Conceptualizing criminal justice theory

Abstract (Summary)
In an attempt to advance the development and generalization of criminal justice theory as a whole, Bernard and Engel propose a framework for classifying specific criminal justice theories. They also present an interpretive history of the academic field of criminal justice to demonstrate how the field can be organized within that framework.

FULL TEXT
In an attempt to advance the development and generalization of criminal justice theory as a whole, we propose a framework for classifying specific criminal justice theories. We then present an interpretive history of the academic field of criminal justice to demonstrate how the field can be organized within that framework. We conclude by describing some lessons about theory learned in the field of criminology and applying these lessons to the field of criminal justice.

The academic field of criminal justice originated in the 1950s with observational research on criminal justice organizations. The diversity and complexity of those organizations led to diversity and complexity in the resulting theories, which then did not easily generalize into something that could properly be called “criminal justice theory.” After 50 years, a great deal of research has accumulated along with an increasing number of relatively specific theories to interpret that research. Despite some movement to generalize the theories, little progress has been made in formulating criminal justice theory per se (e.g., Hagan 1989).

It may be that the criminal justice system as a whole is so diverse and complex that meaningful theories spanning the system are impossible to formulate. It may even be that its major components (police, courts, corrections, and juvenile justice) are too diverse to be the subject of meaningful theories. Alternatively, something called “criminal justice theory” may be not only possible but also necessary for the further progress of criminal justice as an academic and scientific discipline. In this paper we explore the possibilities for developing such theory.

Rather than proposing a new and comprehensive theory that attempts to describe and explain the entire criminal justice system, our methodology is to examine the history and development of the academic field of criminal justice itself. To determine what “criminal justice theory” is, we examine the actual theories that have developed within the field of criminal justice since it was founded. First, we briefly describe earlier attempts to classify and generalize criminal justice theories, discuss the problems accompanying each of these attempts, and assess the reasons why they generally have not been successful as frameworks for organizing the field. We then propose a new conceptual framework for classifying and generalizing criminal justice theories. Next we present an interpretive analysis of the historical development of the academic field of criminal justice in order to demonstrate how the field can be organized within that framework. We conclude with a description of lessons about theory learned in the field of criminology and apply these lessons to the field of criminal justice.

ORGANIZING AND CLASSIFYING CRIMNAL JUSTICE THEORIES

Earlier Classifications of Theories

The field of criminal justice has accumulated a large body of knowledge based on much empirical testing; it has also accumulated a variety of rather specific theories. Some attempts have been made to organize and classify those theories. The three most common bases for classification are (1) type of organization within the criminal justice system (e.g., police, courts, corrections); (2) under– lying theoretical assumptions (e.g., consensus, conflict); and (3) predictor variables (e.g., individual, situational, organizational, community). Although each of these classification schemes presents a way to conceptualize the research, each has weaknesses.

The simplest and most straightforward way to organize theory in criminal justice is to categorize it by component of the criminal justice system. Dividing criminal justice theory into police, courts, and corrections has become the standard way to organize the field. Most introductory criminal justice textbooks take this approach, and many criminal justice researchers identify themselves primarily with one of these three areas.

This approach to organizing material impedes theoretical development in criminal justice for a number of reasons. First, some topics (e.g., juvenile justice, gun control, systemic discrimination) do not fit neatly into any one particular category. Second, this classification limits comparisons across components, hindering development of theory regarding the relationships between organizations and how they work together to accomplish their respective tasks. Furthermore, it is difficult to describe and explore the similarities and differences among the components. Finally, it is difficult to conceptualize the criminal justice system as a single entity when research is continually divided into three areas. Although the criminal justice system is marked by decentralized and fragmented parts with inconsistent goals (some observers argue that it is not even a “system” in the true meaning of the word), the relationships that do exist among and between these parts are obscured by such a classification.

A second type of classification focuses on the underlying assumptions and propositions of particular theories. Scholars generally move toward this type when they attempt to break out of categorization by criminal justice component. For example, Scheingold (1984) acknowledged that there are many individual differences among theories but suggested that they can be classified most appropriately as belonging to one of three groups: mainstream, Marxist, or conflict. Similarly, Hagan (1989) categorized criminal justice theory and research as based on either consensus or conflict theory. He then noted the limitations of having only two predominant theories of criminal justice available, and suggested an alternative approach for theoretical development.

This approach presents certain problems. Such a broad classification obscures differences among theories and research while doing little to increase clarity. Unlike other “pure” academic disciplines, the field of criminal justice is multidisciplinary, applying theoretical propositions from sociology, criminology, economics, political science, psychology, and anthropology. Describing theories on the basis of such general theoretical assumptions further limits theory that should consider the propositions and assumptions of many of these disciplines. Therefore we believe that classifying theories according to their underlying theoretical assumptions is not useful.

A third attempt to organize the literature involves grouping research by dependent and independent variables. Unlike classification by underlying theoretical propositions, which generally is an attempt to move “upward” from classification by system components, classification by dependent and independent variables generally is a move “downward” in the sense that it has occurred only in particular components of the system. The leading example is Sherman’s (1980) review of research on police behavior. Sherman described the different types of police behavior (e.g., arrest, use of force, detection, and service) that were the dependent variables. Primarily, however, he focused on the conceptual grouping of the independent variables that explained these various types of police behavior (e.g., individual, situational, legal, organizational, and community). Riksheim and Chermak (1993) later updated this review, and Worden (1996) adopted a similar approach to categorize research on police use of force.

This approach provides a useful way of organizing the vast amount of research on police behavior, and is a significant improvement over the previous classification schemes. It has not yet been used, however, to organize research across the system components. Indeed, we are aware of such classification only in regard to police, and do not know whether it has been applied to courts or corrections. In addition, because this approach focuses on one particular type of dependent variable, namely police behavior, the classification scheme concentrates on organizing the independent variables.

Our Proposal for Organizing Criminal Justice Theory

We believe that a meaningful categorization of criminal justice theories must cut across the components of the criminal justice system: police, courts, and corrections. It cannot simply organize theories and research within each separate component. We also believe that such a classification must be able to incorporate the similarities in content found in theories across those components, and to facilitate generalization and competitive testing of the theories. Finally, we believe that categories must be able to include considerable detail and specificity in order to be useful; they cannot be so broad that the entire field is divided into only two or three categories.

We propose that criminal justice theories should be grouped or classified first on their dependent variables, and then according to their independent variables. This approach is essentially similar to Sherman’s (1980) and Riksheim and Chermak’s (1993) approach to classifying empirical police research, although much broader. As stated above, Sherman took police officers’ individual behavior as the dependent variable. That general category included a number of more specific dependent variables: arrest, use of force, detection, and service. After grouping the theories on their dependent variables, Sherman grouped them on the basis of their independent variables: individual, situational, legal, organizational, and community. He reported research findings on the basis of independent variables, whereas Riksheim and Chermak, in their update of Sherman’s review, reported findings on the basis of dependent variables.

We propose expanding this approach to the entire field of criminal justice. In our interpretation of the history of criminal justice, as described below, we suggest that there are three broad types of dependent variables; the first two encompass most of the theory and research in the field. The first type focuses on the individual behavior of criminal justice agents (e.g., the behavior of police officers, courtroom officials, correctional officers). The second focuses on the behavior of criminal justice organizations (e.g., the behavior of police departments, court organizations, correctional organizations). The third type of dependent variable focuses on the characteristics of the overall criminal justice system and its components (e.g., police killings, “get tough” sentencing, incarceration rates). This type operates at the aggregate level with theories and research that examine the relationship between societal-level characteristics and the rates and distributions of criminal justice system behaviors.

Each general type of dependent variable would include a variety of more specific dependent variables. And for each dependent variable, whether more general or more specific, a variety of independent variables would be expected to have at least some causal effect, especially because multivariate models now are used routinely in criminal justice research that incorporate different types of independent variables within one model. By categorizing all theory and research on the basis of the dependent variable, one can compare and contrast research on the basis of the perceived influence of various independent variables. This approach allows generalization and competitive testing of criminal justice theory and research across system components.

Our scheme does not classify theories according to underlying assumptions, such as consensus versus conflict theories. In our view, these underlying assumptions represent types or categories of independent variables. Thus classification on these variables would occur only after the theories had been classified on their dependent variables. When classified on their dependent variables, most conflict-type theories are found to be aggregate-level: the dependent variables are either the entire criminal justice system or an entire system component such as policing or corrections. At that point, the issue of consensus versus conflict is empirical rather than theoretical: how much variation in the dependent variable is explained by conflict-type independent variables, and how much is explained by consensus-type independent variables.

Finally, in the interpretive history presented below, we argue that criminal justice theories and research historically have incorporated, either explicitly or implicitly, prescriptive ideals that are used to assess the legitimacy of what criminal justice agents and organizations should be doing. This idealism underlies the reformist tendencies that have always been present in criminal justice as an academic field, in which the reality of criminal justice practices is moved closer to the prescriptive ideal. We suggest that, for all three types of dependent variables, prescriptive ideals in the form of discussions of legitimacy should be made explicit rather than left implicit. This step forms the basis of any reformist implications for criminal justice theory and research.

CRIMINAL JUSTICE AS AN ACADEMIC FIELD

The following description of the history of criminal justice as an academic field is presented as a concrete representation of how our proposed framework can organize the academic field of criminal justice so as to facilitate both the competitive testing and the generalization of theories. This interpretive history, however, should not be taken too literally. Any attempt to sort material into conceptual boxes is always a matter of opinion, and no classification system is completely fair to its subject matter. In addition, although we present a particular sequence that suggests progression over time, the time periods overlap considerably. Nevertheless, and despite whatever flaws might be apparent, we believe that our proposed framework is useful for organizing the academic field of criminal justice.

The ABF Studies: A Shift in the Attribution of Legitimacy

Before 1960, all but a very few empirical studies of criminal justice organizations were undertaken in the context of periodic “reform commissions” (Remmington 1990; Walker 1992). These commissions did not define their task as purely descriptive research, but rather in language taken from the critical legal studies movement, as bringing the “law in practice” into greater conformity with the “law on the books.” That is, they examined actual criminal justice practices to determine how much those practices deviated from some prescriptive ideal. The deviation itself was assigned a negative moral evaluation: it was said to reflect corruption and/or incompetence. The goal of the reform commissions was to minimize deviations between real criminal justice practices and the prescriptive ideal, and thereby maximize the legitimacy of those practices.

In the 1950s, the American Bar Foundation sponsored a series of observational studies that spanned the criminal justice system. These studies originally were conceptualized in much the same way as the earlier reform commissions, with the same goal as those commissions. Yet although they observed an astounding array of incompetence and corruption in actual criminal justice practices, the authors of the ABF studies interpreted those practices in terms of pervasive discretion inherent in the enterprise. This shift in interpretation eventually resulted in an early termination of the ABF study and delayed the publication of its final volumes until 1969 (Walker 1992). The shift, however, also was the single event most responsible for the development of criminal justice as an academic field (Remmington 1990; Walker 1992).

The change in interpretation essentially was a shift in the attribution of legitimacy. The pre-ABF reform commissions attributed legitimacy entirely to the prescriptive ideal (the “law on the books”). The ABF studies were controversial and significant because they attributed some (but not all) legitimacy to actual criminal justice practices, even when those practices deviated from the ideal. The studies therefore posited two simultaneous and largely incompatible sources for legitimacy of criminal justice practices: the “law on the books” and pervasive discretion. The positing of a second source of legitimacy generated hostility and resistance, but it also opened the door to the study of criminal justice practices in their own right.

Closing the Gap Between Ideal and Real

Despite the shift in attribution of legitimacy, significant continuities existed between the pre-ABF reform commissions and the post-ABF academic field of criminal justice. The most significant was the overall goal of closing the gap between the ideal and the real in criminal justice.

As stated above, the goal of the pre-ABF reform commissions was to bring the real into greater conformity with the ideal. The ideal was viewed as an independent and stable reference point that defined legitimacy; thus the real was viewed as illegitimate insofar as it deviated from the ideal. Thus, for the pre-ABF reform commissions, closing the gap between the ideal and the real was to be accomplished entirely by changing the reality of criminal justice practices to make them more like the ideal. In contrast, the ABF interpretation of real criminal justice practices implied that the “law in practice” could have its own legitimacy and could deviate from the “law on the books” without necessarily being illegitimate and needing reform. Thus the “ideal” in criminal justice no longer was defined solely in terms of the “law on the books.” Rather, it incorporated elements of the real world of criminal justice, the actual practices of criminal justice organizations.

Although the goal of the ABF studies still was to close the gap between the ideal and the real, they attempted to accomplish this by changing both the ideal and the real in order to make each more like the other. In part, like the earlier reform commissions, the ABF studies attempted to change the reality of criminal justice practices by bringing them into greater conformity with the ideal. Also in part, however, these studies attempted to change the prescriptive ideal to fit more closely the reality of criminal justice practices. That ideal was broadened to include the legitimate exercise of discretion, a concept taken from the reality of criminal justice practices. The gap between the ideal and the real would close when the two eventually met in the middle.

Polarization Theories: Official Conceptions Versus Actual Practices

In the 50 years since the ABF studies were conducted, the field of criminal justice has focused strongly on the real in criminal justice, and the ideal was faded into the background. But at least initially, in the years following the ABF studies and at least partially in response to the studies, the focus remained more on the gap between the real and the ideal, or on the relationship between the two, rather than on the real itself.

Specifically, in a number of theories that followed the ABF studies, the prescriptive ideal was transformed into something like “official conceptions” of how criminal justice organizations are supposed to operate. These conceptions more or less described how criminal justice agents and administrators, as well as politicians, described criminal justice practices in public discourse. Therefore they also described, more or less, how the public thought criminal justice agents and organizations were supposed to behave. “Official conceptions” diverged markedly from the real practices of those agents and organizations: the real practices were concealed from the public and shrouded in secrecy. Criminal justice agents and politicians who publicly presented these “official conceptions” generally knew all about the real practices, so the secrecy had elements of a conspiracy.

A number of early theories that followed the ABF studies maintained the ABF position that both “official conceptions” and the reality of criminal justice practices had their own legitimacy. In addition, however, these theories suggested that criminal justice as a social phenomenon could be fully understood only in terms of the complex interaction, and even the conflict, between these two aspects. Therefore the theories, in contrast to later studies that focused more directly on the reality of criminal justice practices, always contained some form of polarization at their center; often this was revealed by the use of the abbreviation vs. in shorthand descriptions. In these theories, the argument was that both of the polarized conceptions were necessary for an understanding of criminal justice as a social phenomenon. Compared with the ABF studies, these studies placed greater emphasis on understanding the two sides of criminal justice and the gap between them, and less emphasis on the overt goal of closing that gap, despite an implicit sense that a thorough understanding eventually could lead to reduction or closure of the gap.

The most important of these early polarization theories was Packer’s (1968) argument about due process versus crime control. This classic theory provided a framework for explaining the day-today operations in the process of justice, based on two different philosophical orientations. Packer’s concept of due process did not diverge much from the earlier conception of the “law on the books,” but his conception of crime control went far beyond the ABF conception of pervasive discretion. In particular, Packer’s conception of crime control suggested a legitimate goal, separate from and even in conflict with the due process goal found in the “law on the books.” The legitimacy of crime control as a goal therefore generated legitimacy for many real practices of real criminal justice organizations that could not be legitimated in terms of due process. Nevertheless, Packer did not attribute legitimacy solely to the reality of crime control in criminal justice organizations. Rather, in his view, both due process and crime control had legitimacy. True understanding of criminal justice as a social phenomenon required an understanding of both.

Similarly, Feeley (1973) explained criminal justice in terms of “rational goal” versus “functional systems” models. He assessed the state of criminal justice research at that time and reported that many empirical studies examined criminal justice within a “rational goal” model. In contrast, the functional systems approach emphasized how the nature of organizations and informal norms influenced the behavior of criminal justice officials. It sought to explain behavior through emphasis on “cooperation, exchange, and adaptation” rather than adherence to formal rules and regulations, and could be applied throughout the criminal justice system. In Feeley’s argument, the “rational goal” model is more descriptive of some “official conception” of how criminal justice organizations are supposed to operate, while the “functional systems” model is more descriptive of the “reality” of how criminal justice organizations operate.

In another theory that contains a polarization, Lipsky (1980) described street-level bureaucrats, including police officers, as policy makers because of their relatively high degree of discretion and autonomy from organizational authority. He then explained streetlevel bureaucrats’ patterns of practice as a response to the conditions of their working environment: high degrees of discretion, regular interaction with nonvoluntary clients, inadequate resources, increased demand for services, ambiguous work objectives and goals, and inadequate performance measures. Like Lipsky, Prottas (1978) theorized about individual behavior within “street-level bureaucracies,” suggesting that individual street-level bureaucrats are powerful in part because they make decisions to transform citizens into clients. Although they are low in the organizational hierarchy, these workers have high levels of influence and power, which produce structural conflict within the organizations. Lipsky and Prottas contrasted the “reality” of street-level bureaucrats as policy makers with an “official conception” that policy is the product of the formal policy-making procedures within the organizations.

Finally, Skolnick’s (1966) analysis of the difference between “law” and “order” can be described as the difference between an “official conception” of criminal justice and the “reality” in criminal justice organizations, much like Packer’s earlier analysis of due process versus crime control. In addition, the contradictory directives in complex organizations described by Cressey (1959) arise from the realities of correctional organizations, which contrast with the official conceptions of what those organizations are supposed to be doing. In each of these theories, both points of the polarization are necessary if one is to understand the full complexity of criminal justice as a social phenomenon, even though one of the two poles is said to be more descriptive of the “reality” of criminal justice practices.

Observational Studies of Criminal Justice Organizations

“Polarization” theories were fairly global: they tended to encompass either the entire criminal justice system or at least entire component parts (e.g., police). The academic field of criminal justice moved rather quickly from these more global theories and began to focus on the more particular. The emphasis shifted from broad explanation to close observation and description. With this shift in focus, the goal of closing the gap between the ideal and the real faded into the background, and attention moved strongly toward understanding the extraordinary complexity of criminal justice. Usually somewhere in the background was an unspoken assumption that a thorough understanding of the reality of criminal justice would lead to a reconciliation of the real with the ideal.

Initially most of these investigations were qualitative, observational, descriptive case studies of individual criminal justice organizations. The first two were Clemmer’s ([1940] 1958) study of Menard Penitentiary and Westley’s (1953, 1970) study of the Gary Police Department. Both of these predate the ABF studies, but both gained prominence afterward when the focus shifted from reformism to observation and description.

In the years following the ABF studies, researchers published many additional case studies of criminal justice organizations. Skolnick (1966), Wilson (1968), and Rubinstein (1973) conducted early observational studies of the police. Blumberg (1967), Sudnow (1965), and Eisenstein and Jacob (1977) made early observational studies of the courts. Sykes (1958), Goffman (1961), Toch (1977), and Jacobs (1978) conducted early observational studies of prisons. In some ways, these researchers viewed the prescriptive ideal–”official conceptions” conveyed in public discourse on what was legitimate for criminal justice agents and organizations–as a greater problem than the reality of criminal justice practices. The prescriptive ideals made unrealistic demands on criminal justice agents and organizations, and therefore generated the secrecy that caused many of the problems they faced. In a sense, these observational studies documented the extreme complexity of the reality of criminal justice while simultaneously attacking the simplicity of prescriptive ideals. Criminal justice scholars came down on the side of the real as opposed to the ideal, thus reversing the stance of the pre-ABF reform commissions.

Yet the prescriptive ideal did not entirely disappear in these qualitative studies, as might be expected in a purely scientific endeavor. Rather, it remained in the background as a standard against which to hold the complex descriptions of the reality. In addition, these studies often concluded with comments on what might be appropriate criteria for evaluating legitimacy for real criminal justice agents and organizations. That is, the studies neither eliminated the prescriptive ideal nor assumed that it was a stable and unmoving reference point by which to evaluate real criminal justice practices. Instead they brought the prescriptive ideal into an interactive relationship with reality. The studies both examined the reality of criminal justice practices and considered appropriate standards for legitimacy, given the complexity of that reality. Thus the field of criminal justice became increasingly complex both in its descriptions of reality and in its discussions of the prescriptive ideal.

Early Theories: Structure/Culture/Socialization Versus Individual Typologies

The descriptive studies of criminal justice organizations introduced themes and posed questions that were the source of much later research and theory in criminal justice. One particular theoretical argument that emerged from these early observational studies can be described as a structure/culture/socialization argument. This argument proposed a structural explanation of organizational culture. The organizational culture in turn explained the characteristics of the criminal justice organization, while socialization into the culture explained the behavior of criminal justice agents.

This argument essentially explained why the reality of criminal justice practices diverged so much from the “official conceptions” of those practices. The answer, according to the argument, lay in the structured conditions of the work environment. Organizational realities faced by individual criminal justice agents tended to generate work practices that differed from the “official conceptions” but were similar to practices of other criminal justice agents. Those organizational realities and the resulting work practices generated similar patterns of thought among criminal justice agents; these then were shared and transmitted interpersonally, and became subcultural. The result was a cohesive community in which secrecy was used to conceal the gap between actual work practices and the “official conceptions” of those practices. Finally, socialization processes developed for incorporating new criminal justice agents into this secrecy-based community. These structure/culture/ socialization arguments typically offered vague suggestions about how to close the gap, but primarily they described and explained the reality of criminal justice practices.

The first two qualitative observational studies, by Westley (1953, 1970) and by Clemmer ([1940] 1958), included theoretical explanations that can be described in terms of structure/culture/socialization. Later research making similar theoretical contributions included Brown’s (1981) study of the Los Angeles Police Department, Reuss-Ianni and Ianni’s (1983) arguments about the two cultures of policing, Van Maanen’s (1974) description of the socialization processes for police, and Herbert’s (1998) update on police culture. Lipsky’s (1980) arguments about street-level bureaucrats, as described above, also falls into this category. In the courts, Cole (1970), Eisenstein and Jacob (1977), Eisenstein et al. (1988), and Church (1985) offered essentially a structural/cultural explanation of the courtroom work group, while Heumann (1978) described the socialization process for courtroom actors. Even Blumberg’s (1967) earlier work can be interpreted along these lines. Goffman’s (1961) analysis of asylums included a structural/cultural/ socialization explanation, as did various descriptions of the occupational culture of prison officials (Duffee 1975; Hepburn 1985; Lombardo 1981; Toch and Klofas 1982).

Although structure/culture/socialization arguments were initially the most common, a contrasting theory soon arose. In 1963, Irwin and Cressey argued that prison culture was brought into the prison by inmates. This “importation” theory contradicted structure/culture/socialization theories such as that of Clemmer ([1940] 1958), who argued that prison culture was generated by the structure of the prison and that inmates were socialized into it (“prisonization”). Similarly, Westley (1953, 1970) explained police attitudes and behavior largely in terms of what was essentially a structure/culture/socialization theory, but some later theorists focused on individual differences among police officers to explain at least some police attitudes and behaviors. These included White (1972), Broderick (1977), Muir (1977), and Brown (1981). Similar typologies were proposed to explain the behavior of attorneys (Carter 1974; Jack and Jack 1989; Mather 1979) and correctional officials (Irwin 1980; Kauffman 1988; Rosecrance 1988).

Quantitative Studies of Criminal Justice Agents’ Behavior

Over time, criminal justice research became systematic and quantitative. Much larger numbers of observations were collected more systematically, often from several similar organizations (e.g., several police departments), and the data was analyzed quantitatively rather than interpreted in qualitative, descriptive fashion.

The shift from qualitative to quantitative research was accompanied by a subtler and less obvious shift in the focus of the studies. Earlier qualitative studies had focused on the entire criminal justice organization as a functioning entity; individual agents were viewed as part of that entity. With systematic observation and quantitative analysis, the focus shifted to the behavior of the individual criminal justice agent, particularly decision making. Agents’ behavior, of course, is much more quantifiable than criminal justice organizations’ functioning. The research task now was to describe the complexity of the contextual influences on individual decision making. The criminal justice organization then became one of several types of variables that influenced the individual agent’s behavior. This was a considerably narrower and more precise question than had been addressed in the earlier and more free-ranging qualitative, observational studies.

These quantitative studies became the “bread and butter” of criminal justice research. Larger studies that systematically collected and analyzed original observational data of police behavior included the Black and Reiss study in Chicago, Washington, and Boston in 1966 (Black 1980; Black and Reiss 1970), the Police Services Study of officers in 24 departments in 1977 (Caldwell 1978; Whitaker 1982), and the Midwest City study (Sykes and Clark 1975). In each of these studies the investigators adopted the methodical approach of systematic observation, a technique that combines the rich detail of participant field research with the objectivity of systematic survey data collection (Reiss 1971). Many other studies focused on particular questions or addressed issues raised by the earlier studies, and relied on secondary analysis of data that already had been collected (e.g., Brereton and Casper 1981; HI(linger 1994; Smith and Visher 1981; Worden 1989; Worden and Shepard 1996).

These quantitative studies seemed to abandon all “official conceptions” of what criminal justice agents should be doing, and to focus entirely on describing and explaining what they actually were doing. Therefore they seemed to be purely descriptive and to bear little relation to the reformism that characterized the field at its beginning. Even so, prescriptive ideals about what criminal justice agents should be doing remained in the background of these studies, providing implicit value judgments about the legitimacy of what was observed and described. For example, quantitative research focused quite strongly on the effect of nonlegal variables on criminal justice agents’ behavior, including race, gender, class, and demeanor (Engel, Sobol, and Worden 2000; Kleck 1981; Klinger 1994; Radelet 1981; Smith and Visher 1981; Spohn 1990; Steffensmeier, Ulmer, and Kramer 1998). Eventually, reviews summarized and organized some of this research (Gottfredson and Gottfredson 1981, 1988; Riksheim and Chermak 1993; Sherman 1980; Simpson 1989; Walker, Spohn, and DeLone 1996). These nonlegal variables typically explained a relatively small portion of the variance in comparison with legally relevant variables, but they received a great deal of attention.

The reason for so much attention, as everyone knows, is that nonlegal variables should not influence the behavior of criminal justice agents. That is, the studies contained a prescriptive ideal against which criminal justice agents’ actual behavior was evaluated. In addition to explaining the agents’ behavior itself, the studies focused on explaining deviations of that behavior from an implied prescriptive ideal that defined legitimacy. That second focus implied that an understanding of the sources of agents’ behavior might help to close the gap between the ideal and the real, and thus to increase the legitimacy of real criminal justice practices. Therefore, even while these studies continued, at least partially, to focus on the gap between the reality of criminal justice practices and some prescriptive ideal, they also revealed the continuing (even if implicit) goal of reform in quantitative criminal justice research.

Also important for our purposes, these studies reveal a continuing but quite subtle development in the complexity of the prescriptive ideal. The ABF studies had set that ideal in motion, so that it no longer was a static and unchanging “law on the books,” but instead interacted with the reality of criminal justice practices. In the later polarized theories about the two sides of criminal justice and in the later observational studies of criminal justice organizations, the prescriptive ideal was made even more complex and more firmly grounded in reality, and yet (it was hoped) still capable of attributing legitimacy in public discourse. The systematic quantitative research studies on criminal justice agents’ behavior continued that trend. While describing real criminal justice practices, the researchers asked, at least implicitly, “In the light of the complex realities they face, what should criminal justice agents and organizations be doing? What constitutes legitimacy in criminal justice?” The answers became increasingly complex.

The Explicit Return to Reformism

Armed with a large number of systematic observations of criminal justice agents’ real behavior, as well as with complex and sophisticated interpretations of criminal justice organizations’ functioning, criminal justice scholars proposed various entirely new conceptions of criminal justice practice that differed from the “official conceptions” and from the actual practices that were current at the time. These scholars essentially argued that their reform proposals were both realistic and legitimate. The reforms, they said, were realistic approaches to criminal justice that actually could be implemented. Criminal justice agents no longer would be influenced by structural workload conditions to act in one way, and then pressured by highly divergent “official conceptions” to hide behind the secrecy of the occupational subculture. At the same time, these reforms were said to be legitimate enough to function as “official conceptions” for the purposes of public discourse by criminal justice agents and administrators, as well as politicians. Thus the new proposals were offered as ways to close the gap between the ideal and the real. They were offered as a way to make criminal justice practices more legitimate.

In the area of police, Bittner (1974, 1975) offered a new definition of police professionalism, Goldstein (1977, 1990) defined problem-oriented and community policing, Wilson and Kelling (1982) suggested the “broken windows” approach, and Sherman et al. (1989) described a “hot spots” approach. In regard to the courts, academics produced ideas to transform the structure of sentencing (Rossi and Berk 1997; Ulmer and Kramer 1998) and introduced new ideas such as selective incapacitation (Gottfredson and Gottfredson 1994). In corrections, earlier, more naive proposals for reform (e.g., Morris 1974) gave way to more sophisticated and more realistic proposals that more fully took into account the complex realities of criminal justice (Morris and Tonry 1990).

These new proposals were quite broad; they called for reorganizations of entire areas of the criminal justice system. Much more specific and sharply focused topics also were addressed, particularly through the introduction of evaluation research. At least some of this evaluation research was based on an experimental design with random assignment of criminal justice actions, in order to determine the effectiveness of those actions as measured by crime– related outcomes.

Initially, evaluation research was applied to existing criminal justice practices (Gendreau, Little and Goggin 1996; Gendreau and Ross 1987; Kelling et al. 1974; Larson 1975; Martin and Sherman 1986; Martinson 1974; Sherman 1992), and the results of these studies exerted at least some influence over subsequent policies in the organizations in question. Evaluation research also was applied to the reforms that had been proposed by academics on the basis of earlier observational and quantitative research. These studies attempted to determine how fully the reforms achieved their purpose of closing the gap between ideal conceptions and real practices. That is, the researchers examined the implementation of the reforms to determine whether the actual criminal justice practices conformed to the new official conceptions of what should be happening. Some found a reasonable implementation of the reforms; thus the reforms themselves largely accomplished their goals. Others found that real criminal justice practices tended to “eat” reforms wholesale and to continue unchanged.

This development eventually led to “hydraulic” theories of criminal justice (e.g., McCoy 1984) and to arguments about “criminal justice thermodynamics” (Walker 1998). According to these arguments, which were primarily structural, workload pressures generate and maintain certain work practices even if the criminal justice agents engaged in those practices disagree with them. In these arguments, culture and socialization tend to drop out of the structure/culture/socialization theories, leaving only the structural pressures that exist in the workplace. Studies associated with these structural arguments typically found that reforms result in an array of unintended consequences. When reforms attempt to control workload pressures in one location, the pressures are displaced into another location as if in a “thermodynamic” or “hydraulic” system. In this view, the reforms fail to change criminal justice agents’ actual work practices because they do not change the organizational realities that generated those practices. Instead the reforms generate new “official conceptions” of criminal justice, which then require a new organizational subculture of secrecy in order to hide the organization’s true practices. Thus the reforms do not close the gap between the ideal and the real, between “official conceptions” and actual practices. Ultimately these studies focus on why reforms fail.

In the area of policing, for example, “professional” policing (in the original meaning of the term) was a reform that originated earher in the twentieth century. Several studies found that it failed to accomplish its intended purpose, and instead generated new secrecy (Brown 1981; Walker 1977). A later reform, community policing, also has been evaluated. Some researchers found that it largely accomplishes its purposes (Mastrofski, Worden and Snipes 1995). Others, however, argued that it largely fails to change actual practices; instead it creates further secrecy and this opens a new gap between the ideal and the real (Bayley 1988; Klockars 1988; Manning 1988). In regard to the courts, structured sentencing was found to have varying results (D’Alessio and Stolzenberg 1995; Heumann and Loftin 1979; McCoy 1984; Stolzenberg and D’Alessio 1994). Similarly, rape law reform generated some unanticipated consequences (Horney and Spohn 1991). Feeley (1983) made an argument based largely on structural workload to explain why court reforms fail.

Other arguments explaining the failure of reforms were primarily cultural. Studies associated with these arguments typically found no relationship between workload pressures and work practices. Instead the explanations for failure focused on the social norms of the occupational work groups. For example, several researchers found that plea bargaining is unrelated to workload pressures (e.g., Alschuler 1979; Carns and Kruse 1992; Feeley 1982). The scholars then proposed essentially cultural arguments explaining the persistence of these practices in terms of the shared norms of courtroom work groups. Although less popular than the structural theories of workload pressure, these accounts include an explanation of why criminal justice organizations tend to absorb reforms and to continue unchanged.

Aggregate-Level Studies of the Criminal Justice System and Its Components

Almost from the founding of the discipline, some studies in criminal justice have linked societal-level characteristics to the rates and distributions of criminal justice behaviors. This is a very different type of study than those which focus either on the behavior of individual criminal justice agents or on the behavior of the criminal justice organization as a functioning entity.

In policing, for example, a recent study examined the relationship between police killings and density, overcrowding, and racial and economic inequality (Jacobs and O’Brien 1998). This study does not focus on the organizational behavior of particular police departments nor on the individual behavior of particular officers. Rather, it explains an aggregate-level characteristic associated with policing in the United States–rates of police killings–by using various aggregate-level characteristics of the larger society, such as racial and economic inequality.

Similarly, various arguments have been made to explain the emergence of correctional systems or their later transformations by relating them to larger societal developments (e.g., Foucault 1979, Garland 1990; Rusche and Kirchheimer 1939). Other studies linked imprisonment rates to various aggregate-level economic issues such as unemployment (Chiricos and Bales 1991) or labor surplus (Chiricos and Delone 1992). Imprisonment rates also have been linked to various aggregate-level racial distributions (e.g., McGarrell 1993; Myers 1990). All of these studies explain overall characteristics of the correctional system in terms of societal characteristics. In this way they differ markedly from studies that focus on correctional agents’ individual behavior or on the behavior of correctional organizations.

In regard to the courts, a number of studies use current social conditions to explain recent changes in sentencing practices associated with increasing sentence severity (e.g., Streiker 1999) and with the decline in the use of youthfulness as a criterion for leniency (e.g., Tonry 1999). Here again, the dependent variable is not the individual behavior of prosecutors or judges, nor the behavior of courts as functioning criminal justice organizations. Rather, the theories attempt to explain characteristics of sentencing in the United States in terms of characteristics of the larger society.

At an even broader level, many theories have focused on characteristics of the entire criminal justice system rather than on components of that system. Chambliss and Seidman (1971) presented a theory of “the law in action,” using a conflict perspective. They suggested that law “can be understood only by considering the total social milieu of the persons whose behavior is supposed to conform to the rules and thus achieve the higher goals implicit in the rules” (1974:4). Black’s (1976) theory operates at the same level: he used five groups of aggregate-level independent variables (stratification, morphology, culture, organization, and social control) to explain societal-level characteristics of law (its quantity and style).

Other aggregate-level theoretical arguments focus on criminal justice policy. In an examination of street crime and public policy, Scheingold (1984) described the influence of politics over policy making within the criminal justice system. He argued that divisions in society produce conflicts of both interest and value, and that these conflicts generate cleavages, resulting in the development of subcultures. Within the criminal justice system, Scheingold’s theory explains the political nature of policy making by describing organizational differences, changes in the political environment, and differences in criminal justice actors’ attitudes and beliefs. In his later (1991) work, Scheingold provided a theory explaining how the political culture of the criminal process influences crime control policy.

In similar work, Gordon (1990) attempted to explain specific criminal justice policies (e.g., “get-tough” policies and the growing emphasis on observation as a crime control technique in the 1980s) through an examination of historical and current ideological responses. Reiman (1984) argued that the criminal justice system can be explained by its failures: the failure to identify acts by the powerful as crimes, the failure to implement policies that actually will reduce crime, and the failure to eliminate economic bias in the system. Each of these theorists uses aggregate-level societal characteristics, particularly the political environment, to predict and explain overall criminal justice policies.

Also operating at the aggregate level but with independent variables that are more community-based, Duffee (1980:5) suggested that criminal justice research should focus on the complexity of public organizations and communities. He argued that scholars should attend to larger questions concerning “relationships of the criminal justice system to society, rather than relationships between two or more aspects or behaviors in the criminal justice system” (emphasis in original). Making arguments similar to those of Feeley (1973), Duffee acknowledged that the criminal justice system is not a unitary system serving a single function, but that it serves multiple functions in different places. He suggested that criminal justice organizations cannot deter crime because different localities want and expect different things from their criminal justice organizations. These observations led Duffee to suggest that criminal justice operations should be examined in a theoretical framework “in which it is assumed that criminal justice operations are determined by community political and social structure” (1980:137). Duffee’s community theory of criminal justice advances a fourfold typology of communities, to be used to describe behavior of the criminal justice system.

Also operating at the aggregate level but proposing even more complex, independent variables, Hagan (1989) suggested that the lack of theory in criminal justice was due in part to the nature of the criminal justice system, which he and other scholars characterized as “loosely coupled” (Glassman 1973; Meyer and Rowan 1978). Hagan further suggested that theories of criminal justice should focus on the structure of criminal justice organizations and the environments in which they operate. He provided a framework for a theory of criminal justice system operations that stresses the importance of examining political power and organizational forms through a set of premises and propositions.

Similar arguments have been made about the overall characteristics of criminal justice in other countries. Tanner (1999) argued that the large number of abuses in the Chinese criminal justice system in the early 1980s can be explained by China’s lack of material resources combined with “legal instrumentalism”: the law as a tool used by the government and by the ruling Communist Party to accomplish social goals.

Although all of these studies appear to be descriptive, they exist against the backdrop of prescriptive ideals that define legitimacy in one way or another. Studies linking imprisonment rates to aggregate-level economic characteristics such as unemployment or labor surplus are interesting precisely because such a linkage has the appearance of being illegitimate. Thus variables are interesting even if they explain relatively small portions of the variance. The prescriptive ideal is that imprisonment rates should be determined solely by legally relevant variables such as crime rates.

A similar point can be made about the relationship between police shootings and inequality, density, and overcrowding. The point is not that these are the most important variables explaining rates of police shootings, but that such a relationship appears to violate some implicit prescriptive ideal that defines legitimacy for police discharge of weapons. Perhaps further study will demonstrate that no actual problem exists; at the present, however, this study is interesting precisely because police behavior is being measured against a prescriptive standard, which itself is used to define legitimacy.

DISCUSSION

Classifying theories first on the dependent variable and then on the independent variables increases conceptual clarity in discussions of criminal justice theory, allows different theories to be compared and contrasted with each other, and facilitates theory building through generalization. Some theories explain criminal justice agents’ behavior by focusing on the individual characteristics of the agents themselves, including police officers (Broderick 1977; Muir 1977; White 1972), attorneys (Carter 1974; Jack and Jack 1989; Mather 1979), and correctional agents (Irwin 1980; Kauffman 1988; Rosecrance 1988). These theories are comparable across criminal justice system components because one particular type of dependent variable (the behavior of individual agents) is explained by one particular type of independent variable (individual characteristics). Other theories explain individual criminal justice agents’ behavior as socialized adaptations to the work environment (Blau 1974; Blumberg 1967; Cole 1970; Feeley 1973; Lipsky 1980; Prottas 1978; Skolnick 1966). Again, these theories are comparable because one type of dependent variable (individual agents’ behavior) is explained by one type of independent variable (socialization into a work environment).

This type of classification facilitates competitive testing and generalization across system components for these types of dependent and independent variables. In addition, competitive testing could address the relative explanatory power of the two types of independent variables: individual characteristics versus work socialization. Finally, such testing enables various kinds of theory building. For example, Gottfredson and Gottfredson (1981, 1988) described criminal justice as composed of a series of critical decision-making points of diagnosis, classification, and prediction by criminal justice actors. They suggested that criminal justice operations may be viewed as individual decisions made by officials, and described decision making as having three main components: a goal to be achieved, alternative choices for achieving that goal, and information to guide in choosing among the alternatives. These scholars were able to construct a theory of criminal justice because they focused on one type of dependent variable (individual decision making) and argued that the independent variables possess commonalities across the entire system.

The second type of dependent variable in our proposed organizational scheme is the behavior of criminal justice organizations as functioning entities. Classifying theories on this type of dependent variable allows comparisons based on the independent variables. For example, many theories explain criminal justice organizations’ behavior in terms of the occupational cultures in those organizations. Police scholars generally recognize the work done in the area of police culture (Brown 1981; Herbert 1998; Manning 1977; Skolnick 1966; Westley 1953, 1970), but this work is rarely linked to similar developments in courtroom work groups (Church 1985; Eisenstein and Jacob 1977) or to work describing the occupational culture of prison officials (Duffee 1975; Hepburn 1985; Lombardo 1981; Marquart 1986; Philliber 1987; Toch and Klofas 1982). In all of these theories, occupational cultures are said to arise from structural factors such as workload pressures or work conditions. Essentially these cultures are used to explain the behavior of organizational units such as police departments or courtroom work groups. Categorizing theories by dependent and independent variables, as opposed to the standard categorization by system components, would allow these theories to be generalized across system components, and would facilitate competitive testing of specific independent variables within this overall category.

Other scholars have proposed more specific dependent variables within this category. Several different organizational styles have been postulated: for example, Wilson’s (1968) description of watchman, legalistic, and service police departments, Levin’s (1977) description of two urban criminal courts, and DiIulio’s (1987) description of prison systems in three states. These organizational styles are dependent variables within the general category of criminal justice organizations’ behavior. These variables then are explained with a variety of more specific independent variables including political, community, and organizational factors.

Theories that explain organizational behavior in terms of occupational cultures obviously are similar to individual-level theories that explain the behavior of individual criminal justice agents in terms of socialization into the occupational culture. The difference between these two types of theories lies in differences in the dependent and the independent variables rather than in any contradiction or competition between the theories. In the individual-level theory, the dependent variable is the behavior of the individual agent; socialization into an occupational culture is the independent variable. In the organization-level theory, the dependent variable is the behavior of the organizational unit, the intervening variable most often is the occupational culture (although sometimes this is the independent variable), and the independent variable most often is the set of structural conditions associated with the work.

Particularly with these two types of theories, the attempt to create a comprehensive picture of criminal justice realities leads some scholarly works to move back and forth among the various dependent variables. Although this practice is intuitively appealing, it can lead to confusion when the researcher is attempting to compare, competitively test, or generalize theories. In Wilson’s (1968) study of police organizations and Dilulio’s (1987) examination of prisons, for example, organizational behavior seems to be the primary dependent variable. At the same time, both of these theories explain criminal justice agents’ behavior as a consequence of working for particular organizations. These theories also have implications about the overall characteristics of criminal justice components at the societal level, and to that extent they include aggregate-level as well as individual- and organization-level dependent variables. Recognizing that some theories can be applied across two or more types of dependent variables adds conceptual clarity and allows more specific comparison of research findings across theories and system components.

Additional confusion arises when the dependent variable is not distinguished clearly from the independent variable. In several theories that are described as theories of criminal justice, for example, criminal behavior is the dependent variable and criminal justice is the independent variable (e.g., Gorecki 1979; Gross 1971). We exclude these theories from the present analysis because we regard them as theories of criminal behavior rather than as theories of criminal justice.

Another example of confusion between dependent and independent variables is present in the exchange between Black (1979) and Gottfredson and Hindelang (1979a, 1979b, 1980). In their research attempting to test Black’s (1976) theory of the behavior of law, Gottfredson and Hindelang found that seriousness was the dominant explanation for police behavior. Thus they treated police behavior as the dependent variable, and legal seriousness as an independent variable explaining the behavior. Black essentially responded that legal seriousness (the quantity of law) rather than police behavior was his dependent variable. Confusion arose because some of the factors that Black listed as explaining legal seriousness were the same factors that Gottfredson and Hindelang listed as explaining police behavior. The major problem, however, was that they were discussing different dependent variables.

In general, our proposed organization captures and organizes most of the theory and research described above in the interpretive history of criminal justice as an academic field. A continuing theme throughout that history, however, is the role of the prescriptive ideal in defining legitimacy for criminal justice. This ideal is more explicit in the earlier studies, but even in the later studies it exists in the background as a standard for measuring real criminal justice practices. Criminal justice scholars, we recommend, must recognize that their discipline inevitably includes a prescriptive dimension, which itself is always undergoing change in an interactive relationship with real criminal justice practices. Thus we suggest that this dimension be included explicitly for each type of dependent variable.

By acknowledging and making explicit the prescriptive dimension, we can incorporate into the field of criminal justice various other theories that focus exclusively on that dimension. For example, numerous scholars have advanced theories that describe and analyze the concept of justice as a philosophical orientation and explain the moral and social components of justice within societies (i.e., Barry 1989; Gross 1971; Rawls 1971, 1999). These theories provide the grounding for prescriptive ideals related to aggregatelevel characteristics of the criminal justice system overall. Similarly, theories of the philosophy of punishment (e.g., Hudson 1996; Tunick 1992) focus on the nature and content of prescriptive ideals for correctional systems.

CONCLUSION: LESSONS FROM CRIMINOLOGY

We believe that the best way to organize criminal justice theory is to categorize it on the basis of the dependent variable. There are three general types of dependent variables: the individual behavior of criminal justice agents, the organizational behavior of criminal justice organizations, and the aggregate behavior of the criminal justice system and its components as a whole. Once theories have been classified on the basis of their dependent variables, it is possible to sort and organize by independent variables. This organization of theory promotes conceptual clarity, allows generalization across system components, and permits competitive testing of theories in criminal justice; all of these benefits will advance criminal justice as a scientific enterprise.

Our approach is similar to the approach to criminology theory taken by Bernard and Snipes (1996; also see Vold, Bernard and Snipes 1998). According to Bernard and Snipes, the scientific process provides that theory will always exist in a context of research, just as research always exists in the context of theory. In this view, the ultimate criterion for theory is usefulness to research, just as the ultimate criterion for research is usefulness to theory. Theory can achieve utility for research more successfully by conceptualizing its goal as identifying variables and recognizing the order and organization among these variables. In a more policy-oriented phrase, the utility of theory is best achieved by identifying the sources of independent variation for precise dependent variables.

Bernard and Snipes therefore argue that the most important variable is the dependent variable, and that it should be the basis for sorting and classifying theories of criminology. Conceptual confusion arises when criminologists classify or categorize theories by the independent variables (e.g., biology, psychology, economics, social disorganization, strain, control, learning), even though this approach is intuitively appealing. The problem is that classifying theories on the independent variables implicitly assumes a single dependent variable, whereas in fact criminology involves multiple dependent variables. Bernard and Snipes describe three types of dependent variables in criminology: individual criminal behavior, the rates and distributions of criminal behavior in social units, and the behavior of criminal law. Each type includes many quite specific dependent variables, all of which can be linked to particular independent variables and to particular orders or organizations of those variables. For each dependent variable, multiple causal factors are described by multiple theories that compete with one another over explained variation. This approach is highly integrative as compared with standard interpretations, in which different theories contradict each other, so that only one theory can survive competitive testing while the others are falsified on the basis of statistical significance. In our approach, because the competition concerns explained variation, theories are not falsified, although they may be found to possess so little utility (account for so little explained variation) that they can be discarded.

All of these points, in our view, apply to criminal justice theory. We believe, however, that criminology and criminal justice theories differ in two major respects. First, criminology originated as sweepingly discursive general theories (Gibbs 1985) and has worked its way, with difficulty, towards phrasing those theories in such a way that research could test them adequately. Criminal justice, on the other hand, originated as descriptive research, in which theory, insofar as it existed at all, was highly specific and focused on the particular topic at hand. Over the last 50 years, criminal justice has been working its way, with difficulty, towards theory.

The second difference, we believe, is that criminal justice theories retain implicit or explicit prescriptive ideals. Criminal justice scholars should acknowledge explicitly that their work exists against a backdrop of attributions of legitimacy. This legitimacy standard is not a disembodied, static entity but is constructed in a constant, interactive relationship with the reality of criminal justice as revealed by our research. Our field has been both descriptive and prescriptive since it was founded; it cannot be understood fully without an acknowledgment of its prescriptive elements.
[Footnote]
* An earlier version of this paper was presented at the annual meetings of the American Society of Criminology, held in Toronto on November 17-20, 1999.

REFERENCES

Alschuler, A.W. 1979. “Plea Bargaining and Its History.” Law and Society Review 13:211-46.

Barry, B.M. 1989. Democracy, Power and Justice: Essays in Political Theory. Oxford: Oxford University Press.

Bayley, D.H. 1988. “Community Policing: A Report From the Devil’s Advocate.” Pp. 225-38 in Community Policing: Rhetoric or Reality, edited by J.
Greene and S. Mastrofski. New York: Praeger.

Bernard, T.J. and J.B. Snipes. 1996. “Theoretical Integration in Criminology.” Pp. 301-48 Crime and Justice: A Review of Research, edited by M.
Tonry. Chicago, IL: University of Chicago Press.

Bittner, E. 1974. “Florence Nightingale in Pursuit of Willie Sutton: A Theory of the Police.” Pp. 17-43 in The Potential for Reform of Criminal Justice, edited by H. Jacob. Beverly Hills, CA: Sage.
1975. The Functions of the Police in Modern Society: A Review of Background Factors, Current Practices, and Possible Role Models. New York: Aronson.

Black, D. 1976. The Behavior of Law. New York: Academic Press.
. 1979. “Common Sense in the Sociology of Law.” American Sociological Review 44:18-27.
1980. The Manners and Customs of the Police. New York: Academic Press. Black, D. and A.J. Reiss, Jr. 1970. “Police Control of Juveniles.” American Sociological Review 35:63-77.

Blau, P.M. 1974. On the Nature of Organizations. New York: Wiley.

Blumberg, A. 1967. “The Practice of Law as a Confidence Game.” Law and Society Review 1:15-39.

Brereton, D. and J.D. Casper. 1981. “Does It Pay to Plead Guilty? Differential Sentencing and the Functioning of Criminal Courts.” Law and Society Review 16:4570.

Broderick, J.J. 1977. Police in a Time of Change. Morristown, NJ: General Learning Press.

Brown, M.K. 1981. Working the Street: Police Discretion and the Dilemmas of Reform. New York: Russell Sage Foundation.

Caldwell, E. 1978. “Patrol Observation: The Patrol Encounter, Patrol Narrative, and General Shift Information Forms.” Police Services Study Report MR-02, Workshop in Political Theory and Policy Analysis, Indiana University, Bloomington, IN.

Cares, T.W. and J.A. Kruse. 1992. “Alaska’s Ban on Plea Bargaining Reevaluated.” Judicature 75:310-17.

Carter, L.H. 1974. The Limits of Order. Lexington, MA: Lexington Books. Chambliss, W.J. and R.B. Seidman. 1971. Law, Order, and Power. Reading, MA: Addison-Wesley.

Chiricos, T. and W.D. Bales. 1991. “Unemployment and Punishment.” Criminology 29:701-24.

Chiricos, T. and M.A. DeLone. 1992. “Labor Surplus and Punishment.” Social Problems 39:421-46.

Church, T.W., Jr. 1985. “Examining Local Legal Culture.” American Bar Foundation Research Journal 1985:449-518.

Clemmer, D. [1940] 1958. The Prison Community. New York: Holt, Rinehart and Winston.

Cole, G.F. 1970. “The Decision to Prosecute.” Law and Society Review 4:313-43. Cressey, D. 1959. “Contradictory Directives in Complex Organizations: The Case of the Prison.” Administrative Science Quarterly 4:1-19.

D’Alessio, S.J. and L. Stolzenberg. 1995. “The Impact of Sentencing Guidelines on Jail Incarceration in Minnesota.” Criminology 33:283-302.

DiIulio, J.J., Jr. 1987. Governing Prisons: A Comparative Study of Correctional Management. New York: Free Press.

Duffee, D. 1975. Correctional Policy and Prison Organization. Beverly Hills, CA: Sage. 1980. Explaining Criminal Justice: Community Theory and Criminal Justice Reform. Prospect Heights, IL: Waveland.

Engel, R.S., J.J. Sobol, and R.E. Worden. 2000. “Further Exploration of the Demeanor Hypothesis: The Interaction Effects of Suspects’ Characteristics and Demeanor on Police Behavior.” Justice Quarterly 17:235-58.

Eisenstein, J., R.B. Flemming, and P. Nardulli. 1988. The Contours of Justice: Communities and Their Courts. Boston, MA: Little, Brown.

Eisenstein, J. and H. Jacob. 1977. Felony Justice: An Organizational Analysis of Criminal Courts. Boston, MA: Little, Brown.

Feeley, M.M. 1973. “Two Models of the Criminal Justice System: An Organizational Perspective.” Law and Society Review 7:407-25. 1982. “Plea Bargaining and the Structure of the Criminal Process.” Justice System Journal 7:338-55. 1983. Court Reform on Trial: Why Simple Solutions Fail. New York: Basic Books.

Foucault, M. 1979. Discipline and Punishment: The Birth of the Prison. New York: Vintage.

Garland, D. 1990. Punishment and Modern Society. Chicago, IL: University of Chicao Press.

Gendreau, P., T. Little, and C. Goggin. 1996. “A Meta-Analysis of the Predictors of Adult Offender Recidivism.” Criminology 34:575-607.

Gendreau, P. and R. Ross. 1987. “Revivification of Rehabilitation: Evidence From the 1980s.” Justice Quarterly 4:349-407.

Gibbs, J. 1985. “The Methodology of Theory Construction in Criminology.” Pp. 23-50 in Theoretical Methods in Criminology, edited by R.F. Meier. Beverly Hills, CA: Sage.

Glassman, R.B. 1973. “Persistence and Loose Coupling in Living Systems.” Behavioral Science 18:83-98.

Goffman, E. 1961. Asylums: Essays on the Social Situation of Mental Patients and Other Inmates. Garden City, NY: Doubleday.

Goldstein, H. 1977. Policing a Free Society. Cambridge, MA: Ballinger. .1990. Problem-Oriented Policing. New York: McGraw-Hill.

Gordon, D.R. 1990. The Justice Juggernaut: Fighting Street Crime, Controlling Citizens. New Brunswick, NJ: Rutgers University Press.

Gorecki, J. 1979. A Theory of Criminal Justice. New York: Columbia University Press.

Gottfredson, M.R. and D.M. Gottfredson. 1981. Decision Making in Criminal Justice: Toward the Rational Exercise of Discretion. New York: Plenum. 1988. Decision Making in Criminal Justice: Toward the Rational Exercise of Discretion. 2nd ed. New York: Plenum. 1994. “Behavioral Prediction and the Problem of Incapacitation.” Criminology 32:441-74.

Gottfredson, M.R. and M.J. Hindelang. 1979a. “A Study of The Behavior of Law.” American Sociological Review 44:3-18.
1979b. “Theory and Research in the Sociology of Law. American Sociological Review 44:27-37. 1980. “Trite But True.” American Sociological Review 45:338-40.

Gross, H. 1971. A Theory of Criminal Justice. New York: Oxford University Press. Hagan, J. 1989. “Why Is There So Little Criminal Justice Theory? Neglected Macroand Micro-Level Links Between Organization and Power.” Journal of Research in Crime and Delinquency 26:116-35.

Hepburn, J.R. 1985. “The Exercise of Power in Coercive Organizations: A Study of Prison Guards.” Criminology 23:145-64.

Herbert, S. 1998. “Police Subculture Reconsidered.” Criminology 36:343-69. Heumann, M. 1978. Plea Bargaining: The Experiences of Prosecutors, Judges, and Defense Attorneys. Chicago, IL: University of Chicago Press.

Heumann, M. and C. Loftin. 1979. “Mandatory Sentencing and the Abolition of Plea Bargaining: The Michigan Felony Firearms Statute.” Law and Society Review 13:393-430.

Horney, J. and C. Spohn. 1991. “Rape Law Reform and Instrumental Change in Six Urban Jurisdictions.” Law and Society Review 25:117-53.

Hudson, B.A. 1996. Understanding Justice. Buckingham, UK: Open University Press.

Irwin, J. 1980. Prisons in Turmoil. Boston, MA: Little, Brown.

Irwin, J. and D.R. Cressey. 1963. “Thieves, Convicts, and the Inmate Culture.” Social Problems 10:142-55.

Jack, R. and D.C. Jack. 1989. Moral Vision and Professional Decisions: The Changing Values of Women and Men Lawyers. Cambridge, UK: Cambridge University Press.

Jacobs, D. and R.M. O’Brien. 1998. “The Determinants of Deadly Force: A Structural Analysis of Police Violence.” American Journal of Sociology 103:837-62. Jacobs, J.B. 1978. Statesville. Chicago, IL: University of Chicago Press.

Kaufmann, K. 1988. Prison Officers and Their World. Cambridge, MA: Harvard University Press.

Kelling, G.L., T. Pate, D. Dieckman, and C.E. Brown. 1974. The Kansas City Preventive Patrol Experiment: A Summary Report. Washington, D.C.: Police Foundation.

Kleck, G. 1981. “Racial Discrimination in Criminal Sentencing: A Critical Evaluation of the Evidence With Additional Evidence on the Death Penalty.” American Sociological Review 46:783-805.

Klinger, D. 1994. “Demeanor or Crime? Why ‘Hostile’ Citizens Are More Likely to Be Arrested.” Criminology 32:475-93.

Klockars, C.B. 1988. “The Rhetoric of Community Policing.” Pp. 239-58 in Community Policing: Rhetoric or Reality, edited by J. Greene and S. Mastrofski. New York: Praeger.

Larson, R.C. 1975. “What Happened to Patrol Operations in Kansas City? A Review of the Kansas City Preventive Patrol Experiment.” Journal of Criminal Justice 3:267-97.

Levin, M.A. 1977. Urban Politics and the Criminal Courts. Chicago, IL: University of Chicago Press.

Lipsky, M. 1980. Street-Level Bureaucracy: Dilemmas of the Individual in Public Services. New York: Russell Sage Foundation.

Lombardo, LX. 1981. Guards Imprisoned: Correctional Officers at Work. New York: Elsevier.

Manning, P.K. 1977. Police Work: The Social Organization of Policing. Cambridge, MA: MIT Press.
1988. “Community Policing as a Drama of Control.” Pp. 27-46 in Community Policing: Rhetoric or Reality, edited by J. Greene and S. Mastrofski. New York: Praeger.

Marquart, J.W. 1986. “Prison Guards and the Use of Physical Coercion as a Mechanism of Prisoner Control.” Criminology 24:347-66.

Martin, S.E. and L.W. Sherman. 1986. “Catching Career Criminals: Proactive Policing and Selective Apprehension.” Justice Quarterly 32:171-92.

Martinson, R. 1974. “What Works? Questions and Answers About Prison Reform.” The Public Interest 35:22-54.

Mastrofski, S.D., R.E. Worden, and J.B. Snipes. 1995. “Law Enforcement in a Time of Community Policing.” Criminology 33:539-63.

Mather, L.M. 1979. Plea Bargaining or Trial? The Process of Criminal-Case Disposition. Lexington, MA: Lexington Books.

McCoy, C. 1984. “Determinate Sentencing, Plea Bargaining Bans, and Hydraulic Discretion in California.” Justice System Journal 9:256-75.

McGarrell, EY 1993. “Institutional Theory and the Stability of a Conflict Model of the Incarceration Rate.” Justice Quarterly 10:7-28.

Meyer, J. and B. Rowan. 1978. “Institutionalized Organizations: Formal Structures as Myths and Ceremony.” American Journal of Sociology 83:340-63.

Morris, N. 1974. The Future of Imprisonment. Chicago, IL: University of Chicago Press.

Morris, N. and M. Tonry. 1990. Between Prison and Probation: Intermediate Punishments in a Rational Sentencing System. New York: Oxford University Press. Muir, W.K., Jr. 1977. Police: Streetcorner Politicians. Chicago, IL: University of Chicago Press.

Myers, M.A. 1990. “Economic Threat and Racial Disparities in Incarceration.” Criminology 28:627-56.

Packer, H.L. 1968. The Limits of the Criminal Sanction. Stanford, CA: Stanford Univerity Press.

Philliber, S. 1987. “Thy Brother’s Keeper: A Review of the Literature on Correctional Officers.” Justice Quarterly 4:9-37.

Prottas, J.M. 1978. “The Power of the Street-Level Bureaucrat in Public Service Bureaucracies.” Urban Quarterly Affairs 13:285-312.

Radelet, M.L. 1981. “Racial Characteristics and the Imposition of the Death Penalty.” American Sociological Review 46:918-27.

Rawls, J. 1971. A Theory of Justice. Cambridge, MA: Belknap. . 1999. A Theory of Justice. Rev. ed. Cambridge, MA: Belknap.

Reiman, J.H. 1984. The Rich Get Richer and the Poor Get Prison: Ideology, Class, and Criminal Justice. 2nd ed. New York: Wiley.

Reiss, A.J., Jr. 1971. “Systematic Observation of Natural Social Phenomena.” Pp. 333 in Sociological Methodology, edited by H.L. Costner. San Francisco, CA: Jossey-Bass.

Remmington, F. 1990. “Development of Criminal Justice as an Academic Field.” Journal of Criminal Justice Education 1:9-20.

Reuss-lanni, E. and F. Ianni. 1983. “Street Cops and Management Cops: Two Cultures of Policing.” Pp. 251-74 in Control in the Police Organization, edited by M. Punch. Cambridge, MA: MIT Press.

Riksheim, E. and S.M. Chermak. 1993. “Causes of Police Behavior Revisited.” Journal of Criminal Justice 21:353-82.

Rosecrance, J. 1988. “Maintaining the Myth of Individualized Justice.” Justice Quarterly 5:235-56.

Rossi, P.H. and RA. Berk. 1997. Just Punishments. New York: Aldine. Rubinstein, J. 1973. City Police. New York: Ballantine.

Rusche, G. and 0. Kirchheimer. 1939. Punishment and Social Structure. New York: Russell and Russell.

Scheingold, S.A. 1984. The Politics of Law and Order: Street Crime and Public Policy. New York: Longman. 1991. The Politics of Street Crime: Criminal Process and Cultural Obsession. Philadelphia, PA. Temple University Press.

Sherman, L.W. 1980. “Causes of Police Behavior: The Current State of Quantitative Research.” Journal of Research in Crime and Delinquency 17:69-100. 1992. Policing Domestic Violence. New York: Free Press.

Sherman, L.W., P.R. Gartin, and M.E. Buerger. 1989. “Hot Spots of Predatory Crime: Routine Activities and the Criminology of Place.” Criminology 27:27-55. Simpson, S.S. 1989. “Feminist Theory, Crime and Justice.” Criminology 27:605-31. Skolnick, J.H. 1966. Justice Without Trial: Law Enforcement in a Democratic Society. New York: Wiley.

Smith, DA. and C.A. Visher. 1981. “Street-Level Justice: Situational Determinants of Police Arrest Decisions.” Social Problems 29:167-77.

Spohn, C. 1990. “The Sentencing Decisions of Black and White Judges: Expected and Unexpected Similarities.” Law and Society Review 24:1197-1216.

Steffensmeier, D., J. Ulmer, and J. Kramer. 1998. “The Interaction of Race, Gender
and Age in Criminal Sentencing.” Criminology 36:763-97.

Stolzenberg, L. and S.J. D’Alessio. 1994. “Sentencing and Unwarranted Disparity: An Empirical Assessment of the Long-Term Impact of Sentencing Guidelines in Minnesota.” Criminology 32:301-10.

Streiker, C.S. 1999. “Death, Taxes, and-Punishment? A Response to Braithwaite and Tonry.” UCLA Law Review 46:1793.

Sudnow, D. 1965. “Normal Crimes: Sociological Features of the Penal Code in a Public Defender’s Office.” Social Problems 12:255-76.

Sykes, G.M. 1958. The Society of Captives: A Study of a Maximum Security Prison. Princeton, NJ: Princeton University Press.

Sykes, R.E. and J.P. Clark. 1975. “A Theory of Deference Exchange in Police-Citizen Encounters.” American Journal of Sociology 81:584-600.

Tanner, H.M. 1999. Strike Hard! Anti-Crime Campaigns and Chinese Criminal Justice, 1979-1985. Ithaca: Cornell University Press.

Toch, H. 1977. Living in Prison: The Ecology of Survival. New York: Free Press. Toch, H. and J. Mofas. 1982. “Alienation and Desire for Job Enrichment Among Correctional Officers.” Federal Probation: A Journal of Correctional Philosophy and Practice 46:35-44.

Tonry, M. 1999. “Rethinking Unthinkable Punishment Policies in America.” UCLA Law Review 46:1751-91.

Tunick, M. 1992. Punishment: Theory and Practice. Berkeley, CA: University of California Press.

Ulmer, J.T. and J.H. Kramer. 1998. “The Use and Transformation of Formal Decision-Making Criteria.” Social Problems 45:248-67.

Van Maanen, J. 1974. “Working the Street: A Developmental View of Police Behavior.” Pp. 83-129 in The Potential for Reform of Criminal Justice, edited by H. Jacob. Beverly Hills, CA: Sage.

Vold, G.B., T.J. Bernard, and J.B. Snipes. 1998. Theoretical Criminology. 4th ed. New York: Oxford University Press.

Walker, S. 1977. A Critical History of Police Reform: The Emergence of Professionalism. Lexington, MA: Heath. 1992. “Origins of the Contemporary Criminal Justice Paradigm: The American Bar Foundation Survey, 1953-1969.” Justice Quarterly 9:47-76.
1998. Sense and Nonsense About Crime and Drugs: A Policy Guide. 4th ed. Monterey: Brooks/Cole.

Walker, S., C. Spohn, and M. DeLone. 1996. The Color of Justice. Belmont, CA: Wadsworth.

Westley, W.A. 1953. “Violence and the Police.” American Journal of Sociology 59:3441. 1970. Violence and the Police. Cambridge, MA: MIT Press.

Whitaker, G. 1982. “What is Patrol Work?” Police Studies 4:13-22.

White, S.O. 1972. “A Perspective on Police Professionalization.” Law and Society Review 7:61-85.

Wilson, J.Q. 1968. Varieties of Police Behavior: The Management of Law and Order in Eight Communities. Cambridge, MA: Harvard University Press.

Wilson, J.Q. and G.L. Kelling. 1982. “Broken Windows: The Police and Neighborhood Safety.” The Atlantic Monthly (March):29-38.

Worden, R.E. 1989. “Situational and Attitudinal Explanations of Police Behavior: A Theoretical Reappraisal and Empirical Assessment.” Law and Society Review 23:667-711. 1996. “The Causes of Police Brutality: Theory and Evidence on Police Use of Force.” Pp. 23-51 in Police Violence: Understanding and Controlling Police Abuse of Force, edited by W.A. Geller and H. Toch. New Haven, CT: Yale University Press.

Worden, R.E. and R.L. Shepard. 1996. “Demeanor, Crime, and Police Behavior: A Reexamination of the Police Services Study Data.” Criminology 34:83-105.

THOMAS J. BERNARD**
ROBIN SHEPARD ENGEL***
The Pennsylvania State University

** Thomas J. Bernard is a professor of criminal justice and sociology at The Pennsylvania State University. His main academic interests are in criminology theory and juvenile justice.
*** Robin Shepard Engel is an assistant professor of crime, law, and justice at The Pennsylvania State University, and received a PhD in criminal justice from the University at Albany in 1999. Her current research involves theoretical and empirical explorations of police supervision, patrol officers’ behavior, policing special populations, and criminal justice theory.