the agency from overtime rates. At first police resisted the idea. “Their numbers were down in the department,” explains Fred DeBruhl, KCDC’s executive director. “They didn’t have the manpower.” But the KPD’s recent growth spurt changed that situation: “As time went on [and] their numbers increased,” DeBruhl explains, “[Keith] got to a point where we felt like we could do that.” In any case, although the COPS hiring grants were not specifically targeted for the new KCDC team, it seems likely that they played an important role in making it possible.
Beyond Hiring

Other COPS programs presented less of a dilemma for the KPD, and it aggressively pursued almost every available program that did not involve hiring. Some of these grants have already been described: For example, the COPS DEMO grant (originally out of BJA, which funded five “model community policing programs” around the nation), was precisely the sort of seed program that Knoxville favored, and the department used the money to fund consulting help and other work for its crime control plan, as well as pieces of other initiatives like its community corrections efforts. Knoxville also viewed the regional training institute grant as a seed program: As described above, the department hopes to use the initial funding to get the program running, but it has hired a director for the institute and given him an explicit mandate to find long-term financing for the effort.

In addition to these seed programs, the department’s other main thrust was in technology—equipment-based expenditures of the sort that Knoxville had long looked to grant money for financing. Even the COPS Domestic Violence grant had a significant technology component that involved building a database to track victims. But COPS MORE, which focused on technology and civilianization, fed most directly into the KPD’s growing concern with information systems—an area that had been a priority since the department’s 1987 strategic plan, but which many felt was still lagging behind. Part of the MORE money that Knoxville received funded three civilian crime analysts (the initial idea was to provide one analyst for each precinct, but one of the positions became the analyst for investigations). But the bulk of Knoxville’s $786,000 in MORE grant money went for computer equipment. First of all, the department is purchasing 10 personal computers to develop a computer lab that can be used for training purposes; as it has begun to computerize many departmental functions, the KPD has found that many officers lack the necessary skills. Second, the department plans to purchase some 220 ruggedized laptop computers that officers will use in their cars.

The laptop initiative aims to extend an earlier effort funded by state grant money, in which the department bought 175 laptops for officers to use in their cars. But that endeavor faced several limitations: First of all, at the time the only ruggedized models available were the expensive military models, so the department had to settle for off-the-shelf 486 computers, which had shorter battery lives and more fragile construction than the department would have preferred. More important, the systems ran aground of implementation problems. For example, the computers were intended to essentially serve as substitutes for the MDTs that Knoxville officers have never had: Officers would be able to tap into central computers for information like vehicle registrations and warrants, and they would also be able to write up and send reports directly to the records department. But for various reasons, the department has not yet been able to secure a radio channel for these data transfers, and although officers could write their reports on their computers, they had to download them manually at the end of each shift. Unfortunately, some bugs in the original software design, together with the usual difficulties of teaching employees a new system, led to various degrees of “lost reports.”

In any case, the 486 computers were taken back from the officers, and they will be replaced by the new MORE-funded ruggedized models. Those computers will be phased-in, starting with a smaller “test group” of day-shift officers—an arrangement designed to allow information systems personnel to work closely with the officers to troubleshoot any problems that arise. The 486 computers will then be given to other department personnel who have better access to power than officers in the field do, such as investigators or administrators. The result, the department hopes, will be a fully-computerized department. “Our intent,” Martin explains, “is that when every officer comes through the academy, when he gets his weapon, his vest, and his radio, he also gets a lap top. He’s got more things to carry than he knows what to do with.”


Ten years of change have offered considerable opportunity for movement in the KPD, and department members maintain that their agency has transformed itself dramatically. Keith, for example, argues: “I can tell the difference in core competencies today in everything from putting together something as simple as a newsletter.”

The outcomes of many departmental reforms have already been suggested in the catalogs of how they were accomplished. But here briefly consider today’s KPD through the same lens we used to view its past: The department’s relationship to its environment, its operational and administrative systems, and its management.

1. Relationship to the Environment

The Authorizing Environment

Coker’s hamburger stand metaphor suggests the way Knoxville police think about setting priorities—namely, by trying to foresee which services their community desires. To do so, the KPD tries to listen closely to both elected officials and local community groups.

Structural changes in Knoxville government are one indication of the department’s growing openness to community direction: Whereas previously the chief of police was a somewhat insulated civil service appointment, Ashe has since changed city law to make the position at-will. Keith himself was grandfathered into to the old system, so formally he has the same autonomy as his predecessors. But it is clear that he and the mayor collaborate closely on major policy decisions (such as the department’s response to the Daniels shooting); Ashe, for example, maintains that the two have formal meetings weekly to discuss police issues and that they talk informally on the telephone nearly every day.

A good example of the way these police-government interactions play out comes from the city’s approval of the KPD COPS grants: As Martin explains it, the KPD developed its “wish lists” on its own, but with council’s perspective firmly in mind. The result was a set of tempered proposals that received easy approval—a pattern that she describes as the norm in Knoxville:

In this way, a close police-government relationship does not necessarily mean that the one directly commands the other. Each side respects the other’s autonomy, but on the understanding that both will make reasonable demands.

The police-community relationship, reviewed extensively above, follows a similar pattern: Police do not necessarily run individual policy decisions through formal community boards like the CAC, even if that does happen occasionally. But the department takes pains to learn about community priorities through its many links to the community—including the advisory board, neighborhood organizations, neighborhood watch groups, and the district roundtables (and to some extent the Mayor’s annual budget surveys, which solicit community opinion on a variety of topics). Sensitized to community opinion through these conversations, the department then makes whatever decisions it faces—from making policy to identifying community problems—with its own sense of that opinion in mind. Not all officers or even administrators seem to have equal enthusiasm for attending community meetings and responding to community input, but the proportion seems large21 and has almost certainly grown in the last 10 years, if for no other reason than that the community itself seems to be more vocal.

Communications with Knoxville’s black community are, however, problematic, as suggested by the outcry over the Daniels incident. To be sure, Keith insists that the department enjoys considerable support among many African-Americans, and that even after Daniels was shot, hundreds of blacks wrote in to express support for the police. Nevertheless, officers, public officials, and black leaders report that the relationship between police and black residents is often strained, particularly in the city’s public housing developments: Residents often do not wish to cooperate with police, whether to testify in criminal trials, to give input at community meetings, or simply to talk with officers who respond to calls in their neighborhoods. Many Knoxvillians interpret these problems as signs not of fundamental distrust of police but of fear of retaliation for “snitching.” But either way, there is apparently a serious wall of silence separating officers from some segments of the black community. The department does seem concerned about potential distrust of police: For example, after maintaining that many blacks went out of their way to express support for the KPD after the Daniels incident, Keith goes on to admit: “Now, that doesn’t mean we don’t have problems. We do. I think it’s just an operational issue that you have to keep on top of the table all the time.” In this connection, the chief points particularly to the problem of poor supervision, as he feels that supervisors often fail to rein in the minority of officers who do not act professionally and respectfully.

The Task Environment

Community oversight of Knoxville police on the decisionmaking side is matched by interorganizational collaboration on the service side—much of which has emerged out of the various repeat offender programs, as chronicled above. Criminal justice agencies, in particular, have begun to work more closely together, as parole, the juvenile court, and the district attorney’s office regularly share all manner of information and coordinate their priorities. There are, of course, notable exceptions to this rule: Many officers express frustration with the correctional system, complaining that local jail crowding has hampered their ability to control offenders. And relations between the KPD and the Knox County Sheriff are apparently poor, owing to a decade-long history of strife that has been repeatedly reported in local newspapers. Many KPD members argue that those tensions are restricted to administrative levels, maintaining that officers and deputies have no difficulty working together on the streets. Nevertheless, the tensions do sometimes have destructive impacts on KPD strategies. For example, the Sheriff’s department is reportedly tight-lipped with arrest data and suspect photographs, releasing only what is legally mandated, and rarely making it convenient for city police to retrieve them. But in any case, this difficult relationship is apparently the exception to a rule of strong interagency collaboration in the regional justice system. And even in the exception, the problems have not been fatal, as police and the County Sheriff have been able to work effectively together on the repeat offender projects.22

Though less visible, partnerships outside the criminal justice world have recently grown as well. The repeat offender task force, for example, has added a preventative focus on at-risk youth to its agenda, thereby bringing social service providers and mental health agencies more firmly into a group whose central focus has been on criminal justice strategies. Outside this family of initiatives, partners like code enforcement, traffic engineering, KCDC, and private property managers have developed collaborative approaches to tackling public safety problems with police—particularly through the strategy of Crime Prevention through Environmental Design, an area in which the KPD has sponsored repeated training sessions both for its own staff and for outside agencies.

2. Operations

Inside the KPD, change has touched many aspects of the department’s front-line operations, which divide into the patrol division and the criminal investigations division (CID). Consider each division in turn.


The KPD’s patrol division employs over three-quarters of the department’s sworn personnel and spends 64% of its budget (up from 49% in 1987—though the jump is partly an artifact of tangential developments, like moving 911 out of the KPD). Patrol divides into many smaller units, including a traffic services section, the repeat offender unit, the crime analysis unit, and a patrol support services section, which oversees functions like crime prevention and the K-9 detail. But the bulk of the patrol force is assigned to Knoxville’s three police districts, which in their few short years have bred strong loyalty and competitiveness (for example, there is apparently a strong norm against transferring out of a district).

There is some variation among the districts in how their patrol forces are organized, but all three work around a series of lettered squads. For the most part, officers in squads A through E have regular beat and shift assignments, and they have wide-ranging responsibilities for their beats during their tours of duties. F and G squads, which start work in the afternoon, serve as a reserve force for taking 911 calls when A-E officers are tied up—whether for other assignments (like directed patrols) or simply for other calls (and indeed, the supplementary squads mostly work during periods of high call volume). Finally, the department has recently added an H squad with variable hours that takes on special assignments, such as traffic patrol or special event policing. In any case, the new squads work within a geography-driven command structure that set them apart from the old detachments: Each district is headed by a captain, and each captain oversees his own group of squad officers.

Officers have many specific duties, including call response, random patrol, and problem-solving. Underlying all of them, however, is an expectation that officers will work hard, efficiently, and intelligently. A mundane but clear example of what this work ethic means is the awareness among officers of one thing they should not do: Pile up “backup” cars on a single nonserious call. “That,” one officer explains, “is the sort of thing the Chief gets really mad about”—apparently because it is seen as a waste of departmental resources. Similarly, purely “random” patrol seems to rank low on the KPD’s totem pole of priorities, since much of the “free time” patrol officers have between calls gets programmed through the directed patrol system. Some officers resist this rationalization of their work: “I know that you want to get your money’s worth and you want to operate things like as business,” one argues. “But I still think there are certain cars that ought to be out there answering calls and just looking around . . . . Some of the best things I’ve ever gotten into I’ve run across; I haven’t been assigned to go look for them in other words.” But this opinion certainly does not dominate in the KPD, which seems driven by push for productivity.

As far as working intelligently is concerned, the department has in recent years fed much more information into patrol work. Officers regularly receive updates about their beats, including information about the calls that arose during their absences as well as the addresses of any parolees who live nearby. Moreover, dispatch now gives premises history information on all domestic violence calls in order to better inform officers about the situations they are assigned to handle. On the other hand, some officers are still frustrated with their inability to get needed information. In particular, since officers lack MDTs, they must call into the radio room to request information on vehicles and suspects, and some of them feel that this system does not always work effectively.

In any case, the two major claims on officer time are emergency calls and problem-solving. Although calls for service still clearly dominate officer attention,23 problem-solving has apparently made significant inroads as an accepted part of patrol work, primarily through the directed patrol system (which is reportedly dominated by traffic, prostitution, and suspicious person problems that are concentrated in particular areas).24 In one district, officers on the A-E squads report that they typically have at least one DP on any given shift; and in all three districts it is unusual for an officer to go a week without doing a DP. Of course, it is difficult to ensure that every DP is taken seriously (just as it is difficult to ensure that every emergency call is): Officers report that many DPs must be done between calls, as time does or does not allow, and that even when they are taken out-of-service, a moderately urgent call can pull them away from their DP.

Indeed, 911 apparently trumps other types of activity as well. For example, when one CID unit tried to involve patrol officers more intensively in its investigations, it reportedly found that the officers who participated were often called off the project to take emergency calls. But by the same token, officers also report that DPs have taken them away from emergency response—which is simply to say that the competition between patrol’s two major duties runs in both directions. As one officer summarizes the matter: “Sometimes the thing that happens with those [DPs] is that we get a lot of them going and have problems catching up with [calls]. And some days we don’t get to do things that we want to be doing because of the call volume.”25


Problem-solving and community policing have been centered in the KPD’s patrol force, but the department’s more general efforts to improve organizational focus have touched on its Criminal Investigations Division as well.26 Policies and procedures have been tightened up in many CID subunits, from criminalistics to organized crime, but the division’s overarching Managing Criminal Investigations (MCI) initiative is perhaps the preeminent example of the way in which the KPD has tried to rationalize detective work. Under MCI, all cases are thoroughly reviewed for solvability factors (including things like the presence of witnesses and types of physical evidence), which are in turn used as the basis for prioritizing and assigning all cases. The system has evolved considerably over the years, as new solvability factors have been articulated, investigators have been more intensively trained in applying them, and as patrol officers have been better-schooled in taking down the preliminary reports that lie at the foundation of the entire process.

As in patrol, new information systems have also changed the way CID does its work. Most investigations apparently still follow the traditional model, relying heavily on interviews with witnesses and victims. But in difficult cases where leads are few, it appears that investigators increasingly turn to computerized databases: In particular, investigators of many stripes report using crime analysis databases on repeat offenders and parolees to identify potential suspects (as one detective puts it, the strategy is to identify a “known offender” in the vicinity of a crime wave “who has the potential for committing the types of crimes you’re looking at”). Similarly, the department has also pushed to expand its pawnshop database, which allows investigators to crosscheck all pawned items against incoming theft and burglary reports.

Beyond these efforts to work smarter, CID has also developed diverse initiatives for working in a more community-oriented way—building ties with both community members themselves and the patrol officers who are the acknowledged center of community policing. Community focus is in evidence in many new policies and procedures, including diversity training for investigators, new victim services for domestic violence cases, and a growing sense of obligation to keep victims informed about their cases. It is also evident in the division’s case priorities—notably in the organized crime unit, whose staff maintain that street-level dealing is an important priority because “those cases [are] the most obvious to the public” (something it has learned from its anonymous tipline as well as frequent personal appearances at community meetings). High-level dealers, by contrast, may be essentially unknown in the wider community.

Finally, CID’s attempts to engage the patrol force have taken many forms, from case-by-case joint investigations, to weekly meetings with district commanders, to the aforementioned effort to decentralize property crimes investigations (in its final form, this effort created mixed teams of detectives and patrol officers who work together to investigate all property crimes in a district). Some KPD members report mixed success in these initiatives, pointing to a long history of strong separation between the patrol force and the elite detectives division that has proven difficult to overcome. But in any case, community policing seems to have led to a growing sense in the KPD of the interdependence between its two operational divisions.

3. Administrative Systems

Underlying both sides of the department’s front-line operations is a newly-invigorated body of administrative systems. In other words, many department members maintain that policies and procedures regarding personnel, information, budgeting, and many more mundane things are better-tailored to their stated goals, and that in any case they are more explicit and thorough.

Planning, for example, has become an elaborate production in the KPD: Every fall, the 60-person management team leaves Knoxville altogether to go on a two- to three-day retreat, where participants hammer out the department’s goals and objectives for the coming year and assign responsibility for accomplishing the associated tasks. These annual retreats, moreover, are followed up with regular progress reports: Each month, the team meets again to give updates on the status of their various initiatives, and the department’s command staff also meets weekly to discuss current concerns. The entire planning system, of course, has been further bolstered by extensive training and practice regarding how to plan, with a special emphasis on developing measurable goals and objectives in order to ensure that plans can be used to establish accountability. The result, KPD members argue, is a comprehensive system that takes planning seriously—a condition that did not necessarily hold a decade ago.

Other administrative systems have apparently been tightened up in similar ways. The KPD’s successful drive for national accreditation triggered many administrative reforms, leading to more explicit policies and procedures regarding things like officer pursuits. But more fundamental systems have also been revamped. Information systems, for example, are currently in the middle of a massive overhaul that touches many areas of the department outside of the operational areas that have already been described: Budget data are now available on the department’s computer network; inventory is now exhaustively tracked by computer; and information on emergency calls, response time, and DPs are now available to patrol commanders to help them make deployment decisions. Moreover, the department is currently revamping its records management system in order to coordinate the many decentralized databases that have emerged throughout the KPD and to make it easier to access information in the first place.

4. Management

Administrative rationalization has not, however, meant absolving department managers of their personal responsibility to make decisions and to guide and facilitate employee work. “You can’t rely on that computer to be your decision maker,” Keith explains, pointing to one of the department’s ubiquitous CAD computer screens. In other words, administrative systems are not intended to codify department members’ work, but to facilitate it.

For example, many years of leadership development initiatives have been directed towards areas like integrity and values as much as the technical business of how to deploy manpower. This sort of training is apparently intended to instill a sense of departmental mission, which KPD managers are expected to carry forward. Often that expectation means holding employees accountable for following stated departmental policies and procedures; but it can also mean encouraging them to think beyond regular routines and to respond creatively to individual situations.

To that end, the department has provided considerable flexibility at low levels of the hierarchy. In patrol, for example (though similar examples exist in other units), district captains have a fair amount of autonomy from headquarters, which allows them to use their allotted resources as they see fit. In order to facilitate unpredictable projects like problem-solving efforts, district commanders and even their underlings have the flexibility to do things like juggle shifts by allowing officers to trade scheduled days for unscheduled ones. This authority is not unlimited: For example, districts are apparently discouraged from using overtime as a resource for scheduling unanticipated problem-solving efforts. But district captains argue that they have essentially all the authority of a police chief for their areas of responsibility. They are then expected to use that authority to achieve agreed-on departmental goals.27

In this philosophy, managers are seen as resource providers who encourage discretion (even if they ultimately insist on accountability to an overall mission). One sergeant, for example, exclaims: “You know pretty much what we are? We work when the officers put us to work. We help the officer with analyzing the problem, we help them when they come to us [after] identifying the problem, [and] we help them with developing resources toward the solution of the problem.” For example, in discussions of particular problem-solving efforts they have been involved in, other sergeants report helping officers to contact outside agencies for their troops, or referring them to coworkers who have tackled similar problems for potential ideas.

In Keith’s eyes, this is an approach that must transcend traditional “problem-solving” efforts to embrace every aspect of departmental work: Officers are not to be held accountable for slavish attention to procedure, but for achieving the department’s public safety mission. As an example, he describes a hypothetical traffic stop and how he hopes an officer might handle it:

Creative patrol work like this is partly encouraged through administrative systems that make room for officers to spend the necessary time—notably call diversion systems that reduce the 911 burden. But according to Keith, it also demands supervisors who encourage their employees to exercise discretion intelligently: Neither abusing it by ignoring the larger purposes at issue nor avoiding it by sticking to formal procedure.28

Ever the perfectionist, Keith admits that the department has much work to do in creating a climate where this sort of response will be the norm. In particular, he believes that instilling a broad sense of organizational mission is still a major frontier for improving management in the KPD:

Though framed as a self-criticism, this idea reveals much about what KPD management does emphasize—namely, an overall departmental mission that will guide all levels of decisionmaking; or in other words, a strong sense of organizational focus.

1 At one time the department maintained its own dispatch unit under the Communications Section of its administrative bureau, but in 1987 this unit was replaced by an autonomous Enhanced 911 agency that managed all emergency calls for the Knox County area.

2 This was apparently about the only link between the KPD’s two major divisions: Except for a brief experimental period in the late 1970s, when patrol officers were sometimes assigned to investigative teams, there was reportedly little interaction between the patrol force and CID. Indeed, many managers responding to a 1988 survey described the relationship between the two units with terms like “estranged,” “uncooperative,” or even “miles apart.” Organized crime, in particular, was a problem. The unit was physically located outside police headquarters—it was referred to as “the other police department”—, and officers often did not know who worked in the unit at all.

3 Budget allotments suggest that the department simply did not value training very highly: For example, in 1987, the department spent a total of only $749 on its line items for meetings and training and for education and training reimbursement; by 1997, the figure was $18,915.

4 Indeed, Ashe would go on to change the civil service system after appointing Keith. “My philosophy is the mayor should be able to appoint his or her department heads,” Ashe explains. “Now, with the police and fire there is room for argument and legitimate debate, just due to the nature of the work. . . . [But] my thought has been that the mayor ought to have that discretion. And if the mayor makes a good decision, that will be fine; [if he] makes a lousy decision, that will be a factor in the mayor’s reelection.” In any case, Ashe successfully made this argument to the civil service board, which approved his request to make the Chief’s job a direct mayoral appointment (subject to some basic job qualifications) by a four-to-one vote.

5 Compare, for example, Keith’s description of his strategic issues with Herman Goldstein’s program for police reform—which essentially argues against the “organizational development” model; e.g., Herman Goldstein. “Improving Policing: A Problem-Oriented Approach,” in Crime and Delinquency, vol. 25 (April 1979), pp. 236-258.

6 The grant also funded other community policing activities, like training officers in a new problem-solving process, and building on the department’s repeat offender initiatives.

7 As quoted in Jesse Tinsley. “Police Department to Form Advisory Panel,” Knoxville News-Sentinel, July 11, 1994.

8 In the first year of planning, all of these teams were obviously new creations. But in subsequent years (the city has repeated the entire process annually since 1994 and is now on its fourth plan), goals and their associated implementation teams might be carried over from the previous plan, and new goals might be assigned to an existing team if they overlapped with its work. Conversely, some teams have chosen to dissolve, either because of inactivity or because they felt they accomplished what they set out to do (for example, the housing team disbanded after completing a joint project with Habitat for Humanity to build low-income housing in East Knoxville).

9 For example, Keith argued in a newspaper interview that other cities’ review boards are used mainly by officers seeking to overturn their verdicts (Don Jacobs. “KPD Chief, Fraternal Order Oppose Civilian Review Board,” in Knoxville News-Sentinel, October 22, 1997; p. A6).

10 The corollary to this view, of course, is that times of crisis represent ripe occasions for action; and indeed, in his first management team meeting after the Daniels incident, Keith implored the somewhat downtrodden group: “Don’t treat this as a gloom day but as an opportunity”—particularly because of the extraordinary degree of community focus that had been generated.

11 Knoxville’s Fraternal Order of Police (which represents officer opinion but not as a formal collective bargaining unit) flatly opposed the review board proposal; Knoxville FOP president Ed Mitchell’s response to the idea was that “we already have a civilian review board; it’s called the grand jury.” As quoted in Jacobs, op. cit.

12 Lockmiller argues that such high-level agreement is crucial in order to demonstrate seriousness of purpose, and in the GOCAP effort he spearheaded, the group went so far as to invite the media to the signing: “We had a sit-down, media covered signing of that interagency agreement by the head of each of those units—not whoever they were sending to the meetings necessarily, but the actual head of that department, whether it was the sheriff, or the attorney general, or the superintendents of schools or whoever. . . . We had all those people in one room that signed that document while their underlings, their seconds and thirds in commands looked on. So, in effect, they’re saying we give you our mantle of protection and authority to do this. So they couldn’t say, ‘Well, I didn’t sign that. Somebody five rungs down signed it.’” As it turned out, the media was more than willing to attend the event: “When you can get 12 heads of agencies together in the same room, you’ve done something,” Lockmiller explains. “I saw more power in that room that day in an hour than I normally see anywhere.

13 Information-sharing with other agencies faced greater obstacles: For example, the school system balked at providing truancy information until federal legislation was passed that effectively required them to do so. More generally, arriving at a single release of information form was reportedly one of the most difficult negotiations the group went through. The form was intended to make it easier for participating agencies to share information about habitual offenders by asking the offenders to sign a blanket release form up-front. (Many agencies also felt the new form would speed up service delivery for clients, who in order to receive services from a new provider would no longer have to make a special appointment to sign yet another release form.) But as the idea became a concrete proposal about what the form would actually say, agency lawyers chimed in with innumerable objections. At that point, the task force invited in a Pepperdine University professor who was considered a national expert on information-sharing to explain how the effort could work legally. The group finally settled on a mutually-acceptable document that is now in use, but only after several more iterations in which each agency got the chance to amend the proposed form.

14 The local juvenile court judge could not, of course, meet all the county’s desires unilaterally. For example, the SHOCAP task force felt that young habitual offenders should all be sent to the same strict juvenile institution, but any formal arrangement of that sort would have to be approved at the state level. The state, however, argued that it could not legitimately single out one county’s juvenile offenders for special treatment, and it also rebuffed the county’s proposal to take the SHOCAP system state wide, arguing that Tennessee simply could not afford to start up a brand-new institution for habitual offenders. In response, the task force has taken a bottom-up approach to creating a statewide program by trying to help other Tennessee counties start SHOCAPs one-by-one. Unfortunately, after eight years of trying, no other Tennessee county has been able to replicate SHOCAP (though several have replicated GOCAP), largely because of problems getting a foundational interagency agreement.

15 The KPD would later apply the same model to detectives, developing an in-house training program for new investigators so that it could incorporate topics like diversity training and case management—topics that touched on important pieces of the department’s new mission but which outside training programs did not necessarily cover.

16 This “sponge” model of innovation seems to be a common one in the KPD, but in some initiatives, department and community opinion has been supplemented by informal evaluation studies. Most notably, when the KPD decided to decentralize its property crimes investigations, it initially allowed its three districts to choose their own way of doing so: One chose to create a team of three officers and two detectives that investigates all property crimes in the district; another chose to assign one detective to investigate all burglaries in the district, leaving other property crimes to the investigative division; and the third opted to leave things as they were, sending all property crimes to the investigative decision. After running this “experiment” for several months, the department decided to adopt the team approach for all three districts, in part because it seemed to lead to a higher proportion of solved cases.

17 That is not, of course, to say that they were unwelcome; Keith, for one, explains that such serendipity plays an important role in his philosophy: “[The] process of change is like a lot of other things: It’s dynamic in nature, and it’s going to be constantly changing. And if you’re real rigid as an organization, and I’m a real rigid leader, then I may miss opportunities.”

18 Interestingly, much of Moyers’ expertise about “what works” in problem solving is apparently not a matter of learning which interventions work in general; instead it is a matter of having learned about the differences among Knoxville’s many neighborhoods. In response to a question about what his evaluations entail, Moyers quickly explains that different strategies will be appropriate to different areas, both because the crime problems vary and because the social landscape changes: “[In the business areas], as you come from the inner city to the outer city limits, your problems change: You don’t have the vacancy problem. You don’t have the traffic congestion problems. You don’t have this and you don’t have that. . . . You get out to the county and things such as traffic control devices and pedestrian flow and traffic flow are more important than they in the uptown area, where they are controlled better. So you have to look at how they affect each one of the neighborhoods. [And in] the residential areas, [it may be that] you [successfully] used a technique or strategy in one district where you had a lot of community involvement and community activity, [but] if you move to another side of town, those community organizations may not be that strong. They may not want to get involved that deeply—they’re just kind of like a window dressing sometimes; they want to do a little bit but they don’t want to do a lot. So you have to look and see which one of the communities are strong.” Thus in his mind, becoming a good judge of problem-solving plans is at least as much a matter of learning about the city as it is a matter of evaluating abstract interventions.

19 To be clear, Martin maintains that the department never completely dropped any of its grant-funded programs, whether personnel-based or not; but she admits that some programs drop down to a lower level of activity once outside money dries up.

20 The city used a different strategy to deal with the phase-out in its domestic violence grant, putting the position into the city budget at the program’s outset even though it would be temporarily funded by the grant. “That way at the expiration of the grant they will already be funded positions in the budget,” Martin maintains. “So it’s not like at the end of three years, or two years, or whatever that you’re going to have to lay those people off and shut the program down.”

21 In a 1996 employee survey, only 17% of respondents disagreed with the statement “in our community, both the public and our employees understand that the people we ‘protect and serve’ are our customers;” unfortunately, comparable information does not exist for previous years.

22 Randy Lockmiller, who spearheaded the GOCAP initiative, speculates that self-interest motivates such cooperation, asking rhetorically: “Does he want to know about gang activity? Do the gang members just stay inside the city or do they go out into the county?” As a result, his strategy for bringing potentially uncooperative players to the table has been to lay out their benefits clearly: “You have to dial the right radio station,” Lockmiller quips. “WII-FM, What’s In It For Me?”

23 According to departmental statistics for 1996, Knoxville’s 178 patrol officers logged 26,320 hours of directed patrol activity, or about 8% of their total working hours. During the same period, they responded to 237,561 calls for service.

24 Other forms of problem-solving in the patrol force include the problem-solving kits described above; district roundtables which bring officers and community members together to discuss the problems of a particular neighborhood; and the recently-imported COMPSTAT process, in which patrol division commanders meet with each other and with other units (such as CID and the crime prevention unit) to discuss problems and potential strategies for dealing with them.

25 Keith himself sees the pressures of 911 as a central challenge police agencies face: “That to me is the most crucial mistake law enforcement makes today, because we’re not controlling the demand level. You can’t become real proactive if all you are doing is reacting to the number of times that phone rings.” To manage that demand, the department has instituted a call-diversion system that Keith intends to expand still further. But he recognizes that this strategy creates its own backlash from citizens who have become accustomed to automatic responses to their calls: “Chiefs around the country will say, ‘You know, you're asking me to take a risk, and I can just keep on doing it this way, and kind of trot along. Or I can take a risk—[but if] it turns south, I get fired.”

26 On the other hand, it is interesting to note that while patrol’s share of departmental resources has grown, CID’s has remained almost constant: In both 1987 and 1997, CID (encompassing investigations, the investigative support section, and organized crime) spent 16% of departmental resources. If 911 had not been shifted outside of the department, CID’s claim on total KPD resources would actually have fallen. These blunt figures perhaps reflect the general trend of “community policing” departments towards de-specialization.

27 The KPD recently institutionalized this expectation by importing New York City’s COMPSTAT process. Under that system, district captains and the heads of other divisions meet biweekly to discuss the problems each district faces and develop strategies for dealing with them. In this way, diverse organizational resources are made available to meet emerging needs in each district; and the responses that result are monitored through bottom-line evaluation: Has the department’s mission in each area actually been achieved?

28 Moreover, Keith argues, supervisors themselves must become comfortable with discretion and ambiguity: “Supervisors and managers a lot of times develop lazy skills, and they want things that they can count—things that they can measure relatively simply and [that] eliminate any gray. It’s like a lot of thinking in what I call Generation X now: There is no in-between; it’s either this way or this way. Sadly enough, that is not the way the world evolves, [but] we’ve got some supervisors who fall back on that.”