The representative function of bureaucracy; public administration in constitutive perspective.

THE REPRESENTATIVE FUNCTION OF BUREAUCRACY

Public Administration in Constitutive Perspective

The dominant hierarchical and pluralist views of the proper role for an active, policy-making bureaucracy in a liberal democracy, which are grounded in concepts of representation, do not adequately justify rule by bureaucrats. Moreover, the idea of representative bureaucracy does not compensate for the deficiencies in hierarchical and pluralist arguments. An alternative defense of the legitimacy of the administrative state must draw on a constitutive rather than an instrumental conception of public administration. An active, policy-making bureaucracy can best be justified by understanding administrative agencies as helping to fulfill the representation and separation-of-powers doctrines of the Constitution, and as representing ideas and values that form a critical part of what constitutes the American regime.

The bicentennial of the U.S. Constitution unleashed a virtual flood of scholarship on the connections between the modern administrative state and the founding of the American republic. Much of this scholarship was intended specifically to establish once and for all a more secure foundation for public administration in the face of what many perceived as a continuing crisis of legitimacy (see, for example, Marion, 1986; McSwain, 1985; Mitchell & Scott, 1987; Rohr, 1986; Rourke, 1987; Sedgewick, 1987). Important conceptual and prescriptive elements have now been distilled (Kass & Catron, 1990; Wamsley et al., 1990), and scholars have turned to cultivating the rich soil that has been laid down (see, for example, Farazmand, 1989; Terry, 1990).

To join in this effort I turn my attention to the concept of representation. I contend that the principal arguments legitimating a role for bureaucracy in the American system -- a hierarchical view that depicts administrative organizations as merely agents of the citizenry and of political institutions under direct popular control, and a pluralist view that portrays administrative agencies as fulfilling the need for functional representation in the American system -- are grounded in concepts of representation. A seminal article written by Norton Long (1952) presaged this approach to dissecting the problem fo legitimacy in the administrative state (also see Burke, 1984). In that article, Long sought to escape from "the view of bureaucracy as instrument and Caliban" (Long, 1952, p. 810) imbedded in the hierarchical argument. Along the path of his escape. Long stakes a claim for the representativeness of bureaucracy in functional or pluralist terms. Ultimately, Long, and more recently John Rohr (1986), place considerable weight on the demographic make-up of the career service to legitimate bureaucracy as a governing institution.

It is clear that both Long and Rohr seek more than the single anchor of descriptive representation to justify bureaucratic rule. Rohr draws multiple connections between the modern career service and the founding of the republic. Long carves out a niche for bureaucracy in the "working" constitution. But neither scholar fully breaks free of the constricted understanding of public administration as a political institution associated with hierarchical, pluralist, and even representative bureaucracy arguments, and the concepts of representation on which they are based.

My objective in this article, then, is to examine systematically the concepts of representation that support the long dominant normative arguments about bureaucratic legitimacy. To address the deficiencies in legitimating public administration my analysis reveals, I attempt to reconceive administrative legitimacy on the basis of the distinctive qualities bureaucracy possesses as a political institution: stability and expertise. I draw on a constitutive perspective on political institutions, on the representation and separation-of-powers doctrines of the framers as interpreted by Tullis (1987), and on a Burkean understanding of the representation of "unattached interests." I argue that bureaucracy derives its authority to rule from the Constitution and not from its presumed expertise in administration. Expertise and stability are central, however, to defining a vital representative role for bureaucracy, in which public administrators further the cause of effective governance and help to constitute, not just serve, the American regime.

A DEMOCRATIC HIERARCHY

Herman Finer (1978) offered perhaps the most succinct and rhetorically dramatic rendition of the view that the executive establishment within American government is only legitimate if it is under the strict control of the public's elected representatives.

Are the servants of the public to decide their own course, or is their course

of action to be decided by a body outside themselves? My answer is that

the servants of the public are not to decide their own course; they are to be

responsible to the elected representatives of the public, and these are to

determine the course of action of the public servants to the most minute

degree that is technically feasible. . . . This kind of responsibility is what

democracy means. (pp. 411-412) The responsibility Finer defines does not mean merely deference to the will of the public and its representatives, but carrying out the will of the public and being held to account if this is not done by way of "an arrangement of correction and punishment even up to dismissal both of politicians and officials" (Finer, 1978, p. 410).

Although he states them in the extreme, the core elements of Finer's argument may be discerned even in more moderate and broadly based descriptions of bureaucracy's role. This statement from Meier (1979) provides an example:

Essentially, responsiveness to other political institutions requires that

bureaucrats realize that bureaucracy was created as a subordinate

institution designed to help political institutions perform their functions more

adequately. . . . Responsiveness to political institutions and the law means

that the [bureaucracy] should recognize that it is not the primary

representative institution, that it was created to be subservient to the president and

Congress, and that it should be responsive . . . to those institutions as

institutions and to the Constitution and the law. (pp. 112-113) Moreover, a formal theoretical statement of the hierarchical view, or what has come to be known as overhead democracy, can be found in various "principal-agent" models of the American system. Here, for example, is how Moe (1984) describes democratic politics:

Citizens are principals, politicians are their agents. Politicians are

principals, bureaucrats are their agents. Bureaucratic superiors are principals,

bureaucratic subordinates are their agents. The whole of politics is therefore

structured by a chain of principal-agent relationships . . . on down the

hierarchy of government to the lowest-level bureaucrats who actually

deliver services directly to citizens. (pp. 765-766)

Gruber (1987) has recently explored the dilemmas of controlling bureaucracies posed by the hierarchical view. She begins with the simple stance that democracy means popular rule, either by citizens themselves or through elected representatives. The problem that bureaucracies pose, then, is that they "short-circuit electoral channels of public control" of governmental decisions (Gruber, 1987, p. 11). A range of controls on the discretion of bureaucrats to make policy decisions, from very tight to very loose, is conceivable. Nevertheless, the structuring of a democratic regime, and the placement of the bureaucracy in such a regime, must adhere to the principle that the bureaucracy is distinctive by its subservience to both the rulers and the ruled. "The issue [for modern democratic theory] is not merely the proper distribution of power between the rulers and the ruled, but also the relationship between the rulers and the ruled together on the one hand and the bureaucrats on the other" (Gruber, 1987, p. 32).

After analyzing in depth the relative costs and benefits of various forms of control, the attitudes of bureaucrats and thus their receptivity or aversion to popular control, and the resources bureaucrats have to fend off efforts at control, Gruber modulates the heavy authoritarianism of the hierarchical view into an "exchange" theory of control. "In an exchange model, control results not from political actors telling bureaucrats what to do but from constructing conditions in which bureaucratic behavior is constrained in exchange for resources that bureaucrats seek" (Gruber, 1987, pp. 210-211). The fundamentally hierarchical relationship remains, however. "Control by exchange, of course, operates against a backdrop of authority" (Gruber, 1987, p. 211).

Meier argues that the normative foundation for the hierarchical view is Woodrow Wilson's prescription for a separation of politics and administration. "Because the political branch of government (in Wilson's view Congress) was in charge of politics and policy making, the bureaucracy should limit its concerns to administration. The dichotomy between politics and administration quite naturally led to the prescription that the bureaucracy should be subordinate to the will of the politicians" (Meier, 1979, p. 131; also see Burke, 1984, pp. 180-181). Similarly, Gruber argues that by adopting "the Weberian description of the impersonal bureaucrat working with equal diligence for a succession of masters," Wilson and other proponents of the separation of politics and administration could argue that "democratic control of appointed administrators was assured through their subservience to elected officials" (Gruber, 1987, pp. 5-6).

The Wilsonian dichotomy, principal-agent theory, and the general idea of overhead democracy all rest on the bedrock of a more fundamental normative argument, however: the Hobbesian answer to the problem of connecting sovereignty and representation. Hobbes was the first to address this problem with a fully developed theory of state sovereignty and political representation (Pitkin, 1967; Schwartz, 1988). Hobbes's solution was not only the sovereign leviathan, but also the formalistic …