A number of writers have observed that legalism is a social phenomenon which is found only in certain kinds of legal systems and then only in certain historical periods. Perhaps it was Weber who first gave the concept a theoretical importance by associating a legalistic or “formally rational” legal system with the development of capitalism. Later writers such as Selznick have seen legalism as a complex ideal embracing certain kinds of standards to which legal systems might aspire. (Selznick, 1961). Most recently, Friedman has commented on the decline of legalism (Friedman, 1966). While the contributions of these writers are useful in elaborating the meaning of the concept of legalism, the phenomenon itself has received virtually no empirical analysis. It is the purpose of this paper to show that one possible component of legalism, viz., predictability, is by no means unique to legalism and that a nonlegalistic adjudicatory system can be as predictable if not more predictable than a legalistic system would be.

The subject of the police is a neglected issue among political scientists. The police are rarely viewed from perspectives natural to political science, nor are they studied comparativly among countries. The few studies that have been done are written largely from the point of view of public administration; they tend to be wholly descriptive and to deal largely with matters of formal organization and management. The neglect of the police is not unique to political scientists; the record of other social sciences is hardly better. Even in sociology there has been surprisingly little. Historical monographs on countries rarely refer to the police at all; inspection of indexes of basic histories of most foreign countries will reveal very few references to the police. Unrest in American cities and violence between police and minorities, as well as between police and students, is beginning to convince the scholarly community that the police are crucial social actors. The impetus so far for empirical study has come largely from government, in the form of various national study commissions (President's Commission on Law Enforcement and Administration of Criminal Justice, 1967; National Advisory Commission on Civil Disorders, 1967; National Commission on the Causes and Prevention of Violence, 1969). The kindling of interest in the police is new; it may be dated from 1965.

During the past two decades students of public law have amassed a considerable amount of evidence which indicates that the administration of federal justice is not uniform throughout the United States. First, there are significant differences in judicial administration among the federal district courts. Kenneth Vines and others have demonstrated how important these differences can be with regard to race relations cases (Vines, 1963; Vines, 1964; Peltason, 1961; Steamer, 1962). In his study of the federal trial courts and urban affairs, Kenneth Dolbeare concluded that “many of Vine's findings with respect to race relations policy making have parallels in the area of urban public policy generally.” (Dolbeare, 1969: 376). Also, there is some evidence that sentencing behavior in criminal cases varies significantly from one federal trial judge to another.

One serious obstacle to the achievement of social change in America has been the intransigence of large bureaucracies toward changes in goals, procedures, and styles. Although programs ad infinitum to remedy evils such as poverty, segregation, crime, and pollution are proposed and even written into law, the day-to-day operation of our political institutions often fails to reflect these innovations. Thus we are unable to evaluate the merits of alternative solutions to such ills because they are never given the test of full-fledged implementation.

In this paper I am going to suggest various measures to modify the behavior of one important political institution which has been staunchly resistant to change—the police. Our under standing of police operations is still quite limited (due partially to the secretive nature of most departments), but as David Easton (1969) has argued, social scientists have an obligation to use whatever intellectual tools are at their disposal to obtain “quick, short-run answers” to questions of pressing social concern. To this end, what follows is an attempt to apply some principles of learning theory (developed largely in highly-structured psychology laboratories on infra-human subjects) to the chaotic environment of the cop on the beat.

In a critique of our paper, “Severity of Formal Sanctions as a Deterrent to Deviant Behavior” (Law and Society Review, 1970: 21-40), Professor Paul E. Meehl has questioned our interpretation of the data particularly in Table 3 of the original paper. On the basis of that evidence, we concluded that severity of formal sanctions show no direct deterrent effect on the deviant behavior under consideration, but do appear to have an indirect deterrent effect through their impact on the normative climate of the campus (p. 37). In challenging this interpretation of the data, Professor Meehl expressed concern that non social science readers of the Review might be misled by our use of “causal language” in the interpretation of these tabulations from cross sectional data.

A generation ago, one professor at Harvard Law School used to greet his students at the beginning of the semester with an offer to debate them on any subject so long as he was allowed to “state the question.” Any review of the crime, law enforcement, and criminal justice literature produced during the last ten years indicates that there is still as much disagreement over how the “relevant” questions should be stated as there is over the answers to these questions. Despite a strong public demand to “stop crime,” there is little consensus within the research community as to the questions (if any) which, if answered, would assist in the prevention of crime or the improvement of police, court, and correctional processes. Much of the literature “states the question” in terms of the narrowly defined “efficiency” of law enforcement and criminal justice agencies; a second group asks questions concerning the functions assigned to the criminal justice system; and a third set of critics are concerned with the implications of decisions made by the police, judges, and correctional officials for a democratic system of government. While I cannot attempt to review the literature which has been generated under these three models, I will describe each briefly and suggest some of the problems that have risen in research on law enforcement and criminal justice.

In the 1920s and early 1930s the Marxist legal theorist, Eugene Pashukanis, advanced the thesis that the achievement of communism would produce not only the dissolution of the state but of the legal system as well. Briefly, he argued that law was a direct result of economic exchange, that economic exchange was an integral part of bourgeois society, that all law was hence bourgeois law, and, finally, that the demise of bourgeois society would carry law with it. The fascination Pashukanis exerts is a direct product of the purity of his conclusion. The Revolution does not bring about a revision of the legal system, nor (except for a transitional period) does it even substitute one legal system for another. To the extent that the Revolution succeeds, it obliterates law qua law altogether.

Students of the judicial process have focused considerable attention on the relationship between judges' background characteristics and their voting behavior on collegial courts. The argument for such a focus is that background characteristics are usually indicative of particular socialization processes and that these processes produce certain attitudes which are often responsible, directly or indirectly, for the judges' votes. Thus these characteristics can be viewed, in the aggregate, as predictive of judges' behavior — at least in certain types of cases.

Virtually unstudied and largely unobserved, the police employee organization has evolved over the last fifty years into a strong economic and political institution. The rapid growth of militant police unionism as a new political and economic force in the society has raised serious problems for the police agency administrator in the exercise of his professional responsibilities in the area of law enforcement and his executive responsibilities in the area of personnel management. It has also raised serious public policy questions as to whether the protected right to organize and to bargain collectively which is being extended to all other public employees ought to be extended to the police without limitations. Underlying all these questions is the basic issue of whether official sanction should be extended to another entrant in the competition for control of local police operations.

Today, few would deny that some relation exists between firearms and violent death and crime:

Newton and Zimring of the Task Force on Firearms point out that 63% of all homicides, 37% of all robberies, and 21% of all aggravated assaults involve the use of a gun (Newton and Zimring, 1970: 39). In turn, 76%) of gun homicides are committed by handguns; similarly, 86% of aggravated assaults involving guns and 96% of robberies involving guns are committed by handguns (Newton and Zimring, 1970: 49). In short, although approximately 27% of all firearms in the United States are handguns, they are the predominant firearm used in crime (Newton and Zimring, 1970: 49).

The question of who supports the law is commonly approached within a framework of social stratification analysis. Pressure to pass certain laws and willingness to assist the agents of law enforcement are often related to social status and political power. Marx and Engels (1947) were among the first to see the law and support for the law in terms of the interests of various groups in the stratification system, arguing that the law is a tool employed by the bourgeoisie to oppress and exploit the proletariat. In another historical study, Ranulf (1964) concluded that the “distinterested tendency to inflict punishment” is concentrated in the petty bourgeoisie or lower-middle class, suggesting that it is in that level of the social hierarchy where one will find the greatest support for the law, or at least the punitive aspects of the law. Not too different from this is the conclusion reached by Marshall (1968) which is that middle-class values have long been critical in determining what types of social behavior will be labeled criminal.

The federal trial judge works in a court system which is rapidly modernizing by developing central administrative structures and procedures to monitor and expedite his tasks. The new structures are creating new expectations for the federal judicial role. For the incumbent judge the demand to change his role definition and his pattern of action breeds tension and resistance. The Chandler case (1970), involving the sanctioning of a trial judge by his circuit judicial council, which Justice Douglas characterized as “the most controversial contest involving a federal judge in modern United States history,” epitomizes the stress upon a trial judge, nurtured in a period of mild bureaucracy, finding himself at the end of his career in an organization with new norms and stronger instruments of enforcement.

There has recently been a growing concern within political science with the public's support for governmental institutions. To date, the research in the judicial realm — focusing extensively on the United States Supreme Court — has measured support primarily in terms of the public's evaluation of court outputs and has attempted to explain support through variables believed to function as important bases or references for these evaluations, e.g., policy orientations, political party identification, and race (Murphy and Tanenhaus, 1968a and 1968b; Dolbeare and Hammond, 1968; Dolbeare, 1967; Hirsh and Donohew, 1968). Yet the public is not especially cognizant of court decisions and the low “visibility” of judicial bodies (Dolbeare, 1967; 198-201; Barth, 1968; Jacob, 1966) would suggest that the basis of support might well be more diffuse than specific. An analysis of diffuse support might be concerned primarily with the public's general willingness to comply with judicial decrees or with its willingness to maintain judicial institutions as bodies appropriate for resolving conflicts, and only secondarily with outputs.

One of the more interesting questions that political scientists are likely to overlook in their analyses of public policy, the delivery of services, etc., is the continually changing mix of public and private provision of such policies and services. Present concern with the criminal justice system, in general, and the police in particular, is no exception. Political scientists are interested in questions of police organization, public accountability of police activity, implications of greater involvement in local police activity by state and federal agencies, professionalization of local police officers, etc. But we have tended to concentrate on the political-governmental provision of police services and to ignore both the traditional and the contemporary role of private police as a very important part of the total panoply of police activities in American society.