Chapter 1

COMPLIANCE AS A COMPARATIVE BASE

A DEFINITION OF COMPLIANCE

Compliance is universal, existing in all social units. It is a major element of the relationship between those who have power and those over whom they exercise it (Simmel, 1896). Despite its universality, it has been chosen as a base for this comparative study because it is a central element of organizational structure. The emphasis on compliance within the organization differentiates the latter from other types of social units. Characteristics of organizations such as their specificity, size, complexity and effectiveness each enhances the need for compliance. And in turn, compliance is systematically related to many central organizational variables.

Compliance refers both to a relation in which an actor behaves in accordance with a directive supported by another actor's power, and to the orientation of the subordinated actor to the power applied.

By supported we mean that those who have power manipulate means which they command in such a manner that certain other actors find following the directive rewarding, while not following it incurs deprivations. In this sense, compliance relations are asymmetric (or "vertical"). But it is not assumed that the subordinates have no power, only that they have less.

The power-means, manipulated to support the directives, include physical, material, and symbolic rewards and deprivations. Organizations tend to allocate these means systematically and strive to ensure that they will be used in conformity with the organizational norms.

The orientation of the subordinated actor can be characterized as positive (commitment) or negative (alienation). It is determined in part by the degree to which the power applied is considered legitimate by the subordinated actor, and in part by its congruence with the line of action he would desire. We refer to this orientation, whether positive or negative, as involvement in the organization. In sum, there are two parties to a compliance relationship: an actor who exercises power, and an actor, subject to this power, who responds to this subjection with either more or less alienation or more or less commitment.

The next task is to use compliance as here defined to develop an analytical base for the classification of organizations. This is done in three steps. First, three kinds of power are differentiated; then, three kinds of involvement are specified; and finally, the associations of kinds of power with kinds of involvement are indicated. These associations -- which constitute compliance relationships -- then serve as the basis of our classification of organizations.

THREE KINDS OF POWER: A COMPARATIVE DIMENSION

A Classification of Power

Power is an actor's ability to induce or influence another actor to carry out his directives or any other norms he supports. Goldhamer and Shils state that "a person may be said to have power to the extent that he influences the behavior of others in accordance with his own intentions." (p. 171). Of course, "his own intentions" might be to influence a person to follow others' "intentions" or those of a collectivity. In organizations, enforcing the collectivity norms is likely to be a condition determining the power-holder's access to the means of power.

Power positions are positions whose incumbents regularly have access to means of power. Statements about power positions imply a particular group (or groups) who are subject to this power. For instance, to state that prison guards have a power position implies the subordination of inmates. In the following analysis we focus on power relations in organizations between those higher and those lower in rank. We refer to those in power positions, who are higher in rank, as elites or as organizational representatives. We refer to those in subject positions, who are lower in rank, as lower participants.

Power differs according to the means employed to make the subjects comply. These means may be physical, material, or symbolic.

Coercive power rests on the application, or the threat of application, of physical sanctions such as infliction of pain, deformity, or death; generation of frustration through restriction of movement; or controlling through force the satisfaction of needs such as those for food, sex, comfort, and the like.

Remunerative power is based on control over material resources and rewards through allocation of salaries and wages, commissions and contributions, "fringe benefits," services and commodities.

Normative power rests on the allocation and manipulation of symbolic rewards and deprivations through employment of leaders, manipulation of mass media, allocation of esteem and prestige symbols, administration of ritual, and influence over the distribution of "acceptance" and "positive response." (A more eloquent name for this power would be persuasive, or manipulative, or suggestive power. But all these terms have negative value connotations which we wish to avoid.)

There are two kinds of normative power. One is based on the manipulation of esteem, prestige, and ritualistic symbols (such as a flag or a benediction); the other, on allocation and manipulation of acceptance and positive response (Parsons, 1951, p. 108). Although both powers are found both in vertical and in horizontal relationships, the first is more frequent in vertical relations, between actors who have different ranks, while the second is more common in horizontal relations, among actors equal in rank -- in particular, in the power of an "informal" or primary group over its members. Lacking better terms, we refer to the first kind as pure normative power, and to the second as social power. Social power could be treated as a distinct kind of power. But since powers are here classed according to the means of control employed, and since both social and pure normative power rest on the same set of means -- manipulation of symbolic rewards -- we treat these two powers as belonging to the same category.

From the viewpoint of the organization, pure normative power is more useful, since it can be exercised directly down the hierarchy. Social power becomes organizational power only when the organization can influence the group's powers, as when a teacher uses the class climate to control a deviant child, or a union steward agitates the members to use their informal power to bring a deviant into line.

Organizations can be ordered according to their power structure, taking into account which power is predominant, how strongly it is stressed compared with other organizations in which the same power is predominant, and which power constitutes the secondary source of control.

Neutralization of Power

Most organizations employ all three kinds of power, but the degree to which they rely on each differs from organization to organization. Most organizations tend to emphasize only one means of power, relying less on the other two. Evidence to this effect is presented below in the analysis of the compliance structures of various organizations. The major reason for power specialization seems to be that when two kinds of power are emphasized at the same time, over the same subject group, they tend to neutralize each other.

Applying force, for instance, usually creates such a high degree of alienation that it becomes impossible to apply normative power successfully. This is one of the reasons why rehabilitation is rarely achieved in traditional prisons, why custodial measures are considered as blocking therapy in mental hospitals, and why teachers in progressive schools tend to oppose corporal punishment.

Similarly, the application of renumerative powers makes appeal to "idealistic" (pure normative) motives less fruitful. In a study of the motives which lead to purchase of war bonds, Merton pointed out that in one particularly effective drive (the campaign of Kate Smith), all "secular" topics were omitted and the appeal was centered on patriotic, "sacred" themes. Merton asked a sample of 978 people: "Do you think that it is a good idea to give things to people who buy bonds?"

Fifty per cent were definitely opposed in principle to premiums, bonuses and other such inducements, and many of the remainder thought it a good idea only for "other people" who might not buy otherwise. (1946, p. 47)

By omitting this [secular] argument, the authors of her scripts were able to avoid the strain and incompatibility between the two main lines of motivation: unselfish, sacrificing love of country and economic motives of sound investment. (Ibid., p. 45)

It is possible to make an argument for the opposite position. It might be claimed that the larger the number of personal needs whose satisfaction the organization controls, the more power it has over the participants. For example, labor unions that cater to and have control over the social as well as the economic needs of their members have more power over those members than do unions that focus only on economic needs. There may be some tension between the two modes of control, some ambivalence and uneasy feeling among members about the combination, but undoubtedly the total control is larger. Similarly, it is obvious that the church has more power over the priest than over the average parishioner. The parishioner is exposed to normative power, whereas the priest is controlled by both normative and remunerative powers.

The issue is complicated by the fact that the amount of each kind of power applied must be taken into account. If a labor union with social powers has economic power which is much greater than that of another union, this fact may explain why the first union has greater power in sum, despite some "waste" due to neutralization. A further complication follows from the fact that neutralization may also occur through application of the "wrong" power in terms of the cultural definition of what is appropriate to the particular organization and activity. For example, application of economic power in religious organizations may be less effective than in industries, not because two kinds of power are mixed, but because it is considered illegitimate to use economic pressures to attain religious goals. Finally, some organizations manage to apply two kinds of power abundantly and without much waste through neutralization, because they segregate the application of one power from that of the other. The examination below of combat armies and labor unions supplies an illustration of this point.

We have discussed some of the factors related to the tendency of organizations to specialize their power application. In conclusion, it seems that although there can be little doubt that such a tendency exists, its scope and a satisfactory explanation for it have yet to be established.

THREE KINDS OF INVOLVEMENT: A COMPARATIVE DIMENSION

Involvement, Commitment, and Alienation

Organizations must continually recruit means if they are to realize their goals. One of the most important of these means is the positive orientation of the participants to the organizational power. Involvement refers to the cathectic-evaluative orientation of an actor to an object, characterized in terms of intensity and direction.

The intensity of involvement ranges from high to low. The direction is either positive or negative. We refer to positive involvement as commitment and to negative involvement as alienation. (The advantage of having a third term, involvement, is that it enables us to refer to the continuum in a neutral way. Actors can accordingly be placed on an involvement continuum which ranges from a highly intense negative zone through mild negative and mild positive zones to a highly positive zone.

Three Kinds of Involvement

We have found it helpful to name three zones of the involvement continuum, as follows: alienative, for the high alienation zone; moral, for the high commitment zone; and calculative, for the two mild zones. This classification of involvement can be applied to the orientations of actors in all social units and to all kinds of objects. Hence the definitions and illustrations presented below are not limited to organizations, but are applicable to orientations in general.

Alienative Involvement -- Alienative involvement designates an intense negative orientation; it is predominant in relations among hostile foreigners. Similar orientations exist among merchants in "adventure" capitalism, where trade is built on isolated acts of exchange, each side trying to maximize immediate profit (Gerth and Mills, 1946, p. 67). Such an orientation seems to dominate the approach of prostitutes to transient clients (K. Davis, 1937, pp. 748-49). Some slaves seem to have held similar attitudes to their masters and to their work. Inmates in prisons, prisoners of war, people in concentration camps, enlisted men in basic training, all tend to be alienated from their respective organizations.

Calculative Involvement -- Calculative involvement designates either a negative or a positive orientation of low intensity. Calculative orientations are predominant in relationships of merchants who have continuous business contacts. Attitudes of (and toward) permanent customers are often predominantly calculative, as are relationships among entrepreneurs in modern (rational) capitalism. Inmates in prisons who have established contact with prison authorities, such as "rats" and "peddlers," often have predominantly calculative attitudes toward those in power (Sykes, 1958, pp 87-95).

Moral Involvement -- Moral involvement designates a positive orientation of high intensity. The involvement of the parishioner in his church, the devoted member in his party, and the loyal follower in his leader are all "moral."

There are two kinds of moral involvement, pure and social. They differ in the same way pure normative power differs from social power. Both are intensive modes of commitment, but they differ in their foci of orientation and in the structural conditions under which they develop. Pure moral commitments are based on internalization of norms and identification with authority (like Riesman's inner-directed "mode of conformity"); social commitment rests on sensitivity to pressures of primary groups and their members (Riesman's "other-directed"). Pure moral involvement tends to develop in vertical relationships, such as those between teachers and students, priests and parishioners, leaders and followers. Social involvement tends to develop in horizontal relationships like those in various types of primary groups. Both pure moral and social orientations might be found in the same relationships, but, as a rule, one orientation predominates.

Actors are means to each other in alienative and in calculative relations; but they are ends to each other in "social" relationships. In pure moral relationships the means-orientation tends to predominate. Hence, for example, the willingness of devoted members of totalitarian parties or religious orders to use each other. But unlike the means-orientation of calculative relationships, the means-orientation here is expected to be geared to needs of the collectivity in serving its goals, and not to those of an individual.

As has been stated, the preceding classification of involvement can be applied to the orientations of actors in all social units and to all kinds of objects. The analysis in this book applies the scheme to orientations of lower participants in organizations to various organizational objects, in particular to the organizational power system. The latter includes (1) the directives the organization issues, (2) the sanctions by which it supports its directives, and (3) the persons who are in power positions. The choice of organizational power as the prime object of involvement to be examined here follows from a widely held conception of organization as an administrative system or control structure. To save breath, the orientation of lower participants to the organization as a power (or control) system is referred to subsequently as involvement in the organization. When other involvements are discussed, the object of orientation -- for example, organizational goals -- is specified.

Organizations are placed on the involvement continuum according to the modal involvement pattern of their lower participants. The placing of organizations in which the participants exhibit more than one mode of involvement is discussed in a later chapter.

COMPLIANCE AS A COMPARATIVE BASE

A Typology of Compliance

Taken together, the two element -- that is, the power applied by the organization to lower participants, and the involvement in the organization developed by lower participants -- constitute the compliance relationship. Combining three kinds of power with three kinds of involvement produces nine types of compliance, as shown in the accompanying table.

The nine types are not equally likely to occur empirically. Three -- the diagonal cases, 1, 5, and 9 -- are found more frequently than the other six types. This seems to be true because these three types constitute congruent relationships, whereas the other six do not.

The Congruent Types -- the involvement of lower participants is determined by many factors, such as their personality structure, secondary socialization, memberships in other collectivities, and so on. At the same time, organizational powers differ in the kind of involvement they tend to generate. When the kind of involvement that lower participants have because of other factors and the kind of involvement that tends to be generated by the predominant form of organizational power are the same, we refer to the relationship as congruent. For instance, inmates are highly alienated from prisons; coercive power tends to alienate; hence this is a case of a congruent compliance relationship.

Congruent cases are more frequent than noncongruent ones primarily because congruence is more effective, and organizations are social units under external and internal pressure to be effective. The effective application of normative powers, for example, requires that lower participants be highly committed. If lower participants are only mildly committed to the organization, and particularly if they are alienated from it, the application of normative power is likely to be ineffective. Hence the association of normative power with moral commitment.

Remuneration is at least partially wasted when actors are highly alienated, and therefore inclined to disobey despite material sanctions; it is also wasted when actors are highly committed, so that they would maintain an effective level of performance for symbolic, normative rewards only. Hence the association of remuneration with calculative involvement.

Coercive power is probably the only effective power when the organization is confronted with highly alienated lower participants. If, on the other hand, it is applied to committed or only mildly alienated lower participants, it is likely to affect adversely such matters as morale, recruitment, socialization, and communication, and thus to reduce effectiveness. (It is likely, though, to create high alienation, and in this way to create a congruent state.)

The Incongruent Types -- Since organizations are under pressure to be effective, the suggestion that the six less effective incongruent types are not just theoretical possibilities but are found empirically calls for an explanation. The major reason for this occurrence is that organizations have only limited control over the powers they apply and the involvement of lower participants. The exercise of power depends on the resources the organization can recruit and the license it is allowed in utilizing them. Involvement depends in part on external factors, such as membership of the participants in other collectivities (e.g., membership in labor unions); basic value commitments (e.g., Catholic versus Protestant religious commitments); and the personality structure of the participants (e.g., authoritarian). All these factors may reduce the expected congruence of power and involvement.

A Dynamic Hypothesis -- Congruent types are more effective than incongruent types. Organizations are under pressure to be effective. Hence, to the degree that the environment of the organization allows, organizations tend to shift their compliance structure from incongruent to congruent types and organizations which have congruent compliance structures tend to resist factors pushing them toward incongruent compliance structures.

Congruence is attained by a change in either the power applied by the organization or the involvement of lower participants. Change of power takes place when, for instance, a school shifts from the use of corporal punishment to stress on the "leadership" of the teachers. The involvement of lower participants may be changed through socialization, changes in recruitment criteria, and the like.

Because the large majority of cases falls into the three categories representing congruent compliance, these three types form the basis for subsequent analysis. We refer to the coercive-alienative type as coercive compliance; to the remunerative-calculative type as utilitarian compliance; and to the normative-moral type as normative compliance. Students of organizational change, conflict, strain, and similar topics may find the six incongruent types more relevant to their work.

Compliance and Authority

The typology of compliance relationships presented above highlights some differences between the present approach to the study of organizational control and that of studies conducted in the tradition of Weber. These studies tend to focus on authority, or legitimate power, as this concept is defined. The significance of authority has been emphasized in modern sociology in the past, in order to overcome earlier biases that overemphasized force and economic power as the sources of social order. This emphasis, in turn, has led to an overemphasis on legitimate power. True, some authority can be found in the control structure of lower participants in most organizations. True, authority plays a role in maintaining the long-run operations of the organization. But so does nonlegitimated power. Since the significance of legitimate power has been fully recognized, it is time to lay the ghost of Marx and the old controversy, and to give full status to both legitimate and nonlegitimate sources of control.

Moreover, the concept of authority does not take into account differences among powers other than their legitimacy, in particular the nature of the sanctions (physical, material, or symbolic) on which power is based. All three types of power may be regarded as legitimate by lower participants: thus there is normative, remunerative, and coercive authority (differentiated by the kind of power employed, for instance, by a leader, a contractor, and a policeman.) But these powers differ in the likelihood that they will be considered legitimate by those subjected to them. Normative power is most likely to be considered legitimate; coercive, least likely; and remunerative is intermediate.

Finally, it is important to emphasize that involvement in the organization is affected both by the legitimacy of a directive and by the degree to which it frustrates the subordinate's need-dispositions. Alienation is produced not only by illegitimate exercise of power, but also by power which frustrates needs, wishes, desires. Commitment is generated not merely by directives which are considered legitimate but also by those which are in line with internalized needs of the subordinate. Involvement is positive if the line of action directed is conceived by the subordinate as both legitimate and gratifying. It is negative when the power is not granted legitimacy and when it frustrates the subordinate. Involvement is intermediate when either legitimation or gratification is lacking. Thus the study of involvement, and hence that of compliance, differs from the study of authority by taking into account the effects of the cathectic as well as the evaluative impact of directives on the orientation of lower participants.

LOWER PARTICIPANTS AND ORGANIZATIONAL BOUNDARIES

Before we can begin our comparisons, the following questions still remain to be answered. Why do we make compliance of lower participants the focus of the comparison? Who exactly are "lower participants"? What are the lower boundaries of an organization? In answering these questions, we employ part of the analytical scheme suggested above, and thus supply the first test of its fruitfulness.

Why Lower Participants?

Compliance of lower participants is made the focus of this analysis for several reasons. First, the control of lower participants is more problematic than that of higher participants because, as a rule, the lower an actor is in the organizational hierarchy, the fewer rewards he obtains. His position is more deprived; organizational activities are less meantingful to him because he is less "in the know," and because often, from his position, only segments of the organization and its activities are visible. Second, since we are concerned with systematic differences among organizations (the similarities having been more often explored), we focus on the ranks in which the largest differences in compliance can be found. An inter-organizational comparison of middle and higher ranks would show that their compliance structures differ much less than those of the lower ranks (see Chapter XII).

Who Are Lower Participants?

Organizational studies have used a large number of concrete terms to refer to lower participants: employees, rank-and-file, members, clients, customers, inmates. These terms are rarely defined. They are customarily used to designate lower participants in more than one organization, but none can be used for all.

Actually, these terms can be seen as reflecting different positions on at least three analytical dimensions. One is the nature (direction and intensity) of the actors' involvement in the organization. Unless some qualifying adjectives such as "cooperative" or "good" are introduced, inmates implies alienative involvement. Clients designates people with alienative or calculative involvement. Customers refers to people who have a relatively more alienative orientation than clients; one speaks of the clients of professionals but not ordinarily of their customers. Member is reserved for those who have at least some, usually quite strong, moral commitment to their organization. Employee is used for people with various degrees of calculative involvement.

A second dimension underlying these concrete terms is the degree to which lower participants are subordinated to organizational powers. Inmates, it seems, are more subordinated than employees, employees more than members, and members more than clients. A study in which subordination is a central variable would take into account that it includes at least two subvariables: the extent of control in each area (e.g., "tight" versus remote control); and the scope of control, measured by the number of areas in which the subject is subordinated. Such refinement is not required for our limited use of this dimension.

A third dimension is the amount of performance required from the participants by the organization: it is high for employees, low for inmates, and lowest for clients and customers.

Using concrete terms to designate groups of participants without specifying the underlying dimensions creates several difficulties. First of all, the terms cannot be systematically applied. Although "members" are in general positively involved, sometimes the term is used to designate lower participants with an alienative orientation. Archibald, for instance, uses this term to refer to members of labor unions who are members only pro forma and who see in the union simply another environmental constraint, to which they adjust by paying dues.

Most workers entered the yards not merely ignorant of unions, but distrustful of them....They nonetheless joined the unions, as they were compelled to do, with little protest. They paid the initiation fees, averaging not more than twenty dollars, much as they would have bought a ticket to the county fair: it cost money, but maybe the show would be worth the outlay. As for dues, they paid them with resignation to the principle that all joys of life are balanced by a measure of pain. (1947, pp. 131-32)

The term customers suggests that the actors have no moral commitments to their sources of products and services. But sometimes it is used to refer to people who buy from cooperatives, frequent only unionized barbers, and remain loyal to one newspaper -- that is, to people who are willing to suffer some economic loss because they see in these sources of service something which is "good in itself" -- people who, in short, have some moral commitments.

Any moral commitment on the part of mental patients, designated as inmates, is viewed either with surprise or as a special achievement of the particular mental hospital; on the other hand, members of labor unions are "expected" to show moral commitment and are labeled "apathetic" if they do not. The fact that some mental patients view their hospital as their home, and thus are positively involved, whereas labor union members may see their organization as a secondary group only, is hidden by the terminology employed. The same point could be made for differences in performance and in subordination.

Although the use of such concrete terms leads to overgeneralization, by implying that all lower participants of an organization have the characteristics usually associated with the label, they can also impede generalization. An illustration is supplied by studies of parishioners. Many of these studies focus on problems of participation, such as "apathy," high turnover, and declining commitment. But rarely are comparisons drawn, or insights transferred, from the study of members of voluntary associations and political organizations. Actually, all these organizations are concerned with the moral commitment of lower participants who have few performance obligations and little subordination to the organization.

Another advantage of specifying the analytical dimensions underlying these concepts is that the number of dimensions is limited, whereas the number of concrete terms grows continuously with the number of organizations studied. Thus the study of hospitals introduces patients; the analysis of churches brings up parishioners; and the examination of armies adds soldiers. Following the present procedure, we can proceed to characterize the lower participants of additional organizations by the use of the same three dimensions.

Specifying the underlying dimensions enables us not only to formulate analytical profiles of a large variety of lower participants, but also to compare them systematically with each other on these three dimensions. For instance, "soldiers" (in combat) are high on all three dimensions, whereas inmates are high on subordination and alienation but low on performance; employees are medium in involvement and subordination, but high on performance obligations. The import of such comparisons will become evident later.

Finally, whereas concrete terms tend to limit analysis to participants at particular levels, analytical terms such as alienative, calculative, and moral can be applied equally well to participants at all levels of the organizational hierarchy.

Ideally, in a book such as this, we should refer to lower participants in analytical terms, those of various degrees of involvement, subordination, and performance obligations. Since this would make the discussion awkward, the concrete terms are used, but only to refer to typical analytical constellations. Inmates are lower participants with high alienation, low performance obligations, and high subordination. The term will not be used to refer to other combinations which are sometimes found among lower participants in prisons. Members is used to refer only to lower participants who are highly committed, medium on subordination, and low on performance obligations; it is not used to refer to alienated lower participants in voluntary associations. Similarly, other terms are used as specified below.

Lower versus Higher Participants

Higher participants have a "permanent" power advantage over lower participants because of their organizational position. Thus, by definition, higher participants as a group are less subordinated than lower participants. Often, though not in all organizational types, they are also more committed, and have more performance obligations (if we see decision making and other mental activities as performances). Thus the three dimensions which serve to distinguish among various types of lower participants also mark the dividing line between lower and higher participants. These very dimensions also enable us to suggest a way to delineate the organizational boundaries -- that is, to distinguish between participants and nonparticipants.

Organizational Boundaries

Students of organizations must often make decisions about the boundaries of the unit they are studying: who is a participant, who an outsider. March and Simon, for example, take a broad view of organizational boundaries: "When we describe the chief participants of most business organizations, we generally limit our attention to the following five major classes: employees, investors, suppliers, distributers, and consumers." (1958, p. 89)

We follow a narrower definition and see as participants all actors who are high on at least one of the three dimensions of participation: involvement, subordination, and performance. Thus, students, inmates, soldiers, workers, and many others are included. Customers and clients, on the other hand, who score low on all three criteria, are considered "outsiders."

We should like to underscore the importance of this way of delineating the organizational boundaries. It draws the line much "lower" than most studies of bureaucracies, which tend to include only persons who are part of a formal hierarchy: priests, but not parishioners; stewards, but not union members; guards, but not inmates; nurses, but not patients. We treat organizations as collectivities of which the lower participants are an important segment. To exclude them from the analysis would be like studying colonial structures without the natives, stratification without the lower classes, or a political regime without the citizens or voters.

It seems to us especially misleading to include the lower participants in organizational charts when they have a formal role, as privates in armies or workers in factories, and to exclude them when they have no such status, as is true for parishioners or members. This practice leads to such misleading comparisons as seeing the priests as the privates of the church and teachers as the lowest-ranking participants of schools, in both cases ignoring the psychological import of having "subordinates." One should not let legal or administrative characteristics stand in the way of a sociological analysis. However, the main test of the decision to delineate the organization as we have chosen follows: it lies in the scope, interest, and validity of the propositions this approach yields.

SUMMARY

Compliance patterns were chosen as the basis for our comparative study of organizations because compliance relations are a central element of organizational structure. It distinguishes organizations from other collectivities because organizations require more compliance than other collectivities do, and it is systematically related to many other organizational variables.

Compliance refers both to a relation in which an actor behaves in accordance with a directive supported by another person's power and to the orientation of the subject to the power applied. There are three kinds of power: coercive, remunerative, and normative; and three kinds of involvement: alienative, calculative, and moral. There are, therefore, nine possible types of compliance. Three of these types (congruent types) are more effective than the other six; they are also empirically much more frequent. These three types form the basis of our comparative study.

Each organizational rank has its own compliance structure. We focus on the compliance structure of lower participants, first because their compliance is more problematic than that of higher participants, and second because organizations can be most fruitfully distinguished from each other at this level.

Lower participants are actors who are high on at least one of the three dimensions of participation: involvement, performance obligations, and subordination. An examination of concrete terms often used to refer to different groups of lower participants shows that they can be seen as positions on these three analytical dimensions, which also enable us also to delineate systematically the boundaries of organizations. We are now ready to engage in the first major substantive step: classification of organizations according to their compliance structures.

Copyright © 1975 by Amitai Etzioni