I Theories of OrganizationsPeter M. Blau
II Organizational GoalsCharles Perrow
III Effectiveness and Planning of ChangeChris Argyris
IV Organizational IntelligenceHarold L. Wilensky
V Methods of ResearchAllen H. Barton
The articles under this heading review the field of organizational sociology. Directly related are the entries Administration; Bureaucracy.Discussions of particular types of organizations may be found under Cartels and trade associations; Cooperatives; Interestgroups; International Organization; Labor Unions; Medical Care; Parties, Political; Penology; Social Movements;Voluntary Associations; and in the articles listed in the guide under Industrial Organization.The biographies of Barnard; Mayo; Michels; andWeber, Maxshould also be consulted.
An organization comes into existence when explicit procedures are established to coordinate the
activities of a group in the interest of achieving specified objectives. The collective efforts of men may become formally organized either because all of them have some common interests or because a subgroup has furnished inducements to the rest to work in behalf of its interest. Factory workers organize themselves into unions to bargain collectively with management, and management has organized the workers’ tasks for the purpose of producing goods marketable for a profit. Unions and factories exemplify formal organizations, as do government bureaus and political parties, armies and hospitals.
There are two basic principles that govern social life, and organizations manifest one of these. Social structures may emerge as the aggregate result of the diverse actions of individuals, each pursuing his own ends, or they may reflect the joint endeavors of individuals pursuing commonly accepted ends. Thus, as individuals and groups in a community compete, enter into exchange relations, and use their resources to exercise power over others, an economic system and a class structure develop, which reveal organized patterns of social conduct, although nobody has explicitly organized the endeavors of individuals. The government of a society and a football team, on the other hand, are social structures deliberately established to achieve certain objectives, and the regularities observable in them reflect deliberate design. The distinction is essentially the one made by William Graham Sumner between “crescive” and “enacted “institutions. Social systems produced by formally enacted procedures, rather than merely emergent forces, are organizations. The distinction is an analytical one, since crescive and enacted forces typically interact in their effects on social systems. Nevertheless, the distinction finds concrete expression in the difference between organizations and other collectivities.
Whenever groups of men associate with one another, social organization develops among them, but not every collectivity has a formal organization. The defining criterion of a formal organization—or an organization, for short—is the existence of procedures for mobilizing and coordinating the efforts of various, usually specialized, subgroups in the pursuit of joint objectives. If all relations among the members of organizations and all their activities were completely predetermined by formal procedures, however, organizations would evidently not pose meaningful problems for scientific inquiry, because everything about them could be ascertained by simply examining the official blueprints and procedure manuals. Actually, social interaction and activities in organizations never correspond perfectly to official prescriptions, if only because not all prescriptions are compatible, and these departures from the formal blueprint raise problems for empirical study. Paradoxically, therefore, although the defining characteristic of an organization is that a collectivity is formally organized, what makes it of scientific interest is that the developing social structure inevitably does not coincide completely with the pre-established forms.
Max Weber’s theory of bureaucracy (1922a;1922b) addresses itself to both the problem of the changing social organization of modern society and that of the typical features of the formal organizations that pervade it. On the one hand, Weber was concerned with the increasing bureaucratization of ever wider aspects of contemporary social life. On the other hand, he presented the classical analysis of the typical social structure of the complex, large organization. Accordingly, his writings have given rise to two distinct traditions in the social sciences—historical studies of bureaucratization in societies and empirical research on complex organizations.
Weber ([1922b] 1957, pp. 124132) emphasized that the foundation of a legitimate social order is a common value orientation that effects social control and compliance with authoritative commands in a society. The authority structure, in other words, is the core of the social organization, and authority structures can be differentiated in terms of the value systems that legitimate them. On the basis of this criterion Weber distinguished three main types of authority. First,traditional authority is legitimated by the belief in the sanctity of tradition, according to which a given person or group, usually defined by heredity, is preordained to rule over the rest, as exemplified by the doctrine of the divine right of kings. Second,charismatic authority rests on a value orientation that considers the pronouncements of the leader of a social movement to be inspired by supernatural powers, and this conviction prompts the converts to his cause willingly to obey the leader’s commands. Legal authority, finally, is legitimated by a formalistic belief in the supremacy of a body of legal norms, whatever their specific content;these impersonal principles include the requirement to follow directives originating from designated official positions, regardless of the persons who occupy these positions. Legal authority is characteristic of bureaucratic organizations and, generally, of modern societies. Weber ([1922a] 1946, pp. 204-216) traced the historical conditions that promoted bureaucratization, such as a money economy, the absence of slavery, the large size of a collectivity, the complexity of administrative tasks, and the growth of capitalism, and his historical analysis convinced him of the inevitable trend toward increasing bureaucratization in the modern world.
The importance bureaucracy assumes in modern life led Weber to give its analysis a central place in his theory of society. Although he examined primarily government bureaus, his principles apply to complex formal organizations in general. The basic question he asked was how collective endeavors must be organized to rationalize complex responsibilities that require the joint efforts of many men. In short, what are the principles of rational administration on a large scale?It is not entirely clear whether his answer to this question was intended as a normative theory, indicating what a large organization should be like to maximize efficiency, or as a descriptive theory, outlining what the prototype of existing complex organizations actually is. In any case, his theory specifies the typical characteristics of bureaucracies and analyzes the way in which these characteristics depend on each other.
Large collectivities confronted by complex administrative tasks are most likely to become bureaucratic ally organized. Such formal organization involves an extensive division of labor, with specialized responsibilities being assigned to trained experts as their circumscribed professional duties. A separate administrative apparatus tends to develop, whose members are responsible for maintaining the organization itself, including notably its lines of communication, rather than for contributing directly to attaining its objectives. Another distinctive attribute of bureaucracy is that official positions are organized into a hierarchy of authority, the scope of which is precisely defined by impersonal rules. Operations too are governed by a consistent system of rules and regulations. Impersonal detachment is expected to prevail in the performance of duties and in official relations. The members of bureaucracies are full-time employees and have stable careers with some advancement in the organization, since explicit personnel policies protect them from arbitrary dismissal and promotions are based on objective criteria, such as merit and seniority. These are the major traits of bureaucracy, although Weber noted others, for example, the proliferation of written documents and the ease with which personnel can be substituted.
Weber presented an implicit functional analysis of the interdependence of the characteristics of bureaucracy, with rational efficient administration as the criterion of function. Effective accomplishment of complex administrative tasks on a large scale requires that they be subdivided into specialized responsibilities, which can be readily handled by individuals, and that professionally qualified experts be appointed to discharge these responsibilities. This pronounced division of labor creates serious problems of coordination, particularly in a large organization. A special administrative staff is needed to maintain channels of communication and coordination, and a strict hierarchy of authority serves to effect the coordination of diverse tasks in the pursuit of organizational objectives by enabling superiors on successive levels to guide, directly or indirectly, the performance of increasingly wider circles of subordinates. But close supervision of all decisions is inefficient and produces strains. The system of official rules is designed to standardize operations and restrict the need for direct supervisory intervention largely to extraordinary cases. Professional training and official rules notwithstanding, however, strong emotions and personal bias are likely to interfere with the ability to make rational decisions. The emphasis on impersonal detachment has the function of precluding the intrusion of such irrational factors into official decisions. Lest the impersonal discipline in the hierarchical bureaucracy alienate its members, secure careers lessen this burden and promote loyalty to the organization.
In brief, the problems created by one condition in the organization stimulate the development of another condition to deal with them. A number of interdependent processes of this kind give rise to the constellation of features characteristic of the typical bureaucracy, as conceptualized by Weber. He held that these features of an administrative organization and, especially, their combination are “capable of attaining the highest degree of efficiency” ([1922b] 1957, p. 337).
Political and industrial organizations
While Weber’s is the sociological classic on the subject, theories of organizations have been advanced by scholars in other fields. To be sure, since formal organizations are largely a modern phenomenon and exist only in rudimentary form, if at all, in simple societies, anthropologists have been little concerned with them. In anthropological studies “organization” refers generally to the kinship structure, which is not a formal organization as here defined, although it is a highly formalized social structure in these societies. It makes sense, however, in studying almost any society, to speak of complex political and economic organizations: and theories of organizations have been developed in these fields.
The political government is the most encompassing formal organization in any society, and political scientists as well as philosophers have tfeeorized about the state for more than two thousand years. But the unique characteristics of organized government—notably its encompassing scope and its monopoly of force—discourage comparison with other organizations. The case is different for particular government agencies, on which Weber centered attention, and also for political parties.
The “iron law” of Michels
In his famous study of socialist parties and unions in Germany, Michels (1911) proposed an “iron law of oligarchy,” with which he accounted for transmutations in these organizations. Political parties and labor unions, according to Michels, invariably become bureaucratized, regardless of how egalitarian their ideology. The reason is that centralized control with the help of an administrative staff is essential for success in elections or bargaining. The experience leaders gain and the victories they achieve make them virtually indispensable. Since a strong organization is important for implementing the radical program, the leadership becomes increasingly concerned with fortifying the organization and protecting it against attacks from conservative forces, even if doing so requires abandoning the original radical ideals for more moderate ones to pacify the powerful opposition. In sum, Michels held that inevitable organizational processes transform egalitarian parties or unions into centralized bureaucracies dominated by officials and turn their radical goals into modest reform programs.
Conditions in the International Typographical Union challenge Michels’ thesis, inasmuch as an internal two-party system in this union prevented the development of a self-perpetuating bureaucratic oligarchy. Research revealed that a major factor that preserved democratic processes was the existence of many semiautonomous organizations within the union’s boundaries—diverse associations and clubs of union members as well as strong union locals (Lipset et al. 1956). These “intermediate associations” protected the democratic system by mobilizing membership participation in union affairs, providing a training ground for leadership, and serving as organizational bases for opposition to the incumbent leaders. The conclusions of this study do not so much negate the thesis that effective unions become bureaucratized as refine it by indicating that intermediate organizations can stem the trend toward centralization of power, which is otherwise inherent in bureaucratization.
Political parties and the party system
Duverger’s theory of political parties (1951) analyzes how their organization is affected by the political system in which they operate. Election by a simple majority promotes a two-party system, inasmuch as third parties suffer severe handicaps under it, while proportional representation encourages the development of many parties. If there are only two parties, each derives advantages from moving toward the center, because by doing so it can gain some support from moderates without losing much from extremists, who have nowhere else to go. In a multiparty system, by contrast, parties are inclined to adopt extremist programs and, in the case of minor parties, even demagogic policies, as each seeks to differentiate itself from the others. Duverger’s theory is of general interest as one of the few systematic attempts to relate the attributes of organizations to the nature of the broader system in which they function.
The theory of administration
Whereas students of organizations have concentrated on their internal structure and paid little attention to their interrelations, economists have intensively investigated the relations among firms in the economy and inquired little into their internal organization, just as most political scientists have been more concerned with the larger political system than with the internal operations of government agencies. The specialized fields of business and public administration evolved to study these organizations, with the primary aim of helping to solve practical problems of management. An early example is Taylor’s “scientific management” (1911).It has been called a “physiological organization theory” (March & Simon 1958), because it stresses the influence of the capacities of the human organism on physical performance, although it also emphasizes pay incentives as well as time-andmotion studies.
The traditional administrative theory, which was developed between the two world wars, addresses itself more directly to the principles that govern complex organizations. Thus, Luther Gulick specified purpose, process, clients, and place as four basic factors to be considered in organizing an administrative establishment; M. P. Follett suggested principles for effecting coordination in a large organization; and V. A. Graicunas derived from a mathematical model of social relations his well-known axiom of limited span of control (see Gulick & Urwick 1937). Unlike Weber’s general theory, these principles are oriented toward the practical problems of administration, but like his theory, they conceive of an organization as a rational instrument for implementing objectives and policies.
Criticisms of this conception of organization stimulated subsequent theoretical developments. The rational model has been attacked as an abstraction that ignores actual human relations and behavior in organizations and, in particular, the nonrational elements in human conduct and their implications for operations. Moreover, exclusive concern with the formally instituted conditions in the organization has been criticized for failing to take into account the informal patterns that emerge and modify social relations and the performance of tasks. Finally, the implicit assumption that organizations are functional systems in equilibrium has been assailed as misleading, because it tends to neglect dysfunctions, conflict, and change. While some of these criticisms were directed primarily by students of business management against traditional administrative theory and others by sociologists against Weber, both trends converged in a growing concern with informal organization.
Earlier students of organizations realized, of course, that actual practice did not follow the formal blueprint in every detail. Later researchers discovered, however, that these departures from official procedures are not idiosyncratic but become socially organized. These social patterns, informally organized by the participants themselves, complement those formally organized for them by management. Furthermore, as Chester Barnard pointed out (1938, pp. 115-123), the informal organizations that always arise in formal organizations are essential for operations.
Simultaneously with Barnard, a research team arrived at the same insight from their study of industrial workers in an electrical equipment factory. Roethlisberger and Dickson (1939) found that the informal relations in a work group assumed a distinctive structure, with subgroups and status differences, and that informal norms emerged that regulated the performance of workers. Specifically, workers were expected by their fellows neither to produce too fast nor to produce too slowly, and deviations from these group norms were penalized by ridicule, loss of status, and, eventually, ostracism. In other words, workers informally organized themselves to control output, and an informal status structure complemented this organized social control. The informal organization, therefore, had important implications for operations.
Research on “human relations in industry,” as it came to be called, proliferated widely after this pioneering study. For example, the effect of group cohesion and of the relation between supervisor and subordinates on productivity were intensively and repeatedly investigated at the University of Michigan. One main conclusion of these studies was that close supervision lowers productivity, whereas an interest in the welfare of subordinates raises it (Kahn & Katz 1953). In this research tradition, the study of the social relations among the members of the organization came to replace any concern with the character of the organization itself [see IndustrialRelations, article on HumanRelations].
Merton’s concept of goal displacement
In contrast to the students of human relations, whose findings have implications for Weber’s theory, although they did not directly address themselves to it, Merton (1940) explicitly called attention to shortcomings of Weber’s analysis. He showed that bureaucratic procedure, although designed to further administrative efficiency, often has unanticipated consequences that actually impede it. His central concept is that of displacement of goals, which refers to a notion implicit in Michels. The emphasis on strict discipline and adherence to rules in bureaucracies, intended to assure consistency and impartiality, leads officials to think of formal procedures not simply as means for accomplishing certain objectives but as ends in themselves. The rigid conformity resulting from this displacement of goals prevents adaptation to new situations and generally interferes with effective operations. Bureaucratic arrangements have dysfunctions as well as functions for rational administration, since their unanticipated consequences recurrently create problems—a phenomenon that Weber consistently neglected.
A number of case studies of bureaucracies have attempted to refine Weber’s theory by following Merton’s lead and by applying the insights of industrial research. Thus, a government agency in a hostile local environment was found to adjust to this situation through informal cooptation of opposition elements into its leadership structure, but this method of organizational adjustment gave rise to fundamental revisions in official policies (Selznick 1949). Gouldner’s study of an industrial organization found that managerial suecession promotes bureaucratization (as implied by Weber’s theory), because a new manager, unfamiliar with informal practices, is constrained to rely on official procedures to implement his directives. But such exercise of authority that rests on bureaucratic rules and discipline should be distinguished from professional authority that rests on technical expertness, a distinction Weber failed to make explicit (Gouldner 1954).
Bureaucratization and professionalization are alternative means for rationalizing action by creating order out of disorder. If bureaucratic routines have been established to handle most problems in an organization, the professional expert capable of coping with the remaining problems that have not been routinized is likely to acquire more actual power than his formal position prescribes (Crozier 1963). Impersonal evaluation of performance on the basis of statistical criteria in bureaucracies may engender anxiety and competition, which impede effective operations, but unofficial practices tend to develop, such as informal consultation among colleagues, that relieve these difficulties and restore operating efficiency (Blau 1955).
The general conclusion that emerges from these studies is that procedures formally instituted for specific purposes in organizations recurrently create disturbances in other respects, and the informal patterns that typically arise to cope with these disruptions often produce a basic reorganization of operations. Whereas the focus on informal practices and relations permitted some refinement of Weber’s theory, it also led investigators increasingly away from the study of the fundamental structural features of complex organizations. This is most evident in theories of administrative decision making.
Organizations as equilibrium systems
Barnard (1938) conceived of organizations as equilibrium systems of coordinated effort in which participants make contributions in return for inducements, and this concept has been elaborated by Simon (1947; cf. March & Simon 1958). The two major parts of the theory are concerned with the “inducement-contribution balance” and with the structure of decision making in organizations.
The problem of motivation
An organization must furnish sufficient inducements to motivate individuals to participate in and contribute to it. The utility of the individual’s contributions can be defined in terms of the alternatives foregone by making them. A direct measure of the utility of the inducement-contribution balance is the individual’s satisfaction with his participation. Zero satisfaction, however, does not necessarily result in the decision to leave the organization, since the latter is a function of both the desire to move and the perceived ease of moving. Provided certain assumptions about the utility functions are made, independent indirect measures of the inducementcontribution utilities are also possible.
The problem of rationality
Simon considers that the primary function of administrative organization is to provide a framework for rational decision making. It is beyond a human being’s capacity to make the complex decisions required for implementing a large organization’s goals rationally, because value as well as factual judgments are involved and because no man can weigh the innumerable possible alternatives. The administrative structure is designed to furnish every member of the organization with both the factual information and the value premises necessary to enable him to make rational decisions among limited alternatives. No rational choice among various ends is possible; only the choice among different means for a given end can be rational. The hierarchical organization of responsibility defines the ends for every official and thereby restricts all decisions to choices among means for given ends. The decisions on one level in the hierarchy specify the responsibilities for the next lower level, so that all officials have the ends of their tasks set for them by their superiors and can rationally choose among the appropriate means for reaching these ends.
Simon’s theory is a rational model of organization, as is Weber’s, and both have been criticized for their rationalistic assumptions. Otherwise, however, the two theories are diametrical opposites. Whereas Weber analyzed the attributes of organizations and was little concerned with the behavior of individual members, Simon and March concentrated on the motivation of individuals in organizations and paid little attention to the various characteristics of bureaucratic structure. Indeed Simon’s theory is really not one of organizations but a theory of human behavior in the context of organization. He takes for granted the conditions in the organization and raises questions about human motivation, while the central problem of an organizational theory is the social structure of organizations rather than the psychology of decision making. The reaction to Weber and traditional administration theory for ignoring actual conduct in formal organizations has led all the way to an organizational theory that deals with such conduct to the exclusion of the very characteristics of organizations themselves!
Parsons’ model of organizations
Parsons has presented a different equilibrium model of organizations (1960), which treats them as natural social systems, and which contrasts with both Weber’s rational model (see Gouldner 1959) and Simon’s individual-centered model. Parsons’ general theory stipulates that all social systems must solve four basic problems: adaptation to the environment, goal achievement, integration of subunits into the larger system, and latency, that is, maintenance of the value patterns over time. Since formal organizations are the typical mechanism for implementing social objectives in the modern world, they are part of the goal-achievement system of the society. Furthermore, organizations can be classified by their functions as belonging in part to one of the four systems; economic organizations, for example, are partly elements of the adaptation system. Finally, as social systems, organizations themselves must meet the four generic system problems by developing appropriate subsystems. Thus, subsystems become differentiated for mobilizing resources to meet problems of adaptation, for organizing “production” to achieve the organization’s goals, for winning the loyalty of members and coordinating their efforts in order to integrate the system, and for relieving tension and maintaining consensus on legitimating values. Such treatment of organizations as natural systems evidently minimizes their distinctive traits.
Prospects of organizational analysis
Three foci can be distinguished in the study of organizational life; they may be called “role analysis,” “group analysis,” and “organizational analysis.” In the first case, the focus is on the behavior of individuals in their specific roles as members of organizations, as illustrated by studies of career patterns, work satisfaction and productivity, attitudes of soldiers toward combat, and administrative decision making. The second focus is on the structure of social relations in work groups—for example, the significance of group cohesion for performance, the differentiation of informal status in groups, the exchange of advice for prestige among colleagues, and generally patterns of informal organization. Third, organizational analysis proper focuses upon the system of interrelated attributes that characterize the organization itself rather than its component parts, such as its size, the division of labor in it, its bureaucratization, and the degree of centralization of control. To be sure, these are closely related factors, and those on one level cannot be investigated without reference to those on the others; but neither can systematic inquiry center on all of them simultaneously. The crucial question is the problem focus of the theory, that is, which patterns it seeks to explain—role behavior, group structures, or organizational systems—and which ones are taken as given or as accepted principles in providing the explanation.
The focus of organizational analysis
Social systems constitute “nesting series,” as Parsons has noted, with each system typically being encompassed by wider systems and having, in turn, subsystems within its boundaries. Organizations are the environment in which work groups develop their social structures, and these group structures provide the social setting for the role behavior of individual members. At the same time, organizations function within the context of communities, and these exist in the wider framework of societies and cultures. While human conduct is the fundamental datum for all studies of social as well as psychological systems, the theoretical principles abstracted from discrete empirical observations have an explicit or implicit focus on one of their systems of relationships.
The three types of analysis differ in respect to the system referent of the variations they seek to explain and the variables they treat as problematical rather than given. Although role analysis is often concerned with the significance of different bureaucratic conditions or informal relations, it considers them as given stimuli and tries to explain how they affect behavior, for instance, decision making. Group analysis similarly treats the formal characteristics of organizations as given conditions and investigates their influence on the informal relations and group structures that develop. Neither of them questions why the observed constellation of organizational attributes has come into existence, whereas organizational analysis does precisely that. Theories of role behavior and group structures in the context of formal organizations should be distinguished from genuine theories of organizations, which explain how differences in some organizational characteristics lead to differences in others. Only the comparative focus on variations among organizations can produce such theories.
A theory of organizations should not only establish the prevailing relationships between organizational features but also elucidate them in terms of the underlying processes that produce them. It is not enough to indicate, as Weber did, that bureaucratization facilitates substitution of personnel or that impersonal performance standards and impartial treatment of clients tend to occur together in bureaucracies. It is also important to know whatprocesses account for the association between these factors.
The contribution of case studies of bureaucracies is that they have intensively investigated these intervening processes, which Weber’s theory had neglected. Case studies consequently help to explicate the association between characteristics by showing, for example, that bureaucratic procedures and easy substitutability of personnel are related because new officials unfamiliar with informal practices usually must resort to formal procedures, or that impersonal criteria for evaluating performance promote impartiality by causing officials to be so concerned with achieving specific results that they set aside their personal feelings about clients. Despite these contributions of case studies to organizational theory, however, its foundations must be supplied by the comparative analysis of many organizations.
Need for comparative research
Systematic comparisons of many organizations are necessary not only to test the propositions of organization theory but also initially to establish the relationships the theory is designed to explain. The major limitation to case studies of organizations is not that they cannot demonstrate the validity of theoretical propositions; Weber’s analysis could not furnish such demonstrations either, since he lacked the empirical data required for this purpose. Their major limitation is rather that research in a single organization diverts attention from organizational analysis proper, and directs it instead to the analysis of roles and group processes in the organizational context. But for such intensive analysis of internal processes to make its potential contribution to organizational theory, it should be complemented by comparative studies that determine the concomitant variations of organizational characteristics. The beginning of a research trend in this direction can be observed, as exemplified by a series of studies on the relationship between the size of organizations, their complexity, and the relative size of their administrative apparatus (summarized in Blau & Scott 1962, pp. 225-227). The results of such comparative research promise to supply the empirical foundations needed to refine Weber’s conceptions and build a systematic theory of organization.
Peter M. Blau
Barnard, Chester I. (1938) 1962 The Functions of the Executive. Cambridge, Mass.: Harvard Univ. Press.
Blau, Peter M. (1955) 1963 The Dynamics of Bureaucracy: A Study of Interpersonal Relations in Two Government Agencies. Rev. ed. Univ. of Chicago Press.
Blau, Peter M.; and Scott, W. Richard 1962 Formal Organizations: A Comparative Approach. San Francisco: Chandler. → Contains an extensive bibliography.
Crozier, Michel (1963) 1964 The Bureaucratic Phenomenon. Univ. of Chicago Press. → First published as Le phenomene bureaucratique.
Duverger, Maurice (1951) 1962 Political Parties. 2d English ed., rev. New York: Wiley; London: Methuen.↑ First published in French.
Etzioni, Amitai 1961 A Comparative Analysis of Complex Organizations: On Power, Involvement, and Their Correlates. New York: Free Press.
Gouldner, Alvin W. 1954 Patterns of Industrial Bureaucracy. Glencoe, 111.: Free Press.
Gouldner, Alvin W. 1959 Organizational Analysis.Pages 400-428 in Robert K. Merton et al. (editors),Sociology Today. New York: Basic Books.
Gulick, Luther; and Urwick, Lyndall(editors) (1937)1954 Papers in the Science of Administration. New York: Columbia Univ., Institute of Public Administration.
Kahn, Robert L.; and Katz, Daniel(1953) 1960 Leadership Practices in Relation to Productivity and Morale. Pages 554-570 in Dorwin Cartwright and Alvin Zander (editors), Group Dynamics: Research and Theory. Evanston, 111.: Row, Peterson.
Krupp, Sherman 1961 Pattern in Organization Analysis: A Critical Examination. Philadelphia: Chilton.
Lipset, Seymour M.; Trow, Martin A.; and Coleman,James S. 1956 Union Democracy: The Internal Politics of the International Typographical Union.Glencoe, 111.: Free Press. → A paperback edition was published in 1962 by Doubleday.
March, James G.; and Simon, Herbert A. 1958 Organizations. New York: Wiley. -→ Contains an extensive bibliography.
Merton, Robert K. (1940) 1957 Bureaucratic Structure and Personality. Pages 195-206 in Robert K. Merton, Social Theory and Social Structure. Rev. ed. Glencoe, III.: Free Press.
Michels, Robert (1911) 1959 Political Parties: A Sociological Study of the Oligarchical Tendencies of Modern Democracy. New York: Dover. → First published as Zur Soziologie des Parteiwesens in der modernen Demokratie.
Parsons, Talcott 1960 Structure and Process in Modern Societies. Glencoe, III.: Free Press. → See pages 16-96 on the analysis of formal organizations.
Roethlisberger, Fritz J.; and Dlckson, Wllliam j.(1939) 1961 Management and the Worker: An Account of a Research Program Conducted by the Western Electric Company, Hawthorne Works, Chicago.Cambridge, Mass.: Harvard Univ. Press. →A paperback edition was published in 1964 by Wiley.
Selznick, Philip 1949 TVA and the Grass Roots: A Study in the Sociology of Formal Organization. University of California Publications in Culture and Society, Vol. 3. Berkeley: Univ. of California Press.
Simon Herbert (1947) 1957 Administrative beavior: A Study of Decision-making Processes in Administrative Organization. 2d ed. New York: Macmillan.
Taylor, Frederick W. 1911 The Principles of Scientific Management. New York: Harper.
Weber, Max (1922a) 1946 Bureaucracy. Pages 196 -244 in Max Weber, From Max Weber: Essays in Sociology. Translated and edited by H. H. Gerth and C. Wright Mills. New York: Oxford Univ. Press. → First published in German.
Weber, Max (1922b) 1957 The Theory of Social and Economic Organization. Edited by Talcott Parsons. Glencoe, 111.: Free Press. → First published as Part 1 of Wirtschaft und Gesellschaft.
In a modern society, where large organizations have acquired unprecedented importance, social scientists have increasingly sought to understand the nature of organizational goals—what they are, what shapes or determines them, what their impact is upon the organization and environment, and how they change. This concern with the properties of goals themselves is to be distinguished from the many organizational studies that utilize the concept of goal only to set the stage for inquiry into other matters. Thus, in studies of morale or productivity, from the explorations of Elton Mayo and his followers to the more sophisticated work of such contemporary social scientists as Rensis Likert (1961), we do not find the question of organizational goals treated as problematical. For example, it is generally taken for granted that the only managerial goal with which the researcher need concern himself is increased productivity, from which it follows that the principal research problem is to find out how subordinates can be made to conform to or share this goal. In this article, however, we shall be concerned with the work of authors who have made considered assertions about the properties of organizational goals.
Three recent traditions have contributed to the literature on goals. In the 1940s a number of economists questioned the basic assumption of microeconomic theory that the goal of the firm was to maximize profit or return on investment; the theory was criticized as overrational, and it was suggested that other goals might be at least as important as profit (see the review in Cyert & March 1963). The debate continues; but while some interesting concepts have emerged, they have remained largely untested by empirical research.
Another tradition has stemmed from those sociologists who tend to see formal organizations as complex institutions with evolving goals and a “character” of their own, rather than as rational instruments designed to pursue a single, rational goal. The most influential of these sociologists are Everett Hughes, Robert K. Merton, and Philip Selznick; all of them have acknowledged the importance of the observations of Robert Michels (1911) on the evolving character of European political and trade-union movements. The body of theory that originated with these authors is sometimes referred to as “institutional” sociology.
The teaching and research of the institutional sociologists have provided the stimulus for a great many studies of such “people-changing organizations” as mental hospitals and prisons, both of which, since they are concerned with treatment as well as custody, are obvious examples of organizations pursuing two goals at once. Since the implicit aim of many of these studies has been to change the goals of the organization, this tradition has paid some attention to clarifying and operationalizing the concept of goal (see Zald 1963; Perrow 1961a; Cressey 1961; Conference Group 1960).
A variable concept
None of the three traditions in the literature on goals has succeeded in developing rigorous or even very self-conscious conceptual tools. In part this is because almost any sequence of behavior can be cut into small or large pieces, each of which can be said to be goaldirected. What is designated as a goal, as well as what is used as evidence for the existence of a goal, depends for the most part on the purpose of the study being made.
Since there is only a relative distinction between means and ends and since, therefore, any end or goal can be seen as a means to another goal, one is free to enter the “hierarchy of means and ends” (Simon  1961, pp. 63-66) at any point. Moreover, a goal may be said to exist on the basis of the statements of officials, questionnaire data from employees, observations of behavior, or inferences about contributions to a larger social system. Accordingly, there is a very wide range of organizational behavior that can be interpreted as goaldirected, and the scope of general statements about the effect of goals on an organizational structure, changes in goals, goal displacement, and so forth, will inevitably vary with the main interest of the researcher.
Thus, to say, as Talcott Parsons does (1960, p. 17), that an organization is defined as pursuing a specific goal is perhaps as true as the much more frequent observation that organizations have multiple goals. Indeed, much of the dispute about the role of profit in economic organizations hinges on differing conceptions of goals. For those concerned with macroeconomic theory (the theory of the market place), the assumption that profit is the only important goal is quite useful; for those concerned with studying individual firms or sectors of the economy, it may be quite misleading. When the National Foundation for Infantile Paralysis shifted from an exclusive concern with polio to a concern for a variety of diseases, it was, from one point of view, a case of “goal succession” (Sills 1957, pp. 253-265), but not from other perspectives, such as the view that the goal is research on disease, or conserving the organization, or even (following Michels) maintaining positions of prestige and power for the top executives. In these views, only the means had changed.
Categories of goals
To provide some order in this subject area and to illustrate the work on goals that has been done, we shall distinguish six categories of goals, recognizing that the number could be smaller or larger. Three have external referents—society, the public in contact with the organization, the investors—and will be referred to, respectively, as societalgoals, output goals, and investor goals. The other three have internal referents, that is, to the organization and its members. They are system goals (survival, growth, etc.), -product goals (the defining characteristics of the product such as quality, availability, styling), and the somewhat residual category of derived goals (those which make use of the power the organization generates in the pursuit of the other goals).
The relationships between these categories may, in some cases, be those of means to ends, but not necessarily. To assume a means-end hierarchy is to assume that behavior is completely integrated and that all behavior is therefore functionally indispensable. By contrast, we assume, not that all things contribute to a single ultimate end, but that organizations may pursue a variety of goals, that these goals may be incompatible or in competition with each other, and that an organization may attend to its different goals in sequence rather than all at once.
Goals as societal functions
At the most abstract level the goal of an organization (or more usually, a type of organization) is expressed in terms of its function for society. One such scheme, advanced by Talcott Parsons and others (1953, pp. 183-186), posits four functions that must be fulfilled in any social system [see Systems Analysis,article onSocial Systems]. This scheme enables one to classify organizations according to their characteristic functions. For instance, a steel firm is in the “goal-attainment” sector, presumably concerned with material production, while a correctional institution is in the “integrative” sector, concerned with integrating deviants into society. A similar classification is that of Etzioni (1961), who posits economic, order, and cultural goals for organizations. Little analysis of concrete organization can be done at this level; but these typologies have the merit of causing us to ask how organizations are legitimated by society and how changes in a society’s values will affect the growth of types of organizations. Such questions can lead to analysis of broad changes in organizational goals; for instance, the development of a religious sect into a church can be seen as a case of an organization’s transition from an “integrative” function to a “latency” or “pattern-maintenance” function (Clark 1948). Similarly, an analysis of the British East India Company might focus upon the shift from trade and transport to political domination in the interests of British foreign policy. Organizations may even develop from fulfilling a single, basic societal function, such as the production and marketing of agricultural products, to providing for all the functional needs of a social system (Diamond 1958).
Goals as output
At a less abstract level it is convenient to distinguish organizations according to output or general type of product; the referent is the public constituency of the organization. Shifts in the application of resources are analyzed at this level, as when the emphasis in deviance control shifts from incarceration to probation or preventive work. Peter Blau and Richard Scott (1962, pp. 42-44) use an essentially output category of goals when they distinguish organizations in terms of the question, Cui bono? (Who benefits?). They distinguish four types of beneficiaries:membership (for example, the members of voluntary associations); owners (covering most business organizations); clients (such as the clients of a welfare agency); and society (a category filled almost exclusively by governments). This approach draws our attention to the central problems of each type of organization; thus, membership organizations have the problem of maintaining a democratic form of administration, and owner organizations are chiefly concerned with efficiency.
A number of studies focus mainly on organizational output and can thus be easily related to the schema outlined above. James Thompson and Frederick Bates (1957; compare Thompson & McEwen 1958; Pittsburgh, University of, 1959) examine how output goals are affected by the technology employed: thus mining organizations are tied to a static output, while hospitals have adaptable technologies and can shift or multiply their outputs (for example, by branching out into research or education). However, most case studies of organizations tend to range over several categories of goals. In Selznick’s study (1949) of the Tennessee Valley Authority (TVA) it appears that the output goals of that organization were flood control and fertilizer production; with this perspective, we can see how the need to maintain these outputs influenced the organization’s ability to pursue specific “product” goals, such as assistance to farmers and the development of recreation areas.
A more specific referent for goals is the return to investors. Whether investors provide capital, labor, or legitimation, at least some organizational efforts must be directed toward this source of support. In the case of business and industrial firms return on investment is the goal most frequently cited, even in periods when capital is generated internally, and stockholders therefore play a passive role. Profit alone is not a sufficient characterization, for it may be a long-run or shortrun profit, with differing consequences for the organization. More importantly, profit may be distributed to groups other than the investors, as when it is reinvested, or held for reasons of organizational security.
In the case of public welfare organizations the legislature or authority that provides funds and legitimation may simply be content to have the organization exist as evidence of some effort on its part and take no interest in the organization as long as there are no important complaints about the way funds are spent or about the adequacy of the effort; under such circumstances the organization is free to pursue other goals. The same is often true of economic organizations. In private welfare organizations investors may demand that the organization engage in activities that bring public visibility or confer prestige upon the donors. Such demands have manifold consequences for the organization (Perrow 196l→). In Selznick’s TVA study it is clear that investor goals included the promotion of public power, thus placing limits on the types of product goals that could be pursued. For example, support on the public power issue had to be bought by channeling help and resources to the relatively well-to-do farmers rather than to the needy ones. Another study, by Burton Clark (1960), shows how investor goals constituted the “administrative web” that seriously weakened a junior college’s efforts to cope with the educational demands of its students.
Turning to internal referents, the system goal is concerned with the goals established by key executives (and perhaps investors, if they take an active role) regarding the size, growth, stability, market share, domain, and so forth, of the organization. These are characteristics of the collectivity as a whole, independent of other referents, such as output or social function. The common assertion that the primary task of an organization is to survive refers to this goal category. The concept of organizational equilibrium (Barnard 1938; compare Simon 1947), where inputs balance outputs, is also based on a survival goal. While many propositions at this level are highly tautological and tell us little, this goal category is useful. To know what an organization is willing to pay for survival, security, or growth, or what it considers to be the domain upon which others shall not encroach can be crucial for understanding its behavior. For instance, some economic organizations are generally considered to be highly security conscious, and others are risk takers; likewise, some business organizations appear to be more concerned with growth than with profits, however the latter are distributed. William Baumol (1959) notes that an increase in sales is a frequent goal of business organizations even at the expense of profits, while the du Pont corporation is fond of announcing that it prefers a high return to stockholders over sales growth. Herrymon Maurer (1955, pp. 35-39) mentions that in the early history of the United States there was a point beyond which some prominent owners would not expand, both because a larger business would take too much of their time and because their businesses already brought in sufficient return. Attitudes toward growth and size are examined at length in David Granick’s report (1962) on European businessmen; goals at this level in Europe, it would appear, differ markedly from those in the United States. This may be because of differences in the way in which elites are selected, as well as the predominance in Europe of closely held firms.
The behavior of voluntary associations can often be explained in terms of survival goals. For instance, the leaders of the Townsend organization, a group promoting the interests of older people, were apparently willing to go along with a change in output goals (from political action to recreation) rather than insist upon the ideological purity which would have further weakened what was in any case a dying organization (Messinger 1955). The Woman’s Christian Temperance Union did not choose to reactivate broader welfare goals when their brief prohibition victory turned into a rout; thus the organization appears to be dying because growth, or even survival, is less important than ideological purity (Gusfield 1955; 1957). These examples certainly suggest that the system category of goals should figure heavily in studies of organizational behavior; however, although excellent work has been done in a variety of organizational settings, little systematic attention has been paid to the topic.
Goals as product characteristics
Perhaps the most fruitful goal category for the study of specific organizations is that of product characteristics. Here the referent extends from key executives to the leaders of such functional groups as sales, production, engineering in economic organizations, or the clinical staff, child supervisors, and work staff in correctional institutions. Product goals are a subcategory of output goals; they are most useful for comparisons between similar organizations. Such dissimilar organizations as a steel firm and a correctional institution, for instance, have output goals that admit comparison only in the most general terms; the goals and associated behavior of two or more steel firms, however, can be compared in terms of actual product characteristics.
The product goal includes such things as the type, quality, quantity, cost, styling, and availability of the goods or services produced. Thus, a recent study of several juvenile institutions (Street et al. 1966) contrasts the consequences for each organization of adherence to such different product goals as bringing about extensive changes in character and personality, inculcating obedience, or merely deterring, punishing, or containing inmates. Economic organizations have not been systematically studied in these terms, but casual evidence from a variety of sources suggests that they differ widely in their “character” because of different commitments to different types of products. This category of goals is also heavily emphasized by Philip Selznick (1957), who gives examples from business and military organization.
“Succession of goals,” a phenomenon identified by Robert K. Merton and intensively discussed in a study by David L. Sills (1957, pp. 253-265), generally involves changes in product characteristics rather than in categories of output. Evidence of this may be found in a study of the Young Men’s Christian Association, which distinguishes various points at which the organization took up new goals while maintaining the same general “output” of serving individuals (Zald&Denton 1963). Changes in product characteristics are also discussed in an analysis of a general hospital by Charles Perrow (1963): as control of the organization, in response to changes in technology and consumer demand, passed from trustees to doctors, and then from doctors to administrators, the characteristics of the product goals also changed, together with investor goals and derived goals. Thus, although the output goals (as measured by official statements) changed only gradually, the “operative” goals (measured in terms of effective policies) showed several very pronounced changes, including a shift from emphasis on serving indigent Jews to one on serving middle-class Jews and gentiles, an increase in research as opposed to teaching, and the development of programs that competed with non-Jewish hospitals, whereas previously such programs had been avoided.
In the course of producing goods and services and meeting such threats to survival as competitive products, dissatisfied investors, or changes in demands of the consuming public, an organization generates a good deal of power that is not directly related to these requirements. It has the power to hire and fire, relocate its plant, develop or enfeeble its employees, invest in times of imminent recession, donate to educational or other community purposes, support ideological positions, and so on. Sometimes these efforts are minor or trivial and have few apparent consequences for the organization or the environment. However, they can also represent goals that are essential for an understanding of the organization and its impact. We shall call them derivedgoals, since the ability to pursue them is derived from the existence and behavior of the organization but is not considered essential to its conduct. Thus, a firm may sacrifice dividends, executive salaries, sales, stability, and growth because it engages in a protracted effort to avoid unionization, even though similar firms do not. Some firms may distribute profits to employees in the form of benefits far beyond those which similar firms find necessary. One firm may hire minority groups, while its equally successful competitor does not. Indeed, the whole question of the “social goals” of business organizations, evident since the industrial revolution but receiving increasing attention today, is primarily a question of derived goals: How is the power the organization generates to be used or controlled? Concern ranges from the psychological development of employees to impact on foreign policy. For instance, Chris Argyris argues that a great service the industrial world can offer to our society at large is to develop “fully functioning human beings who aspire to excellence” (1962, p. 5). Organizations have the power to enrich the lives of their members in other senses too; Michels and many other commentators (see, for instance, Lipset 1960, chapter 12) have noted the ability of organizations to raise the status of their leaders and offer them perquisites that are hard to abandon. Pursuit of status and financial rewards, derived from the existence of the enterprise, has led, it is asserted, to an innate conservatism that causes deflection of the original goals, because the organization’s ability to pursue more precarious ends, such as political or economic militancy, is thereby weakened.
Studies of the impact of informal social relations upon organizations generally provide examples of derived goals. Pursuit of derived goals may be the occasion for goal succession, so that what was once incidental to the aim of the organization becomes its product goal. Thus, for members of the Townsend organization the derived goal of fellowship supplanted the product goal of political action (Messinger 1955).
The phenomenon of goal displacement, probably the most widely noted characteristic of organizational behavior (Sills 1957, pp. 62-69), is often a case of pursuing derived goals. Essentially it is asserted that adherence to prescribed means on the part of employees may interfere with goal achievement; the means become ends in themselves. Indeed, many of the examples in the literature cited by Sills are not concerned with goals at all, but result from inflexible means that should have been changed to meet new situations, or from inefficient or inappropriate means that did not contribute to the goal of the organization, or from the application of an unrealistic performance standard on the part of the researcher (for which, see Etzioni I960; Gouldner 1955). Furthermore, many organizations would fall into disorder if employees were free to discard rules and regulations and themselves attempt to interpret, from their own truncated perspective, what the goals of the organization are and how they should be met.
Nevertheless, subgroup or personal goals, which are derived from participation in the organization but are not essential for the achievement of goals other than derived goals, can certainly contribute to goal displacement and thus alter the character of the organization. As William Starbuck (1965, pp. 475-476) points out, in young organizations commitment to goals rather than to tasks or social relations is likely to be high; but with time, satisfactions are afforded simply by participating in the social relations that grow up. Commitment to goals and concern about changes in goals tend to diminish, together with concern about the manipulation of organizational resources that displace goals.
Goals and interest groups
These two levels—product goals and derived goals—require an analysis of the organization as a coalition or set of interacting interest groups, rather than as an integrated entity. In fact, most discussions of conflict over goals view the organization in these terms and utilize the product category or the derived category. One statement of this view focuses upon shifting coalitions of groups with parochial interests and notes that each may have a “file drawer” of goals from which it selects particular ones at particular times (Cyert & March 1963, pp. 26-43). These goals may, of course, be attended to sequentially and therefore need not be integrated. The same authors (1963, pp. 36-38) propose the notion of “organizational slack”—what the economist would term “economic rent”—which is a surplus of resources that encourages the pursuit of a variety of goals, a pursuit that would be curtailed in less prosperous times.
The development of group interests within the organization is generally taken to be a negative factor in much of the human-relations literature, since the interests are felt to be uncontrolled by top management (for example, see the critical review by Krupp 1961, especially chapter 3). However, it has been pointed out that groups such as sales, auditing, personnel, or production represent latent sources of energy and commitment and are expected to put forth their views forcefully and protect the integrity of their operation; if they fail in this responsibility, the organization loses the benefits of their perspective (Selznick 1957). Acknowledged conflict over product characteristics (“Is reliability more important than styling?”) or over derived goals (“What is our responsibility to the community?”) openly serves to determine “the fundamental prize” in organizations—the nature of the organizational character (Selznick 1949, p. 181).
Goal analysis—a continuing problem
As this discussion of goal categories has suggested, the sources of goals and the consequences of goals, both for organizations and for society, are manifold. Broad social changes set the stage for technological developments, which in turn determine, within broad limits, the range of goals possible in types of organizations (Perrow 1965). Changes in one category of goals can produce changes in others, as when derived goals become denned as essential, or investor goals require shifts in product goals. More specific causes abound, such as changes in the source of recruitment of personnel (Selznick 1957, pp. 47-111), government regulation (Keller 1963), or pressures for forward integration (e.g., distributing and marketing functions added to productive functions, as discussed in Chandler 1962). The variety of sources asserted to generate or change goals reflects, in part, the variety of goal categories utilized. But it also reflects the essentially dependent nature of goals. Organizations are not born with a fixed structure or a stable internal guidance system directing them to some precise end; they are subject to countless internal and external forces, and both their competencies and liabilities change with a changing environment. Nor do these changes occur at a uniform rate; societal and output goals may change at a glacial pace, but the others may often fluctuate with the seasons.
If goals are essentially dependent phenomena, they may be no less dependent than other elements of organizations, such as task structure and social structure. Technology and the structure of tasks necessary to produce changes in the raw material may alter goals or be altered by goals. The social structure, or systems of relationships that surround the work effort, may also influence as well as be influenced by goals. This perceived fluidity of organizations, coupled with, or abetted by, our crude and ad hoc conceptual tools and the relativity of means and ends, poses a nearly insuperable problem for goal analysis. The serious student of organizational goals finds the matter so complex and categories and concepts so interdependent that there is no certainty about what should be labeled a goal, where it comes from, how it changes, and what impact it has. Nevertheless, the best work done in attempting to understand complex organizations has had to confront the question, To what variety of ends is organizational behavior patterned and motivated? The future theories of organization will inevitably founder unless they do the same.
Some reference to organizational goals is included in almost every important work on organizations. The following list indicates studies with more or less explicit attention to goals for a variety of types of organizations. Governmental agencies: Selznick 1949, 1957; Simon 1947; Merton 1940. Communist party: Selznick 1952. Military: Janowitz 1960. Corrections and prisons: Zald 1960, 1963; Street et al. 1966; Conference Group on Correctional Organization 1960; Cressey 1961. Hospitals: Perrow 1961b, 1965; Freidson 1963. Unions: Lipset 1960. Business and industry:Granick 1960, 1962; Cyert & March 1963; Dent 1959.Schools: Clark 1960. Voluntary associations: Sills 1957; Zald & Denton 1963; Clark 1948; Messinger 1955; Grusky 1959. Sills 1957 also has an extensive review of the literature on goal succession and goal displacement. The goals of individuals in organizations, or a psychological view of goals, are represented in Simon 1964 and Katona 1951(chapter 9). A representative discussion of “social goals” for business organizations is found in three collections: Mason 1960; Anshen & Bach 1960; and Cheit 1964. March 1965contains some discussions of goals in chapters dealing with particular types of organizations and in chapters on general topics. Eisenstadt 1958 provides a good international bibliography, with the focus on bureaucratization.
Anshen, Melvin L.; and Bach, G. L. (editors) 1960 Management and Corporations, 1985. New York: Mcgraw-Hill.
Argyris, Chris 1962 Interpersonal Competence and Organizational Effectiveness. Homewood, III.: Dorsey.
Barnard, chester I. (1938) 1962 The Functions of the Executive. Cambridge, Mass.: Harvard Univ. Press.
Baumol, William J. 1959 Business Behavior, Value and Growth. New York: Macmillan.
Blau, Peter M.; and Scott, W. Richard 1962 Formal
Organizations: A Comparative Approach. San Francisco: Chandler. → Contains an extensive bibliography. Chandler, Alfred D. JR. 1962 Strategy and Structure: Chapters in the History of the Industrial Enterprise.Cambridge, Mass.: M. I. T. Press.
Cheit, Earl F. (editor) 1964 The Business Establishment. New York: Wiley.
Clark, burton R. 1960 The Open Door College. New York: McGraw-Hill.
Clark, Samuel D. 1948 Church and Sect in Canada. Univ. of Toronto Press.
Conference Group On Correctional Organization 1960 Theoretical Studies in Social Organization of the Prison, by Richard A. Cloward et al. New York: Social Science Research Council.
Cressey, Donald R. (editor) 1961 The Prison: Studies in Institutional Organization and Change. With contributions by Johan Galtung and others. New York: Holt.
Cyert, Richard M.; and March, James G. 1963 A Behavioral Theory of the Firm. Englewood Cliffs, N.J.: Prentice-Hall.
Dent, James K. 1959 Organizational Correlates of the Goals of Business Management. Personnel Psychology12:365-393.
Diamond, Sigmund 1958 From Organization to Society: Virginia in the Seventeenth Century. American Journal of Sociology 63:457-475.
Eisenstadt, Shmuel N. 1958 Bureaucracy and Bureaucratization: A Trend Report and Bibliography. Current Sociology 7, no. 2.
Etzioni, Amitai 1960 Two Approaches to Organizational Analysis: A Critique and a Suggestion. Administrative Science Quarterly 5:257-278.
Etzioni, Amitai 1961 A Comparative Analysis of Complex Organizations: On Power, Involvement, and Their Correlates. New York: Free Press.
Freidson, Eliot (editor) 1963 The Hospital in Modern Society. New York: Free Press.
Gouldner, Alvin W. 1955 Metaphysical Pathos and the Theory of Bureaucracy. American Political Science Review 49:496-507.
Granick, David 1960 The Red Executive: A Study of the Organization Man in Russian Industry. Garden City, N.Y.: Doubleday. → A paperback edition was published in 1961 by Doubleday.
Granick, David 1962 The European Executive. Garden City, N.Y.: Doubleday.
Grusky, Oscar 1959 Organizational Goals and the Behavior of Informal Leaders. American Journal of Sociology 65:59-67; 302-304.
Gusfield, Joseph R. 1955 Social Structure and Moral Reform: A Study of the Woman’s Christian Temperance Union. American Journal of Sociology 61:221232.
Gusfield, Joseph R. 1957 The Problem of Generations in an Organizational Structure. Social Forces 35:323330.
Janowitz, Morris 1960 The Professional Soldier: A Social and Political Portrait. Glencoe, 111.: Free Press.
Katona, George 1951 Psychological Analysis of Economic Behavior. New York: McGrawHill. → See especially Chapter 9. A paperback edition was published in 1963 by McGraw-Hill.
Keller, Morton 1963 The Life Insurance Enterprise, 1 885-1910: A Study in the Limits of Corporate Power.Cambridge, Mass.: Harvard Univ. Press.
Krupp, Sherman 1961 Pattern in Organization Analysis: A Critical Examination. Philadelphia: Chilton.
Likert, Rensis 1961 New Patterns of Management. New York: McGraw-Hill.
Lipset, Seymour M. 1960 Political Man: The Social Bases of Politics. New York: Doubleday.
March, James (editor) 1965 Handbook of Organizations. Chicago: Rand McNally.
Mason, Edward S. (editor) 1960 The Corporation in Modern Society. Cambridge, Mass.: Harvard Univ. Press.
Maurer, Herrymon 1955 Great Enterprise: Growth and Behavior of the Big Corporation. New York: Macmillan.
Merton, Robert K. (1940) 1957 Bureaucratic Structure and Personality. Pages 195-206 in Robert K. Merton, Social Theory and Social Structure. Rev. ed. Glencoe, 111.: Free Press. → First published in Social Forces, Volume 18, pages 560-568.
Messinger, Sheldon L. 1955 Organizational Transformation: A Case Study of a Declining Social Movement.American Sociological Review 20:3-10.
Michels, Robert (1911) 1959 Political Parties: A Sociological Study of the Oligarchical Tendencies of Modern Democracy. New York: Dover. → First published as Zur Soziologie des Parteiwesens in der modernen Demokratie.
Parsons, Talcott 1960 Structure and Process in Modern Societies. Glencoe, 111.: Free Press. → See pages 16-96 on the analysis of formal organizations.
Parsons, Talcott; Bales, Robert F.; and Shils, Edward A. 1953 Working Papers in the Theory of Action.Glencoe, 111.: Free Press.
Perrow, Charles 1961a The Analysis of Goals in Complex Organizations. American Sociological Review 26: 854-866.
Perrow, Charles 1961b Organizational Prestige: Some Functions and Dysfunctions. American Journal of Sociology 66:335-341.
Perrow, Charles 1963 Goals and Power Structures: A Historical Case Study. Pages 112-146 in Eliot Freidson (editor), The Hospital in Modern Society. New York: Free Press.
Perrow, Charles 1965 Hospitals: Technology, Structure and Goals. Pages 910-971 in James March (editor), Handbook of Organizations. Chicago: Rand McNally.
Pittsburgh, University OF, Administrative Science Center 1959 Comparative Studies in Administration. Univ. of Pittsburgh Press.
Selznick, Philip 1949 TVA and the Grass Roots: A Study in the Sociology of Formal Organization. University of California Publications in Culture and Society, Vol. 3. Berkeley: Univ. of California Press.
Selznick, Philip (1952) 1960 The Organizational Weapon: A Study of Bolshevik Strategy and Tactics.Glencoe, 111.: Free Press.
Selznick, Philip 1957 Leadership in Administration: A Sociological Interpretation. Evanston, 111.: Row, Peterson.
Sills, David L. 1957 The Volunteers: Means and Ends in a National Organization. Glencoe, 111.: Free Press.
Simon, Herbert A. (1947) 1961 Administrative Behavior: A Study of Decision-making Processes in Admillan.
Simon, Herbert A. 1964 On the Concept of Organizational Goal. Administrative Science Quarterly 9:1-22.
Starbuck, William 1965 Organizational Growth and Development. Pages 451-533 in James March (editor),Handbook of Organizations. Chicago: Rand McNally.
Street, David; Vinter, Robert D.; and Perrow, Charles 1966 Organization for Treatment. New York: Free Press.
Thompson, James D.; and Bates, Frederick L. 1957 Technology, Organization, and Administration. Administrative Science Quarterly 2:325-343.
Thompson, James D.; and Mcewen, William J. 1958 Organizational Goals and Environment: Goal-setting as an Interaction Process. American Sociological Review 23:23-31.
Zald, Mayer N. 1960 The Correctional Institution for Juvenile Offenders: An Analysis of Organizational “Character.” Social Problems 8:57-67.
Zald, Mayer N. 1963 Comparative Analysis and Measurement of Organizational Goals: The Case of Correctional Institutions for Delinquents. Sociological Quarterly 4:206-230.
Zald, Mayer N.; and Denton, Patricia 1963 From Evangelism to General Service: The Transformation of the YMCA. Administrative Science Quarterly 8:214234.
Concepts of organizational effectiveness typically focus on the degree to which the organization accomplishes its objectives. A business firm is effective if it makes a profit, a welfare agency if it provides professional service of high quality, the Department of State if it achieves its substantive goals related to foreign policy. This concept has led scholars like Blau and Scott (1962) to suggest that organizational effectiveness be denned in terms of “who benefits.” Thus, a business organization is effective if it reduces costs (because the owners have profit as their objective). Unions are effective if they have strong rank-and-file participation (because the union is created to serve the needs of the membership).
The limitations of this view can be seen when we realize that presently there are more large business firms than trade unions concerned about increasing rank-and-file participation and designing work to be more meaningful (which increases costs). Trade unions, on the other hand, are becoming more conscious of cost and efficiency, more worried about “marketing” procedures (Argyris 1964), and less concerned about their internal democracy. Recently, officials of the U.S. Department of State, after a searching study, concluded that one of the most significant factors contributing to organizational ineffectiveness is the subservience of the administrative activities to the substantive activities. Guest (1962) found that an organization which for years had focused primarily on making a profit, to the neglect of its human resources, had to pay more attention to the effectiveness of its internal human system. Argyris (1960) found that an organization which for years had overemphasized its “people-centeredness” had to reduce this emphasis or face the prospect of destroying itself.
These illustrations point up the fact that the moment one considers the “ongoing” property of organizations, one must add two dimensions to make the concept of organizational effectiveness more valid. They are, first, the organization’s capacity to maintain its internal system in working order so that it can solve problems effectively and, second, the relative cost of transforming the human energy inputs into useful organizational outputs. Organizational effectiveness, then, is the balanced or optimal emphasis upon achieving objectives, problem-solving competence, and human energy utilization. It may be possible, in the short run, to emphasize one of these activities at the expense of the others, but eventually all must be optimally expressed.
How should an organization be constructed if it is to be effective? Max Weber suggested creating a system characterized by specialization of effort, a clearly defined authority structure, formal position descriptions, and a sense of impersonality. “Scientific management” and early public administration theorists, plus military organizational specialists, would agree with Weber but would say it somewhat differently. They would emphasize such factors as work specialization, unity of command, unity of direction for any given unit, span of control, and employee loyalty.
The causes of dysfunctional activities
Toward the end of World War II a number of social scientists began to study organizations by observing the actual behavior of the participants. As expected, they did find people highly specialized (especially at the lower levels), reporting to one supervisor, and organized as members of a unit that had one primary function. But they saw more. They observed some people who reported to several superiors (Shartle 1956); some who bypassed their superiors (Arensberg & McGregor 1942); some who enlarged their work much beyond the prescribed specialization (Richardson & Walker 1948); and some who fought their work and their superiors (Dalton et al. 1946), the incentive system (Whyte et al. 1955), and managerial controls such as budgets (Argyris 1952). In other words, they found an informal system in addition to the formal one. Indeed, the more they studied the behavior in organizations, the more they became impressed with the scope and potency of the informal activities, especially the dysfunctional ones (Gouldner 1959).
What causes these dysfunctional activities? Why do people seem to fight the system? McGregor (1960) and Likert and his associates (1961) found some of the problems to be related to the frequent use by management of production-centered, dependency-producing, submission-oriented leadership styles. Whyte (1961) agreed substantially with this finding, although he conceptualized it as the initiation of action downward being much greater than the initiation of action upward.
Later, the thesis was developed that all the major components of organization—structure, leadership, managerial controls, incentive systems, and technology—tend to have a similar impact on the employees, that is, to place them in a dependent, submissive position where they can use few of their central abilities (Argyris 1964). It was predicted that this tendency should be more pronounced as one goes down the administrative hierarchy and as the demands of the job control the individual to a greater degree. The opposite should be the case as one rises in the hierarchy and as the individual has greater control over the job demands made upon him. (The existence of large stock options for high executives may blunt the first part of this proposition, because the executive may be willing to become dependent and submissive in order not to lose the lucrative rewards available to him at the end of the stipulated period.) From this point of view, the underlying causal factors of dysfunctionality are the basic managerial principles, or what some sociologists call the “organizational imperatives”: work specialization, chain of command, unity of direction, etc. The impact of such factors as technology, organizational structure, managerial controls, and leadership is seen as a manifestation of these principles. It is not true, therefore, as some have suggested, that this approach to organizations ignores the part played by technology; rather, it refuses to consider technology apart from management.
It is also important to point out that this approach is not limited to the lower levels of employees and thereby to a particular social class (Strauss 1963). It has been successfully applied to scientists in 11 university, industrial, and governmental research laboratories (Farris 1962), to a national sample of nearly two thousand managers ranging from presidents to first-level supervisors (Porter 1962), to a study of a thousand managers in five firms (Carnarius 1962), as well as to a national sample of lower-level employees (Gurin et al. 1960), two extensive reviews of the literature (Blauner I960; Argyris 1957), and a depth study of nearly five hundred employees at the lower levels (Kornhauser 1962)—to mention only a few of the published studies. Moreover, the relevance of social class is questioned by the recent work of Mann and Williams (1962). They found that employees whose social class remained constant while their work was altered to provide them with greater control and use of their abilities changed their attitudes about their work in the direction of desiring more challenge and more responsible work. The frequently cited example of Dubin’s results (1956) that employees do not place work as their central life interest is not a disconfirmation of this point of view; indeed, I would predict such results. It may be that such employee apathy is a reaction to the very conditions that are described above.
Nor does this point of view make predictions about employee happiness. Its only prediction is that as one goes down the organizational hierarchy, the employees should report less control over their work world. Even Blauner (1964), who found that most employees are satisfied, confirmed the fact that decreasing control over work is reported by employees as one goes down the hierarchy; this finding was also confirmed by Inkeles (1960) and Baldamus (1961) in studies of employees in other countries.
This viewpoint also predicts that the dysfunctional attitudes and informal activities that have been so thoroughly documented—apathy, absenteeism, trade unionization, decreasing emphasis upon human values and increasing emphasis upon material values (Erich Fromm’s “market orientation”),decreasing motivating power of wages, and increasing alienation—will tend to arise under two conditions. Such reactions will occur when individuals do not want to be primarily dependent, submissive, and confined to a position where they can use few of their abilities, but the organization requires them to be so; or when individuals do want to be dependent and submissive, and their organization requires that they use more of their abilities and take on more responsibility.
On the other hand, these informal activities should not tend to occur when individuals do want to have more control and responsibility and to use more of their abilities, and the organization permits it; or when individuals do not want to take on responsibilities and do not want to have control over their work, and the organizational context does not require it. (Apathy, indifference, and alienation may be exceptions, in the last case, because these qualities represent the employees’ needs.) If data can be produced to show that the informal activities listed above do occur under the latter conditions or do not occur under the former conditions, then this viewpoint would be in trouble.
Nor does this view make the assumption that all people want to “actualize” themselves, in the sense of expressing some higher-order needs. It merely states that if people want to actualize their higher-order needs and if their job requirements make them dependent and submissive and minimize their control, then certain difficulties will occur. However, if self-actualization is taken to mean the expression of any need (e.g., autonomyor dependence), then all that we are claiming is that whenever employees’ needs are frustrated, the consequences will be the ones described above. One would also predict that the greater the control, the responsibility, and the variety of individual experiences involved in the worker’s job, the less the frustration, dissatisfaction, and alienation he will experience; this proposition is supported, for example, by Blauner’s finding (1964) that the degree of alienation and dissatisfaction decreased as the job became more challenging and permitted greater control. In terms of understanding organizational effectiveness, the main point is that most of these informal activities (with the exception of trade unionism ) tend to inhibit problem-solving effectiveness and consume much human energy in nonproductive activities.
Assumptions and values of management
According to McGregor (1960), the principles of organization as described by Weber, the scientific management school, and the military theorists in reality imply certain assumptions about employees: (1) the average person has an inherent dislike for work and will try to avoid it; (2) because of their dislike for work, most people have to be coerced to put forth an adequate day’s work toward company goals; and (3) the average worker wants to be closely supervised and to avoid responsibility. Some of the protagonists of traditional organizational theory have questioned these assumptions.
Recently, in both a national sample of executives (Porter 1964) and an international study of 3,000 managers in 14 countries (Haire et al. 1963), the researchers found little belief in the capacity of others for initiative, individual action, and leadership. In all but three countries managers viewed the ordinary individual as preferring to be directed, wishing to avoid responsibility, and possessing little talent for leadership.
Employees feel these basic assumptions and, as is pointed out above, tend to adapt by behaving in an apathetic manner and seeking less responsibility; thus, the worker adapts by making the managers’ assumptions self-fulfilling prophecies. Management, to put it dramatically, tends to create a world that makes its own administrative strategy self-defeating.
Before we can turn to the question of how to change the situation, we need to look more closely at the upper levels of organization. My own suggestion is that the Weberian-scientific management concepts of organization imply three basic values of effective human relationships at the upper levels and that, to the extent these values are internalized, they will tend to create an executive subsystem that carries on many unintended and self-defeating dysfunctional activities.
The first basic value is that the important human relationships are those which are related to achieving the organization’s objective. One study found, for example, that in over 265 different types and sizes of meetings, the executives always tended to focus their behavior on “getting the job done” (Argyris 1966). Out of literally thousands of units of behavior, almost none were spent in analyzing and maintaining group effectiveness. This was true even though in many meetings the group’s effectiveness “bogged down” and the objectives were not being reached because of interpersonal factors. When the executives were interviewed and asked why they did not spend some time in examining the group operations or process, they replied that they were there to get a job done, and besides, “If the group isn’t effective, it is up to the leader to get it back on the track by directing it.”
The second value is the emphasis on cognitive rationality and the de-emphasis on the rationality—indeed, the existence—of feelings and emotions. This value influenced the executives to see cognitive, intellectual discussions as “relevant,” “good work,” and so on. The emotional and interpersonal discussions tended to be viewed as “irrelevant,” “immature,” “not work,” and so on. As a result, when emotions and interpersonal variables became blocks to group effectiveness, all the executives reported (and were observed doing so) that they should not deal with them. For example, if there was an emotional disagreement, they would tell the members “to get back to facts” or “keep personalities out of this.” In other words, the value that effectiveness was primarily a cognitive dimension tended to coerce the men to suppress and deny the emotional and interpersonal variables that influenced their group’s effectiveness.
The third value suggests that human relationships were most effectively influenced through unilateral direction, coercion, and control, as well as by rewards and penalties that sanctioned all three values. This value of direction and control was implicit in the chain of command and also in the elaborate managerial controls that had been developed within the organizations.
The impact of these values can be considerable. For example, to the extent that individuals dedicate themselves to the value of intellectual rationality and “getting the job done,” they will tend to be aware of and emphasize the cognitive, intellectual aspects of work and consciously or unconsciously to suppress the interpersonal and emotional aspects, especially those which do not seem relevant to achieving the task. As the interpersonal and emotional aspects of behavior become suppressed, organizational norms will tend to arise that coerce individuals to hide their feelings. Their interpersonal difficulties will either be suppressed or be disguised and brought up as cognitive, technical, intellectual problems. Under these conditions the individual may tend to find it very difficult to develop competence in dealing with feelings and interpersonal relationships. Also, the individual may build defenses, both psychological and organizational.
Another way to prevent individuals from violating the organizational values and from embarrassing one another is to block out, that is, refuse to consider consciously or unconsciously, ideas that, if explored, could expose suppressed feelings. Such a defensive reaction in the organization eventually inhibits the effectiveness of the search process involved in decision making. The participants tend to limit themselves to those ideas and values which are not threatening and do not violate organizational norms. Consequently, the members decrease their openness to new ideas and values. As the degree of openness decreases, the capacity to experiment also decreases, and the fear to take risks increases. As the fear to take risks increases, the probability of experimentation decreases. This, in turn, decreases the range or scope of openness, which, in turn, decreases risk taking. We have another closed circuit, which becomes an additional cause of organizational ineffectiveness.
Moreover, the same factors help to keep the true state of affairs from the awareness of those who could change the situation. This can result in a sense of resignation among the people below (i.e., “the top will never change”), which, in turn, inhibits full and open discussion during the more innovative decision-making sessions and consequently decreases internal commitment to the more innovative decisions, increases subordinate and superior “gamesmanship,” and increases executive blindness about their relationships with their subordinates (Argyris 1962).
Optimizing organizational effectiveness
Formal organizations may have built into their designs the seeds for many nonproductive, dysfunctional, energy-consuming activities at all levels which tend to result in increasing organizational rigidity, organizational defensiveness, and intergroup conflict, as well as less effective decisionmaking processes (especially at the upper level). In short, organizations have a built-in tendency toward ineffectiveness or disorganization (as is apparently the case in physical nature).
Can this tendency toward ineffectiveness be decreased and the tendency toward effectiveness strengthened? What changes would this require at the different levels of the hierarchy? Beginning with the lower levels, we need to redesign the technology, the managerial controls such as budgets, and the financial incentives, as well as the jobs themselves, in order to increase the degree to which the employees use their more important abilities, have control over their work, and develop feelings of essentiality toward it.
How may these objectives begin to be accomplished? First, jobs can be enlarged so that individuals have more responsibility and use more of their central abilities. In spite of the increased training and redesign costs, apparently the longrange financial returns can be attractive (Conant & Kilbridge 1965). Some very preliminary research suggests that feelings of essentiality can be effectively produced if the organizational unit is “optimally undermanned” (Barker 1962; Sweetland & Hay thorn 1961). For example, branch banks that were designed to be small models of the main office were found to be more productive, and employee commitment and cooperation in them was found to be much higher, as compared with the main office, partially because the branches had been optimally undermanned (Argyris 1954).
Gillespie (1948) has shown that employees can conduct their own work studies that lead to effective work design. Likert and Seashore (1963) have presented examples of how budgeting can be altered to provide employees more control over the budgets, which, in turn, can lead to decreased costs as well as to increased employee commitment to further cost reduction. Lawler (1966) has shown that pay can be an effective incentive if employees see a clear connection between good job performance and their pay. When pay is seen as a reward for good performance, it is a particularly potent incentive because it becomes a form of recognition and a mark of achievement, thereby satisfying several important needs. McKersie (1963) has suggested that by including greater employee participation in the setting of performance standards and by having standards that are simple and stable and provide quick rewards, the incentive plan of the future can create the perception that pay is a reward for good performance. Of course, employees who do not seek to use their abilities, who are not motivated by autonomy needs, and who do not make work a central life interest may be threatened by these changes (see Strauss 1963).
Turning to the upper levels of the organization, technology, controls, and structure become less important and interpersonal relationships among the executives become more important. Executives with power can alter the components of formal organization if they wish. However, they tend to have much more difficulty in altering their own behavior, especially the conformity and mistrust within their executive subsystem, as well as the less effective decision-making activities (Argyris 1966). According to my analysis, change at the upper levels should begin with changing the values and the behavior of the executives. Lewin and Grabbe (1945) have pointed out that changing values and behavior is not an easy process and that it should begin by “unfreezing” and re-educating the individual. Schein (1963), following Lewin, has proposed that effective change requires several elements : the involvement of the people in the mutual determination of goals, a reliance on publicly shared data, a voluntary mutual interaction of the client and the change agent, the freedom of the client and consultant to influence each other, and an emphasis on the importance of making informed decisions (Bennis 1963).
Methods of organizational change. How can some of the above changes be instituted in organizations? Four approaches have been developed, to date, and used in various combinations. Mann and Likert (1952) have attempted organizational change by feeding back the research results to the members at all levels. The individuals are encouraged to disagree with and modify the findings and to develop their own strategies for change based upon those parts of the diagnosis which they accept. Each group is free to ask for further data and to feed information to, and to try to influence, those above as well as those below them. Seashore and Bowers (1963) published a systematically documented study of this type of change process, which resulted in a significant improvement in employee satisfaction; machine waste and absentee rates also improved, but not significantly.
The second approach focuses on two-man and small-group relationships. Argyris (1965), Sofer (1961), and Rice (1965) have published illustrations of three relatively different variations of this approach. The basic thrust of Sofer and Rice is to rely on tensions released through anxiety reduction: that is, they try to help the executive become aware of his real anxieties and then help him to overcome them. Argyris focuses more on helping executives to increase their interpersonal competence, which, in turn, should decrease their defensiveness and increase the effectiveness of their problem-solving and decision-making activities. Rice is willing to focus on unconscious material and Sofer is less so, while Argyris explicitly attempts to stay away from such material.
The third approach is concerned with the reduction of intergroup rivalries and the management of conflict. Presently organizations are so structured that competing groups are created and the powerful overcome the weak in political warfare. However, organizations could be structured so that each group can maintain its identity and seek resolution of conflicts through functional relations. This means creating new organizational policies and developing executives who are capable of bringing the competing groups together as groups. Each group discusses the problem and seeks, in private, to agree on its perceptions and attitudes toward the other groups and toward itself. Then representatives of the groups talk together in the presence of other members of the organization. This joint meeting is followed by further private discussion within each group regarding the way the group is perceived by others. Thus the groups develop an understanding of the discrepancy between their views (Blake et al. 1965).
The fourth method for organizational change is the laboratory method of education, at the core of which is the T (for training)-group. Briefly, the T-group experiences are structured so that individuals are placed in a learning experience where they are in control of their learning and can focus upon their behavior in a relatively isolated but accepting “culture.” In this training group they can give and receive feedback about the impact of their own behavior as well as the behavior of others. Thus, it is hoped that they will develop a deeper awareness and acceptance of themselves and others (as well as of the dynamics of small groups) and, if the experience lasts long enough (or if they learn fast enough), increase their interpersonal competence (Schein & Bennis 1965).
Variants of the laboratory method of education have been used with encouraging results. For example, Blake and Mouton conducted a change program at all levels of an organization (Blake et al. 1964). L. B. Barnes and L. E. Greiner conducted an independent evaluation of this program (see Blake et al. 1964). Although the degree of methodological rigor possible was limited by the field situation, the authors provide quantitative and qualitative data to show significant increases in productivity, profits, and employee commitment to organizational effectiveness. Argyris (1965) reports a change in the values and behavior of a top management group as a result of a laboratory program; the change is known to have lasted for nearly a year (unfortunately, further follow-up data were not available). Harrison (1962) conducted an independent evaluation of the same program and reported similar but not conclusive results, while Argyris (1965) later found that the laboratory method altered both the values and the behavior of a board of directors. These behavioral and attitude changes, as reported in both quantitative and qualitative terms, were still evident after 16 months, at the time the study ended. Increases in organizational effectiveness were also observed, ranging from the development of new products through the change of top officers who were ineffective, to a decrease in interpersonal tension among the board members.
Evaluating strategies for change
In the mid-1960s a number of large organizational change programs are in process, each of which has a built-in research activity to study its effectiveness. For example, Herbert Shepard and his associates are involved in the organization change program of one of the largest space laboratories in the United States, and a separate research team is studying the change. Friedlander’s preliminary results (1966) show that this program is beginning to have important positive effects. The National Training Laboratories is conducting a two-year change program in a very large industrial research organization; again, a team of social scientists (half of whom are not members of that network) are studying the results. The coupling of systematic research with each of these programs should help us to evaluate more effectively the results of the changes as well as the differential effectiveness of the several change strategies. However, much more research is needed, especially studies of a longitudinal nature.
To date, all the organizational changes influenced by social scientists have been in the direction of creating more “organic” organizations. This approach has been variously called “participative group” (Likert), “problem solving” (Bennis), “open system” (Louis B. Barnes), “human relations” (Eugene Litwak), “Theory Y” (McGregor), and “flexible organization” (Rice). The organic organization is characterized by decentralization of decision making; an emphasis on mutual dependence and cooperation based on trust, confidence, and high technical or professional competence; a constant pressure to enlarge tasks and interrelate them so that the concern for the whole is emphasized; the decentralization of responsibility for, and use of, information, rewards, and penalties; the responsibility of participants at all levels for developing and maintaining loyalty and commitment at as high a level as possible; and an emphasis on status through contribution to the whole and through intergroup and interindividual cooperation. Such an organization assumes that people are capable of being responsible, committed, and productive and that they desire a world in which the rationality of feelings and interpersonal relationships is as valued as cognitive rationality.
Although I have been involved in helping some organizations become more organic, I do not believe that the organic organization is the final answer. I believe that organizational effectiveness can be optimized by using any one of at least four discrete organizational strategies or structures, each of which would have a set of decision rules guiding the administrators in their use. The first is the pyramidal structure; the second, a Likert-type “link-pin” structure; the third, a structure where people obtain power according to their functional contributions; and fourth, a structure where all members have equal power and may not abdicate their basic responsibilities.
In addition, I believe that there are four different types of leadership patterns, ranging from the traditional authoritarian leadership to a more “reality-centered” leadership (which Likert calls “ultural” and “supportive”). In reality-centered leadership the power, rewards and penalties, con ditions of membership, and control of information can reside with one leader, or they can be increasingly shared by all the members. Mann (1965) has identified three basic skills in reality-centered leadership: administrative, human relations, and technical skills. The importance of each of these skills varies with the level of supervision, the technology involved, the degree of professional work involved, and situational changes, such as a change to electronic data-processing equipment.
It is too early to evaluate the usefulness of these views in regard to the organization of the future. Much more research is needed. The only available “evidence” is from a few industrial firms that are experimenting in limited fashion. Brown (1960) has reported that profits and problem-solving effectiveness have increased in firms that have created such substructures as a representative system to represent employees’ views, a legislative system that allows every member to participate in formulating and implementing policy, and an executive system in which employees have significant influence on the present and future operations of the firm. Non-Linear Systems, an electronic manufacturing firm near Los Angeles, reported an increase of 30 per cent in gross income by giving employees more responsibility in designing the product, in enlarging their jobs, and in controlling most of the accounting process (Kuriloff 1966). Gomersall and Myers (1966) reported significant financial gains through decreased costs and increased productivity achieved by giving the employees greater control over the design of their work. Union Carbide and General Electric have been so favorably impressed with the results of the change processes outlined above that they have trained change agents and created departments for organizational diagnosis and change. Here again, studies of whole organizations over a long period of time are needed. Understandably, it is not easy to get executives to agree to permit behavioral scientists to experiment with the very core components of their organization. Perhaps, in addition to encouraging partial experiments in organizations, organizational simulation may provide a vehicle by which more of the totality and complexity can be studied. As the findings in this field become better validated, we can expect greater theoretical precision, methodological sophistication, and, for the practitioner, the probability of less risk and greater returns.
Arensberg, Conrad M.; and Mcgregor, Douglas 1942 Determination of Morale in an Industrial Company.Applied Anthropology 1, no. 2:12-34.
Argyris, Chris 1952 The Impact of Budgets on People. New York: Controllership Foundation.
Argyris, Chris 1954 Organization of a Bank: A Study of the Nature of Organization and the Fusion Process.New Haven: Yale Univ., Labor and Management Center.
Argyris, Chris 1957 Personality and Organization: The Conflict Between System and the Individual. New York: Harper.
Argyris, Chris 1960 Understanding Organizational Behavior. Homewood, 111.: Dorsey.
Argyris, Chris 1962 Interpersonal Competence and Organizational Effectiveness. Homewood, 111.: Dorsey. ARGYRIS, CHRIS 1964 Integrating the Individual and the Organization. New York: Wiley.
Argyris, Chris 1965 Organization and Innovation. Homewood, 111.: Irwin.
Argyris, Chris 1966 Interpersonal Barriers to Decision Making. Harvard Business Review 44, no. 2:8497.
Baldamus, W. 1961 Efficiency and Effort: An Analysis of Industrial Administration. London: Tavistock.
Barker, Roger G. 1962 Big School-Small School: Studies of the Effects of High School Size Upon the Behavior and Experiences of Students; Final Report.Lawrence: Univ. of Kansas, Midwest Psychological Field Station.
Bennis, Warren G. 1963 A New Role for the Behavioral Sciences: Effecting Organizational Change. Administrative Science Quarterly 8:125-165.
Blake, Robert R. et al. 1964 Breakthrough in Organization Development. Harvard Business Review 42, no. 6:133-155.
Blake, Robert R.; Shepard, Herbert A.; and Mouton, Jane S. 1965 Managing Inter group Conflict in Industry. Houston, Tex.: Gulf.
Blau, Peter M.; and Scott, W. Richard 1962 Formal Organizations: A Comparative Approach. San Francisco: Chandler.
Blauner, Robert 1960 Work Satisfaction and Industrial Trends in Modern Society. Pages 339-360 in Walter Galenson and Seymour M. Lipset (editors),Labor and Trade Unionism: An Interdisciplinary Reader. New York: Wiley.
Blauner, Robert 1964 Alienation and Freedom: The Factory Worker and His Industry. Univ. of Chicago Press.
Brown, Wilfred B. 1960 Exploration in Management. New York: Wiley.
Carnarius, Stanley 1962 Motivating Managers. Opinion Research Corporation, Vol. 20, No. 8, 464-S. New York: The Corporation.
Conant, Eaton H.; and Kilbridge, Maurice D. 1965 An Interdisciplinary Analysis of Job Enlargement: Technology, Costs, and Behavioral Implications. Industrial and Labor Relations Review 18:377-395.
Dalton, Melville; Collins, Orvis; and Roy, Donald 1946 Restriction of Output and Social Cleavage in Industry. Applied Anthropology 5, no. 3:1-14.
Dubin, Robert 1956 Industrial Workers’ Worlds: A Study of the “Central Life Interests” of Industrial Workers. Social Problems 3:131-142.
Farris, George F. 1962 Congruency of Scientists’ Motives With Their Organizations’ Provisions for Satisfying Them: Its Relationship to Motivation, Affecting Job Experiences, Styles of Work and Performance. Unpublished manuscript, Univ. of Michigan.
Friedlander, Frank 1966 Performance and Interactional Dimensions of Organizational Work Groups.Journal of Applied Psychology. 50:257-265.
Gillespie, James 1948 Free Expression in Industry: A Social-psychological Study of Work and Leisure. London: Pilot.
Gomersall, Earl R.; and Myers, M. scott 1966 Breakthrough in On-the-job Training. Harvard Business Review 44, no. 4:62-72.
Gouldner, Alvin W. 1959 Organizational Analysis. Pages 400-428 in Robert K. Merton et al. (editors),Sociology Today. New York: Basic Books.
Guest, Robert H. 1962 Organizational Change: The Effect of Successful Leadership. Homewood, III.: Dorsey.
Gurin, Gerald; Veroff, Joseph; and Feld, Sheila 1960 Americans View Their Mental Health: A Nationwide Interview Survey. Joint Commission on Mental Illness and Health, Monograph Series, No. 4. New York: Basic Books.
Haire, Mason; Ghiselli, Edwin E.; and Porter, Lyman W. 1963 An International Study of Management Attitudes and Democratic Leadership. Pages 101-104 in International Congress for Scientific Management, 13th, New York, 1963, Human Progress Through Better Management. New York: Council for International Progress in Management.
Harrison, Roger 1962 Impact of the Laboratory on Perceptions of Others by the Experimental Group. Pages 261-271 in Chris Argyris, Interpersonal Competence and Organizational Effectiveness. Homewood, 111.: Dorsey.
Inkeles, Alex 1960 Industrial Man: The Relation of Status to Experience, Perception, and Value. American Journal of Sociology 66:1-31.
Kornhauser, Arthur 1962 Toward an Assessment of the Mental Health of Factory Workers: A Detroit Study. Human Organization 21:43-46.
Kuriloff, Arthur 1966 Reality in Management. New York: McGraw-Hill.
Lawler, Edward E. 1966 The Mythology of Management Compensation. California Management Review9, no. 1:11-22.
Lewin, Kurt; and Grabbe, Paul (1945) 1948 Conduct, Knowledge, and Acceptance of New Values. Pages 56-68 in Kurt Lewin, Resolving Social Conflicts: Selected Papers on Group Dynamics. New York: Harper. → First published in Volume 1 of the Journal of Social Issues.
Likert, Rensis 1961 New Patterns of Management. New York: McGraw-Hill.
Likert, Rensis; and Seashore, Stanley E. 1963 Making Cost Control Work. Harvard Business Review41, no. 6:96-108.
Mcgregor, Douglas 1960 The Human Side of Enterprise. New York: McGraw-Hill.
Mckersie, Robert B. 1963 Wage Payment Methods of the Future. British Journal of Industrial Relations1.191-212.
Mann, Floyd 1965 Toward an Understanding of the Leadership Role in Formal Organization. Pages 68 -102 in Leadership and Productivity: Some Facts of Industrial Life, by Robert Dubin et al. San Francisco: Chandler.
Mann, Floyd; and Likert, Rensis (1952) 1960 The Need for Research on the Communication of Research Results. Pages 57-66 in Richard N. Adams and Jack J. Preiss (editors), Human Organization Research.Homewood, 111.: Dorsey. → First published in Volume 11 of Human Organization.
Mann, Floyd; and Williams, Lawrencex R. 1962 Some Effects of the Changing Work Environment in the Office. Journal of Social Issues 18, no. 3:90-101.
Porter, Lyman W. 1962 Job Attitudes in Management: I. Perceived Deiiciencies in Need Fulfillment as a Function of Job Level. Journal of Applied Psychology46:375-384.
Porter, Lyman W. 1964 Organizational Patterns of Managerial Job Attitudes. New York: American Foundation for Management Research.
Rice, Albert K. 1965 Learning for Leadership: Interpersonal and Intergroup Relations. London: Tavistock.
Richardson, Frederick L. W. JR.; and Walker, Charles R. 1948 Human Relations in an Expanding Company: A Study of the Manufacturing Departments in the Endicott Plant of the International Business Machines Corporation. New Haven: Yale Univ., Labor and Management Center.
Schein, Edgar H. 1963 Forces Which Undermine Management Development. California Management Review 5, no. 4:23-34.
Schein, Edgar H.; and Bennis, Warren G. 1965 Personal and Organizational Change Through Group Methods: The Laboratory Approach. New York: Wiley.
Seashore, Stanley E.; and Bowers, David G. 1963 Changing the Structure and Functioning of an Organization: Report of a Field Experiment. Institute for Social Research, Survey Research Center, Monograph No. 33. Ann Arbor: Univ. of Michigan.
Shartle, Carroll L. 1956 Executive Performance and Leadership. Englewood Cliffs, N.J.: Prentice-Hall.
Sofer, Cyril 1961 The Organization From Within: A Comparative Study of Social Institutions Based on a Sociotherapeutic Approach. Chicago: Quadrangle
Strauss, George 1963 Some Notes on Power-equalization. Pages 34-84 in Harold J. Leavitt (editor), The Social Science of Organizations: Four Perspectives.Englewood Cliffs, N.J.: Prentice-Hall.
Sweetland, Anders; and Haythorn, William W. 1961 An Analysis of the Decision-making Functions of a Simulated Air Defense Direction Center. Behavioral Science 6:105-116.
Whyte, William F. 1961 Men at Work. Homewood, III.: Dorsey.
Whyte, William F. et al. 1955 Money and Motivation: An Analysis of Incentives in Industry. New York: Harper.
To gather, process, interpret, and communicate the technical and political information needed in decision making is to fulfill the intelligence function. The staff and consultants involved are typically men of specialized knowledge and training, who serve as experts. Many of them are, or once were, “intellectuals.” This article draws on a military analogy and brings together the scattered literature on the intelligence function in complex organizations in order to offer hypotheses about the determinants of the uses of intelligence and about the structural and ideological roots of intelligence failures.
Increased availability of intelligence has been assured by the so-called knowledge explosion. This may be seen in the exponential expansion of scientific knowledge, knowledge-producing professions and industries, research and development units, information services and machines, and the media of communication themselves. It is also evident in the growing emphasis on education, not just in the school but in the home, on the job, in the church, and in the armed forces (Machlup 1962). It is symbolized in the spectacular spread of electronic computers and automatic control systems. In industry and politics, as in the armed forces and the agencies of national security, the greater ease of obtaining and processing information has induced more information consciousness among executives, who are now more willing to use both overt sources of intelligence, such as government data and private surveys, and covert sources, such as electronic snoopers.
Information is more available; so is manpower. The swift growth of the World War II OSS, of its successor, the CIA, of the FBI, of the intelligence units of both the armed forces and the police, of research in every agency of government—a trend common to every rich country—has yielded a pool of potential manpower for industry, labor, political parties, indeed, for every type of organization. Many intelligence specialists in such agencies find their skills transferable; a free labor market insures their use. Another growing source of manpower, tapped mainly for covert intelligence, is maintenance technicians and engineers, men preferred as undercover operators in industry because they can move about unnoticed.
In short, the opportunity to gather, process, transfer, and deliver information has expanded; the relevant manpower is available; and there has been a corresponding increase of information consciousness among officials and executives.
Determinants of the use of intelligence
Organizations vary a great deal in the extent to which they use staff services; the types of experts they use also vary. To explore these variations, I shall delineate types of experts found in all complex organizations (Wilensky 1956, pp. 33-108) and relate their incidence to general propositions about the reasons for large budgets for intelligence.
There are four major determinants of the manpower, time, and money that a formal organization allocates to the intelligence function: (1) the degree of its conflict or competition with the environment—typically related to the extent of its involvement with, and dependence on, government; (2) the degree of dependence on internal support and unity; (3) the degree to which internal operations and external environment are believed by the organization’s managers to be capable of rationalization, that is, characterized by predictable uniformities and therefore subject to planned influence; and affecting all of these, (4) the size and structure of the organization—the heterogeneity of its membership, the diversity of its goals, and the centralization of its authority system. On all counts we can expect that the typical formal organization in modern society, whatever its products or services and whatever its cultural context, will make increasing use of experts.
The more an organization is in conflict with its social environment or depends on it for the achievement of its central goals, the more resources it will allocate to the intelligence function, and the more of those resources will be spent on experts we might call “contact men.” The contact man supplies political and ideological intelligence the leader needs to find his way around modern society; he mediates the relations of the organization with the outside world. He is primarily concerned with, and skilled in, dealing with facts about, and techniques of, changing the thoughts, feelings, and conduct of men through persuasion and manipulation. He is valued for his knowledge of the political and social topography of the containing society, for his “contacts,” which are so well developed that they become nontransferable, and for his skills in exploiting these contacts —skills in private inquiry, consultation, negotiation, mediation. Typical titles in this area include general counsel; lobbyist, Washington representative, legislative representative, or trade-association representative; public-relations, community-relations, or press-relations man. The backgrounds of such specialists are heterogeneous; most have held jobs in the bureaucracies with which their employers deal. Their contacts are used mainly for keeping information channels open for the speedy transmission of accurate political intelligence, typically between competing staffs (Milbrath 1963, pp. 256 ff., 269, 339; Wilensky 1956, pp. 61-79; Wildavsky 1964). In addition to serving as a listening post, the contact man interprets his employers or clients to important publics, real or imagined, and those publics to the organization he represents.
A second major force shaping the use of experts is the problem of internal control. The more the organization depends on the unity and support of persons, groups, factions, or parties within its membership for the achievement of its central goals, the more resources it will devote to the intelligence function, and the more of those resources will be spent on experts we might call “internal-communications specialists.”
The internal-communications specialist supplies political and ideological intelligence the leader needs to maintain his authority. He transmits policies to the membership or reports to the leader on membership sentiments and opinion, or he does both; he helps induct new members and train activists and leaders. He is primarily concerned with, and skilled in, changing the thoughts, feelings, and conduct of members. He is valued for his propaganda and group work skills, his ability to gauge group reactions, and his knowledge of the politics and personalities of the employing organization. Typical titles include: editor of the house organ, education, recreation, or training director, personnel- or industrial-relations officer, safety director, security man, community- (i.e., race) relations man. Some auditors or controllers also belong in this list. Backgrounds of such specialists, while diversified, typically include administrative, ideological, or political jobs that provide a grasp of the internal operations and structure of the organization in question.
Heterogeneous in origin and activity, these men fulfill similar functions. They build executives’ prestige and facilitate their control over the members; they supply information about the performance, politics, and morale of various organizational units; they locate and train new leaders; and they allay minority group pressures that threaten stability. Occasionally they function to maintain morale in crisis situations (Wilensky 1956, pp. 80 ff.; Myers & Turnbull 1956; Controllership Foundation 1954). Political parties, unions, voluntary associations—formally democratic organizations that cannot assume member support but must win it—are especially prone to use these specialists. In general, the more directly accountable the leader is to the rank and file, the more he needs to know about their state of mind. Further, the more the leader depends on the rank and file’s compliance with organizational rules and goals, the more he needs to know about the quality of the rank and file’s performance.
A third major source of variation in the number and types of experts used is the uneven rationalization of social life. The more an organization sees its external environment and internal operations as rationalized—that is, as subject to discernible, predictable uniformities in relationships among significant objects—the more resources it will devote to the intelligence function and the more of those resources will be spent on experts we might call “facts-and-figures men.” The facts-and-figures man supplies technical economic, legal, or scientific intelligence that helps the leader build his case in relations with outsiders and members, fend off attacks, and compete with rival organizations for markets, power, and prestige. He is primarily concerned with, and skilled in handling, empirical data; he is expected to produce quick, simple answers to complex technical questions as well as to make judgments of the power and intent of competitors and enemies. His humanrelations skills are less prominent. Typical titles of facts-and-figures men include: director of research, of planning and analysis, of reports and estimates, or simply, of intelligence; economist, statistician, accountant, actuary, pension and insurance consultant; industrial engineer, human engineer; and the ubiquitous “management consultant.” Included here are those lawyers who specialize in court litigation, case building for negotiations and arbitration, drafting contracts and briefs, or appearances before administrative agencies; it also includes those lobbyists who accent “research.” The backgrounds of facts-and-figures men are mainly in the law, college teaching, accounting, market and consumer research, operations or systems research, engineering and the physical sciences, and in the social sciences. On the average, they have more formal education than other intelligence specialists.
Facts-and-figures men function both to supply information and to mobilize support. In internal policy deliberations and decision making generally, they introduce a “rational-responsible” bias—a more conscious examination of alternatives, of relevant factors beyond the organization’s power, even of long-range consequences. In external relations they build pressure on outside parties via the government and the public, imparting a tone of public responsibility to executive rhetoric. They persuade and impress legislative committees, executive agencies, and judicial or quasi-judicial boards, promoting a general reorientation toward data. They strengthen the morale and conviction of their employers, and so of the members, thus boosting bargaining power with rivals (Wilensky 1956, pp. 39-60).
We see here a kind of dialectic of expertise. Facts-and-figures men are preoccupied with rational argument and criteria; their technical competence compels opposing parties to be more careful or honest in their use of information, to match each other expert for expert, fact for fact.
When the facts count
Among intelligence specialists, no conviction is stronger than the notion that their main function, whatever their intentions, is “backstopping,” or rationalizing established policy. And administrative leaders who hire experts tend to derogate technical intelligence and accent the importance of political and executive skills they themselves possess.
There is no doubt that much policy is improvised and later, with the help of experts, made to appear planned. Nor is there doubt that all experts, including facts-and-figures men, contribute to the selfesteem of their employers by lending them prestige and “fronting” for them with various publics, “taking the heat” when mistakes are made, and, in general, creating the verbal environment of the organization (Wilensky 1956, pp. 33-38). But such observations obscure both the conditions under which the “facts” do count and the significance of “mere slogans.”
A study of the interplay of technical economiclegal considerations and political-ideological considerations in union decision making found that the area in which union staff experts have their highest sustained influence in the entire decisionmaking process is that of problems in public relations and governmental relations—issues such as taxes, tariffs, welfare legislation, and world politics, which, while objectively important to the members, are seen by top officers as far from the core function of the union. But some problems are so technical that the specialized knowledge of facts-andfigures men becomes overriding even at the core of union functioning, despite the presence of political considerations (ibid., pp. 181-195).
Similarly, a study of governmental budget making—an eminently “political” process—suggests that a good “case” often counts. While precedent and political support are the major determinants of who gets what the government has to give, substantial departures from previous budgets and new programs with no precedent receive careful scrutiny and have a better chance if accompanied by well-documented arguments. Moreover, the cultivation of political support and official confidence are themselves partly a product of how well the homework is done. U.S. Bureau of the Budget examiners depend for their information on the agency they are assigned to investigate; the agency often converts the examiner into an advocate of particular programs by a sensible, even open, flow of information. Slippery statistics, grossly inaccurate reports are avoided; the expert from the Bureau of the Budget is referred to agency experts whose data he can trust (Wildavsky 1964, pp. 3, 13, 39, 56, 62-67, 136, 147, 169). If trends in federal budgeting practices of the 1960s continue—that is, if there is a further shift from “ceiling budgets” or from “incremental budgeting” based on previous decisions to a “planning-programming budgeting system” (PPBS) based on calculations of the cost effectiveness of alternative-program outputs—then reliance on technical data will increase [seeBudgeting, article onBudgeting as a political process, and compareHitch 1965].
This story is repeated in industry, where an enterprise may use “share of the market” as an operational guide to simplify its calculations and yet at the same time be forced to resort to special cost studies, carefully scrutinized, whenever there is deviation or innovation. And an auditor, investigating the operations of a local plant, can more easily be converted into its advocate at central headquarters if he learns to trust the local sources of his information (Cyert & March 1963; Controllership Foundation 1954).
Facts-and-figures men who command technical intelligence obviously are given more discretion where the problems are technical. Less obviously, they also carry more weight when the organization is weak in grass-roots political resources. Among Washington lobbyists, for instance, representatives of small organizations with limited political resources, such as humanitarian organizations and specialized trade associations, accent research in their lobbying strategy, in contrast to large-membership organizations, such as farm groups, veterans groups, and labor unions, who incline toward grass-roots campaigns and publicity (Milbrath 1963, pp. 227 ff., 343, 348). Paradoxically, this may give an advantage to the weak, whose case, if strong and technical, can count for something [SeeLobbying].
Technical and political intelligence
World War II saw an accelerated use of social science by government and the military as well as the emergence of freewheeling experts who combined training in physical science with some understanding of social and political life. Through the devices of the standing advisory committee, contract research, and the support of semiautonomous operations research institutes, the government began to tap more generalized advice from the academic community (Price 1954). A model for the nonmilitary uses of such advice can be derived from the best of the military examples: small, flexible, loosely organized, interdisciplinary groups in RAND, an advisory corporation engaged in systems analysis for the United States Air Force (B. Smith 1964; for similar examples from Britain, see Jones 1947). There is an urgent demand for broad policy advice on those political or administrative issues that rest heavily on technical intelligence. To cope with such issues, administrative leaders are turning to a new breed of experts who can interpret the work of separate disciplines, combine the functions of contact men with those of internal-communications specialists and facts-and-figures men, supply a blend of technical and ideological intelligence, and in general, enrich the verbal environment of elites. Needless to say, there is more demand for than supply of such experts.
Intelligence and structural complexity
Within a context of increased availability of information and of growing demand for generalized policy advice, it may be hypothesized that (1) an organization in conflict with its environment, especially one deeply involved with government, will turn toward contact men; (2) an organization heavily dependent on internal solidarity will hire internal-communications specialists; (3) an organization that sees its external relations and internal operations as rationalized will depend on facts-andfigures men.
These general determinants of the amount and type of resources devoted to the intelligence function are reflected in more familiar attributes of structure. Briefly, the more complex the structure, the more use of experts. The main locus of the managerial revolution is the large organization with many specialized units to coordinate, great heterogeneity of membership (manifested in a diversity of goals), and much centralization. Other things being equal, the larger the size of an organization the greater is its public impact, the more intense are its problems of internal control, and the more resources are available to it for the intelligence function. The more specialization there is, the more interdependent the specialized parts become, and the greater is the cost if any one part fails. Greater specialization thus calls for greater resources for each part of the organization to keep track of the other parts, and for more staff at the center to coordinate the whole. Moreover, greater specialization increases the difficulty of recruitment and training and the amount of effort needed to secure information about morale and performance. Finally, the more heterogeneous the membership or constituents of an organization, the more ambiguous, diffuse, diversified, and numerous its purposes or products; this more complex structure of goals means more variables to consider and a more urgent problem of coordination. In other words, size, specialization, centralization, heterogeneity of membership and goals—all of them attributes of structure—generate the need for experts insofar as they increase the number and diversity of social units in the environment that must be taken account of, aggravate the problem of internal control, and intensify the search for uniformities, thereby multiplying formal rules. The law for the computer age is: Programmed activity either drives out unprogrammed activity or drives it underground.
The quality of intelligence
To explain the great expansion in intelligence and major variations in the employment of experts is not to understand how their work is organized. Nor does that explanation tell us much about variations in the quality of intelligence. The knowledge explosion intensifies an old problem: how to draw from a highly compartmentalized body of knowledge the messages—whether in the form of questions, of insights, or of evidence—relevant to policy and get them into the room where decisions are made. Sources of failure are legion. Even if the initial message is accurate, timely, and relevant, it may be translated, condensed or completely blocked by personnel standing between the sender and the intended receiver; if it does get through, it may be only in distorted form. If the receiver is in a position to use the message, he may screen it out because it does not fit his preconceptions, because it has come through a suspicious or poorly regarded channel (because he is in the wrong), because it is embedded in piles of inaccurate or useless messages (what information theorists call excessive “noise” in the channel) or simply because too many messages are being transmitted to him (“information overload”) [seeInformationTheory]. When the relevant message is not in the system or when the intended receiver is not in a position to act on it (he is clearly constrained by forces beyond his control), we cannot call the instance an intelligence failure.
Pearl Harbor is alleged to be an intelligence lesson burned into the minds of general staffs and top planners of national strategy. The failure to take the final, crucial step of communicating to commanders the urgent warnings about Japan’s intentions supplied by Far Eastern code analysts is said to have been caused by the lack of a central intelligence organization (Ransom 1958, p. 58; see also ibid., pp. 41, 54-56; and Wohlstetter 1962, pp. 125 ff., 311 ff., 395). Yet 25 years later, after a reorganization of the intelligence function along these lines, two American presidents were led into disastrous military adventures based largely on miscalculation by intelligence agencies communicating at the top misleading pictures of the situations in Cuba, the Dominican Republic, and Vietnam. The vast expansion of the American government’s intelligence apparatus has not prevented successes in the uses of intelligence; neither has it prevented a string of fateful failures. (For examples of both, see Ransom 1958, pp. 60 ff.; Snyder & Paige 1958.)
Intelligence failures are rooted in structural problems that cannot be fully solved; they express universal dilemmas of organizational life that can be resolved in various ways at varying costs. In all complex social systems, hierarchy, specialization, and centralization are major sources of distortion and blockage of intelligence. The quality of intelligence is also shaped by the prevailing concepts of intelligence, the types of problems to be confronted, the stages of growth of the organization, and the cultural or ideological contexts of decision.
Insofar as the problem of organizational control is solved by rewards of status, power, and promotion, the problem of obtaining accurate, critical intelligence is intensified. For information is a resource that symbolizes status, enhances authority, and shapes careers. At every level, hierarchy is conducive to concealment and misrepresentation. When subordinates are asked to transmit information that can be used to evaluate their performance, their motives for making it look good are obvious.
As for the opportunity to distort information, middle-level managers, and even lower-level employees, sometimes have a near monopoly of wisdom about feasible alternatives. For instance, observers of man-paced factory jobs have noted the ingenuity of machinists who invent and hide cutting tools that do the work more efficiently than the prescribed tools and thereby permit more worker control of the pace. Although automation and centralization may change this, first-line supervisors still have indispensable practical knowledge of both unofficial work behavior and “bugs” in the technical system. Local plant managers in multiplant systems know the limits of their productive capacities far better than does central headquarters (Whyte et al. 1955; Sayles 1958; Dalton 1959; Ward 1967).
Matching the motive and opportunity of the subordinate to remain silent are the superior’s motive and opportunity to close his ears. The common belief that staff experts should be “on tap, not on top” functions to maintain line authority and reduce the status of the staff. Thus, if an organization has many ranks and in its administrative style and symbolism it emphasizes rank, it will display the greatest distortion and blockage in the upward flow of information. Where the pyramid is tall and narrows sharply at the top, providing a long promotion ladder for only a few, there are many timeservers at lower ranks who have neither information nor the motive to acquire it. In the middle, among the nonmobile, there are many defensive cliques that restrict information to prevent change, many mutual aid and comfort groups that restrict information because of resentment of more ambitious colleagues, and many coalitions of ambitious men who share information among themselves but pass on only the portion that furthers their careers. For the purposes of intelligence, the optimal shape of the hierarchy would be flat (since an organizational structure with few ranks makes for more accurate and more speedily diffused information) with a bulge in the middle (that is, there would be more specialists who had information, and many potential managers motivated to command it).
Whatever the shape of the hierarchy, however, to extract information from those who have it typically requires the bypassing of regular organizational structures. Efforts to resolve the dilemma of hierarchy versus need for intelligence include team or project organization, devices for communicating out of channels, machinery for investigation and inspection, performance checks, and reliance on informed outsiders.
The higher in the organizational hierarchy one goes, the less can the kinds of problems encountered be translated into terms that are part of the existing structure of specialized knowledge and the less, therefore, can a decision be programmed. Only at the lower levels of policy deliberation can the specialized expert tackle a specialized problem with a chance of solving it by the precise methods of science (Price 1954, p. 164). Further, the role of the expert at any level is self-changing: if he is successful within his sphere of competence, if his advice is taken on matters where his specialized knowledge is relevant, he is likely to be chosen for tasks outside that sphere of competence, where his specialized knowledge will very likely be irrelevant (Wilensky 1956, pp. 209 ff.). This is why there is so urgent a demand for generalized advisors at the top. It is also why an appropriate form of organization for industrial research and development or, more broadly, for the use of experts anywhere is the task force—a team or project of diverse specialists who are brought together to solve a limited range of problems and are then reassigned when the task is done. Several swiftly growing corporations based on sophisticated technology, such as the IBM Corporation, have adopted this form (Shepard 1956; Goldner 1965; Kornhauser 1962, pp. 50 ff.). It is possible that the task force is best for crisis situations, higher policy deliberations, policy-oriented research, and technical development, while other forms of organization are appropriate for basic research.
Additional defenses against the information pathologies of hierarchy include various reporting services and statistical controls. These include data on the cost variances of individual departments, trend reports on “invisible” operations (machine performance or the consumption of operating supplies), accounting statements for “profit centers,” and special “problem-solving” studies done by teams drawing on accounting information, engineering estimates, and industrial engineering standards (Controllership Foundation 1954). In interpreting such data, top executives even in very hierarchical organizations are not entirely defenseless: they have built up a sense of reasonableness regarding performance reports, and they can check submitted data for consistency. Finally, there is a universal check in every type of economy: the quality of performance of one organization is known by other organizations that receive its products. Thus, a Soviet metal fabricator who must work with defective steel will complain no less than his counterpart in the United States (Ward 1967).
Perhaps the most fruitful and common response to hierarchy is communication out of channels: contact men, on salary or retainer, keep in touch with outsiders who have a detached or critical view of the organization; internal-communications specialists, some of whom combine close ties to local leaders with loyalties to central headquarters, report on local performance and morale. A variety of marginal men at points along the organization’s boundaries supply supplementary and often crucial intelligence.
Modern military services evidence all of these responses to hierarchy. Like executives everywhere, military elites have moved from coercion and command to persuasion and manipulation. Instead of a pyramidal structure, they now favor a diamond bulging in the middle; instead of ritualistic close order drill, an accent on the training mission; instead of a mechanical assimilation of the army way, a reduction in symbols of rank; instead of standing operating procedures (SOP), group command conferences and informal briefings, bypassing normal channels (Janowitz 1960, pp. 38 ff.).
As a source of information blockage and distortion, specialization may be more powerful than hierarchy. The organization of the armed forces and industry alike encourages rivalry and restriction of information. Lines of organization become lines of loyalty and secrecy; each department restricts information that might advance the competing interests of the others. While information can also be used to persuade potential allies and facilitate accommodation with rivals, it is more commonly hoarded for selective use in less collaborative struggles for power and position.
The dilemma of intelligence versus specialization is twofold: specialization is essential to the efficient command of knowledge but antithetical to the penetrating interpretation that bears on high policy. Specialization and its concomitant, interunit rivalry, frequently block the sharing of accurate information, but if problems of upward communication can be solved, rivalry can result in the specification of clashing alternatives and the presentation of opposing cases, both of which are great gains. The main cost of specialization in intelligence is parochialism, or the production of misleading or irrelevant information—a result of the familiar limitations of the expert (Laski 1931). The professionally biased intelligence producer remains too distant from the intelligence user and too ignorant of policy needs; he is also forced to compete with other producers for the support and guidance of the user (Kent 1949, pp. 81, 94 ff.). The gain from constructive rivalry is another matter; it depends on administrative styles and structures that expedite the free flow of rival perspectives and solutions to the responsible executives and their general advisers.
To resolve this structural dilemma, especially in organizations dependent on technical intelligence, administrative leaders use the following devices: they recruit managers from professional or scientific staff; they tie specialists in the field closely and informally to the home staff via rotation, frequent conferences, and career lines that lead from the field to central headquarters; they expose themselves systematically to intelligence, usually by examining multiple sources firsthand but sometimes by stimulating competition between sources. Franklin D. Roosevelt was a master of the last two techniques (Schlesinger 1959, pp. 522-528; Neustadt 1960, pp. 154 ff.). However, if rival departments are forced to a consensual decision in committee, the gains of calculated competition cannot be secured (Sorensen 1963, pp. 62-63; Possony 1941, pp. 226-227; Schlesinger 1965, p. 209).
The firmest generalization that can be made in this area is that the greater the number of ranks and the greater the number of organizational units involved in a decision process, the more the distorting influence of rank and jurisdiction and the greater the chance of an intelligence failure.
Related to the information pathologies of hierarchy and specialization is the dilemma of centralization. If intelligence is lodged at the top, too few officials and experts with too little accurate and relevant information are too far out of touch and too overloaded to cope. On the other hand, if intelligence is scattered in many subordinate units, too many officials and experts with too much specialized information may engage in dysfunctional competition, may delay decisions while they warily consult each other, and may distort information as they pass it up (Wiles 1962, pp. 141 ff.; Ward 1967; Kent 1949, p. 81).
Centralized intelligence presents the same dangers as specialized intelligence: it keeps the collection of data too far from their use in policy; it encourages agreed-on estimates that may conceal strong disagreement and that in any case do not reveal the weights of diverse opinions (these disadvantages are only partly offset by such devices as the dissenting footnote, which themselves result from horsetrading and logrolling); and it competes with its own subsidiaries for scarce personnel and documentation facilities (Ransom 1958, pp. 81, 94, 197 ff.). Centralization contributes one additional danger peculiar to itself: the acquisition of unnecessary responsibility (“empire building”) is always accompanied by cries of duplication and inefficiency among the units to be absorbed or eliminated. After administrative reform, a more unified, centralized intelligence agency producing a unified consensual judgment fosters an illusion of security and reliable intelligence that can, as in the case of the Bay of Pigs invasion, conceal fantasies at the highest level.
In the minds of political, military, and industrial elites the advantages of centralized intelligence have apparently tended to outweigh these dangers. Most executives have been less concerned with preserving the independence and objectivity of their experts than with controlling them. For their part, the experts, seeing a chance for greater influence, have not been loath to secure guidance from the top. And in the reorganization of national strategic intelligence in the United States, the image of Pearl Harbor—the signals scattered, no central interpretive agency alert to them—has been commanding.
This debate about departmental or local versus central or national intelligence has obscured two fundamental points. First, there is need for interpretive skills at every point where important decisions are taken; the ideal is the close integration of data collection and evaluation in every department and division, and within the limits of budget and personnel, at every level (Wasserman I960; Kent 1949, p. 81). Second, various degrees and kinds of decentralization are appropriate for various purposes and measures of effectiveness. For instance, a study of the controller’s department in seven multiplant companies—all of which aimed to provide high-quality information services at minimum cost and, incidentally, to facilitate the training of accounting and operating executives—found that the effectiveness of decentralization depended on the type of information to be provided (Controllership Foundation 1954). If the main aim is to keep score (“Am I doing well or badly?”) and to direct attention (“What problems should I look into?”), then it is best to hire competent professional men who look to the controller’s department for promotion, attach them to local factories and district offices, and put them under the factory manager. The formula for success is much geographical decentralization, some decentralization of formal authority, but very limited decentralization of loyalty. For the purpose of keeping score, the decentralization of location and authority makes it easier to gain access to resource documents and to know the reliability of source records. For the purpose of directing attention, such decentralization gives operating executives direct and active contact with accounting people and thus more confidence in the operating standards and performance reports going to the boss. For the purpose of preserving the quality of data available to central headquarters for special studies and big policy, it is equally important to maintain loyalty to the controller’s department through recruitment of professionally committed men who look there for promotion. Ideally, the formal organization puts accountants close enough to the situation to be accepted by factory management and obtain roughly accurate information; at the same time, career-line loyalty and professional identification provide enough social support for them to apply standards.
In short, there is a balance to be struck among various kinds of decentralization—an optimal, or at least “workable,” set of gains and losses depending upon the purposes to be pursued. This is an area where research has barely begun.
Doctrines of intelligence
Concepts of the intelligence function deeply affect the organization of intelligence, its recruitment base, and hence its quality. Administrative leaders often evince antiintellectualism, narrow empiricism, and a demand for secrecy in odd combination with a demand for scientific prediction and quantitative estimates. This combination leads to foolish ideas about the proper organization and possibilities of intelligence.
If users of intelligence think of it as “facts” rather than “evaluated facts,” if they are imbued with the notion that more raw “information” to “fill in gaps” is their great need, they are likely to make intelligence a distinct, subordinate function, to separate sharply collection of data from interpretation, and to exclude experts from policy deliberations. They will attract either crude empiricists or conformists content to “backstop” the preconceptions of policy makers. Thus, intelligence administrators in government and in the military have typically been lawyers or soldiers trained in school and in practice as empiricists—suspicious of plans and predictions, hostile to generalizations and abstractions. Similarly, intelligence staffs have been composed mainly of historians, lawyers, journalists, and soldiers (Kendall 1949, pp. 550-551; Hilsman 1956, pp. 81, 96; Millikan 1959, p. 163; Wasserman 1960). The most influential Washington lobbyists and union staff experts have been lawyers (Wilensky 1956, pp. 230 ff.; Milbrath 1963, p. 157). The background and outlook of the intelligence producer often match those of his boss; both sometimes display a pervasive anti-intellectualism, a general resentment of the outsider, and an exaggerated belief in practical experience. In response, students of intelligence, like the writers of introductory textbooks in social science, still tend to belabor the points that the facts never speak for themselves and that information should be gathered in relation to questions and categories for analysis. The main questions remain open: What types of experts, with what background and training, make the best interpreters, and what concepts and organization of intelligence attract and motivate them?
The common distinction between “intelligence” (information gathering) and “operations” (clandestine action) also has consequences for recruitment. While fact gathering appeals to naive realists, whose interpretive capacities are limited, the excitement of operations attracts the kind of adventurers and activists who guided the American government into the Bay of Pigs. Two related concepts of intelligence further complicate the problem of organization and reduce the quality of information services—the accent on short-run prediction, reflected in a pressure for direct, simple answers to immediate questions, and the accent on secrecy. Given the familiar difficulties of prediction in social science and politics, intelligence estimates, like stock market forecasts, are heavily hedged, as in: “There is a serious possibility that Red China will intervene, but there are some grounds for optimism.” Intelligence users and operators, incongruously demanding from researchers “all the fact” at the same time as they demand short, speedy, journalistic estimates of future developments, use the inevitable hedging and the frequent failures of prediction to justify their antiintellectualism, their naive separation of “facts” from “know-how,” and their denigration of research. Faced with repeated disconfirmation of short-run predictions, they are susceptible to the “cry-wolf syndrome,” which discounts all predictions and warnings whatsoever (Wohlstetter 1962, p. 52). The notion that secret sources are intrinsically superior is reflected in loyalty-security systems and, in the case of government, the segregation of clandestine operations from research. Students of intelligence agree that while covert, or “black,” intelligence, like overt intelligence, has expanded in every sphere of modern life (U.S. Congress . . . 1965; R. Smith 1956), policy makers grossly exaggerate its importance and reliability. They also agree that an emphasis on secrecy impairs critical judgment in the production and interpretation of intelligence and dulls the sense of relevance. At the same time, secrecy blinds the executive to the increasingly available open sources (Kent 1949, p. 168; Schlesinger 1965, pp. 248, 258).
The distorting effects of secrecy can be discerned both in the type of personnel attracted to secret operations and in the peculiar conditions for intellectual work that secrecy imposes. The more secrecy, the smaller the intelligent audience, the less systematic the distribution and indexing of research, the greater the anonymity of authorship, and the more intolerance for deviant views and styles of life (B. Smith 1964, p. 78; Merton 1957). Such a restrictive atmosphere reduces the incentives for participation by the best scholars, whose ways of life and thought are not always orthodox and who work best with critical appraisal and acclaim and with wide-open access to the best sources. Thus, the security regulations and loyalty investigations of the McCarthy era in the United States permitted cautious mediocrities, dedicated to the most impoverished cliches of the cold war, to rise to the top of the foreign service (Schlesinger 1965, p. 411; Kent 1949, pp. 69, 74, 146-147; Ransom 1958, p. 178).
If we consider the motives of secret agents, informers, stool pigeons, and other spies, as well as the stigma under which they labor, the limitations of covert intelligence become obvious. In order to protect their sources, clandestine agencies want complete control of data collection and operations; they therefore have a natural penchant for the kind of compliant people who make good security risks. The motives of such people have been noted by students of police work, of military intelligence, of foreign affairs: fear, insecurity, revenge, envy, remorse, money—a shaky basis for objective reporting (Harney & Cross 1960, pp. 33-39; Bontecou 1953, pp. 131-135; Hilsman 1956, pp. 21 ff.; Schlesinger 1965, pp. 230 ff.). While the classic spy of fiction and practice struggles successfully to obtain startling information, his home office, knowing that in sensitive areas the enemy always leaks false information, is likely either to discount accurate information (Moyzisch 1950; Seabury 1954) or fall for fabrications (Ransom 1958, pp. 168-169).
In the end, the most reliable intelligence sources for competing organizations are open; the best data, seldom secret, are the actions of the other party (compare George 1959). Yet, at the very time that open sources multiply and communications technology speeds the processing and flow of information, we may be elevating the position of the spy. Since the onset of World War n, the reservations of diplomats and administrative leaders about spying have weakened (Vagts 1956, pp. 65, 70-71); in government and industry alike, the budgets and influence of spies have expanded (R. Smith 1956; U.S. Congress . . . 1965; International Union . . . 1964, pp. 323).
An emphasis on secret information not only threatens individual privacy; it can demoralize an organization. Such practices as prying into the personal lives of employees to fend off the efforts of rivals to obtain secret information, or conducting disguised surveys to uncover prounion sentiment, can build resentment, provoke strikes, increase labor turnover, and reduce efficiency.
This area of study is rife with sensationalism; systematic research is rare. The conditions that would give substance to an Orwellian nightmare are only dimly understood.
The nature of the problem
The nature of an executive decision itself shapes the uses and quality of intelligence because it affects the number, kinds, and organization of experts called to serve. To illustrate, I will suggest a few hypotheses using the variables of urgency, cost, innovation, and certainty.
First, when the decision is urgent, the distortions of hierarchy, specialization, centralization, and doctrine are minimized. Ironically, “hasty” decisions made under pressure may on the average be better than less urgent ones.
Compare the two Cuban crises that dominated the Kennedy years. In the Bay of Pigs affair, planning, begun in the Eisenhower administration, was lengthy, discussion was rank-oriented, the range of serious exploration of alternatives and consequences limited, and the outcome disastrous. In the more urgent missile crisis, discussion was egalitarian; protocol, seniority, and rank counted for little; a wide range of alternatives was explored in depth (Sorensen 1965, pp. 679 ff.); and, by the standards of the Bay of Pigs, the decisions, in both process and consequence, were superior. Urgency functioned to overcome many information pathologies. Similarly, President Truman’s rapid decision to intervene in Korea, with the modest objectives of restoring the status quo ante and preserving Allied unity (Snyder & Paige 1958), was clearly more sensible than the later series of more leisurely decisions culminating in a new war aim of a unified noncommunist Korea—an aim which provoked Chinese intervention (Neustadt 1960, p. 139).
The greater the costs, risks, and uncertainty, and the more significant the changes in methods and goals involved, the more intense is the search for information, but the stronger is the weight of established policy and vested interests. Decisions involving many people, much money, great uncertainty or great risks, and major innovations evoke action and advice from every specialized unit at every level of the hierarchy, thereby increasing the dangers of paralyzing delays and of distortion or blockage of communication.
Perhaps the soundest hypothesis we can derive from these diverse considerations is that only those “big” (that is, costly, risky, or innovative) policy decisions that are also very urgent are likely to activate intelligence of a high quality, because in such cases deliberation moves out of channels toward men of generalized wisdom, executives and experts alike, who can then communicate informally and effectively at the top. But when crises become routine, the effects of urgency on intelligence may be modified; if there are too many turning points, too many critical junctures, the usual response to the extraordinary event—the cutting of new channels of intelligence—is inhibited. (For opposing hypotheses, see March & Simon 1958, pp. 168-175, who offer a theory that derives social organization from individual psychology.)
Any analysis of the determinants of the quality of intelligence must be sensitive to the flow of time and talent.
Because little is known about the growth of organizations and hence there is no agreement on possible “stages” of growth, we cannot systematically relate the problem of intelligence to processes of structural change. The literature provides only contradictory clues (see Haire 1959, pp. 292-293; March & Simon 1958, pp. 185-187; Schumpeter 1942, pp. 132 ff. in 1947 edition). Applying our scheme for the analysis of intelligence failures, however, we can hypothesize that stages of organizational growth will affect the quality of intelligence because they affect the types of experts hired and the problems they confront, the doctrines of intelligence, and the salience of structural distortions of intelligence.
Assume a new, swiftly growing organization in a fast-changing environment. Its problems—unroutinized, unprecedented—are mainly those of internal control and external competition and conflict. It must build a social base of members, employees, customers, clients, or constituents; it must select and shape a core of top personnel committed to its mission (compare Selznick 1957). While high degrees of centralization may prevail, it is more likely that hierarchy and specialization will be de-emphasized. And while secrecy may be demanded in intelligence doctrine as in everything else, urgency will shape many decisions and free the flow of information, secret or not. In short, information pathologies are minimized. Aggressive, expansionist organizations in touch with a fluid environment apparently have the best intelligence systems (Ransom 1958, p. 46).
Suppose that the organization becomes institutionalized (“established”) at a slow growth rate and that it negotiates many stable arrangements with its environment (Cyert & March 1963, p. 119). The shift in problems and personnel is not simply from innovation to execution, from idea men to orderly bureaucrats (March & Simon 1958, p. 187; Lazarsfeld 1959, p. 22). It is a more complicated shift from entrepreneurial careerists and dedicated missionaries to professionals (some of whom are bureaucratic in mentality but others of whom are program-minded innovators), with a scattering of enduring old missionaries (Wilensky 1956, pp. 138—153, 170-174; Wilensky 1964, pp. 150-151). It is also a shift from ideological intelligence emphasizing problems in the strategy of conflict to technical intelligence emphasizing problems of negotiation and administration; a shift from less routinized to more routinized problems; and a shift from most urgency to least. Organizational (or national) myths become more fixed, the counter-stereotyping activities of intellectuals more difficult. Hierarchy and specialization become prominent. Insofar as there is also a shift away from secrecy, information pathologies may be reduced, but all the other changes listed would seem to reduce the accuracy, relevance, and timeliness of intelligence.
Related to the growth of organization is the process of succession. In the succession crises characteristic of rapidly changing organizations, there may be greater play for policy-oriented intellectuals. In the routine succession characteristic of large, stable organizations, there may be a strong bias in favor of established policy and official prejudice. The new man, like his predecessor, strongly desires continuity; he must avoid frightening key incumbents and potential supporters below. The outgoing official wishes to avoid being discredited. Moreover, the new man requires briefing, which both the outgoing administration and holdover experts are glad to give.
Sheer frequency of succession exerts an influence apart from the mechanism of turnover. Other things being equal, high turnover of administrative leaders discourages the expression of critical opinion in the short run; the newcomers tend to keep quiet until they learn the lay of the land, build confidence among superiors and subordinates, acquire a political base, or in other ways solidify their positions. Frequent, institutionalized succession is one reason for the reluctance of the United States government to liquidate error in such places as Cuba and Vietnam—a striking continuity of policy through a succession of men so different in viewpoint and style as Eisenhower, Kennedy, and Johnson.
A severe limitation of this analysis of process is that we cannot assume a straight-line trend—that is, an inevitable slowing down of growth rates, increasing stability of the environment, a decline in urgency, a reduction in succession crises, or the like. Until we have a diversified analysis of typical growth curves and their correlates, research in this area will remain primitive.
Cultural and national variations
The remaining piece of the picture is the most impressionistic. Although my analysis has been cast in general terms applicable to complex organizations in every modern state, cultural-ideological contexts obviously condition the uses and influence of intellectuals and experts as well as the quality of intelligence itself. But no comparative studies show how. Moreover, some of these national “cultural” variations turn out to be themselves the structural variations already discussed.
The main distinctions for our purposes are between command and market economies and between totalitarian and pluralist polities. First, economic and political contexts affect the balance between types of intelligence used. The contact man, for instance, is everywhere called to service because of the entanglements of diverse private and public bureaucracies defending and promoting special interests. The greater the public impact of these struggles, the more government intervention and the more indispensable the lobbyist, the publicist, and the lawyer. If labor unions and public utilities deeply involved with government in a market economy must rely on the ideologicalpolitical intelligence of contact men, major industrial firms in a command economy are even more dependent on such information. In a Soviet-type economy, “planners’ tension” (an excess of administratively ordered output over capacity output) necessitates an immense formal apparatus for communicating and enforcing orders, and a complicated system of unofficial expediters and fixers who try to circumvent the orders so that goals may be met—which creates a demand for more planners to keep track of what is going on (Wiles 1962, pp. 132-135; Berliner 1957; Dahl & Lindblom 1953; Devons 1950).
More generally, totalitarian states, with their penchant for ideological indoctrination, allocate a larger share of all intelligence resources to political intelligence. This is true both for contact men and for internal-communications specialists. Finally, totalitarian states tend to obliterate the distinction between internal and external intelligence [seeIntelligence, Political AndMilitary].
Second, economic and political contexts shape the quality of the technical, economic, and legal data that are generated, reported, or both. Command economies and totalitarian political systems strengthen the structural roots of intelligence distortion. Hierarchy, of course, is more prominent, so the temptation and opportunity to conceal, misreport, and mishandle data are enormous (Dahl & Lindblom 1953, pp. 254-271; Wiles 1962, pp. 222 ff.). Moreover, if central planning is emphasized, information-processing costs soar, uncertainty is great, and problems of data collection and interpretation are often insoluble (Ward 1967).
Secrecy, too, is more prominent and the information pathologies associated with it more highly developed in totalitarian states. The national variations here are considerable. It is said that Russian regimes, whether tsarist or communist, have shown a passion for secrecy not found in the West. One has the impression, however, that no governments have ever been more obsessed by fear of espionage or more organized to protect secrets than such modern totalitarian systems as Hitler’s Germany and Stalin’s Russia. Similarly, one could argue that among free societies the distorting effects of secrecy are felt least in the United States and most in parliamentary democracies with aristocratic traditions, such as Great Britain. In the decentralized federalism of the United States, in which the separation of powers is emphasized, officials and experts can act out parochial loyalties by “leaking” secret information to the press, by conducting investigations, or by encouraging exposes of rivals. The publicity consciousness of the American public combines with easygoing libel laws to free newspapermen from normal constraints. Secrets cannot be kept for long. In contrast, administrative or scientific subordinates in Britain owe their main allegiance to the Cabinet, and hardly anything justifies public deviation from its policies, let alone the use of restricted information as a weapon in the struggle for power; French officials also enjoy intensive privileges of secrecy (Price 1954, pp. 138-139).
Unfortunately, the bright light of publicity in the United States may not, on balance, improve the reliability, relevance, and timeliness of intelligence. The pressure of publicity evokes a counter pressure for the safeguarding of secrecy. The opportunity and desire to expose political enemies and their secrets are forces making for extremism, as a result of which truth is assassinated along with the characters of the men who seek it. Shils (1956) has provided a sensitive analysis of secrecy-fearing extremism, as institutionalized in the loyalty-security programs of the 1950s. To design studies of the amount of secrecy, of its effects on recruitment, morale, and the conditions for intellectual work in various countries and organizations would tax the ingenuity of social scientists. But no set of problems in this area is more important.
Related to cultural variations in the incidence and effects of secrecy are variations in ethnocentricism among elites. Ethnocentricism in the modern state is occasionally a major source of intelligence failures (Hilsman 1956, pp. 103 ff.; Kendall 1949, p. 543; Kautsky 1965, p. 15; Ransom 1958, p. 171; Wasserman 1960, pp. 167-168). National stereotypes helped lead American statesmen to faulty interpretations of intelligence in the period before Pearl Harbor, when Far Eastern specialists were systematically kept from communicating accurate judgments not only because of their subordinate positions but because they were believed to be too immersed in “the Oriental point of view” (Wohlstetter 1962, pp. 102, 395). This was also true in the Korean case (Snyder & Paige 1958, p. 241 in 1962 edition). If American leaders in the early 1960s were having a difficult time putting aside their delusions about a unitary “Sinosoviet bloc,” imagine how hard it was for the xenophobic men in Peking to recognize a thaw in the cold war. These examples suggest that elites vary in their ability to transcend ethnocentric molds. The sources, amount, and effects of such variation deserve study.
Insofar as national ideologies proscribe particular academic disciplines or encourage antiexpert or anti-intellectual fervor, they shape the quality of intelligence at its source. During the period of maximum Stalinist terror the Soviet regime wiped out the science of genetics by persecuting Mendelian “deviationists,” pronounced quantum mechanics inconsistent with dialectical materialism, and labeled mathematical economics as bourgeoisidealist. This last step directly affected the technical intelligence needed for economic planning. It is difficult to separate ideological from social conflicts here. Objection to particular intellectual perspectives may rationalize a general mistrust of troublesome experts and technicians, a mistrust inspired by their indispensability, their love of autonomy, and the great social distance between them and party rulers. Within free societies a similar phenomenon is found in labor movements and labor parties (Wilensky 1956, pp. 260-269). Perhaps the surest generalization is that national ideologies accenting either workers’ control or the dictatorship of the proletariat tend to subvert effective organization of the intelligence function by exacerbating universal differences between men of knowledge and men of power.
A final determinant of the use and quality of intelligence that might be called cultural-ideological is the time perspective of elites (Speier 1952, p. 448). The roots of national variations in the range of foresight typically employed by administrative leaders might be sought in the frequency and mechanism of succession. The frequent, routine succession of leaders in stable democracies, while encouraging continuity of past policy, may discourage long-range plans. The repeated succession crises of unstable totalitarian regimes may also permit little long-range planning. If not too frequent, however, succession crises may open up opportunities for long-range planning by executives and staff with training and outlook vastly different from those of their predecessors. It is possible that Mussolini, who ruled for 21 years, and Stalin, who ruled for 27, had longer time perspectives than those of their contemporary counterparts in the Western democracies or the shaky military dictatorships of Latin America. In view of the common assumption that the other fellow has a plan—if not a plan for a Thousand Year Reich, then at least one for next year—it is remarkable that the sociology of complex organizations has so little to say about the conditions that foster the failure of foresight.
Information has always been a source of power, but it is now increasingly a source of confusion. In every sphere of modern life, the chronic condition is a surfeit of information, poorly integrated or lost somewhere in the system. At the same time, available technology and expert staff have made it possible to cope with the problem. Whatever the national variations, elites in every rich country are being moved to overcome the information overload and even to break through comfortable slogans and grasp reality. Their success in understanding internal operations and external environment is affected by the shape of the organizations they command and by their defenses against information pathologies.
Some gains in the quality of intelligence are possible from a reorganization of the intelligence function; for instance, if the hierarchy is too tall, it can be flattened; or experts can be given more autonomy by having them report to fewer bosses. Such redesign of structure, however, meets its limits where it entails large losses of efficiency, coordination, and control, and where structural resistance to change is strong.
Unfortunately, intelligence failures are built into complex organizations. Thus, alert executives are everywhere forced to bypass or diversify the regular machinery and seek firsthand exposure to intelligence sources both inside and outside the organization. They typically move to points along the organization’s boundaries: looking toward the bottom, they rely on internal-communications specialists; looking outward, they rely on contact men. They talk to reporters and researchers investigating their organizations; they establish study commissions or review boards comprised entirely of outsiders, like the members of British royal commissions (Hanser 1965); they institutionalize complaints procedures and thereby subject themselves to systematic, independent criticism from below, as in the case of the ombudsman; they assemble ad hoccommittees, advisory corporations, kitchen cabinets, general advisers, personal representatives. These unofficial intelligence agents may constitute the most important and reliable source of organizational intelligence. They are sufficiently sensitive to the culture of the executive to communicate successfully with him and sufficiently independent to provide detached judgment. They bring to bear the multiple perspectives of marginal men.
Research implications are apparent. We need studies that compare the performance of organizations or governments that rely most and least on (1) unofficial sources and (2) various types of outside sources, both official (study commissions, grievance commissioners, advisory corporations, task forces) and unofficial (journalists, scholars). For instance, is a sophisticated reporter working with open sources better than a secret agent working with top-secret information (Wohlstetter 1962, pp. 130, 169; Schlesinger 1965, pp. 247-248)? The question could not be settled without comparisons of both good and bad estimates among large populations of outsiders (e.g., columnists in all the prestige papers) and insiders (e.g., all top staff in major intelligence agencies) on several policy issues. More generally, estimates of the costs and gains for intelligence of various strategies for bypassing or diversifying official channels might be feasible. An equally important area for research is the effect of the new information technology on the problem of intelligence (for leads, see Simon I960; Shultz & Whisler I960; and Kaufmann 1964).
Given the institutional roots of intelligence failures and the urgency of so many big decisions, what counts is the top executive’s preconceptions—what he has in mind when he enters the room and must act. The role of experts and intellectuals in shaping these preconceptions, both inside and outside the organization, is little understood. For instance, in college and in on-the-job training programs, business executives and government officials are increasingly exposed to social-science perspectives. We do not know the impact of such education on them. If it does not yield direct answers to their immediate questions, perhaps it does break through their cruder stereotypes, enhance their understanding of themselves and their organizations, alert them to the range of relevant variables, and make them more skillful in the use of experts.
Similarly, the symbols that surround the executive in his daily life shape his orientation. In foreign affairs, experts invent slogans that reinforce or diminish policymakers’ adherence to the wartime myth of the single enemy. For example, the image of the Sino-Soviet bloc fostered so much rigidity in the State Department and its publics in the 1950s and early 1960s, that President Kennedy, seeking very minor modifications of America’s China policy in accordance with the dramatic signs of the Sino-Soviet split, felt himself a captive of both his foreign office and the mass media (Schlesinger 1965, pp. 410-430, 483 ff.). Selfrighteous cold war rhetoric, rooted in the facts of past conflicts, served to perpetuate policies long outmoded (Neustadt 1960, p. 139; Wasserman I960; Sorensen 1965; Kautsky 1965). On the other hand, such doctrines of the arms-control experts as “stable nuclear deterrence” and “transitional deterrent” moved debate inside the Kennedy administration toward a disarmament policy culminating in the 1963 test-ban treaty (Schlesinger 1965, pp. 470 ff., 889 ff.)- Thus, the most important function of the general adviser may be to set the tone of policy discourse—to reduce or increase the number of recognized options and to bolster or break up the brittle slogans that blind executive vision.
Harold L. Wilensky
[Directly related are the entries Administration; Budgeting; Bureaucracy; Information Storage And Retrieval; Intellectuals; Intelligence, Political And Military; Lobbying; Privacy; Strategy. Other relevant material may be found inBusiness Management; Decision Making; Information Theory; Knowledge, Sociology OF; Leadership; Military; Organizations, article on THEORIES OF ORGANIZATIONS; Operations Research; Planning, SOCIAL; War. ]
Of all the major ’problems in organizational sociology, the intelligence function has received least systematic treatment. Manuals on strategic intelligence provide little but vacuous homilies (see especially Platt 1957). More analytical works by students of military and governmental intelligence include Hilsman 1956; Kent 1949; Kendall 1949; Pettee 1946; and Ransom 1958. There are a few detailed accounts of intelligence use in national crises, notably Sherwood 1948; Neustadt 1960; Schlesinger 1965;and Sorensen 1965. Outstanding studies of intelligence in military campaigns include Tuchman 1962; and Marshall 1953. Smith 1964 is an excellent analysis of the policy impact of military consultants (in this case, the RAND Corporation). Wohlstetter 1962 is by far the most painstaking and analytical study of an intelligence failure (see alsoSnyder & Paige 1958).
There are many useful discussions of staff-line relations or decision making in management, especially Urwick 1937; Simon et al. 1950; Dahl & Lindblom 1953; Dalton 1959. Polemics about the nature, types, plight, or sins of the intellectual in society (for instance, Baritz 1960) have been endless. But there is only a handful of systematic empirical studies of the relations of men of knowledge and men of power. These include Bailey 1950; Thompson 1950;and Flash 1965 (economists and lawyers in government agencies and congressional committees); Controller ship Foundation 1954 (controllers); Kornhauser 1962 (scientists);Wilensky 1956 (staff experts.in labor unions); Milbrath 1963 (Washington lobbyists); and Wildavsky 1964 (experts in the federal budgetary process). There are also a few indirectly relevant studies of experts, politicians, and pressure groups in the formulation and administration of public policy; see, for instance, Schattschneider 1935 (the tariff); Macmahon et al. 1941 (work relief); Selznick 1949(the Tennessee Valley Authority); and Meyerson & Banfield 1955 (public housing). A sociological treatment of organizational intelligence based on cases from international relations, politics, welfare, and industry is Wilensky 1967.General books on science and the social order includeBarber 1952; Machlup 1962; Gouldner & Miller 1965.
Bailey, Stephen K. 1950 Congress Makes a Law: The Story Behind the Employment Act of 1946. New York: Columbia Univ. Press.
Barber, Bernard 1952 Science and the Social Order. Glencoe, 111.: Free Press.
Baritz, Loren 1960 The Servants of Power: A History of the Use of Social Science in American Industry.Middletown, Conn.: Wesleyan Univ. Press.
Berliner, Joseph S. 1957 Factory and Manager in the U.S.S.R. Cambridge, Mass.: Harvard Univ. Press.
Bontecou, Eleanor 1953 The Federal Loyalty-Security Program. Cornell Studies in Civil Liberty. Ithaca, N.Y.: Cornell Univ. Press.
Brooks, John N. 1963 The Fate of the Edsel and Other Business Adventures. New York: Harper.
Controllership Foundation 1954 Centralization vs. Decentralization in Organizing the Controller’s Department: A Research Study and Report, by Herbert A. Simon et al. New York: The Foundation.
Cyert, Richard M.; and March, James G. 1963 A Behavioral Theory of the Firm. Englewood Cliffs, N.J.: Prentice-Hall.
Dahl, Robert A.; and Lindblom, Charles E. 1953 Politics, Economics, and Welfare: Planning and Politicoeconomic Systems Resolved Into Basic Social Processes. New York: Harper.
Dalton, Melville 1959 Men Who Manage: Fusions of Feeling and Theory in Administration. New York: Wiley.
Devons, Ely 1950 Planning in Practice: Essays in Aircraft Planning in War-time. Toronto: Macmillan; Cambridge Univ. Press.
Flash, Edward S. 1965 Economic Advice and Presidential Leadership: The Council of Economic Advisers.New York: Columbia Univ. Press.
George, Alexander L. 1959 Propaganda Analysis: A Study of Inferences Made From Nazi Propaganda in World War II. Evanston, III.: Row, Peterson.
Goldner, Fred H. 1965 Demotion in Industrial Management. American Sociological Review 30:714-724.
Gouldner, Alvin W.; and Miller, Seymour M. 1965 Applied Sociology: Opportunities and Problems. New York: Free Press.
Haire, Mason 1959 Biological Models and Empirical Histories of the Growth of Organizations. Pages 272306 in Foundation for Research on Human Behavior,Modern Organization Theory: A Symposium. Edited by Mason Haire. New York: Wiley.
Hanser, Charles J. 1965 Guide to Decision: The Royal Commission. Totowa, NJ. : Bedminister Press.
Harney, Malachi L.; and CROSS, JOHN C. 1960 The Informer in Law Enforcement. Springfield, III.: Thomas.
Hilsman, Roger 1956 Strategic Intelligence and National Decisions. Glencoe, 111.: Free Press.
Hitch, Charles J. 1965 Decision-making for Defense. Berkeley: Univ. of California Press.
International Union, United Automobile, Aircraft AND Agricultural Implement Workers OF America(UAW), Nineteenth Constitutional Convention, March 20-27, 1964 1964 Proceedings. Atlantic City. N.J.: UAW.
Janowitz, Morris 1960 The Professional Soldier: A Social and Political Portrait. Glencoe, III.: Free Press. → A paperback edition was published in 1965.
Jones, R. V. 1947 Scientific Intelligence. Journal of the Royal United Service Institution 92:352-369.
Kaufmann, William W. 1964 The McNamara Strategy. New York: Harper.
Kautsky, John H. 1965 Myth, Self-fulfilling Prophecy, and Symbolic Reassurance in the East-West Conflict.Journal of Conflict Resolution 9:1-17.
Kendall, Willmoore 1949 The Function of Intelligence. World Politics 1:542-552.
Kent, Sherman 1949 Strategic Intelligence for American World Policy. Princeton Univ. Press.
Knorr, Klaus 1964 Failures in National Intelligence Estimates: The Case of the Cuban Missiles. World Politics 16:455-467.
Kornhauser, William 1962 Scientists in Industry: Conflict and Accommodation. Berkeley: Univ. of California Press.
Laski, Harold J. 1931 The Limitations of the Expert. London: Fabian Society.
Lazarsfeld, Paul F. 1959 Reflections on Business. American Journal of Sociology 65:1-31.
Machlup, Fritz 1962 The Production and Distribution of Knowledge in the United States. Princeton Univ. Press.
Macmahon, Arthur W.; Millett, J. D.; and Ogden, Gladys 1941 The Administration of Federal Work Relief. Published for the Committee on Public Administration of the Social Science Research Council. Chicago: Public Administration Service.
March, James G.; and Simon, Herbert A. 1958 Organizations. New York: Wiley.
Marshall, S. L. A. 1953 The River and the Gauntlet: Defeat of the Eighth Army by the Chinese Communist Forces, November, 1950, in the Battle of the Congchon River, Korea. New York: Morrow.
Merton, Robert K. 1957 Priorities in Scientific Discovery: A Chapter in the Sociology of Science. American Sociological Review 22:635-659.
Meyerson, Martin; and Banfield, Edward C. 1955 Politics, Planning, and the Public Interest: The Case of Public Housing in Chicago. Glencoe, 111.: Free Press. → A paperback edition was published in 1965.
Milbrath, Lester W. 1963 The Washington Lobbyists. Chicago: Rand McNally.
Millikan, Max F. 1959 Inquiry and Policy: The Relation of Knowledge to Action. Pages 158-180 in Daniel Lerner (editor), The Human Meaning of the Social Sciences. New York: Meridian.
Moyzisch, L. C. 1950 Operation Cicero. New York: Coward-McCann.
Myers, Charles A.; and Turnbull, John G. 1956 Line and Staff in Industrial Relations. Harvard Business Review 34, no. 4:113-124.
Neustadt, Richard E. 1960 Presidential Power: The Politics of Leadership. New York: Wiley. → A paperback edition was published in 1962 by Science Editions.
Pettee, George S. 1946 The Future of American Secret Intelligence. Washington: Infantry Journal.
Platt, Washington 1957 Strategic Intelligence Production: Basic Principles. New York: Praeger.
Possony, Stefan T. 1941 Organized Intelligence: The Problem of the French General Staff. Social Research8:213-237.
Price, Don K. (1954) 1961 Government and Science: Their Dynamic Relation in American Democracy.New York Univ. Press. → A paperback edition was published in 1962.
Ransom, Harry Howe 1958 Central Intelligence and National Security. Cambridge, Mass.: Harvard Univ. Press.
Sayles, Leonard R. 1958 Behavior of Industrial Work Groups: Prediction and Control. New York: Wiley.
Schattschneider, Elmer E. 1935 Politics, Pressures,and the Tariff: A Study of Free Private Enterprise in Pressure Politics, as Shown in the 1929-1930 Revision of the Tariff. Englewood Cliffs, N.J.: Prenticehall.
Schlesinger, Arthur M. JR. 1959 The Age of Roosevelt. Volume 2: The Coming of the New Deal. Boston: Houghton Mifflin.
Schlesinger, Arthur M. JR. 1965 A Thousand Days: John F. Kennedy in the White House. Boston: Houghton Mifflin.
Schumpeter, Joseph A. (1942) 1950 Capitalism, Socialism, and Democracy. 3d ed. New York: Harper; London: Allen & Unwin. → A paperback edition was published by Harper in 1962.
Seabury, Paul 1954 The Wilhelmstrasse: A Study of German Diplomats Under the Nazi Regime. Berkeley: Univ. of California Press.
Selznick, Philip 1949 TVA and the Grass Roots: A Study in the Sociology of Formal Organization. University of California Publications in Culture and Society, Vol. 3. Berkeley: Univ. of California Press.
Selznick, Philip 1957 Leadership in Administration: A Sociological Interpretation. Evanston, 111.: Row, Peterson.
Shepard, Herbert A. 1956 Nine Dilemmas in Industrial Research. Administrative Science Quarterly 1: 295-309.
Sherwood, Robert E. (1948) 1950 Roosevelt and Hopkins: An Intimate History. Rev. ed. New York: Harper.
Shils, Edward 1956 The Torment of Secrecy: The Background and Consequences of American Security Policies. Glencoe, 111.: Free Press.
Shultz, George P.; and Whisler, Thomas L. (editors) 1960 Management Organization and the Computer: Proceedings of a Seminar Sponsored by the Graduate School of Business, University of Chicago and the McKinsey Foundation. Chicago, University of, Graduate School of Business, Studies in Business. Glencoe, 111.: Free Press.
Simon, Herbert A. 1960 The New Science of Management Decision. Ford Distinguished Lectures, Vol. 3. New York: Harper.
Simon, Herbert A.; Smithburg, Donald W.; and Thompson, Victor A. 1950 Public Administration. New York: Knopf.
Smith, Bruce L. R. 1964 Strategic Expertise and National Security Policy: A Case Study. Volume 13, pages 69-106 in J. D. Montgomery and A. Smithies (editors), Public Policy: A Yearbook of the Graduate School of Public Administration, Harvard University.Cambridge, Mass.: Harvard Univ. Press.
Smith, Richard A. 1956 Business Espionage. Fortune 53, no. 5:118-121, 190, 192, 194.
Snyder, Richard C ; and PAIGE, GLENN D. 1958 The United States Decision to Resist Aggression in Korea: The Application of an Analytical Scheme. Administrative Science Quarterly 3:341-378. → Reprinted in Richard C. Snyder et al., Foreign Policy Decisionmaking, 1962.
Sorensen, Theodore C. 1963 Decision-making in the White House: The Olive Branch or the Arrows. New York: Columbia Univ. Press.
Sorensen, Theodore C. 1965 Kennedy. New York: Harper.
Speier, Hans 1952 Social Order and the Risks of War. New York: Stewart.
Thompson, V. 1950 The Regulatory Process in OPA Gas Rationing. New York: King’s Crown Press.
Tuchman, Barbara 1962 The Guns of August. NewYork: Macmillan.
U.S. Congress, Senate, Committee on the Judiciary, Subcommittee ON Administrative Practice AND Procedure 1965 Invasions of Privacy: Government Agencies. Parts 1-3. Hearings held February 18March 3, 1965; April 13-June 7, 1965; and July 13 -August 9, 1956. Washington: Government Printing Office.
Urwick, Lyndall (1937) 1954 Organization as a Technical Problem. Pages 49-88 in Luther H. Gulick et al. (editors), Papers on the Science of Administration.New York: Columbia Univ., Institute of Public Administration.
Vagts, Alfred 1956 Defense and Diplomacy: The Soldier and the Conduct of Foreign Relations. New York: King’s Crown Press.
Ward, Benjamin 1967 The Socialist Economy: A Study of Organizational Alternatives. New York: Random House.
Wasserman, Benno 1960 The Failure of Intelligence Prediction. Political Studies 8:156-169.
Whyte, William F. et al. 1955 Money and Motivation: An Analysis of Incentives in Industry. New York: Harper.
Wildavsky, Aaron B. 1964 The Politics of the Budgetary Process. Boston: Little.
Wilensky, Harold L. 1956 Intellectuals in Labor Unions: Organizational Pressures on Professional Roles.Glencoe, 111.: Free Press.
Wilensky, Harold L. 1964 The Professionalization of Everyone? American Journal of Sociology 70:137158.
Wilensky, Harold L. 1967 Organizational Intelligence: Knowledge and Policy in Government and Industry.New York: Basic Books.
Wiles, P. J. D. 1962 The Political Economy of Communism. Cambridge, Mass.: Harvard Univ. Press.
Wohlstetter, Roberta 1962 Pearl Harbor: Warning and Decision. Stanford Univ. Press.
A formal organization consists of a set of people who are engaged in activities coordinated by the relatively consistent expectations that these people have about one another and about the purposes of the organization. Such expectations define a set of organizational statuses or offices, each status with a set of roles linking it to the other statuses with which it interacts. Organizational research examines how people behave in their organizational roles and how organizations behave as collective units. Here we will consider the types of data collection and measurement, and research design and analysis used in organizational research.
Organizational research uses three main sources of data: qualitative observation and interviewing, surveys of organization members, and institutional records. Qualitative methods of data collection are described elsewhere [seeField work]. Surveys within organizations take three forms: surveys of one stratum only; surveys of two or more strata within the organization (for instance, workers and management, or students and faculty); and “relational surveys” of linked pairs or sets of role partners. In surveys of two or more organizational strata (“multistratum surveys“) individuals are usually sampled at random within status groups and are identified only by the general status each one occupies. In “relational surveys” each individual is identified by the specific others with whom he interacts, and the sample is set up to include these role partners. When a sample of individuals is asked to provide the information on their role partners, who are not themselves interviewed, we have a “pseudo-relational survey.”
Institutional data include records made by organizations for their own use; information from directories, which can be turned into “data banks”; and “institutional questionnaires,” sent to samples of organizations, to befilledout by one or more key informants in each organization.
The formal aspect of measurement that is most significant for organizational research is the relation of the units from which the basic data are gathered to those which are characterized by the measurement. Information can be gathered on the organization as a whole or on component parts of it, such as individuals, interacting pairs and sets, and subgroups of members. Data on component parts can be aggregated in various ways to characterize the organization; and data on the organization can be considered as a “contextual characteristic” of the members. The following types of characteristics of collectives and their members have been distinguished by Lazarsfeld and Menzel (1961), and further discussed by Barton (1961, pp. 2-3 and appendix 1).
Integral or global characteristics of a unit do not derive from aggregation of members, pairs, or sets within that unit, but from properties of the unit as a whole. In the case of an organization, examples of such properties are its physical equipment, formal rules, budgets, programs, collective events, and collective outputs.
Relational characteristics derive from information concerning the relationship between a unit and other units. In the case of an individual, they include popularity, measured by number of choices received on sociometric questions; participation in an occupational community, measured by the number of friends who belong to the same occupation; supportiveness of political environment, measured by the proportion of his associates who vote the same way as the respondent; cosmopolitanism, measured by the number of extraorganizational contacts he has; and so on. For collective units, relational characteristics would include the amount of communication between one unit and various others in its environment, the frequency of cooperation or conflict with other units, the volume of economic transactions between them, and so on (Levine & White 1961; Litwak & Hylton 1962).
The underlying model for relational data is a “who-to-whom” matrix showing the value of the relationship for each pair within a group. There may be different types of relationships (who likes whom, who gives orders to whom, who talks shop with whom, etc.), each of which generates a matrix[SeeSociometry]. Summary scores for individuals are found in the marginal totals of each matrix. Other operations can give us the number of secondorder and higher-order connections which an individual has (through his immediate partners) and the characteristics of his interpersonal environment. Organizations likewise have higher-order connections and an interorganizational environment.
Aggregate measures characterize a collective in terms of distributions of data concerning its members. The simplest aggregate measures are additive properties, such as rates and means. Thus the morale of army units may be measured by the proportion of their men who feel strong loyalty to the unit; the political climate of a college by the proportion of its faculty members who are liberal; the ability level of a school by the mean IQ of its students. In these cases, as we would expect, there is a simple correspondence between the individual and the collective properties; relationships true on the individual level are duplicated on the organizational level. However, this kind of correspondence does not always exist, as discussions of the “ecological fallacy” have shown (Robinson 1950; Duncan & Davis 1953; Goodman 1953). Additive measures can also be used as indicators of more complex organizational characteristics that are not simply the sum of individual traits: for example, the proportion of an organization’s total personnel that is in administrative positions may indicate how bureaucratized the organization is.
A peculiar type of aggregate property arises when researchers ask samples of members for their perceptions of the organization or parts of it, and add up their answers to obtain a kind of aggregate perception. For example, an army company may be classified as having authoritarian leadership if a large proportion of members report that the leader behaves in an authoritarian way (Selvin 1960). A problem here is that different groups within an organization may evaluate the same behavior differently: Halpin (1956) found school principals reporting themselves as high on “consideration” for teachers; the school boards agreed with them, but the teachers themselves did not.
The variance of the distribution of member data may also be used to characterize the collective. Thus the range or standard deviation of the incomes of an organization’s members can be used as a measure of the organization’s equalitarianism, and the same measures, when applied to the ability levels of a high school class or the values of members of a union local, can indicate the degree of heterogeneity in the class or of value consensus in the union. Equality, homogeneity, and consensus are emergent properties on the collective level, without individual counterparts.
Finally, the correlations between variables within a collective can be used as aggregate measures. For instance, the correlation of values with rank in the American army indicates its degree of “value stratification” (Speier 1950). Similarly, the correlation of military rank with external social status measures the degree of ascription as compared with achievement in the organization’s promotion system.
The members of the collective from which data are aggregated need not be individuals; they may be smaller collectives within the larger one. Thus an aggregate measure of the degree of local democracy within national unions might be the proportion of locals within a union that had contested elections.
When we have data from several status groups within a number of organizations, we can obtain aggregate measures for each stratum. An organization might thus be characterized as having high morale in the upper strata but low morale among the rank and file, as having authoritarian top managers and democratic foremen, and so on. Thedifferences between different strata, in rates or mean scores on various attitudes, behaviors, background data, etc., are actually a form of correlational index.
Relational-pattern measures (sometimes termedstructural or sociometric measures) are aggregations of relational data on the members of a collective unit. Group cohesion is often measured in this way by the ratio of in-group to out-group choices when each member of a group is asked whom he chooses or would choose as a close friend; group integration may be indicated by the over-all frequency with which group members communicate with one another [seeCohesion, Social]. More complex measures would include the extent to which relations within a group form self-contained cliques. Patterns of relationship among subunits—for example, the amount of conflict between subgroups in an organization—can also characterize a collective. Just as relational characteristics of individuals are indicated by their row or column scores in a who-to-whom matrix, relational pattern measures for a collective derive from the matrix as a whole—that is, from the distribution or patterning of pair relations.
Contextual measures arise when we characterize a member by the properties of the collective of which the member is a part. Being dissatisfied is an integral property of the individual; being a member of a work group where a high proportion are dissatisfied is a contextual property, derived from the aggregate of individual properties of members. Being a member of a conflict-ridden organization is a contextual property derived from a relational-pattern property of the collective; attending a college with a large library is a contextual property derived from an integral property of the collective.
In a multistratum survey of many organizations, contextual data can also be derived from different strata. For example, aggregate data on university faculty are contextual for students, and characteristics of foremen are contextual for the workers under them.
Organizations, too, have contexts, such as the larger organization of which they are part or the industry or community in which they are located. Locals may be part of a “democratic” or an “undemocratic” national union; business firms may be located in communities with growing or shrinking populations, or belong to competitive or concentrated industries.
In summary, individual members of an organization have integral, relational, and contextual properties. Collectives likewise have integral properties, relational properties (with regard to other organizations), and contextual properties (those of larger collectivities of which they are members). But in addition, collectives have aggregate properties derived from their members’ integral characteristics, and relational-pattern properties derived from their members’ relationships.
We have distinguished five main sources of data: qualitative observation, one-stratum surveys, multistratum surveys, relational surveys, and institutional data; this is the first of three dimensions on which research designs can be classified. A second dimension is the number of organizational units studied: “case studies” of one organization, “comparative studies” of two or several organizations; and “large-sample studies” of enough organizations to make statistical analyses with organizations as units of analysis. (Either a whole organization or a formal subunit can be the focus of study; a study of 100 work groups within a large corporation would be a “large-sample study” of work groups.) The third dimension is whether data are gathered for one point in time or whether comparable data are gathered for each of two or more time periods. Studies over time permit analysis of change and inferences of causal relationships.
A typology of designs for organizational studies is generated if we run these three dimensions against one another (see Table 1, in which each row of six cells derives from a different source of data).
Analysis of data on organizations
The different types of research design outlined in Table 1 permit several basic types of analysis, some qualitative and some quantitative. Each type of analysis can be performed for units at different levels of aggregation—for individuals, for linked sets of individuals, or for organizations. We will examine each type of analysis in turn, indicating its limitations as well as the problems that it permits us to study.
A study of one organization at one point in time, using qualitative methods, provides a very limited scope for analysis. Such analysis usually involves comparing the case at hand with a theoretically derived ideal type—Weber’s model of bureaucracy, for instance, or the economists’ model of a firm composed of rational economic men. Thus, an analysis of bureaucratic patterns in the navy officer corps describes a number of striking behavior patterns (avoiding responsibility, ritualism, insulation from the outside world, and ceremonialism) and concludes: “The military variant of bureaucracy may thus be viewed
|Table 1 - A typology of designs of organizational studies|
|TIME PERIODS||NUMBER OF UNITS STUDIED|
|1||Qualitative case study||Qualitative comparative study||Many qualitative case studies|
|Davis 1948||Coser 1958||Udy 1959|
|Form & Nosow 1958|
|2+||Qualitative case study over time||Qualitative comparative study over time||Many qualitative case studies over time|
|Selznick 1949||Guest 1962||——|
|Roethlisberger & Dickson 1939|
|1||One-stratum one-unit survey||One-stratum comparative survey||One-stratum survey of a large sample of organizations|
|Walker & Guest 1952||Katz & Hyman 1947||Lazarsfeld & Thielens 1958|
|2+||One-stratum one-unit panel survey||One-stratum comparative panel survey||One-stratum panel survey in a large sample of organizations|
|Stouffer et al. 1949||——||——|
|1||Multistratum one-unit survey||Multistratum comparative survey||Multistratum survey of a large sample of organizations|
|Stouffer et al. 1949||Georgopoulos & Mann 1962|
|Illinois, University of 1954|
|Mann & Hoffman 1960|
|2+||Multistratum one-unit panel survey||Multistratum comparative panel survey||Multistratum panel survey of a large sample of organizations|
|1||One-unit relational survey||Comparative relational survey||Relational survey of a large sample of organizations|
|Weiss 1956||——||Kahn & Katz 1953|
|Organizational Stress 1964||Gross etal . 1958|
|Stogdill e t al. 1956|
|2+||One-unit relational panel survey||Comparative relational panel survey||Relational panel survey of a large sample of organizations|
|W.Wallace 1964||Morse & Reimer 1956||——|
|A. F. C. Wallace & Rashkis 1959|
|1||Case study using one-time institutional data||Comparative study using one-time institutional data||Study of a large sample of organizations using one-time institutional data|
|——||Harbison et al. 1955||Faunce 1962|
|2+||Case study using institutional data over time||Comparative study using institutional data over time||Study of a large sample of organizations using institutional data over time|
|Brown 1956||Haire 1959||Lipset et al. 1956|
as a skewing of Weber’s ideal type by the situational elements of uncertainty and standing by” (Davis  1952, p. 384). Similarly, the authors of the famous Western Electric study (Roethlisberger & Dickson 1939) sought to discover, by means of qualitative interviewing and observation, whether factory workers behaved like a set of discrete and rational economic units and concluded, when faced with such phenomena as group production norms and informal leaders, that they did not.
Single case studies thus can disclose the existence of phenomena that raise problems for the theories from which ideal types have been generated. Such studies therefore inspire further research of more complex design to find the conditions under which the ideal-typical or the deviant phenomena occur.
Multivariate comparison of a few cases
When we have data on two or several organizations—whether these data are qualitative, aggregated from surveys, or compiled from institutional records—we can use some form of “quasi-experimental” analysis (also known as “quasi-correlational” analysis). The simplest form is “strategic paired comparison,” a method that calls for two organizations that are formally similar but differ strongly in one respect. The presumed consequences of this difference are then explored in detail. Thus Coser (1958) compared a medical ward with a surgical ward in the same hospital; the wards, though formally similar, presented many differences in the actual exercise of authority and the informal relations of doctors and nurses. Mann and Hoffman (1960), using a survey of top managers, foremen, and workers, compared an automated power plant with a non-automated one. Treating the results as a “quasiexperiment,” they suggest that automation increased the men’s sense of having an influence on plant operations, as well as their interest and satisfaction, but also aggravated tensions related to the job.
In these two examples, the researchers translated a gross institutional characteristic (medical versus surgical, automated versus nonautomated) into several more general sociological characteristics (time pressure on decision making, job enlargement, interdependence of components), which were capable of explaining the observed differences in the dependent variables in terms of general sociological propositions (for example, when there is more time pressure on decision making, there will be less group consultation). Having only two cases, however, they could not test the relative effects of the several proposed explanatory variables or control for the effects of other variables.
When researchers have somewhat more cases, they can locate each organization in a multidimensional classification of explanatory variables. Thus a study of six organizations’ responses to disaster found seven characteristics that differentiated the more effective from the less effective ones; each organization represented a particular pattern of these seven attributes (Form & Nosow 1958). Such an analysis is suggestive but not conclusive, since there is only one case per cell.
Qualitative study of change
Most qualitative case studies that go beyond description of formal structure necessarily examine time sequences of the behavior of organization members. These may disclose the “microprocesses”—the exercise of informal pressures, the ways of getting around rules, the immediate causes of deviance—that maintain normal equilibrium in the organization. Many examples of such microprocesses were observed in the Western Electric study. A worker who went beyond the group norm of output was ridiculed or “binged” on the arm by fellow workers. A researcher entering the room was mistaken for a time-study man, and everyone slowed down. Changes in production often followed off-the-job personal problems (Roethlisberger & Dickson 1939). More recently, it was found that episodes of patient disturbance in a mental hospital followed staff disagreements on the patients’ treatment (Stanton & Schwartz 1954).
The logic of sequence analysis is post hoc, ergo propter hoc; but this kind of reasoning can be misleading in the absence of experimental or statistical controls to eliminate accidental or spurious factors. Researchers therefore try to test their interpretations by observing as many repeated sequences as they can and by locating changes precisely in the time sequence. This procedure can become a qualitative form of time-series experiment [seeExperimentalDesign, article onQuasi-Experimental design; see also Campbell & Stanley 1963].
Qualitative studies over long time periods, particularly if they cover major organizational changes such as succession, growth, and reorganization, permit derivation of relationships between organizational variables. Such studies also test hypotheses about the functions of various parts of the system by observing what happens when one part of it is changed. For instance, McCleery (1957) studied the process of change in a prison from an authoritarian to a more liberal regime. His main argument was that the arbitrary behavior of the old regime created such insecurity that the inmates accepted an exploitative elite of “old cons” who interpreted and negotiated with the authorities, and who controlled disorderly inmate behavior that might have jeopardized their own special privileges. At the same time, this administrative arbitrariness made the inmates hostile to the official programs of rehabilitation. When the new administration created fairer procedures and more communication with the inmates, this hostility was reduced, but the power of the inmate elite was reduced still faster, so that an outburst of disorder resulted. A new equilibrium was finally achieved without an exploitative elite and with more rehabilitative activity.
Several operations can be distinguished here: identifying the “anatomy” of the system—the major formal and informal status groups; identifying the key variables characterizing each group; reporting the values of these variables at several stages in the process of change; deriving causal relationships among them; and locating these causal relationships in a functional model of the whole system (see Figure 1). This model can be analyzed to account for the initial equilibrium, the sequence of changes, and the final equilibrium (Barton & Anderson 1961).
That such a system of relationships was found to exist in a single case is hardly conclusive evidence that it exists in all such cases. Comparative study of several organizations can begin to provide checks against alternative possibilities and to specify
conditions under which the relationships hold. Thus, Guest (1962) compared two studies of succession of a top manager and proposed no fewer than seven variables in an effort to explain why the two men adopted opposite policies.
Multivariate analysis of individuals
A survey of individual attitudes and behavior within a single organization can analyze the correlates of job satisfaction, productivity, mobility aspirations, or any other organizationally related attitudes and behaviors, in exactly the manner of ordinary publicopinion studies, except that detailed information on organizational status and activities can be added to the usual limited background data. For instance, Stouffer and his colleagues (1949) analyzed the relation of rank, combat experience, length of service, and similar organizational-status variables to morale, mobility aspirations, etc., for the army as a whole and for various components.
Multivariate analysis of linked sets
Analysis of characteristics of pairs or sets of role partners is made possible by relational samples. This is easiest when the role set corresponds to an organizational subunit—as in studies of the correlation of supervisor behavior with worker morale or productivity (see, for instance, Kahn & Katz 1953). In Hall’s study (1955) of 40 air crews, behavior of each commander was described for three variables by the aggregate perceptions of his crewmen; each crew’s role prescriptions for their commander were measured for the same three variables; and a measure of the conformity of commander behavior to crew prescriptions was derived. His conformity could then be correlated with other attributes of his crew, such as its cohesion and degree of consensus.
Where the role set does not correspond to an organizational subunit, “snowball sampling” can be used. For example, Kahn and his associates(Organizational Stress . . . 1964) took 53 managers as “focal persons” and interviewed two superiors, two subordinates, and three pairs of each. Tensions of the focal person could then be related to characteristics of his role set. The variety of measures that can be created with such samples is shown by a study by Stogdill and his colleagues (1956), who surveyed 47 role sets of a superior and one or two subordinates in a research agency, measuring role expectations and perceived role behavior for self and other for each of 45 role elements. From these basic data they derived measures of each pair’s consensus on role expectations, conformity of behavior to own and partner’s expectations, agreement on perceptions of behavior, and differentiations or division of labor between the two statuses, both perceived and expected.
Multivariate analysis of organizations
Whenever we have data on a large number of organizations or subunits within organizations, it is possible to apply the same methods of analysis that are used in survey analysis of individual data. The organizational data may derive from a large set of qualitative studies containing comparable data, from surveys that have sufficient samples within a large number of organizations to provide aggregate data, or from institutional data derived from records, “data banks,” and surveys of informants by means of institutional questionnaires. Thus Udy (1959) classified 82 production organizations, using data from anthropological monographs compiled in the Human Relations Area Files. He found a high correlation between the degree of complexity of their technology and the number of levels of authority (see Table 2).
Blau and his colleagues (1966) studied a large sample of public bureaucracies with an institutional questionnaire which measured size, division of labor, professionalization, and centralization of authority. Quite complex conditional relationships were found; for example, in organizations with few professionals, division of labor was positively related to centralization of authority, while in highly professionalized organizations the relationship was the reverse.
|Table 2 —Authority and technological complexity in 82 organizations|
|INDEX OF TECHNOLOGICAL COMPLEXITY|
|Number of levels|
|Source: Udy 1959, p. 584.|
|3 or more||28||3|
|2 or less||5||46|
Very large samples of organizations can be studied by using published directories, data banks of institutional information, or questionnaires to informants in each unit. Thus, Faunce (1962) studied 753 union locals by passing out a questionnaire at the U.A.W. convention, and Douglass (1926) studied 1,000 city churches by using informant questionnaires.
Contextual analysis of individuals
When we have drawn samples of individuals within several or many organizations, contextual analysis becomes possible. The simplest form of this is an examination of the relationship between individual attitudes or behavior and the attributes of the organization of which the individuals in question are members. Thus, students report more cheating at colleges that are large, have many students per faculty member, have most students living off campus, have easy admissions policies, and are coeducational (Bowers 1964). Moreover, it can be shown by means of three-variable “contextual tables” that the relationship between two individual attributes is different in different organizational contexts, and thus that there is an organizational factor influencing the processes of individual behavior. For instance, Lipset and his colleagues, in a study of a large local in the International Typographical Union, surveyed two strata: chapel chairmen and rank-and-file members. Over-all, the chairmen were more interested, knowledgeable, and active in union politics than the rank and file were. However, when union members were separated according to size of the shop, it was found that this difference between chairmen and rank and file appeared only in the larger shops (Lipset et al. 1956, pp. 176-182).
Similar effects were obtained in a study of three prisons by Berk (1966).
Contextual analysis of linked sets
Relationships between characteristics of pairs of role partners or between those of whole role sets may be modified by their organizational context. To study this requires comparing relational surveys in a number of organizations—for example, examining the relation of informal contact to consensus between role partners in organizations of different degrees of bureaucr atization.
Contextual analysis of organizations
The larger setting—type of community, industry, or society—may affect the characteristics of organizations or the interrelations among them (Udy 1965). Crossnational comparisons show, for instance, that the ratio of supervisors to workers may be 1 to 15 in an American steel plant and 1 to 50 in a similar German steel plant (Harbison et al. 1955; see also Evan 1963). Presumably the relationship between plant size and the supervisor-worker ratio is also different in each case. A complex contextual effect is exemplified by the finding that the relationship between secular orientation and effectiveness is positive for YMCAs at nondenominational colleges, but negative for those at denominational colleges (Lucci 1960).
When information is available on several variables within one organizational unit for many points in time, it is possible to search for causal relationships by examining the sequence of changes in these variables. Brown (1956) examined union records covering a period of 50 years. Indicators of intraunion conflict—number of resolutions for constitutional change, challenges to convention delegates, failure of officers to be re-elected, jurisdictional disputes between locals—were found to be inversely related to size year by year during a cycle of growth, decline, and new growth. Tsouderos (1955) used time series of income, expenditure, capital, membership, and number of administrative employees to study the process of growth, bureaucratization, and decline in ten voluntary organizations. His finding was that loss of members after a period of growth led to increased administrative staff, which led to maintenance of income and activities but further loss of members and to eventual decline.
Time series could also be provided by survey data on several organizational strata repeated over many periods. Some large corporations collect data of this kind as a “morale barometer,” but it does not seem to be systematically analyzed in the way that public-opinion time series have been.
Panel analysis of individuals
A design that obtains information on a sample of organization members at two or more points in time permits a much clearer isolation of causal relationships among individual characteristics than does a onetime survey. For instance, Stouffer and his associates found that higher-ranking enlisted men were more often in good spirits, accepted the soldier role, were more satisfied with their army job, and thought the army was well run. To test whether these conformist attitudes were a cause of promotion or only a result of being promoted, samples of privates were surveyed when newly recruited and subsequently followed up after several months during which some had been promoted. Those with more conformist attitudes early in their army careers were more likely to be promoted later on. The difference held good when various background factors related to both promotion and attitudes were held constant (Stouffer et al. 1949, vol. 1, pp. 147-154).
Another study dealt with workers who became foremen or shop stewards, repeating on them and on matched control groups a set of attitude questions given a year previously to all workers in the plant. A third wave was done two years later, when some of the foremen had been demoted because of layoffs and some of the stewards had been replaced in office. Many attitudes had changed considerably in response to changes in status (Lieberman 1956).
A panel study that included many organizations would permit comparison of social processes in different settings. Such studies do not appear to exist. (See, however, Miller 1958 for an example of a “pseudo-panel” study which compares students in different years of college for different types of institutions, suggesting major differences in the socialization process.)
Panel analysis of linked sets
If, as is often said, organizations are systems of interacting parts, the best design for studying them is a relational panel, since this would permit analysis of changes among related individuals and groups over time. But studies that make use of such panels are remarkably rare. We have mentioned the qualitative analysis by Stanton and Schwartz (1954) of patient disturbance and staff disagreement in a mental hospital. One quantitative panel study of staff and patients found no such relationship (Wallace & Rashkis 1959), but another found not only that staff disagreement was associated with subsequent patient disturbance but also that patient disturbance led to staff disagreement (Rashkis & Wallace 1959).
Walter Wallace (1964; 1965) obtained a complete sociogram of contacts among students at one college. He also obtained panel attitude data from freshmen at entry, at midsemester, and at the end of the first semester, and attitude data on all other students at midsemester. He constructed measures of the predominant attitude of each freshman’s “interpersonal environment,” which he related to change in freshman attitudes. Segments of the interpersonal environment, classified by sex and college class, were shown to have differential effects.
Relational panel analysis would be particularly useful in studying role relationships as a system, since it would permit examination of such phenomena as the effects on the whole network of expectations and behavior of a change in one partner’s role expectations. But such studies do not appear to exist.
Panel analysis of organizations
If standardized data were gathered for a sample of organizations over time, the determinants of organizational change could be analyzed by the same methods that are used in analyzing panels of individuals. The basic data might be from any of the sources we have discussed—qualitative studies, surveys, or institutional records—provided they measure the same characteristics of a large number of organizations at several comparable points in time. A near example of this type of analysis is the finding by Stouffer and his colleagues that companies which before D-day had the highest willingness for combat subsequently had the lowest rates of nonbattle casualties. The relationship was much higher for veteran regiments than for nonveteran regiments (Stouffer et al. 1949, vol. 2, chapter 1).
A comparable approach was taken by Lipset and his colleagues, who analyzed time-series data on party voting in the printers’ union, broken down by large versus small locals. They were able to show that a rising opposition normally gains strength first in the large locals, while the small ones support the party in power; when the party supported by the large locals wins, these locals are again the first to go into opposition (Lipset et al. 1956, pp. 373-382). But, once again, such studies are hardly ever undertaken.
For purposes of developing theories of organizational processes at the microlevel, relational panel studies obtaining data on role expectations and behavior would appear to be particularly appropriate. They would provide factual data on which to base development of simulation models and mathematical models, just as panel studies of voting behavior (including data on the perceivednorms and behaviors of friends and associates and of various social groups) did for the development of simulation models of electoral processes [seeSimulation, article onPolitical Processes; see alsoVoting].
For testing theories of organizational change, panel data on organizational characteristics for large samples of organizations would be highly desirable. We now have a good many examples of studies of large samples of organizations, but because they deal with the organizations at only one point in time, they are severely limited in establishing causal relations.
Allen H. Barton
[Directly related are the entries Experimental design; Field work; Observation; Panel studies. Other relevant material may be found in Administration,article on Administrative behavior; Counted data; Industrial relations, article onthe sociology of work; Interviewing, article on Social research; Multivariate analysis; and in the biographies of Mayo; Stouffer.]
Barton, Allen H. 1961 Organizational Measurement and Its Bearing on the Study of College Environments.New York: College Entrance Examination Board.
Barton, Allen H.; and Anderson, bo 1961 Change in an Organizational System: Formalization of a Qualitative Study. Pages 400-418 in Amitai Etzioni (editor),Complex Organizations: A Sociological Reader. New York: Holt.
Berk, Bernard B. 1966 Organizational Goals and Inmate Organization. American Journal of Sociology 71: 522-534.
Blau, Peter M. 1960 Structural Effects. American Sociological Review 25:178-193.
Blau, Peter M.; Heydebrand, Wolf V.; and Stauffer, Robert E. 1966 The Structure of Small Bureaucracies. American Sociological Review 31:179-191.
Bowers, William J. 1964 Student Dishonesty and Its Control in College. Cooperative Research Project No. OE 1672. Unpublished manuscript, Columbia Univ., Bureau of Applied Social Research.
Brown, Julia S. 1956 Union Size as a Function of Intra-union Conflict. Human Relations 9:75-89.
Campbell, Donald T.; and Stanley, Julian C. 1963 Experimental and Quasi-experimental Designs for Research on Teaching. Pages 171-246 in Nathaniel L. Gage (editor), Handbook of Research on Teaching.Chicago: Rand McNally.
Coser, Rose L. 1958 Authority and Decision-making in a Hospital: A Comparative Analysis. American Sociological Review 23:56-63.
Davis, Arthur K. (1948) 1952 Bureaucratic Patterns in the Navy Officer Corps. Pages 380-395 in Robert K. Merton et al. (editors), Reader in Bureaucracy.Glencoe, 111.: Free Press. → First published in Volume 27 of Social Forces.
Douglass, H. 1926 1000 City Churches: Phases of Adaptation to Urban Environment. New York: Doran.
Duncan, Otis Dudley; and Davis, Beverly 1953 An Alternative to Ecological Correlation. American Sociological Review 18:665-666.
Evan, William M. 1963 Indices of the Hierarchical Structure of Industrial Organizations. Management Science 9:468-478.
Faunce, William A. 1962 Size of Locals and Union Democracy. American Journal of Sociology 68:291298.
Form, William H.; and Nosow, Sigmund N. 1958 Community in Disaster. New York: Harper.
Georgopoulos, Basil S.; and Mann, Floyd C. 1962 The Community General Hospital. New York: Macmillan.
Goodman, Leo A. 1953 Ecological Regressions and Behavior of Individuals. American Sociological Review18:663-664.
Gouldner, Alvin W. 1954 Patterns of Industrial Bureaucracy. Glencoe, 111.: Free Press.
Gross, Neal; Mason, Ward S.; and McEachern, Alexander W. 1958 Explorations in Role Analysis: Studies of the School Superintendency Role. New York: Wiley.
Guest, Robert H. 1962 Managerial Succession in Complex Organizations. American Journal of Sociology68:47-53.
Haire, Mason 1959 Biological Models and Empirical Histories of the Growth of Organizations. Pages 272-306 in Foundation for Research on Human Behavior, Modern Organization Theory: A Symposium. Edited by Mason Haire. New York: Wiley.
Hall, Robert L. 1955 Social Influences on the Aircraft Commander’s Role. American Sociological Review 20: 292-299.
Halpin, Andrew W. 1956 The Leadership Behavior of School Superintendents: The Perceptions and Expectations of Board Members, Staff Members and Superintendents. Columbus: Ohio State Univ., College of Education.
Harbison, Frederick H. et al. 1955 Steel Management on Two Continents. Management Science 2:31-39.
Illinois, University OF, Institute OF Labor AND Industrial Relations 1954 Labor-Management Relations in Illini City, by William E. Chalmers et al. Champaign, III.: The Institute.
Kahn, Robert L.; and Katz, Daniel (1953)1960 Leadership Practices in Relation to Productivity and Morale. Pages 554-570 in Dorwin Cartwright and Alvin F. Zander (editors), Group Dynamics: Research and Theory. 2d ed. Evanston, 111.: Row, Peterson.
Katz, Daniel; and Kahn, Robert L. 1966 The Social Psychology of Organizations. New York: Wiley.
Katz, David; and Hyman, Herbert H. 1947 Morale in War Industries. Pages 437-448 in Society for the Psychological Study of Social Issues, Readings in Social Psychology. New York: Holt.
Lawrence, Paul R. 1958 The Changing of Organizational Behavior Patterns: A Case Study of Decentralization. Boston: Harvard Univ., Graduate School of Business Administration, Division of Research.
Lazarsfeld, Paul F.; and Menzel, Herbert 1961 On the Relation Between Individual and Collective Properties. Pages 422-440 in Amitai Etzioni (editor),Complex Organizations: A Sociological Reader. New York: Holt.
Lazarsfeld, Paul F.; and Thielens, Wagner JR. 1958 The Academic Mind: Social Scientists in a Time of Crisis. A report of the Bureau of Applied Social Research, Columbia University. Glencoe, 111.: Free Press.
Levine, Sol; and White, Paul E. 1961 Exchange as a Conceptual Framework for the Study of Interorganizational Relationships. Administrative Science Quarterly 5:583-601.
Lieberman, Seymour 1956 The Effects of Changes in Roles on the Attitudes of Role Occupants. Human Relations 9:385-402.
Lipset, Seymour M.; Trow, Martin A.; and Coleman, James S. 1956 Union Democracy: The Internal Politics of the International Typographical Union.Glencoe, III.: Free Press.
Litwak, Eugene; and Hylton, Lydia F. 1962 Interorganizational Analysis: A Hypothesis on Co-ordinating Agencies. Administrative Science Quarterly 6: 395-420.
Lucci, York 1960 The YMCA on the Campus. Unpublished manuscript, Columbia Univ., Bureau of Applied Social Research.
McCleery, Richard H. (1957) 1961 Policy Change in Prison Management. Pages 376-400 in Amitai Etzioni (editor), Complex Organizations: A Sociological Reader. New York: Holt.
Mann, Floyd C.; and Hoffman, L. Richard 1960 Automation and the Worker: A Study of Social Change in Power Plants. New York: Holt.
Miller, Norman 1958 Social Class and Value Differences Among American College Students. Ph.D. dissertation, Columbia Univ.
Morse, Nancy C.; and Reimer, Everett 1956 The Experimental Change of a Major Organizational Variable. Journal of Abnormal and Social Psychology 52: 120-129.
Newcomb, Theodore M. (1943) 1957 Personality and Social Change: Attitude Formation in a Student Community. New York: Dryden.
Organizational Stress: Studies in Role Conflict and Ambiguity, by Robert L. Kahn et al. 1964 New York: Wiley.
Rashkis, Harold A.; and Wallace, Anthony F. C. 1959
The Reciprocal Effect: How Patient Disturbance Is Affected by Staff Attitudes. Archives of General Psychiatry 1:489-498.
Robinson, W. S. 1950 Ecological Correlations and the Behavior of Individuals. American Sociological Review 15:351-357.
Roethlisberger, Fritz J.; an d Dickson, William J. (1939) 1961 Management and the Worker: An Account of a Research Program Conducted by the Western Electric Company, Hawthorne Works, Chicago.Cambridge, Mass.: Harvard Univ. Press. A paperback edition was published in 1964 by Wiley.
Selvin, Hanan C. 1960 The Effects of Leadership. Glencoe, III.: Free Press.
Sills, David L. 1957 The Volunteers: Means and Ends in a National Organization. Glencoe, III.: Free Press.
Speier, Hans 1950 The American Soldier and the Sociology of Military Organization. Pages 106-132 in Robert K. Merton and Paul F. Lazarsfeld (editors),Continuities in Social Research. Glencoe, 111.: Free Press.
Stanton, Alfred H.; and Schwartz, M. S. 1954 The Mental Hospital: A Study of Institutional Participation in Psychiatric Illness and Treatment. New York: Basic Books.
Stinchcombe, Arthur L. 1961 On the Use of Matrix Algebra in the Analysis of Formal Organization. Pages 478-484 in Amitai Etzioni (editor), Complex Organizations: A Sociological Reader. New York: Holt.
Stogdill, Ralph M.; Scott, Ellis L.; and Jaynes, william E. 1956 Leadership and Role Expectations.Bureau of Business Research, Monograph No. 86. Columbus: Ohio State Univ., College of Commerce and Administration.
Stouffer, Samuel A. et al. 1949 The American Soldier. Studies in Social Psychology in World War II, Vols. 1 and 2. Princeton Univ. Press. Volume 1:Adjustment During Army Life. Volume 2: Combat and Its Aftermath.
Tsouderos, John E. 1955 Organizational Change in Terms of a Series of Selected Variables. American Journal of Sociology 20:206-210.
Udy, Stanley H. Jr. 1959 The Structure of Authority in Non-industrial Production Organizations. American Journal of Sociology 64:582-584.
Udy, Stanley H. Jr. 1965 The Comparative Analysis of Organizations. Pages 678-709 in James G. March (editor), Handbook of Organizations. Chicago: Rand McNally.
Walker, Charles R.; and GUEST, Robert H. 1952 The Man on the Assembly Line. Cambridge, Mass.: Harvard Univ. Press.
Wallace, Anthony F. C ; and RASHKIS, HAROLD A. 1959 The Relation of Staff Consensus to Patient Disturbance in Mental Hospital Wards. American Sociological Review 24:829-835.
Wallace, Walter L. 1964 Institutional and Life-cycle Socialization of College Freshmen. American Journal of Sociology 70:303-318.
Wallace, Walter L. 1965 Peer Influences and Undergraduates’ Aspirations for Graduate Study. Sociology of Education 38:375-392.
Weiss, Robert S. 1956 Processes of Organization. Ann Arbor: Univ. of Michigan, Institute for Social Research, Survey Research Center.
"Organizations." International Encyclopedia of the Social Sciences. 1968. Encyclopedia.com. (August 25, 2016). http://www.encyclopedia.com/doc/1G2-3045000894.html
"Organizations." International Encyclopedia of the Social Sciences. 1968. Retrieved August 25, 2016 from Encyclopedia.com: http://www.encyclopedia.com/doc/1G2-3045000894.html
Formal organizations are ubiquitous in modern society. Most people work in organizations, join organizations, and interact with organizations in their daily lives. The study of organizations includes an examination of the relationship between individuals and organizations, organizations and their external environments, and how corporations and other constellations of organizations exercise power and influence public policy.
Organizational theorists maintain that the emergence of modern large-scale bureaucracy is an outcome of the increased application of formally rational means (e.g., written laws, rules, and regulations) to achieve the substantively rational goals (i.e., large-scale capitalism, democracy) associated with modernity. Although large-scale organizations can be used to increase the productive capability of capitalism and extend democracy to more spheres of social life, they also provide means for elites who control them to exercise power and advance their political and economic interests (Weber  1978). Scholars who examine organization control have shown that at the turn to the twentieth century capitalists hired scientific managers to experiment with techniques to increase control over the labor process. These scientific managers collected information from workers and centralized it in planning departments where engineers used it to establish formal rules to control the labor process (Braverman 1974).
In the middle decades of the twentieth century, decision-making theorists demonstrated that the capacity of elites to control managers in large and complex organizations is constrained by bounded rationality: limits on rationality that are affected by unclear preferences, limited information, and the cognitive abilities of individual decision makers (Simon 1957). These decision-making theorists showed that superordinates attempt to overcome bounded rationality and increase their control over subordinates by establishing premise controls, which limit the search for alternatives in the decision-making process (March and Simon 1958). Other organizational researchers point out that bounded rationality cannot be completely eliminated and that premise controls and other formal controls have unintended consequences (Perrow 1972; Burawoy 1979). Research in this tradition shows that formally rational controls caused widespread inefficiencies in the decision-making process in corporate American during the middle decades of the twentieth century. In response to the capital accumulation crisis in the 1970s and early 1980s, like the earlier period when elites increased their control over the labor process, capitalists and their top managers implemented scientific management principles to increase control over the managerial process. They hired engineers and other technical experts to collect information from production managers, centralize that information into computerized information processing systems, and use it to develop premise and other formal controls that limited the discretion of lower and middle-level managers (Prechel 1994).
Another line of research in organizational sociology examines the relationship between the internal organizational structures and the environment. There are two distinct types of organization-environment theories. The “weak form” of the organization-environment hypothesis stresses how organizations respond to environmental forces. Organizational ecology, which borrows from the biological model of natural selection, conceptualizes organizational populations as aggregates of organizations that are alike in some respect (e.g., newspapers). Organizations are considered part of a larger system that must adapt in order to survive. This perspective stresses the importance of structural isomorphism, which denotes a fit between the organization and its environment (Carroll and Hannan 1995). However, organizational adaptation is limited by inertia, which includes internal politics and increased age, size, and complexity. In the absence of adaptation to the environment, organizations will cease to exist.
Neoinstitutional theory also represents the weak form of the organization-environment hypothesis. This perspective examines organizational fields: a recognized area of institutional life with key suppliers, resource and product consumers, regulatory agencies, and other similar organizations (DiMaggio and Powell 1983). Once an organizational field is established, three mechanisms cause organizations to become homogeneous. First, mimetic behavior occurs when organizational technologies are poorly understood, when goals are ambiguous, or when the environment creates symbolic uncertainty. When these conditions exist, organizations model themselves on the dominant organization in their field. Second, normative behavior entails professionalization, which includes efforts by practitioners of an occupation to define the conditions and methods of their work, control the supply of new producers, and establish a cognitive base and legitimation for their occupational autonomy. Third, coercive pressures emerge from political and cultural influences. This neoinstitutional perspective assumes that an autonomous state creates public policies and laws (e.g., antitrust, product safety) and is capable of making corporations and other organizations comply with those laws.
In contrast, the “strong form” of the organization-environment hypothesis suggests that corporations and other organizations are capable of influencing public policies that are designed to control and regulate them. Resource dependence theory maintains that the environment constrains organizations, and organizations depend on other organizations for resources (Aldrich and Pfeffer 1976; Pfeffer and Salancik 1978). This perspective also maintains that the key to survival is the ability to acquire and maintain resources from other organizations in the environment. To reduce dependencies that limit their prospects for autonomous action and survival, organizations set up strategies to manage their environments.
The strong form of the organization-environment hypothesis is consistent with research by political sociologists who show that business elites use networks of organizations to influence public opinion and public policy (Domhoff 2006), and economic sociologists who maintain that the behavior of individuals and organizations can only be understood by examining the social relations in which they are embedded (Granovetter 1985). Other researchers maintain that the universalizing assumptions in many social theories can be misleading and that the strong form of the organization-environment hypothesis is historically contingent (Prechel 1990). Historical contingency theory of corporate political behavior maintains that although elites are always politically active, the level of elite political activity varies over time. When the condition of economic uncertainty is high (e.g., periods of low or unstable profits), corporations become more politically active, and they pressure state managers to implement policies to attain economic stability and predictability and to preserve the social relations in capitalist society (Prechel 2000). Although capitalists and their top managers do not always have a coherent conception of the relationship between their economic goals and the means to achieve those goals, they engage in political behavior to transform public policies in order to create the conditions that better advance their profit-making agendas.
Historical contingency theory explains the three major corporate form changes in the largest U.S. firms between the 1880s and 1990s. First, in the 1880s and 1890s, business elites lobbied states to pass formal laws to give industrial corporations the right to use the joint-stock holding company. This corporate form holds the majority of the stock (i.e., more than 50 percent) of legally independent subsidiary corporations. The embeddedness of the holding company in an environment with little regulatory control created opportunities for capitalists to consolidate many independent companies into giant companies such as AT&T, General Motors, and U.S. Steel Corporation. By the 1920s, stock overvaluation and other forms of malfeasance were widespread in many holding companies (Berle and Means 1932). To reinvigorate and sustain capitalism after the Great Depression, the Franklin D. Roosevelt (1882–1945) administration set up public policies to limit the use of this corporate form. These policies included a tax on the transfer of capital between subsidiary corporations and the holding company. By increasing the cost to operate the holding company, this capital transfer tax contributed to the deinstitutionalization of this corporate form. After this legislation was passed, corporations began to restructure into a multidivisional form where product lines are organized as divisions inside a single corporation. By 1960, the multidivisional form prevailed in corporate America (Chandler 1962).
In response to the capital accumulation crises of the late 1970s and early 1980s, corporate America mobilized politically to replace the institutional arrangements that were established during and after the New Deal. These market-driven neoliberal policies included: (1) redefining antitrust regulations that prohibited large corporations from merging with others operating in the same market; (2) eliminating the tax on capital transfers from subsidiary corporations to parent companies; and (3) dismantling regulatory control over the banking and financial industry. Now, corporations were allowed to engage in behaviors that were previously illegal or not viable, including using new forms of financing to merge or acquire other large corporations and organize their new and existing corporate entities as subsidiaries in the multilevel-subsidiary form. This multilevel-subsidiary form has a parent company at the top of the corporate hierarchy that operates as a financial management company, with two or more levels of legally separate subsidiary corporations embedded in it (Prechel 2000, p. 12). By 1993, 65 percent of the one hundred largest U.S. industrial corporations used this corporate form (Prechel 2000, p. 244). As occurred during the late nineteenth and early twentieth centuries, this multilevel-subsidiary form provided a means to create giant corporations such as Exxon-Mobil, Chevron-Texaco, and ConocoPhillips. The embeddedness of this corporate form in these new institutional arrangements also created opportunities to overvalue corporate stock and to conceal financial malfeasance (Prechel 2000, p. 265).
Following the 2001 bankruptcy of Enron Corp., civil and legal investigations showed that Enron’s management used the multilevel-subsidiary form to transfer capital between legally independent corporate entities and manipulate its financial statements (Prechel 2003). Self-interest-seeking with guile was not limited to managers in the few corporations that the media focused on. Between January 1995 and May 2005, the Commodity Futures Trading Commission, the Department of Justice, and the Securities and Exchange Commission filed 457 allegations against 151 (i.e., 30%) of the 500 largest U.S. corporations. In 2005 alone, the Federal Bureau of Investigation pursued 405 corporate fraud cases that resulted in an additional 317 convictions (FBI 2005). However, many executives who engaged in corporate malfeasance were not accused of wrongdoing or held responsible for misleading the public because deregulation made many social acts legal that were previously illegal or not viable.
Unlike in the 1920s, when most stock was owned by the upper classes, the expansion of mutual funds created means for the working classes to invest their retirement and savings into corporate securities. As a result, the working classes lost billions of dollars due to corporate chicanery and malfeasance in the 1990s. Estimates suggest that $2.1 billion was wiped out from the employee pension plan at Enron alone and that $40 billion was lost from employee pension funds in general.
Political mobilization by corporations and wealthy individuals was not limited to deregulating corporations. During the same period that public policy was changed in ways that created opportunities for corporate malfeasance, elites financed political action committees, think tanks, and lobbying organizations to transform public policy in ways that lowered taxes on wealthy individuals and corporations. This policy shift also included a reduction in social programs for the poor.
By 1990, family income inequality increased to the highest level since 1947, when the U.S. government began compiling these data, and continued to increase into the twenty-first century. The very rich benefited most in this redistribution of income and wealth. Whereas the top 1 percent of American income earners received 9.3 percent of the total income in 1979, this group received 17.8 percent of the total income in 2000 (Mishel et al. 2005, p. 62). Although the rich have always held a large percentage of the wealth in American society, wealth inequality also increased during the 1980s and 1990s. Whereas the wealthiest 1 percent of the population held 33.8 percent of the nation’s total net worth in 1983, it had increased to 38.1 percent by 1998 (Mishel et al. 2005, p. 282).
The strong form of the organization-environment hypothesis suggests that the power of elites is derived, in part, from their capacity to control organizations and to use organizations to change their environment. Beginning in the 1970s, corporations and wealthy individuals used their wealth and power to decrease government regulation, increase corporate property rights, and extend free markets. These new institutional arrangements resulted in a historic turning point characterized by reduced regulatory oversight, lower taxes on corporations and wealthy individuals, higher rates of corporate chicanery and malfeasance, and increased inequality.
SEE ALSO Organization Theory
Aldrich, Howard, and Jeffrey Pfeffer. 1976. Environments of Organizations. Annual Review of Sociology 2: 79–105.
Berle, Adolf A., and Gardiner C. Means.  1991. The Modern Corporation and Private Property. New Brunswick, NJ: Transaction Publishers.
Braverman, Harry. 1974. Labor and Monopoly Capital: The Degradation of Work in the Twentieth Century. New York: Monthly Review Press.
Burawoy, Michael. 1979. Manufacturing Consent: Changes in the Labor Process under Monopoly Capitalism. Chicago: University of Chicago Press.
Carroll, Glenn, and Michael Hannan, eds. 1995. Organizations in Industry: Strategy, Structure, and Selection. New York: Oxford University Press.
Chandler, Alfred. 1962. Strategy and Structure: Chapters in the History of the Industrial Enterprise. Cambridge, MA: MIT Press.
DiMaggio, Paul J., and Walter W. Powell. 1983. The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields. American Sociological Review 48: 147–160.
Domhoff, G. William. 2006. Who Rules America? Power and Politics, and Social Change. 5th ed. Boston: McGraw Hill.
Federal Bureau of Investigation. 2005. Financial Crimes Report to the Public. http://www.fbi.gov/publications/financial/fcs_report052005/fcs_report052005.htm.
Granovetter, Mark. 1985. Economic Action and Social Structure: The Problem of Embeddedness. American Journal of Sociology 91: 481–510.
March, James G., and Herbert A. Simon. 1958. Organizations. New York: Wiley.
Mishel, Lawrence, Jared Bernstein, and Sylvia Allegretto. 2005. The State of Working America, 2004/2005. Ithaca, NY: Cornell University Press.
Perrow, Charles. 1972. Complex Organizations: A Critical Essay. Glenview, IL: Scott, Foresman. 3rd ed., 1986. New York: McGraw Hill.
Pfeffer, Jeffrey, and Gerald Salancik. 1978. The External Control of Organizations: A Resource Dependence Perspective. New York: Harper.
Prechel, Harland. 1990. Steel and the State: Industry, Politics, and Business Policy Formation, 1940–1989. American Sociological Review 55: 648–668.
Prechel, Harland. 1994. Economic Crisis and the Centralization of Control over the Managerial Process: Corporate Restructuring and Neo-Fordist Decision Making. American Sociological Review 59: 723–745.
Prechel, Harland. 2000. Big Business and the State: Historical Transitions and Corporate Transformation, 1880s-1990s. Albany: State University of New York Press.
Prechel, Harland. 2003. Historical Contingency Theory, Policy Paradigm Shifts, and Corporate Malfeasance at the Turn of the 21st Century. In Political Sociology for the 21st Century, eds. Betty A. Dobratz, Lisa K. Waldner, and Timothy Buzzell, 311–340. Amsterdam: Elsevier Science.
Simon, Herbert. 1957. Administrative Behavior: A Study of Decision-Making Processes in Administrative Organization. 2nd ed. New York: Macmillan.
Weber, Max.  1978. Economy and Society: An Outline of Interpretive Sociology. Eds. Guenther Roth and Claus Wittich; trans. Ephraim Fischoff et al. Berkeley: University of California Press.
"Organizations." International Encyclopedia of the Social Sciences. 2008. Encyclopedia.com. (August 25, 2016). http://www.encyclopedia.com/doc/1G2-3045301831.html
"Organizations." International Encyclopedia of the Social Sciences. 2008. Retrieved August 25, 2016 from Encyclopedia.com: http://www.encyclopedia.com/doc/1G2-3045301831.html