TWO

Hierarchy

The concept of hierarchy was invented in the fifth century by Denys the Areopagite.¹ He used the term to define and describe an organizational structure based on the top-down delegation of power and determination of functions. This system derived from his religious beliefs; for the Greeks, hierarchy meant “sacred rule.”² The word was also adopted to “denote the power given by Christ to his apostles and their successors to form and govern the church.”³ Later, Shakespeare defended the concept of hierarchy in his play Troilus and Cressida with the following famous passage: “Take but degree away, untune that string, and hark what discord follows.”⁴

In the modern literature of organization theory, the term has been defined as “any system in which the distributions of power, privilege and authority are both systematic and unequal.”⁵ Power can be taken here to mean domination, while privilege implies a right, immunity, or benefit enjoyed by a particular person or a restricted group of persons. Authority can be defined as “legitimate” power, in which those subject to domination by others accept that domination as a legitimate arrangement. Given this use of these terms, a more specific definition of hierarchy results: “society arranged according to ‘degree,’ with power, privilege and authority varying together.”⁶

Most other definitions of hierarchy have incorporated some variations of power, privilege, and authority, but have been more specific in describing the exact relationships between them. This has been particularly true of definitions coming from theorists ascribing to the rational systems point of view. Weber’s definition of hierarchy states that each lower office is controlled and supervised by a higher one. Other rational systems theorists, such as James Mooney and A.C. Reiley and Luther Gulick and L. Urwick, emphasize the importance of coordination and specialization. ⁷ Coordination is the “vertical” arrangement, including the scalar principle, unity of command, span of control, and the exception principle. These categories warrant further explanation.

The scalar principle emphasizes hierarchical organization “in which all participants are linked into a single pyramidal structure of control relations.”⁸ Unity of command means that no organization members receive orders from more than one superior. Span of control means that no superior should have more subordinates than can be overseen, and the exception principle means that all routine matters are handled by subordinates, so that the superior is free to deal with “exceptional” situations. This is the distinction that Selznick later describes as that between “routine” versus “critical” decisions.⁹

Specialization refers more to the horizontal division of labor. This concept has two major elements: the departmentalization principle and the line-staff principle. The first is a method of organizing based on homogeneity of activity, whether of purpose, process, or clientele. The line-staff principle means that activities dealing with organization goals are line functions and that the staff is there to provide administrative support for those individuals engaged in those activities. Together, coordination as a vertical element and specialization as a horizontal element create the power, privilege, and authority relations within a hierarchy.

One important element missing from this configuration of hierarchy is communication. Some theorists believe this aspect to be of such importance that they define hierarchy as “centralized communication systems.”¹⁰ Whether hierarchy is defined as a communications system or not, the flow of information up and down the pyramidal structure forms an important aspect of power, privilege, and authority relationships. In most instances, the possession of information or knowledge constitutes power. It is the possession of information, on one hand, and the deprivation of it, on the other that keeps systems of domination in operation.

Certain messages can travel in only one direction: Definitional messages (what is to be done, who is to do it, etc.) travel down the scale, operational messages (what was done, what happened, etc.) travel up the scale. Message priorities are based on positional origin, not content. Further limits on message content are constrictive for lower ranks and permissive for upper ranks. Access to stored information is confined to upper ranks and denied lower ranks, etc.¹¹

The relationship between information and hierarchy takes on interest in light of the question of efficiency. As Blau and Scott point out, “there appears to be a curvilinear relation between information-processing requirements and the utility of hierarchy.”¹² In other words, if there is little information to be transmitted within an organization, there may be no particular advantages to hierarchy and some clear disadvantages, such as the costs of administrative overhead. As the volume of information increases, however, hierarchies can offer benefit in reducing transmission costs and ensuring coordination—up to a certain point. If the volume of information grows excessive, hierarchies become overloaded, and resources at the lowest levels of the organization are underutilized. ¹³

Considering these factors, hierarchy can be defined as a vertical and horizontal system of domination with varying degrees of centralized communication. Within this system of domination, privilege and authority are determined by scalar position within the organization, not by the person holding the position. Once a person has held a position for an extended period of time, however, this distinction between individual and position becomes less clear. Informal leadership characteristics may or may not qualify a certain individual for a certain position within the hierarchy. Thus, a distinction should be made between domination and leadership. Domination can be characterized by the following factors:

a) the position of headship is maintained through an organized system and not by the spontaneous recognition of the individual contribution to the group goal; b) the group goal is arbitrarily chosen by the autocratic head in his own self-interest and is not internally determined; c) there is not really a group at all, since there is no sense of shared feeling or joint action; and d) there is in this process a wide social gap between the group members and the head, who strives to maintain this social distance as an aid to his coercion of the group through fear.¹⁴

The difference in organizational settings between leadership and domination is an important one. This is particularly true with regard to the development of more egalitarian forms of organization. Every organization faces the question of leadership. As mentioned earlier, organizations attempting to avoid hierarchical structure face a dilemma. Such organizations may wish to allow leadership to develop naturally out of the skills and interests of its members. However, there is a danger that certain members may gain positions of power within the organization without those positions being formalized. Once such individuals develop power there is no procedural means of removing them from the position. Yet if positions are formalized, the organization runs the risk of becoming a hierarchy—what it was attempting to avoid. This is a dilemma that will be discussed at greater length with regard to the three women’s organizations studied herein.

In focusing on the question of whether more egalitarian forms of organization can exist, it is necessary to consider the relative costs and benefits of hierarchy. Theorists with the most positive view of hierarchy come primarily from the rational systems perspective. As one would expect, they view hierarchy as the most efficient means of attaining organization goals. The “essential elements” of the rational perspective are goal specificity and formalization. Specification of goals allows the organization to select among desired activities and design a specific organization structure to accomplish those desired tasks. Formalization means that “rules governing behavior are precisely and explicitly formulated.”¹⁵ Such rules attempt to make behavior predictable, in order to foster stable expectations and the conditions for rational planning.

A primary school of thought within this perspective is that of scientific management. This approach, mainly put forth by Frederick Taylor in 1911, sought to rationalize every activity of managers and workers in order to “produce the maximum output with the minimum input of energies and resources.”¹⁶ Taylor was attempting to rationalize the organization from the bottom up, while others, such as Henri Fayol, were developing another school of administrative principles to rationalize the organization from the top down.

Weber’s theory of bureaucracy is also categorized within this rational perspective, though Weber himself was simply describing, not advocating, such a view. Given that coordination and specialization are key elements of hierarchy, it is not difficult to see how well Weber’s position fits with the rational perspective. Hierarchy’s benefits include providing the necessary structure for what is perceived to be efficient management. By specifying roles and coordinating tasks within the organization, more time is spent in working toward goals than in working out power and authority relations. This part of the organization is already a ‘given’.¹⁷

Not only the rational systems theorists see benefits to hierarchy. Natural and open systems theorists emphasize benefits with regard to communication, but, again, these are viewed in terms of efficiency in information processing. Studies indicate that:

The centralized structures more rapidly organize to solve the problems. Participants in peripheral positions send information to the center of the network, where a decision is made and sent out to the periphery. Furthermore, this pattern of organization tends to be highly stable once developed. In less centralized structures the organization problem is more difficult and observed interaction patterns are less stable, as well as less efficient.¹⁸

This statement raises questions concerning the concepts of centralization and decentralization with regard to hierarchy and organization.

CENTRALIZATION, DECENTRALIZATION, AND FRAGMENTATION

In the discussion of hierarchical versus non-hierarchical organization, consideration of top-down versus bottom-up styles of management merits importance. The terms centralization, decentralization, and fragmentation are often used to refer to one or the other of these arrangements. Centralization usually refers to the top-down arrangement; however, confusion often exists with regard to the association of decentralization with the bottom-up style. Decentralization and fragmentation are often confused and warrant further clarification.

Decentralization has commonly been described in the political science literature simply as delegation of authority from the center or top level to lower levels of an organization.¹⁹ The degree to which decision-making authority can be pushed downward may relate directly to the life cycle of an organization. Organizations evolve, beginning as centralized structures and moving toward decentralization.²⁰ They are centralized at first due to the need to develop the path upon which the organization will embark. The determination of policy goals allows the organization to develop more of a similar mindset among members or personnel. This homogeneous outlook helps ensure that when authority is pushed downward, decisions made at lower levels are likely to be compatible with the overall direction of the organization. Centralization and determination of organizational purpose also allow for distinctions to be made as to whether policy decisions are critical or routine.²¹ Critical decisions are those that will most clearly affect survival of the organization. Routine decisions are made on a daily basis and are unlikely to alter significantly the organization’s future.

This distinction between critical and routine decision making is crucial to the implementation and maintenance of decentralization, because it sets up the structure for maintaining authority at the center of the organization while still allowing for delegation of responsibility to lower levels. In a decentralized setting, critical decisions are reserved for those at the top level of the organization, while routine decisions are delegated downward. This is thought to be useful for at least two reasons. First, lower-level personnel who are delegated responsibility deal with problems they are close to and have information about. They are assumed to have the resources needed to make whatever the organization defines as effective decisions at this level. Thus, decentralization allows for some degree of adaptation at lower levels. Second, as Selznick’s work indicates, executives who are freed from routine decision making have more time to consider the larger policy questions that will ultimately determine the direction of the organization. ²²

Yet even in the case of routine decisions, guidelines accompany the delegation of authority. Such guidelines allow for some discretion by personnel at lower levels, but may restrict the options or decisions available. If this occurs to the extent that decisions are absolutely prescribed, the organization remains centralized. Such were the findings of Herbert Kaufman’s study of the Forest Service.²³ Although forest rangers located in various parks were delegated authority to make decisions, the types of decisions and the ways to implement them had already been specified by the central authority. Thus, the organization was centralized.

In the course of events affecting organizations, it is possible for routine decisions to become critical ones. Within a decentralized organization, the upper level or center should retain and exercise the authority to recall responsibility it has delegated. If the upper level is unable to do this, the organization is moving toward fragmentation. Fragmentation can most clearly be identified when critical policy is made at lower levels, shifting the direction of the organization without guidance from upper levels.²⁴ Given these distinctions, decentralization can be defined as the delegation of authority from the center to lower levels, with the following requirements: critical policy decisions are reserved for the center while routine decisions are delegated to lower levels, and the center retains and is able to exercise the power to recall authority it has delegated.

TOWARD A CRITIQUE OF HIERARCHY

Unlike rational systems theorists, natural and open systems theorists perceive some of the costs of hierarchy along with the benefits. As alluded to above, there is a curvilinear relationship between information processing and the utility of hierarchy, in that in the case of only a low level of information or of too high a level, the costs of administration outweigh the benefits. In fact, as Scott points out, “It would be incorrect … to conclude that hierarchies are superior to more decentralized or egalitarian arrangements under all conditions.”²⁵

Natural and open systems theorists have shown that hierarchies impede work and can interfere with tasks presenting complex or ambiguous problems.²⁶ They impede work by stifling interaction among workers, as well as between workers and managers. Hierarchy undermines social support among organization members in that the workplace is formalized to the extent that “human” needs are not considered part of the process. Such a structure often reduces incentives for employees to produce, because they play no direct role in the decision making. When they have little real personal input or involvement in the job, performance means less to them.

These aspects of hierarchy all contribute to a lack of flexibility, which could be important to the survival of an organization faced with complex or ambiguous problems. In such situations the organization needs to have all members engaged in the correction of errors and the creative activity of problem solving. There are exceptions to this problem, as when hierarchies are comprised of subsystems that are interdependent and adaptive. But the development of such subsystems is atypical.²⁷

Thus, the costs of hierarchy can be measured in terms of human potential: loss of worker satisfaction due to low status and lack of control over decision making; the costs to the administration of maintaining communication as well as surveillance and control of participants; and actual threats to the survival of the organization posed by a lack of flexibility or ability to adapt to complex and/or ambiguous situations.

So far, then, the concept of hierarchy has been examined from two perspectives: that of the rational systems theorists, who tend to emphasize the benefits of efficiency, security, and organization survival, and that of the natural and open systems theorists, who weigh these benefits against the economic and human costs. The “human costs” aspect of this critique has come primarily from the human relations school within the natural systems perspective. This school was responsible for bringing attention to the development of informal, more personal aspects of organization life and fostering the idea of participatory management. This approach aims to mediate some of the ill effects of hierarchy by sharing organizational information and decision making with those members or workers at lower levels. This school of thought is characterized by the work of Rensis Likert, whose leadership studies focused on ways in which lower levels of management could relate to higher levels within the hierarchy.²⁸ Chris Argyris’s work focused on methods of reducing alienation by increasing the scope of certain jobs or rotating them.²⁹ In a similar light, Douglas McGregor’s work centered on the restrictive nature of hierarchy with regard to worker input and satisfaction.³⁰

While the idea of participatory management has been around for decades, it is currently receiving renewed attention. For instance, Du Pont has for some time now been stepping up its attempts toward worker participation through the “quality circle” or “team approach.” As one Du Pont executive explains:

The premise is simple—the level of thinking and the ultimate performance of supervisor and supervised can be significantly and permanently improved through communication and joint participation in problem solving.³¹

For Du Pont this has meant an increase in productivity, as well as improvements in manufacturing, marketing, and distribution of products. In addition, the company claims to have been able to reduce the need for direct supervisors due to increased employee participation. They reduced from seven to four the number of supervisory levels in several of their plants. Thus, the company has reduced the cost of its administrative overhead while at the same time increasing productivity. Currently, General Motors is implementing a similar plan.

While there are clearly benefits to such an arrangement, it is important to note that these must still be measured in terms of organization efficiency. While Du Pont was able to reduce a few levels of administration, the overall hierarchy itself was not altered signiflcantly. Thus, the natural systems critique of hierarchy is limited in that it still works within the same boundaries put forth by the rational systems perspective. This problem has led to the following critical perspective critique of human relations:

The history of organization theory may be seen, in part, as a process in which a series of ‘nonrational factors’ have been conjured up only to be subdued by the rationalizing core. Thus in the 1930s human relations theory arose as a champion of the informal structure. The trust of human relations theory, however, was to harness and control the informal in the interest of rationality.³²

Thus, from a natural systems perspective, hierarchy, as a system of domination with efficiency as a primary goal, is still endorsed. This is also true of the open systems perspective, which is concerned with organization adaptation, “fit,” and ultimate survival within its environment. While this perspective acknowledges that hierarchy is not appropriate for every organizational circumstance, it still reflects the dominant dictum of inevitability with regard to hierarchy. Nonetheless, the open systems approach does provide a theoretical bridge to the critical perspective of organizations, by recognizing the role of the environment in determining organization structure or reinforcing hierarchy, and that a kind of “natural selection” enables the survival of those organizations that “fit” with the environment.

The critical perspective operates on a much broader level than the natural and open systems perspectives, however. As explained earlier, this perspective considers how the dominant values in society foster or support broad patterns of organization structure, such as hierarchy. Thus, the critical perspective is as much an examination of dominant values in society as of the prevailing organizational structures. A first step in the development of this perspective was Victor Thompson’s work Modern Organization, in which he argues that hierarchy is culturally determined.³³ Because hierarchy is treated as the only way to “organize,” he states, those organizations that do not reflect this structure are considered to be “unorganized,” even unworthy of the label “organization.” Here Thompson has identified a dominant perspective in operation, one that allows for only hierarchical arrangements.

At the center of the critical perspective critique of hierarchy is the question of social control. The argument is that goal efficiency in organizations is only half the picture. The more pressing problem is that of those in power who wish to maintain their position without threat from those at lower levels. This theory draws on recent Marxist scholarship on the historical evolution of the labor process in the United States and Western Europe.³⁴ Fear of the development of trade unions and worker uprisings gave managers an incentive to seek out more effective means of control over workers. Hierarchy and the development of bureaucracy provided the means.

Detailed division of labor and associated bureaucratic mechanisms encouraged workers to internalize the values and structures of the industrial workplace. To the degree that this occurred, employers were successful in creating a climate of greater social control … capitalism has developed a production process which not only delivers the goods, but also controls its workers.³⁵

From this discussion it is clear that theorists within the critical perspective would define and criticize hierarchy as a system of domination flowing from capitalistic values. They would argue that there is a difference between efforts at efficiency and efforts at efficiency of control, and that hierarchy is more useful for the latter than the former. In fact, they would argue that in terms of efficiency of production, hierarchy fails.

This perspective builds on criticisms of hierarchy already discussed with regard to worker alienation, which suggest that productivity would be dramatically increased if workers were not cut off from decision making, control over their own jobs, and information about the organization. Given their low status and lack of meaningful rewards, workers would rebel against the system were control not exerted from the top of the organization and the larger society. If the prevailing societal values did not reinforce hierarchy, this control would not be possible.

Fundamentally, internal structure is only capable of taking strategic advantage of coercive market forces that already exist. Without a labor market that would make disobeying orders costly, administrative procedures would probably not be very effective in maintaining control.³⁶

The most important aspect of the critical perspective is that it does not take hierarchy to be inevitable. Rather than supporting this dominant view, the critical perspective puts forth a dialectical analysis that “takes the rationalized organization as an arbitrary model unevenly imposed upon events and insecure in its hold.”³⁷ This analysis allows us to examine the societally reinforced, market-related values that determine hierarchy and to consider how changes in those values could lead to other organizational forms. Thus, the critical view places the development of hierarchy in a historical/cultural perspective that allows for consideration of new structural possibilities.

NOTES

1. Larry D. Spence, “Prolegomena to a Communications Theory of Human Organizations” (unpublished manuscript, Department of Political Science, University of California, Berkeley, January 1969), p. 12.

2. Roger Scruton, A Dictionary of Political Thought (New York: Harper & Row, 1982), p. 202.

3. Ibid.

4. Ibid.

5. Ibid.

6. Ibid.

7. Richard W. Scott, Organizations (Englewood Cliffs, N.J.: Prentice-Hall, 1981), p. 65.

8. Ibid.

9. Philip Selznick, Leadership in Administration (New York: Harper & Row, 1957).

10. Scott, p. 147.

11. Spence, p. 15.

12. Peter Blau and W. Richard Scott, Formal Organizations (San Francisco: Chandler, 1962), pp. 116–28.

13. Ibid.

14. Spence, p. 15.

15. Scott, p. 60.

16. Ibid., p. 63.

17. Ibid., p. 61.

18. Ibid., p. 149.

19. Theodore J. Lowi, The Politics of Disorder (New York: Basic Books, 1971), p. 78.

20. Selznick, p. 113.

21. Ibid., p. 56.

22. Ibid., pp. 56–64.

23. Herbert Kaufman, The Forest Ranger (Baltimore: Johns Hopkins University Press, 1967), pp. 210–13.

24. Daniel J. Elazar, American Federalism: A View From the States (New York: Crowell, 1972).

25. Scott, p. 149.

26. Ibid., p. 150.

27. Ibid.

28. Rensis Likert, New Patterns of Management (New York: McGraw-Hill, 1961).

29. Chris Argyris, Personality and Organization (New York: Harper & Row, 1957).

30. Douglas McGregor, The Human Side of Enterprise (New York: McGraw-Hill, 1960).

31. H. Gordon Smith, “A More Open Workplace,” Nations Business May 1986, p. 6.

32. J. Kenneth Benson, “Organizations: A Dialectical View,” in Mary Zey-Ferrell and Michael Aiken, ed, Complex Organizations: Critical Perspectives (Glenview, Ill.: Scott, Foresman, and Company, 1981), p. 270.

33. Victor A. Thompson, Modern Organization (New York: Knopf, 1961).

34. Paul Goldman and Donald R. Van Houten, “Bureaucracy and Domination: Managerial Strategy in Turn-of-the-Century American Industry,” in Zey-Ferrell and Aiken, p. 192.

35. Ibid., p. 193.

36. Kenneth McNeil, “Understanding Organizational Power: Building on the Weberian Legacy,” in Zey-Ferrell and Aiken, p. 51.

37. J. Kenneth Benson, “Organizations: A Dialectical View,” in Zey-Ferrell and Aiken, p. 270.