Baldy Center for Law and Social Policy Working Paper
Working Group on Sociology of Regulation
ADMINISTRATIVE REGULATION AND DEMOCRACY
by Errol Meidinger
Professor of Law
State University of New York
Buffalo, NY 14260
This is a working draft circulated for discussion pending substantial revision and expansion. Comments are most welcome. I gratefully acknowledge earlier comments by Jim Atleson, Bob Bell, Guyora Binder, Barry Boyer, Diane Ciurczak, Allan Freeman, Don Gjerdingen, Fred Konefsky, Jane Mansbridge, Tom McGarity, Jack Schlegel, Margaret Shannon, Joe Tomain and Gerry Wetlaufer.
© 1992 by Errol Meidinger
Contents
I. Introduction
A. The Conventional Critique of Administrative Regulation
B. The Conventional Solution and Some of Its Limitations
C. Reformulating the Problem
II. The Domain of Democracy: Politics versus Rights
III. The Political Community
A. The Bounds of the Polity
1. Outcome Variations
2. Types of Democracy
B. The Delegation Problem
C. The Solution of Representativeness
IV. Self-Interest
A. The Public Interest
B. Deliberation
C. Reprise
V. Power
A. Definition and Sources of Power
1. Social Structure
2. Culture
3. Social Action
B. Good and Bad Power in Theory
1. Legitimate Authority
2. Pure Communication
3. Abstracted Self-Interest and Equal Power
4. Concrete Self-Interest and Constrained Power
C. Conclusion: Good and Bad Power in Practice
1. The Power of Agencies and the Delegation Problem
2. The Power of Regulatees and the Capture Problem
3. The Power of Beneficiaries and the Veto Problem
I. Introduction.
Administrative regulation is a major institutional mechanism for defining and implementing social order in this and most other modern societies. Domains of social life subject to regulation run the gamut from food and drug consumption to pollution production, from toy safety to building design, from commodities trading to ship safety, and so on ad infinitum. Most modern governments practicing extensive administrative regulation base their claims to legitimacy on the ideal of democracy, or "government by the people". (1) Yet, paradoxically, much administrative regulation is subject to significant challenges based on the ideal of democracy. For reasons outlined below, these challenges appear to be at least partially misframed. Moreover, their proposed correctives are often neither particularly democratic nor practical. This paper begins the work of bringing together several strands of normative and empirical research to reconceptualize the relationship between administrative regulation and democratic legitimacy.
A. The Conventional Critique.
The dominant contemporary criticisms of administrative regulation focus on the fact that regulatory officials often seem to have considerable "discretion" in making regulatory decisions. Many such decisions take the form of commands which are legally binding on parties within the agency's jurisdiction. Yet because they involve discretion, the contents of these commands are not dictated by legislative directives. They are therefore arguably not those of the legislature, whose commands are typically portrayed as legitimate under the "government by the people" standard because legislators are elected "by the people". Nor are these commands often those of the elected executive, who frequently has no legal authority to determine regulatory decisions and in any case cannot possibly make the bulk of them.
This situation is portrayed as fostering a number of evils. First, of course, many administrative decisions fail to meet the basic criteria for democratic governance, since they are arguably not government by the people (e.g., Lowi 1969, Freedman 1978). Moreover, in the American system they may be seen as unconstitutionally violating the "separation of powers" structure set out in Articles I and II of the Constitution, which direct the Congress to legislate and the President to execute the laws of the land. Allowing administrative agencies to make binding rules may be viewed as allowing them to legislate in place of Congress.
Administrative regulation has also been characterized as suffering from many other evils due largely to popular non-accountability. These include the political "capture" of agencies by their ostensible subjects (e.g., Huntington 1952, Bernstein 1955), unequal treatment of similarly situated parties (Davis 1969), regulatory patronage (Lowi 1987), ineffective regulation (Schoenbrod 1983), evasion of legislative responsibility (Ely 1980), dilluted rationality (Winter 1985), excessive production of public goods (Coase 1959), excessive production of private goods (Aranson, et al 1983), and so on.
B. The Conventional Solution and Some of Its Limitations
While the accuracy of each of the above problem diagnoses can be seriously disputed (e.g., Mashaw 1985, Pierce 1987), the scholars who make them tend to converge on a single solution: legislatures must specify in statutes the precise rules to be imposed on parties subject to regulation. Unfortunately, and regardless of whether the claims that give rise to it are valid, this solution seems to suffer from such severe problems as to be quite untenable. The problems are numerous and multidimensional, and cannot be discussed fully here. Perhaps the most readily understandable are those of agenda overload and inherent limits to rationality (Stewart, 1987). Agenda overload derives largely from the extreme complexity of modern society. There are just too many potential issue areas, each involving too much detail and too much complexity, for any single legislative body to deal with effectively. No group of human legislators could possibly know enough even about the regulatory problems listed above (food and drugs, pollution, toy safety, construction design, commodities trading, and ship-building) to promulgate wise, detailed, effective rules for their regulation.
The problem of complexity is exacerbated by profound limits to collective rationality. The best known of these, the "paradox of voting", holds that there will often be no single best solution for problems concerning two or more individuals and involving two or more alternatives (Arrow 1951). In addition, even were legislatures to articulate clear rules for policy decisions, those rules would by nature almost always be indeterminate for at least some policy alternatives (e.g. Meidinger 1987). Finally, requiring legislatures to make such comprehensive decisions would impose enormous information costs, require immense analytical resources, and would probably be unwise in many cases anyway, since high levels of uncertainty often indicate that problems should be dealt with on a continuing, incremental basis (e.g., Lindblom 1959). These considerations, together with the fact that democratically elected legislatures continue to create and fund systems of administrative regulation, suggest that there may be important criteria for democratic legitimacy other than the model of legislative rule control relied on by most critics.
C. Reformulating the Problem.
The normative presumption that policy-making is to be done by a legislature and executed by its governmental agents has obscured several of the key empirical characteristics of modern administrative regulation. First, "delegations" of policy making powers are quite widespread and take many forms. Some are to local or regional levels of government. Examples include zoning, insurance regulation, restaurant inspection, and countless others. Other less obvious delegations are to non-governmental institutions, such as professional groups (e.g., lawyers' and doctors' associations), neighborhood organizations, and even religious organizations and business corporations (e.g., Jacobson 1980). As will be elaborated below, all of these institutions govern the lives of their subjects in various ways by promulgating and enforcing codes of conduct, allocating vital resources, and so on.
Second, wherever it is done, regulatory decision making generally does not conform to a simple bureaucratic model. High level officials do not "make" regulatory policies which low level officials implement. Rather, policy making in modern democracies involves complex negotiations not only within governmental agencies, but among agencies, regulated subjects, regulatory advocates ("public interest groups"), legislators and legislative staffs, and often other interested actors such as researchers and journalists (e.g., Sabatier 1988, Jordan and Richardson 1983). Although the formal distribution of authority is important, expertise -- detailed knowledge of the regulatory domain and the likely operation of alternative policies -- is also a major source of regulatory influence. Not surprisingly, membership in regulatory decision making groups tends to vary with the issue involved. There are few common participants, for example, in environmental regulation and commodities trading regulation. Even within a relevant regulatory area, specific policy issues tend to be further delegated to subgroups involving a small number of members from each relevant type of interest (e.g., Hjern and Porter 1981).
Third, although policy actors bring existing, often divergent interests to policy negotiations, their definitions of interest may change significantly during negotiation processes (e.g., Shannon 1989). This can happen as they grasp previously unknown implications of particular policy proposals or as they reflect on their future roles in the system, which they may find appealing or unappealing. It can also happen because regulatory actors are accountable not only to the external "interests" they represent, but also to each other, whose mutual respect they desire and who are often capable rewarding and punishing past behavior (Boyer and Meidinger 1986, Meidinger 1987). Their definitions of self-interest, therefore, are often influenced by their desire for respect and their continuing interdependence with other members of regulatory networks.
Fourth, the most important results of regulatory negotiations are not always encoded as formal rules. Rather, they often take the shape of "regulatory culture". By regulatory culture, I mean shared understandings which can both shape future negotiations and guide behavior (Meidinger 1987). Because such shared understandings are often taken for granted, it is easy to neglect their importance in shaping regulatory practice. Deliberation based on a certain type of rhetoric, cost benefit analysis for example, can have a large influence not only on what policies seem acceptable or natural, but also, as suggested above, on how regulatory actors perceive their interests (Reich 1985). Conversely, of course, this is precisely why so much regulatory interaction is devoted to setting the terms of the debate, rather than to the precise formulation of rules, and ultimately why culture is so important. (2)
Because of their characteristic diversity of domains, negotiation among interdependent groups, interest shaping functions, and reliance on shared understandings, regulatory groups are in many ways comparable to traditional, geographically focused communities. Such communities have long confronted the problems of democratic decision making, and I believe it may be fruitful to analyze regulatory networks in analogous terms. For convenience sake I will call regulatory groupings by the generic term "regulatory communities", but hasten to add that I do not intend to romanticize the situation with the term. Geographical communities as we know them are partly defined by continuous, often bitter conflict.
Overall, then, the system of modern administrative regulation involves a multitude of partially autonomous "regulatory communities" each focused around a specific type of social problem (cf. Moore 1973). Like other communities and municipalities (cf. Frug 1980) these regulatory communities are persistent organizations which occupy an intermediate position between the formal apparatus of the state and that of the economy. They are neither inherently "public" nor "private". While regulatory communities may be answerable in various ways to central legislatures, they are not simply "agents" of those legislatures.
Any democratic legitimacy regulatory communities might possess, therefore, must be based on more than the legitimacy of legislative decision making. I accept as a working postulate that some sort of technically expert, problem specific administrative regulation is necessary in technologically complex, economically differentiated and interdependent modern societies. I also believe that all regulation must be held to the standard of "government by the people". (3) But I do not by any means believe that all modern regulatory communities meet that standard. A critical problem for assessing modern administrative regulation, therefore, is to develop a revised conception of democracy which is actually appropriate to the problem and practice. Doing so requires working between the general features of administrative regulation just sketched out and the available body of normative theory about democracy. In doing so, however, I will assume that the general characteristics of administrative subject to change based on a coherent critique, and may be completely transformable.
Normative discussions of the democratic legitimacy of regulatory communities revolve around several problems central to both regulatory practice and political theory. These include the problems of defining the relationship between rights and politics (Section II), the bounds of the polity (Section III.A.), the appropriateness of regulatory delegations (Section III.B.), the nature of representation (Section III.C.), the appropriate role of self-interest in governance (Section IV), and the appropriate role of power in governance (Section V). These problem definitions are taken by many to imply their own solutions, and I take up the applicability of the most commonly proposed solutions to the problems of administrative regulation in turn.
II.The Domain of Democracy: Politics versus Rights
The ideal of democracy as government-by-the-people has often been taken to mean that the people as a whole, rather than particular groups or individuals, should make important social decisions. A critical problem, however, has been to define the realm of "important social decisions". Even in principle, it is impossible to make every social decision in any complex social order democratically (or collectively at all). At the same time, virtually any particular decision allocating scarce resources or determining acceptable conduct could be made "by the people".
Such possibilities ordinarily raise serious concerns among those who control substantial resources or enjoy considerable security during any given polity. Historically, it has been primarily the wealthy who have been protected by rights limiting the domain of democratic politics (e.g., Nedelsky, 1990). Their concerns were protected by two key cultural and political choices. One was to attach a personal interest requirement -- the people should make important social decisions that affect them -- and then to define a large domain of actions as not affecting the people -- i.e., as not affecting protected property or contract rights. For example, an individual who had accumulated great property resources was not accountable to a community for deciding not to use those resources in a productive way -- so long as he had no contractual or proprietorial obligation to do so. Though such a decision might have a considerable practical effect on other community members, it had no legal effect on them. The other key move, in some ways the flip side of the first, was to define exercises of traditional property or contract rights as outside the domain of democracy, and indeed as protected from it by the due process, takings, and contract clauses of the Constitution. Together, the two moves serve to establish separate, mutually reinforcing spheres of public and private action. To some extent the creation of an autonomous private sphere was justified in terms of democratic theory: material independence has long been portrayed as a necessary foundation for virtuous democratic citizenship in American political theory. (4)
In practice, the boundary between democratic policy making and protected rights has regularly been a major focus of political struggle. It organizes much of constitutional history, and still generates several thousand pages of scholarship each month. Rather than adding much to the volume here, I wish simply to make the point that the issue is still with us. Even when solicitude for one category of protected private rights has waned, that for others has general grown. Fundamentally, most American citizens would be quite uncomfortable with eliminating all spheres of protection from democracy. If we still feel a need to limit the domain of democracy, then a theory of democratic regulation must consider whether there are any types of actions that should be outside its capacity. The old problems of what immunities to provide property, privacy and expression rights are continually joined by new ones. For example, should the Nuclear Regulatory Commission, implementing the purposes of a democratically elected Congress, have the power to force residents of any given local community to live near high level waste repository, or should residents be protected against that sort of democratic decision? (5) As far as I am concerned, the answer is not simple. Although the legal case against such regulatory power is currently weak, I think the question may yet receive serious political deliberation in the society at large. And I believe it possible that repository siting exercises of regulatory power will eventually be defined by cultural political processes to be beyond the capacity of democratic regulation.
III. The Political Community.
A. The Bounds of the Polity.
1. Outcome Variations. Even if we conclude that there should not be a protected right to be free of having a nuclear waste dump located next door, and therefore that nuclear waste siting is an appropriate issue for democratic decision, the tyranny-of-the-majority problem raises its head in another form. What should be the bounds of the polity? Should it be the nation, the state, or the locality? I doubt that a deductive process called democratic theory can offer a lasting answer to this question. It will necessarily be an issue for very broad discussion and argument. At the moment I find it hard to predict any fundamental consensus. A substantial and perhaps growing portion of the citizenry considers it simply wrong to force such burdens on unwilling communities. It refuses to sign off on an "acceptable level of risk" (e.g, Rayner, 198 ). Other parts of the citizenry are more utilitarian, and are willing to impose the burden if the benefits greatly exceed the costs and the risks.
Whether it is democratic or undemocratic to impose nuclear waste repositories on unwilling communities will depend in part on our developing a much better understanding of just what the burden is. But given the problems of uncertainty, we can only go so far in finding out. And of course we face the second order question whether it is more democratic for experts or citizens to decide how risky the burden is (e.g., Dietz, forthcoming). In practice, society will probably play a big role in defining the risk by creating a widely shared image of what having a nuclear waste repository next door means. That will be a cultural process of the form outlined in Section I and detailed more adequately in "Regulatory Culture" (1987). After that process is farther along we may have a better answer to the question of democratic siting. Meanwhile, the representatives of the majority at the national level seem to be betting on a favorable outcome with the currency of power. (6)
2. Types of Democracy. The size of the polity also will have a major effect on thetype of democracy that can feasibly be practiced. In general, the larger the polity, the more remote and adversarial will be the practice of democracy. Conventional wisdom suggests that when a polity attains a certain size, direct democracy (decision making in which all citizens are free to participate and, if necessary, vote) is not possible. Rather, democracy must be implemented by and through elected representatives, as exemplified in American state and national governments. Conversely, in extremely small polities a special type of direct, face-to-face democracy is possible. Jane Mansbridge (1980) describes this kind of polity in her study of a forty-one member cooperatively organized crisis intervention center. Because of its small size and the essential commonality of its members' interests, Mansbridge argues, this form of polity is able to practice a special ("unitary") kind of democracy based on three principles: equal respect for all members, (7) consensus decision making, and regular face-to-face interaction. A third possible type of democracy, located between the poles of unitary and large scale representative democracies, is exemplified by Mansbridge's case of a New England town meeting. Although this form of democracy also allows regular, direct participation by citizens, its larger size and the greater divergence of its members' interests preclude the possibility of consensus decision making on all issues. Consequently it must often operate on the principles of majority rule and equal protection of interests as implemented by secret ballot.
We can thus distinguish three types of democracy, based on the criteria of increasing social scale and divergence of interests: unitary, direct-adversary, and representative-adversary. Structurally, regulatory communities would seem to fall closest to the town meeting type of direct-adversary democracy. Although the number of participants in most regulatory communities is likely to be smaller than the populations of even small New England towns, it is likely to be larger than the number of members in most unitary democracies.
More importantly, although regulatory community members will have some interests in common (e.g., mutual interdependence requiring them to make the system work at least at a minimal level), their interests will typically be as divergent as those of the citizenry of most small towns. Therefore it might be plausible to analogize regulatory communities to direct-adversary forms of democracy. But the analogy encounters two serious obstacles. First, most regulatory communities do not formally provide for either majority rule or consensus decisions. Rather, they provide for hierarchical decision making. Formal authority for final decisions typically resides in the head of the governing agency; practical authority is often exercised by designated staff members. In practice, agencies may often seek decisions resting on as much community consensus as possible, and other actors in regulatory communities sometimes exercise considerable powers to override official decisions. but there are no formal decision mechanisms of the type common in either direct or representative democracies to ensure that officials respond to participants' desires. The governance structures of regulatory communities, therefore, tend to look more like oligarchies than democracies. To this criticism (generally stated in terms of the non-elected status of administrators) the typical response is that administrators are accountable to democratically elected representatives, and that therefore their policies are democratically justified. But this response throws the democratic analysis of regulation into a second obstacle: the delegation problem.
B. The Delegation Problem. The traditional terms of the delegation problem are simple. Although the people themselves do not make laws in a representative democracy, they elect the individuals who do. Laws made by duly elected representatives therefore qualify as legitimate "government by the people". Administrative action, by extension, is legitimate so long as it simply implements the will of the people as expressed by their legislators. Perhaps not surprisingly, regulation was long legitimated in precisely these terms (see Stewart, 1975). This rationale, that administrators were simply executing legislative decisions, was probably never quite believable to anyone who knew or thought about the actual operation of regulatory agencies. The indeterminacy of rules (8) probably means that it is an inherently impossible condition to meet. In any event, it has become increasingly clear with political experience that agencies frequently make policy, sometimes quite controversial policy. Consequently, some critics have argued that because administrative actions are not directly derivable from the decisions of elected legislatures they are indefensible (e.g., Lowi, 1969; Ely, 1980). (9) This conclusion seems questionable to me for several reasons. First, it does not necessarily follow as a matter of formal logic. If rules made by directly elected representatives can meet the government-by-the-people standard, then there is no inherent reason why rules made by the agents of those representatives (whom we might think of as second order representatives), and subject to their supervision and reversal, cannot meet it. The decisions of administrators are in principle no less predictable than those of legislators. Indeed, this defense is often made of regulation, though perhaps less and less frequently by academics.
Second, there are many spheres of social life in which enormous authority is exercised in ways that are not directly derivable from legislative decisions. Because it may confront certain conceptual blinders, this point deserves some elaboration. In light of the definition of democracy as "government by the people" most people probably treat democracy as a problem of the state, the traditional governmental apparatus. If so, that can be taken as some evidence of the success of the long term cultural process of separating public from private spheres of action outlined above. I expect, however, that on reflection most people would agree that democracy is not solely or even perhaps essentially a problem of the state. The ideal and practice of democracy pervade all forms of social organization: friendships, clubs, families, businesses, religious organizations, faculties, and so on. Most of these organizations are not branches of the state. (10)
In any event, whether such organizations operate democratically or not, they often do a significant amount of "governing" by virtue of allocating large quantities of social resources and establishing rules governing the conduct of people who participate in and deal with them. Corporations like General Motors, for example, make decisions ranging from which community will experience the economic growth (and possibly social disorganization) resulting from the construction of a new manufacturing plant, (11) to how long its executives will live in a certain community, (12) to which drugs its employees may ingest. (13) They therefore exercise considerable social authority. Often, their authority is directly conferred or ratified by legislatures and enforced by courts, as in the case of business corporations. But the way they exercise that authority is neither determined by nor predictable from the rules established by legislatures. (14)
In a nutshell, many nominally private organizations in reality exercise large amounts of delegated authority over large spheres of social conduct. Nonetheless, few contemporary democratic theorists rush to argue that such delegated authority is impermissible because it is not directly derivable from democratic decisions of the citizenry or legislature. The failure to do so by proponents of the non-delegation doctrine may not make sense as a logical matter, since the only feature that distinguishes state from private delegations as an a priori matter is the label. Neither type is inherently more democratic than the other, (15) and any defense of private delegations must therefore rely on additional criteria. (16)
The third general problem with the attack on regulatory delegations is practical. Just as societies might reasonably decide to establish and respect protected spheres of conduct, they might also decide to delegate resource management and governance powers over certain types of social problems to social groups with a particular desire or ability to handle them. Indeed, some system of delegation seems an absolute necessity. The reasons include not only the problems of indeterminacy and the paradox of voting, (17) but the practical fact that there seems to be no way any legislature (much less large citizenry) could deliberate and decide about every important resource use and behavioral decision that must be made.
Even in the impossible case that legislatures could muster the time, energy, consensus, and statutory precision to decide all important issues, they might not always be the best fora for doing so. First, the community actually interested in a given regulatory issue might be much smaller than the legislature, or might cross jurisdictional boundaries of existing legislatures. Because the legislature faces a plenitude of issues, it might not decide an issue hotly contested within a partial community with great care. Second, and more importantly, that community might have a much stronger relationship to and greater knowledge of the issue (valuable local culture) than the legislature. Rather than distrusting that relationship, and seeking refuge in pure democratic decisional procedures, we might be better off allowing it to continue and to grow. Third, when issues involve substantial uncertainty, it may make sense, rather than resolving them in one fell swoop, to let the interested regulatory community work out tentative accommodations, then follow the development of the problem, and make adjustments when necessary. (18) The problem for scholars and critics would be to figure out how to make such regulatory communities work better, rather than to call for their abolition. Fourth, there is an old and respectable argument for encouraging regulation by delegated communities as a good thing in itself. Participation in regulatory communities may often serve to educate us as citizens and to improve our understanding of the problems and operation of government in all its forms. (19) It may give us an opportunity to participate in and shape community cultures we care about. Given the complexity and differentiation of modern societies, an equally complex, differentiated web of voluntary regulatory communities may be the only meaningful way to facilitate citizen participation in government. (20) The practical alternative of voting for legislative candidates based on slogans and glittering generalities can hardly be defended as inherently superior.
My argument based on regulatory communities and local culture suggests that we should consider supporting delegations to regulatory communities absent definite reasons to the contrary. I do not think it will be fruitful or feasible to make the locus of decision a conclusive criterion for determining the validity of a decision. Many important decisions in any society are made in subsectors of that society. Moreover, democratic theory provides very little guidance on what the boundaries of the polity should be for any decision, since many decisions affect only a portion of the citizenry, or affect one portion much more than others (e.g., hazardous waste siting), and other decisions (e.g., those about taxes or trade restrictions) have decisions made by interested, informed regulatory communities, although affecting the society at large, are likely to be better informed, better calibrated to the problems at hand, and perhaps more adjustable than those made by legislatures in many cases. Because of the inherent constraints on legislative decision-making (21) and the potential strengths of delegated decision making, there is no inherent reason not to let legislatures decide when they are not the best groups to make particular decisions.
This presumption in favor of delegations seems all the more sensible in light of the participation in regulatory communities of representatives of the parties interested in regulatory outcomes. If the direction of the regulatory decision seems inconsistent with the probable result before a legislature, interested parties generally have the option of taking the issue to the legislature, the executive, or even the public, and they frequently use or threaten to use that option. Therefore, although neither the structure nor the decisional procedures of regulatory communities may meet the formal criteria of democratic theory, their decisions seem likely to operate within parameters set by democratic institutions and to approximate the outcomes desired by those institutions. Given the other general advantages of inclusive regulatory communities, a substantive democratic standard would seem to suggest a weak presumption in favor of their validity absent evidence of nondemocratic outcomes. This, it seems to me, is what most courts in fact generally practice in judicial review. They try to make sure that the policy produced is consistent with the legislature's statute, that the agency pursued a reasoned decision making process, and that all interested parties were reasonably represented and listened to in the process. Even so, there remain at least two potential challenges asserting that even if regulatory policies are consistent with legislative compromises and involve the participation of all interested parties, they may fail to qualify as democratic.
IV. Self-Interest.
One type of democracy-based challenge to contemporary regulation holds that it may fail to be democratic because it responds to private interests. Regulation may fail to be democratic even if it responds to all interested parties and achieves maximum social benefits at minimum social costs, thus simultaneously fulfilling the pluralist and efficiency ideals. The reason politically and economically optimal policies might fail the test of democracy is outlined at the beginning of this section. By simply allocating resources to meet current wants they may reflect a failure to engage in affirmative societal deliberation. Policies should embody collective decisions on desirable social relationships and resource uses -- in short, on the nature of the good. Often, achieving the good will be inconsistent with some or many preexisting preferences or interests. In any event, optimally fulfilling existing wants is at best one vision of the good. To be a valid democratic principle, it should be affirmatively selected in deliberative democratic process open to choosing other alternatives. The problem in applying this criterion of democracy is to know when satisfactory democratic decision-making has occurred and when it has not. Formal criteria, such as universal participation and consensus or majority vote are not satisfactory. All of them may function simply to validate the pursuit of existing private interests.
There are two general approaches to the problem. The first is to look for the existence of a distinctively public interest in regulatory policies. The second is to try to determine whether affirmative, forward-looking deliberation has in fact occurred.
A. The Public Interest. The key difficulty with looking for the existence of a distinctively public interest in policies is to define the public interest. Once it is accepted that the public interest will be a collective creation not reducible to any separate standard, there is no external touchstone or deductive standard against which a given policy can be tested. Accordingly, critics often proceed from what they perceive to be the opposite of the public interest: private interests. Cass Sunstein, for example, argues that a policy fails to meet the public interest standard if it is reducible to private interests (1985:49; 1986:274). Conversely, a policy meets the standard if it entails something beyond private interests.
Although Sunstein's formulation has a certain arithmetical attraction, it is extremely difficult to apply in practice. On one hand, neither Sunstein nor any other legal commentator I know of provides any guidance or even illuminating examples on how to figure out if a policy is reducible to private interests. Moreover, legions of political scientists, sociologists, and economists have spent their careers trying to show how public policies are simply reflections of private interests -- generally without overwhelming success. On the other hand, starting from the alternative premise (that a policy is not reducible to private interests) does not necessarily imply a ringing endorsement of the policy, or demonstrate that it is democratic. The policy could of course be arbitrary or random. It could be irrational. Therefore it is still necessary to figure out whether the policy is in the public interest.
It is possible to come to a conclusion on whether a policy reflects the public interest either by referring to one's understanding of the public interest or by considering the motivations of the policy makers. Sunstein seems to be most fundamentally concerned about motivation when he defines the problem of self-interest. Many of his preferred institutional solutions appear to be based on eliminating self-interested motivation from democratic decision making (1985:77-85). I have indicated my response to this proposal in "Regulatory Culture" (1987). Briefly, social life seems to me almost always to involve a combination of pecuniary interest-pursuit and citizenship. Actors' definitions of their self-interest are generally congealed with a vision of the appropriate social order, of right relationships. For example, the Northwest loggers opposing further protection for the Northern Spotted Owl do so not merely because they want to retain their relatively high paying jobs for a few more years, but also because they believe they have been living in an proper relationship to the forest. Accordingly, they argue that the correct way to manage the forest must reconcile the protection of endangered species with other values supporting the continuation of loggers established ways of life. (22) Further, members of communities often combine efforts to further their interests with efforts to make their communities better. Their definitions of self-interest mix material desires with normative visions. The ways in which they care about their communities are often not very different from the ways in which they care about themselves. Effective, knowledgeable, responsible citizenship, in sum, is most likely to found in conjunction with civic attachment. Reason is most likely to be applied by passion -- in the form of interests. (23)
I therefore have serious doubts about political strategies that appear to depend simply on separating institutions involving self-interest (economy) from institutions involving citizenship (government). Although such a separation might not be intrinsically bad, it courts the assumptions both that virtue is irrelevant to production and distribution, and that self-interest is irrelevant to government. More significantly, it seems to imply a divided structure of individual character. Citizens are to be virtuous in government and venal in business. To allow such a division, beyond being hypocritical, is to make it all the more difficult to hold economic life to any ethical standards. To the degree that American society has moved in that direction, I suggest it has been a cultural movement of the general form outlined above. If so, it is also in principle reversible through cultural politics, albeit with structural and practical constraints.
B. Deliberation. The second approach to the problem of discerning democratic decisions is to check whether deliberation has in fact occurred. But this approach encounters the same problems as the public interest approach. Either it is limited to formal review of whether people went through a process that looked like deliberation, or it examines the products of deliberation to see whether they meet substantive criteria. The available criteria tend to come down to those already noted in the introduction to this section. An important one that has not yet been discussed is rationality. Rationality does not necessarily indicate democratic decision making. Some kinds of instrumental rationality, like efficiency, may fail the tests of democracy as discussed above. Even if they are consistent with other democratic criteria, efficiency and other forms of instrumental rationality are not likely to qualify as positive justifications based on democracy.
There have also been renewed efforts to construct non-instrumental conceptions of rationality recently. Robert Bell, for example, attempts to develop a concept of administrative rationality in which regulatory agencies choose both ends and means. His prescription for how they might do so, however, seems to be essentially a rule-of-law one. In order to achieve rationality agencies must develop a legal culture, which is the ability to interpret their mandates according to extant legal and social authorities (1985:216). Agencies, in other words, must figure out how to operate within the normative systems of the societies in which they act. Although Bell's conception is thus consistent with my own in several basic assumptions, I cannot see how rationality provides an independent criterion for assessing agency actions. Rather, it seems either to dissolve into the concept of culture (what is rational is what is consistent with the overall social framework) or to become circular as discussed above (an outcome is rational if it resulted from deliberation; deliberation occurred if rational outcomes resulted). Moreover, rationality does not provide a means for evaluating or changing the norms that structure the larger social system. Thus while rationality in any of its possible forms may be desirable and important, it seems to have little meaning apart from signifying a workably coherent accommodation of conflicting social purposes within the existing social structure.
C. Reprise. Neither the elimination of self-interested motivation nor the requirement of rationality offer solutions to the problem of achieving democratic deliberation. It seems to me that any social community must deal directly with the problem by culturally defining the forms of self-interest members may legitimately pursue and enforcing those definitions. When, for example, an official advocates a strict regulatory policy in hopes of obtaining a payoff from a regulatee, the community must develop the means to reject the policy and deter officials from pursuing such self-interests. When an official advocates a strict policy in hopes of obtaining a promotion, however, the community cannot reject the policy simply because it is self-interested. The community must develop other means of deciding whether the policy fits with or furthers its purposes. The same is true when an executive or a faculty member seeks a position for a personal friend. Some communities choose (often with little success) to prohibit such behavior. Others choose not to prohibit it because of its potential contribution to the community interest. In any event, such self-interested behavior often imposes special strains on communities, and they accordingly must develop various procedures and norms to constrain and channel it. Banning self-interested behavior entirely seldom works, and pretending it does not exist simply distorts relationships more than the strain of self-interested behavior inevitably does.
In sum, communities have three generic, but interrelated types of solutions for the problems of self-interested behavior. First, they can socialize members to define their interests in certain ways and not in others. Second, they can define certain (but not all) types of interest-pursuit as inherently unacceptable. The legal system attempts to do this by criminalizing political corruption, for example. Third, they can constrain interest-pursuit to certain forms and evaluate political proposals in ways which take account of the interest pursuit they represent. In other words, they can develop institutional structures to channel interests and control self-interested behavior. Overall, as I suggested above, communities can make decisions about the degree of interest-pursuit they will accept. The mix of self- and community-regarding interests is very different in different cultures. (24) And one can hope, as I do, that the quotient of community to self-regarding interest can be increased in modern societies. But the process will be one of cultural politics as outlined above. Structures can be of some assistance, as interdependent people experience incentives to develop cooperative institutions. But changing structures also generally requires pursuing cultural politics. Moreover, as Turnbull's (1972) study suggests, people do not always respond in utility-maximizing ways to material exigencies and interdependency. (25)
V. Power
The problem of power is closely related to the problem of self-interest. Power is a challenge to the ideal of democracy because its existence means that rather than being government by the people as a whole, there might be government of some people by other people. Power is a more serious challenge to democracy than is self-interest because it refers to the capacity of self-interested parties to achieve their goals, rather than to their motivations or incentives.
In this section I examine the implications of power for the democratic practice of regulation. First (subsection A), after offering a preliminary definition of power, I review the general sources of power based on the categories developed at the beginning of this paper and in the companion paper (1987). Because I conclude, among other things, that power is both ubiquitous and unavoidable, I next review four major theoretical approaches to distinguishing good from bad power (subsection B). Finally, after concluding that none of the approaches can offer a usable or realistic standard without referring to understandings evolved in concrete social practices, I review three types of good and bad power described in the conventional wisdom of regulation (subsection C). Finally, because the conventional wisdom is both underdeveloped and theoretically simplistic, I conclude that actual forms and practices of power should be a central focus for future inquiry into regulation.
A. Definition and Sources of Power.
Like the concepts of culture, community, and democracy, power is a prime and sharply contested construct in western social thought. I will again seek to avoid recapitulating the debate (26) by starting with a minimal and conventional definition of the concept: power is the ability of an individual or group to achieve a desired outcome through social action. In the course of the argument that follows, this definition will be gradually modified to better reflect the social bases that create and structure power.
Building on the concepts thus far developed, power can be thought of as deriving from three sources: social structure, culture, and social action.
1. Social Structure. Power is probably most often viewed as deriving from an agent's position in a social structure (and social structure is often but not always legally enunciated). The manager's power over the worker, the regulator's power over the regulatee, the landlord's power over the tenant, the owner's power over the slave, the parent's power over the child, the tenured professor's power over the untenured one, the property owner's power over the vagrant, the teacher's power over the student, all derive from the occupation of a social position by the agent over the object of power. Dependency is the central element in the ideal-typical power relationship. That dependency, however, is ordinarily a reflection of the larger social order. The worker must meet the requirements of the manager because the larger order allows the worker access to the socially indispensable productivity of capital only if she does. (27) Similarly, the tenant must meet the landlord's conditions because the larger order will allow and help the landlord to exclude her otherwise.
Conversely, the larger order may act occasionally to limit the power exercised by managers and landlords -- for example by prohibiting racial or sexual discrimination, or by prohibiting unhealthy working conditions. Moreover, dependency is generally reciprocal to some extent. Managers and landlords are dependent on the cooperation and consent of their workers and tenants. Without it their businesses could be ruined, so they must work constantly to create it (e.g., Burawoy, 1979). Thus workers and tenants also have some power over managers and landlords. During various periods of activism and reaction their relative power has alternately increased and declined somewhat, although the general terms of the relationships seem to have remained relatively stable. Any significant shift in their relationships would require a change in the society at large, and therefore in the shared understanding of the appropriate structure of society.
2. Culture. Culture is a source of power first because it is an indispensable source of structure, which is a source of dependency. (28) Culture also creates power in other ways. Most people realize, for example, that knowledge of culture -- knowledge of a group's traditions, its routines, its shared assumptions, its conflicts -- can be an important source of power. Also, shared understandings can themselves be a source of power, as when one of several individuals moving a piano says "lift" and the others lift (Collingwood, 1942:153) or when a bandleader gives the downbeat and the other members play (Becker, 1982). More importantly, as noted in Section II, culture defines the primary terms of political interaction. If workers want to change their relationships with managers, for example, it is important for them to able to draw upon widely shared cultural images, like the principles of fair reward for labor, security of expectations, and personal dignity, and to apply them to their own situations in pressing their cases in the larger social system. (29) Actors who are good at manipulating cultural principles, moreover, can often exercise more power than those who are not (e.g., Schattschneider, 1960). Finally, culture can also be a source of power in the opposite way, by obscuring issues or making it difficult to get them on the social agenda (e.g., Bachrach and Baratz, 1963). Contemporary examples include the difficulty of seeing private property as a delegation issue and the difficulty of making animal rights a political issue.
3. Social Action. As the postulates of political interaction and partial incoherence indicate, simply taking affirmative action can also be a source of power. By doggedly pursuing objectives, social actors can sometimes effectively create power -- the ability to achieve objectives -- where there was none before. Polsby offers the example of "an unprepossessive Negro Dining-car waiter" who exercised power by being "extremely diligent in pressing claims on renewal planners on behalf of the welfare of his block" (1980:131). Similarly, by acting to organize themselves and pursue mutual action plans, workers and tenants have at times affected their power relationships with managers and landlords. The Nader movement and some environmental organizations provide more recent examples of the creation of power through purposive social action. (30)
In sum, power can exist virtually everywhere, and can be marshaled by many different actors for many different purposes. It is both derivative and constitutive. Power can be used to work for democracy (or any other social value) as well as against it. Power can be used by privileged people and also against them. (31) Therefore it will probably be no more helpful to condemn power per se than to condemn self-interest per se. The threshold problem is to distinguish good power from bad power. Then it might be possible to foster good power and suppress bad power.
B. Good Versus Bad Power in Theory.
This section examines four general approaches to the problem of distinguishing good from bad power. The first, associated primarily with Max Weber, attempts to distinguish good power (legitimate authority) based on voluntary consent. The second, associated primarily with Jurgen Habermas in social theory and applied most thoroughly by Bruce Ackerman in legal theory, attempts to define good power as that which operates through pure communication, or rationality alone. The third, that of John Rawls, attempts to distinguish good from bad power in terms of what free contracting individuals would accept if they did not know their concrete interests. The fourth, recently developed most clearly by Jane Mansbridge, attempts to distinguish good from bad power in terms of the actual self-interests of community members and the social contraints they develop to define and structure their interests in the effort to implement democratic ideals. I conclude that, primarily because communities actually construct and reconstruct individual self-interests, only Mansbridge's approach offers any real possibility of informing a meaningful dialogue about regulatory democracy. However, to be realized, that possibility must be pursued in the actual practice of regulation.
1. Legitimate Authority. Social theory's traditional approach to distinguishing good from bad power has been the concept of legitimate authority. As the basis of legitimate authority shifted from sacred to secular during the Enlightenment, the analytic procedures of moral philosophy and then empirical social science came to the fore. In both cases, the ideal of consent was the pivotal source of legitimacy. Starting with Hobbes, social contract theorists held that the origins of society must lie in some primordial, voluntary agreement among individuals to join together in a society (generally on grounds that everyone would be better off as a result) and to abide by the rules of that society. After thus justifying the existence of social authority, they turned their attention to the very difficult problem of distinguishing legitimate from illegitimate rules, which continues to embroil moral philosophy today.
Gradually, the ideal of consent also took on a growing empirical dimension. In a formulation still central to social science discourse, for example, Max Weber attempted to distinguish legitimate authority from simple power primarily in terms of the existence of an attitude of voluntary consent on the part of the governed (1978:214-217). Just as moral philosophers repeatedly ran up against the fact that social institutions were as likely to be founded on violence as contract, however, social scientists continually found that consent was as likely to be manufactured or forced as voluntary. Indeed, in a world of significant economic and political inequality, it was difficult to see truly voluntary consent in any empirical relationship or bargain. At the same time -- and in significant part as a consequence -- the ideal of voluntarism continued to carry tremendous normative power. Contract law, for example, went through a wrenching series of involutions as its exponents attempted to reconcile the ideal of voluntary consent with the reality of coercion. (32)
On the whole, the academic project of documenting voluntary consent seems to have gone into decline for several decades after World War II. Most sociologists and many political scientists concentrated either on measuring vectors of interest or on demonstrating the lack of voluntariness in most relationships; (33) legal scholars retreated into the infinite regress of proceduralism; (34) and most economists simply treated voluntariness as axiomatic. Thus partly because of the difficulty of documenting its cognitive opposite, liberty, power was generally assumed to be either ubiquitous and inherently illegitimate or nonexistent and nonproblematic. Meanwhile, of course, much social life was devoted to achieving voluntarism and democracy in daily life (e.g., Keniston, 1968; Kramer, 1972; Zablocki, 1971). Not surprisingly, there has recently been a resurgence of interest in the problems of voluntarism and power among academics. Several academic efforts are worth noting here.
2. Pure Communication. Among social science theorists, Jurgen Habermas has shown the most serious and persistent interest in the problems of democracy, legitimacy, and consent. Although the work he has published in English over the last two decades contains complex and potentially inconsistent diagnoses of the problem of legitimacy, (35) Habermas seems to have settled upon the view that the key to political legitimacy is pure, unconstrained communicative action. By unconstrained communicative action, Habermas means first communication which is not instrumental, which is not aimed at getting a particular result in the world but rather at symbolically communicating with other human beings. He draws a sharp line between communicative action (which is expressive) and productive action (which is instrumental). He in fact seeks a complete Kantian separation between economy (which is instrumental, and uses people as means) and culture (which is symbolic, treats people as ends, and is therefore the appropriate forum for legitimate political interaction). Second, communicative action is objective, not subjective. Much like the conception of culture outlined in above, it is a system of shared understandings which can be studied and deciphered through "speech acts" such as talking and writing. Third, unconstrained communication is not only a form of expressive action, but a route to truth, to a situation where everyone involved in a conversation is in agreement:
In contradistinction to purposive-rational action, communicative action is, among other things, oriented to observing intersubjectively valid norms that link reciprocal expectations. In communicative action, the validity basis of speech is presupposed. The universal validity claims (truth, rightness, truthfulness), which participants at least implicitly raise and reciprocally recognize, make possible the consensus that carries action in common" (1979:118). (36)
For Habermas, then, although power might exist in the realm of instrumental action, it may not legitimately exist in the realm of communicative action. Once power is eliminated from communicative action, the unconstrained collective reflection made possible might even lead to the elimination of power from the instrumental realm. In any event, the key to political legitimacy is simple: eliminate constrained communication. Although Habermas' definition of constrained communication is not entirely clear to me, it is clear that he includes communication involving self-interest and that involving power in it. Legitimate political communication cannot involve either power or self-interest. Held up to the terms of analysis I have already developed -- that it is in practice impossible to delineate a sphere of either non-self-interested or power-free behavior -- the conclusion is obvious: unconstrained communication is a practical impossibility.
Even outside the terms of the analysis developed here, writers attempting to specify and apply the ideal of unconstrained communication have not fared well. Forester (1983), for example, ends up accepting that some distorted communication and some power might turn out to be necessary and legitimate in social organizations. So does Denhardt (1981), although not as forthrightly. There is no axiomatic reason in critical theory why unconstrained communication could not result in inequality. (37)
For Habermas, the legitimacy or illegitimacy of any form of power or inequality depends on the consensus reached through unconstrained communication. The critical question then becomes: what is the basis of the consensus? In principle, there seem to be only two basic answers. The process of unconstrained communication can produce consensus either by galvanizing a new convergence of previously divergent values or by leading to a mutual realization of already convergent values. Although Habermas is somewhat vague on this question, and indeed seeks to retain latitude for both answers to be true, he leans heavily on an assumption of underlying consensus, as the quote above suggests.
American legal theorists attempting to follow the path charted by Habermas generally feel a need to be more chary about the assumption of a preexisting, underlying consensus, given its rough treatment at the hands of the pluralists, its contemporaneous discrediting in legal discourse (e.g., Hartz, 1952; Freund 1954), and perhaps simply their experience of the highly conflictual nature of most important legal questions. Bruce Ackerman (1980, 1984), for example, attempts to rest the genesis of consensus much more heavily on method than on shared values. Major policy choices are to be made through a process of dialogue, or "disciplined conversation", which is based on rationality, neutrality, and conversational equality, (38) and from which passion, interests and power are banned. Without detailing exactly how, Ackerman asserts that the process of rational dialogue is capable of producing determinate results for important policy questions. Such a dialogue has never in fact occurred in the real world of politics, and Ackerman undertakes to demonstrate what would result if neutral dialogue ensued. The result is not very different from the system we already have: a moderate mix of constitutionalism and democracy, but with an increased measure of redistribution.
Thus, although the current system is the unclean product of passions, interests, and power, the sanitized product of reason, neutrality, and conversational equality would not look very different. To boot, it would be democratically legitimate, since it is what the people would choose in a process of free, unconstrained, rational communication (e.g., 1984:79). In substituting rational discourse for traditional procedural standards of democracy, Ackerman is probably not doing much more than trading one rationalization of the existing system for another. More importantly, by congealing rationalism and democratic legitimacy he in effect brings the assumptions of underlying consensus in the back door. (39) Unfortunately, he fails either to deal with the potential seriousness of the partial incoherence postulate (Section I above) or to provide any substantive standard for political legitimacy.
In practice, Ackerman appears to vacillate among the ideals of democracy, efficiency, and legalism. When democracy threatens to disturb a settled entitlement he finds desirable, rational-legal or efficiency analysis rises to protect it. When legalism threatens to protect an existing arrangement he finds undesirable, rational efficiency analysis or democratic deliberation rises to do away with it. As I suggested at the beginning of this section, such a compromise between rights, utilitarianism and democracy might in fact be socially workable and sensible, but it has no special claim to rationality. Indeed, the conflicts between the principles combined with their internal ambiguities make determinate conclusions anything but certain. (40)
Ackerman's portrayal of rationality as a product of philosophy, rather than of the rough and tumble of social life, seems to misportray both the true origins of many of the institutions he promotes and the nature of rationality. He does not make the strong deductive claims for rationality that an earlier generation of philosophers attempted (see Quine, 1952). Rather, Ackerman's rationality is essentially a form of cultural dialogue. It draws upon the alternative legitimating principles available in our shared understanding of the world; the uneasy compromise among democracy, efficiency and legalism mirror their uneasy accommodation in modern society. That accommodation was established in significant part by the play of interests and power. Yet Ackerman refuses to take seriously the real-world relationship of culture to interests and power. He assumes those relationships away without offering any strategies for severing them. In the end, therefore, his theory seems to be a retreat from the "new language of power" he seeks to outline. Indeed, one might even argue that it is no more than a fervent, self-interested bid to exercise a modicum of social power. In any event, by failing to address the problems of underlying consensus, the construction of interests, and the different uses of power, Ackerman's vision of neutral dialogue provides little help in distinguishing good from bad power.
3. Abstracted Self-Interest and Equal Power. John Rawls' (1971) modern version of social contract theory shares both strengths and weaknesses with Ackerman's attempt to describe neutral rationality. Like Ackerman, Rawls starts with rational individuals choosing social institutions. Similarly, because the situation he postulates has never existed, Rawls devotes much of his effort to imagining the probable results if it had. Rawls does take self-interest and power more seriously than Ackerman. He corrals them, however, by making two key assumptions. First, the deliberating individuals are in an "original position" of political and economic equality in which no resources have been allocated. Second they choose under a "veil of ignorance" which keeps them from knowing the positions they will occupy in the society they design, the talents they will exercise, or even the values they will eventually hold. Thus while the resulting society reflects self-interest, it is an abstracted self-interest. Deliberating individuals will attempt to design a society which will be as good as possible for them regardless of their positions in it. It will thus be the fairest possible society.
Rawls argues that individuals deliberating in the original position under the veil of ignorance would choose to design a society organized around two principles. The first principle would require equality in the assignment of basic rights and duties. The second would permit social and economic inequalities only when they benefit everyone, particularly the least advantaged members of society. Rawls proceeds to reason from these distributional principles to a vision of a just society. In the end, very much like Ackerman, Rawls concludes that rational individuals would choose to create a society resembling the modern welfare state, but with a slightly higher level of redistribution. On the one hand, therefore, Rawls seems to provide a clearer standard for separating good from bad power. Good power would be allowed by individuals deliberating in the original position because it would be in everyone's interest. Bad power would not be allowed because it would not be in everyone's interest.
On the other hand, Rawls also ends up simply importing his cultural understanding into the definition of rationality and justice. The only concrete visions of society available to Rawls are the ones that exist. The individuals in his original position are completely unreal; they have no values, experiences or concrete interests. Their process of choice under the second principle, therefore, is impossible, since the definition of any social benefit (or interest) is both variable and socially constructed. The individuals in the original position, in other words, can have no way of knowing whether a benefit of inequality is sufficient compensation, since both the cost and benefit of the inequality are contingent on a historical process of social negotiation and definition. Rawls' techniques for valuing them do not exist independent of the inequality and contingency of his own historical situation. Accordingly, because he can have no way of knowing what social institutions individuals in the original position would in fact accept, and since they would have no way of deciding what to accept without shared social experience, Rawls' attractive premises offer little real guidance for assessing acceptable and unacceptable forms of power in concrete social institutions, regulatory or other. Evaluation requires values. Values are social products of a world rife with self-interest and power. There is no reason to believe that Rawls' values were forged only by the forces of good power. Rawls has not found the elusive independent democratic ground for evaluating social institutions.
4. Concrete Self-Interest and Constrained Power. Jane Mansbridge (1980) returns the discussion of democracy to the empirical world. Rather than basing the legitimacy of power on an attitude of consent (Weber), consensus-based rationality (Habermas and Ackerman) or abstracted self-interest (Rawls), she attempts to base it on the concrete self-interests of citizens. When polities are small and their members have common interests, unitary democracy based on equal respect, consensus, and face-to-face interaction is possible.
Mansbridge argues that power is not a major problem in unitary democracies because individuals exercising power will generally be advancing the other members' interests. When members of polities have fundamentally inconsistent interests, adversary democracy is necessary. Decisions based on majority voting by secret ballot confer formal equality on all members' interests. While therefore related to Rawls', Mansbridge's formulation offers a much more concrete means for distinguishing good from bad power. Power is not likely to be bad as long as citizens have common interests. Where interests are divergent, however, the traditional requirement of equal representation must be invoked to constrain power.
Although I think Mansbridge's common/divergent interest formulation has considerably more potential for improving our understanding and practice of democracy than the others I have discussed, it turns out to be more complicated than the above characterization suggests. First, interests are socially constructed. Much interaction at Helpline, the cooperatively organized crisis intervention center she studied, for example, was devoted to promoting some definitions of self and quelling others. While many members undoubtedly joined the center because of the consonance between the organization's and their own values, the identities and relationships they developed in the organization further augmented and tied them to those values. Moreover, regardless of how much they reflected collective definition processes, members also made choices regarding how to define their interests. Certain members, for example, worked for decisions that would cost them their jobs; they "chose to make the good of the whole their own" -- i.e., to have common interests (1980:239).
Second, interests and power were socially constrained. The requirement of consensus decision-making meant that any member had the power to deadlock the community policy process. Conversely, the community appears to have taken a dim view of certain kinds of "power trips", and applied considerable pressure to keep them within defined limits (1980:214,230). Obversely, the community defined the rules under which power could be exercised. Approved means included doing "ground work" and "cultivating" other members before meetings, as well as certain kinds of resorts to personal feelings (1980:201).
In sum, while community political processes at Helpline were consonant with the common interests of their members, they also helped define those interests, which ones were legitimate, and how they could be pursued. Just as significant as the typology of interests and forms of democracy in Mansbridge's work, therefore, is her detailed description of community processes by which those institutions were worked out. Those processes had an important forward-looking, experimental dimension. (41) They required developing a new vision of appropriate organization as well as a culture to implement it.
When Mansbridge asked intervention center staff how they could accept the power inequalities that existed at the center given their shared ideology of equality, a number of them told her they could accept them because a more fundamental kind of equality existed at the center (1980:234). They felt that all the staff members "could" have equal power "if they wanted to". Mansbridge does not pursue those answers in any detail because she considers them somewhat naive and concludes that what was really at bottom of the acceptance of limited inequality was the reality of common interests. Given my argument thus far, however, I suspect there may have been more sense and substance to the feeling of fundamental equality than Mansbridge allows. Helpline staff may indeed have found existing inequalities acceptable because they could, if they wished and were willing to work hard enough, take more powerful roles in defining the community's values and policies. Thus, while the future-oriented dimension of political interaction undercuts deductive efforts to define good and bad power, it also creates possibilities for future equality which, where there is relatively open access to power, may reduce the felt present inequality of communities.
Because a community decides which kinds of interests are valid and which ones are not at the same time it develops an affirmative vision of itself, good power cannot be distinguished from bad power entirely in terms of how it affects interests. The distinction must also be based on how power affects the self-definition and functioning of the community. The story of the crisis intervention center is noteworthy not just because it achieved a form of organization on a conceptual grid, but because it did so at the same time that it was working out an affirmative vision of itself and fulfilling social functions within a larger system. While keeping the lessons of unitary and adversary democracy in mind, therefore, it is important to follow Mansbridge's practice of grounding them in an empirical understanding of how the uses of power relate to a given community's self-definition and operation.
C. Conclusion: Good and Bad Power in Practice.
I have argued that distinguishing good from bad power based on democratic criteria necessarily requires a concrete understanding of the community involved. This is so because an essential feature of democracy is the political interaction whereby community members participate in creating a shared vision of the community. That vision is not reducible to the prior interests or power of community members because community processes shape, reshape, and channel interests and power. The fact that democracy in practice has the "density" of any other cultural phenomenon (i.e., power can be created and defined in a huge, open ended array of possible particular forms) also limits the possibilities for making theoretical distinctions between good and bad power. Participation in creating collective ideals may be structured and protected in many different ways. The affirmative meaning of democracy will necessarily be explicated in social practice.
Nonetheless, this is a theoretical paper, part of a larger effort to synthesize a contemporary understanding of the nature of culture and its relationship to social structure (1987), to describe contemporary administrative regulation in terms of that synthesis (1985; 1987), and to think through the relationship between the concept of regulatory culture and some core issues in democratic theory. As expected, although the last step in this theoretical process has suggested some of the limitations of democratic theory for evaluating regulation, it has not yielded many determinate results regarding the democratic legitimacy or illegitimacy of regulation. On the other hand, this paper cannot illuminate the workings of regulatory communities in the way that Mansbridge illuminates the workings of Selby and Helpline. A series of projects of that sort is underway (see Meidinger, Boyer and Thomas, 1986), but much remains to be done before any useful conclusions can be offered. Still, there already exists a substantial body of scholarship and conventional wisdom on regulation, and I am not by any means suggesting that body of knowledge is useless. Quite the contrary -- it qualifies as valuable local and general culture. In fact, it seems to me that much of the accumulated lore of regulation can be understood as a collective effort to distinguish good from bad power. It has been most successful when it has considered power in practice rather than in theory; ever since the absorption of the progressive movement into the New Deal, prescriptions for banning power wholesale have generally failed. (42) Here I will simply outline the form that I think an understanding of regulatory power informed by the kind of cultural analysis and democratic theory developed thus far might take.
Before proceeding, I should reiterate two key points made in prior sections. First, although the traditional democracy-based analysis underlying the nondelegation doctrine is flawed, it is still useful to think of regulatory communities as delegations of power. In practice regulatory power takes the cultural, structural and interactional forms outlined above. But those forms are practiced within the larger social system. Participants in regulatory communities both exercise power deriving from the constituencies and principles they represent and retain the option of appealing to the larger system. Second, power can be bad for other reasons than democratic ones. Power might counter the efficiency, rationality, accuracy, etc., of a regulatory process. In practice, the arguments against a given type of power generally invoke many or all of these values.
Regulatory power can be classified generally in terms of the type of regulatory actor exercising it: agency, regulatee, or regulatory beneficiary. Each type of power seems to have somewhat different characteristics and to be vulnerable to different types of problems.
1. The Power of Agencies and the Delegation Problem. The problem of power exercised by administrative agencies focuses most of the history of administrative law. Responses to the problem of agency power have been largely legalistic, and are embodied in the received law of judicial review. (43) The role of democratic theory in those responses has been limited, doubtless because democracy was a primary rationalization for agency power and because regulatees generally had little democratic practice to legitimate their complaints. Nonetheless, in the form of the non-delegation doctrine democratic theory had a long period of significant influence, (44) followed by a generation of dormancy, and now by a recent revival.
The problems with most traditional versions of the non-delegation doctrine are discussed above (subsection III.B.). Conventional separation of powers analysis is overly formalistic in first defining legislatures as the only arenas for democratic decision-making and then asking whether a given policy is based entirely on a legislative decision process. A more realistic understanding would ask whether decisions made in regulatory communities are consistent with those that would likely be made in legislatures. This kind of analysis obviously involves asking hypothetical questions. But it is no more difficult than the questions that have to addressed in traditional non-delegation analysis -- for example, what constitutes an "important" or "fundamental" policy issue? (45) Several approaches to it are already evident in judicial doctrine. One is simply to check whether the types of interests and arguments that would likely be decisions. (46) A second, related one is to ask whether the decision falls within the authority of the governing statute and reflects a "reasoned accommodation of manifestly competing interests", as Justice Stevens did in the Chevron decision. (47) A third is to consider what sort of success a complainant could in fact expect if it went to the legislature, as I suspect courts often implicitly do when they assess the fidelity of an administrative action to statutory language and legislative intent. And a fourth is to assess how effectively the agency is dealing with and responding to its policy environment, as I would also argue many courts in fact already do. In practice this can involve prodding an agency to develop creative solutions to problems, as the courts have done in a number of natural resources cases. (48)
There is a second, much more subtle variant of the problem of agency power, a variant also applicable to the other types of actors in the regulatory community. Agency officials may use regulatory power for purposes like forcing regulatees or regulatory beneficiaries to show them more personal deference. Or the officials may use their enforcement powers under legislatively authorized rules to make regulatees comply with unauthorized requirements. These uses of power can be thought of as variants of the traditional problem of graft, wherein officials use their power to force subordinates to provide them with income or other illicit favors. By defining these uses of power as "problems" I am assuming that neither the legislature nor the citizenry would approve of them. Unlike traditional delegation problems, however, these exercises of power generally do not involve documented, formal acts. They are therefore much more difficult to deal with. In general, I suspect the only way they can be reduced is through the development of internal controls and normative standards in regulatory communities. Since regulatory folklore indicates that a considerable amount of power may be exercised in this form, it is important to document the processes and relationships involved, as well as the successes and failures of different regulatory communities in dealing with them.
2. The Power of Regulatees and the Capture Problem. The argument about regulatory power has as often been that agencies exercise too little, as that they exercise too much. My suggestion that the democratic nature of regulation is enhanced by the ability of parties to remove issues to the larger political community can readily be countered by the argument that the larger community is skewed in favor of certain interests. That argument is backed by a substantial body of opinion holding that many agencies have historically been "captured" by -- made to serve the interests of -- the industries that they ostensibly regulated (e.g., Huntington, 1952; Bernstein, 1955; Kolko, 1965; McConnell, 1966). Although widespread acceptance of the capture thesis contributed to substantial changes in the structure of modern regulation, primarily the expansion of participation, judicial review, and enforcement rights by regulatory beneficiaries, there is still a serious argument that those changes have served primarily to legitimate regulation, rather than to alter its effects (e.g., Offe, 1976).
Rather than summarizing or critically evaluating the large, diffuse, and rather inconclusive literature on capture here, I simply want to make three points based on the analysis presented above. (49) First, I generally expect the structure of regulation to be consistent with the structure of power in the larger society. Powerful interests are likely to benefit more on average from regulatory institutions than weak ones. Yet this is not a critique of regulation, per se, but of the power structure of society. Second, however, the larger power structure does not necessarily determine the content of regulatory policy. Regulation is just as much an arena for political interaction as any other social arena. It is a place where existing practices can be challenged, where power can be created and reconstructed, by sustained action and concrete arguments for implementing critical ideals like democracy and fairness. Moreover, changes in local regulatory communities can sometimes be exported into other arenas, as has been the case with "market mechanisms" (Meidinger, 1985), regulatory analysis (McGarity, 1991), and a variety of procedural innovations, the most recent being "regulatory negotiation" (e.g., Susskind and McMahon, 1985). Because regulatory policy is not simply a derivative of the larger system, but provides an arena for critical deliberation and practice, the explication and implementation of ideals can provide concrete models and standards for other communities.
Finally, I hypothesize that although broad participation may appear to be merely a means of making regulatory policy consistent with the structure of power in the larger society, it also has the potential for creating a more vibrant, dynamic regulatory community. This may seem somewhat paradoxical, since expanded participation probably makes the practice of regulation more difficult. However, the more representative a regulatory community is of society at large, the more likely that community is to participate in the ideals of the larger society and to provide examples for it. Thus expanded participation is to be supported not only because it protects certain interests, but because it is also likely to improve the quality of regulatory deliberation and to create the necessary preconditions for a self-policing regulatory community.
3. The Power of Regulatory Beneficiaries and the Veto Problem. Only recently has it become fashionable to ponder whether regulatory beneficiaries might have excessive power. This is relatively understandable, since regulatory beneficiaries have only recently developed a regular presence in regulatory communities; until the 1960s, they were generally expected to limit their activities to legislative arenas. Thus far, however, our understanding of the power exercised by regulatory beneficiaries and their representatives is far less developed than our understanding of agency and regulatee power. At one level, the expanded power of beneficiaries is hardly mysterious. One need only look to the massive expansion in their legal rights to participate in, appeal, and enforce regulatory policies. (50)
The more important question is how the new prerogatives have been working in practice. Although critiques of beneficiaries' power generally refer to its possible theoretical effects on the speed, efficiency, or fairness of regulation, they do suggest some useful lines of approach for addressing the democratic legitimacy of beneficiaries' power. If the question of democratic legitimacy is framed as I have suggested -- the consistency of regulatory policies with the likely decisions of legislatures or the citizenry as a whole if they were to set policies -- there seem to be two significant ways in which beneficiary power could be a problem.
First, the interests and values of regulatory beneficiaries could be fundamentally inconsistent with those of the citizenry as a whole. If so, outcomes reflecting them would be inconsistent with the policies likely to be chosen in formally democratic processes. Although this argument has been made, for example in terms of the middle class backgrounds of environmentalists (Tucker, 1980), I will not consider it seriously here. Criticisms from this perspective can provide no even remotely plausible explanation of why democratically elected legislatures would continue to delegate such significant participation, review and enforcement rights to regulatory beneficiaries if the consequences were fundamentally inconsistent with the outcomes they or the citizenry would choose.
The second possible type of democratically bad beneficiary power is more subtle. It accepts that the results of beneficiaries' actions might often comport with the likely choices of the legislature or citizenry. But it also argues that beneficiaries will use their power to block key decisions often enough to make overall mix of regulatory policy inconsistent with what the legislature or citizenry would choose. As I suggested in the companion paper, this "beneficiary veto" analysis depends on how regulatory beneficiaries define their interests and what sorts of relationships they need to maintain with other members of regulatory communities. (51) In light of the accumulated experience and persistence of regulatory communities, I find it rather implausible as a description of the general case, given the generally long-term, ongoing relationships involved. Still, how regulatory beneficiaries define their interests, maintain relationships, use power, and are or are not constrained by the regulatory community deserves the careful attention of scholars. Not only would we learn how beneficiary power is being used in contemporary regulatory communities, we would learn how regulatory communities are faring in harnessing the interests and powers of their members to the development and pursuit of collective ideals. Then we might be able to talk more cogently about making regulation democratic.
References
Abel, Richard (1985) "Risk as an Arena of Struggle", Michigan Law Review 83:772-812.
Ackerman, Bruce (1984a) Reconstructing American Law. Cambridge, Mass: Harvard.
Ackerman, Bruce (1984b) "The Storrs Lectures: Discovering American Law", Yale Law Journal 93:1013- .
Ackerman, Bruce (1980) Social Justice and the Liberal State. New Haven, CT: Yale.
Ackerman, Bruce and Richard Stewart (1985) "Reforming Environmental Law",Stanford Law Review 37:1333-1365.
Alford, Robert and Roger Friedland (1985) Powers of Theory: Capitalism, The State, and Democracy. Cambridge: Cambridge University Press.
Aranson, Peter, Ernest Gellhorn and Glenn Robinson (1983) "A Theory of Legislative Delegation", Cornell Law Review 68:1- ___.
Arensberg, Conrad M. and Solon T. Kimball (1965) Culture and Community. New York:Harcourt, Brace and World.
Argyris, Chris and Donald Schon (1978) Organizational Learning. Reading, Mass: Addison Wesley.
Aristotle (1946) The Politics of Aristotle (Ernest Barker, ed.) London: Oxford University Press.
Arrow, Kenneth J. (1951) Social Choice and Individual Values. New Haven: Yale.
Atleson, James (1983) Values and Assumptions in Labor Law.
Ayers, Ian and John Braithwaite (1992) Responsive Regulation: Transcending the Deregulation Debate. Oxford: Oxford University Press.
Ayers, Ian and John Braithwaite (1991) "Tripartism: Regulatory Capture and Empowerment", Law and Social Inquiry 16:435-496.
Axelrod, Robert (1984) The Evolution of Cooperation.
Bachrach, Peter and Morton S. Baratz (1963) "Decisions and Non-Decisions: An Analytic Framework", American Political Science Review 57:641-651.
Barber, Benjamin (1984) Strong Democracy: Participatory Politics for a New Age. Berkeley: University of California Press.
Bardach, Eugene and Robert A. Kagan (1982) Going by the Book: The Problem of Regulatory Unreasonableness. Philadelphia: Temple University Press.
Becker, Howard S. (1982) "Culture: A Sociological View", The Yale Review 71:513-27.
Becker, Howard S. and Blanche Geer (1960) "Latent Culture", Administrative Science Quarterly 5:303-313.
Bell, Robert (1985) The Culture of Policy Deliberations. New Brunswick, NJ: Rutgers University Press.
Berger, Peter and Thomas Luckmann (1966) The Social Construction of Reality. Garden City: Doubleday.
Bernstein, Marver (1955) Regulating Business by Independent Commission. Princeton: Princeton Univ. Press.
Binder, Guyora (1988) "Critical Legal Studies as Guerilla Warfare".
Boudon, Raymond (1971) The Uses of Structuralism. London: Heinman.
Boyer, Barry (1981) "Funding Public Participation in Agency Proceedings: The FTC Experience" Georgetown Law Journal 70:51-172.
Boyer, Barry (1983) "Fifty Years of Regulatory Reform in the United States". Paper prepared for the Conference on Regulation in Britain and the United States, Trinity College, Oxford.
Boyer, Barry and Errol Meidinger (1986) "Privatizing Regulatory Enforcement: A Preliminary Assessment of Citizen Suits Under Federal Environmental Laws",Buffalo Law Review 34:834-965.
Brickman, Ronald, Sheila Jasanoff and Thomas Ilgen (1985) Controlling Chemicals: The Politics of Regulation in Europe and the United States. Ithaca, New York: Cornell University Press.
Burawoy, Michael (1979) Manufacturing Consent: Change in the Labor Process under Monopoly Capitalism. Chicago: University of Chicago Press.
Burnheim, John (1985) Is Democracy Possible: The Alternative to Electoral Politics Berkeley and Los Angeles, CA: University of California Press.
Campbell, Donald T. (1965) "Variation and Selective Retention in Sociocultural Evolution", in Barringer, G. et al, eds., Social Change in Developing Areas: A Reinterpretation of Evolutionary Theory. Cambridge, Mass: Schenkman.
Calabresi, Guido and Douglas Melamed, "Property Rules, Liability, and Inalienability: One View of the Cathedral", Harvard Law Review, 85:1089-___.
Chibnik, Michael, "The Evolution of Cultural Rules", Journal of Anthropological Research 37:256-268.
Clune, William (1984) "A Political Model of Implementation and Implications of the Model for Public Policy, Research, and the Changing Roles of Law and Lawyers", Iowa Law Review 69:47-123.
Coase, Ronald H. (1959) "The Federal Communications Commission" Journal of Law and Economics 2:1
Cohen, Morris R. (1923) "Property and Sovereignty" Cornell Law Quarterly 13:8-30.
Collingwood, R.G. (1942) The New Leviathan. Oxford: Clarendon Press.
Dawson, John (194 ) [get cite on contracts critique]
Davis, Kenneth C. (1969) Discretionary Justice. Urbana: University of Illinois Press.
Debnam, Geoffrey (1994) The Analysis of Power: A Realist Approach. London: MacMillan Press.
Denhardt, Roger B. (1981) "Toward a Critical Theory of Public Organization", Public Administration Review ___:628-635.
Dietz, Thomas (forthcoming) "Theory and Method in Social Impact Assessment",Sociological Inquiry, 77:___-___.
Douglas, Mary (1966) Purity and Danger: An Analysis of the Concepts of Pollution and Taboo. New York: Pantheon.
Ely, John Hart (1980) Democracy and Distrust: A Theory of Judicial Review. Cambridge, Massachusetts: Harvard University Press.
Engel, David M. (1984) "The Oven Bird's Song: Insiders, Outsiders, and Personal Injuries in an American Community", Law and Society Review 18:551-582.
Forester, John (1983) "Critical Theory and Organizational Analysis", in Gareth Morgan, ed. Beyond Method: Strategies for Social Research. Beverly Hills, CA: Sage.
Freedman, James O. (1978) Crisis and Legitimacy: The Administrative Process and American Government. Cambridge: Cambridge University Press.
Freeman, Derek (1983) Margaret Mead and Samoa: The Making and Unmaking of an Anthropological Myth. Cambridge, MA: Harvard University Press.
Freund, Ernst (1915) "The Substitution of Rule for Discretion in Public Law," Amer. Pol. Sci. Rev. 9:666-676.
Freund, Paul (1954) "Umpiring the Federal System", Columbia Law Review 54:561-578.
Friedman, Lawrence (1975) The Legal System: A Social Science Perspective. New York: Russell Sage.
Friedman, Milton (1962) Capitalism and Freedom. Chicago: University of Chicago Press.
Gabel, Peter (1977) "Intention and Structure in Contractual Conditions: Outline of a Method for Critical Legal Theory", Minnesota Law Review 61:601- .
Gabel, Peter (1984) "The Phenomenology of Rights-Consciousness and the Pact of the Withdrawn Selves", Texas Law Review 62:1563-1599.
Gabel, Peter and Duncan Kennedy (1984) "Roll Over Beethoven" Stanford Law Review36:1- .
Geertz, Clifford (1973) The Interpretation of Cultures. New York: Basic Books.
Genovese, Eugene (197 ) Roll Jordan Roll
Giddens, Anthony (1984) The Constitution of Society: Outline of the Theory of Structuration. Berkeley and Los Angeles: University of California Press.
Habermas, Jurgen (1975) Legitimation Crisis (T. McCarthy, trns.) Boston: Beacon.
Habermas, Jurgen (1979) Communication and the Evolution of Society. Boston: Beacon.
Habermas, Jurgen (1983) The Theory of Communicative Action. (2 vols.) Boston: Beacon
Hale, Robert (1923) "Coercion and Distribution in a Supposedly Non-Coercive State",Political Science Quarterly, 38:470- ___.
Hartz, Louis (1952) "The Whig Tradition in American and Europe", American Political Science Review 46:989- .
Hawkins, Keith (1984) Environment and Enforcement: The Social Definition of Pollution. Oxford: Clarendon.
Held, David (1987) Models of Democracy. Stanford, CA: Stanford University Press.
Hiller, Rodney (198 ) [piece on the English Peasant Revolt]
Hirschman, Albert O. (1977) The Passions and the Interests: Political Arguments for Capitalism Before Its Triumph. Princeton: Princeton University Press.
Hjern, Benny and David O. Porter (1981) "Implementation Structures: A New Unit of Administrative Analysis", Organization Studies 2/3:211-227.
Huntington, Samuel P. (1952) "The Marasmus of the ICC: the Commission, the Railroads, and the Public Interest", Yale Law Journal 61:467-509.
Jacobson, Arthur (1980) "The Private Use of Public Authority: Sovereignty and Associations in the Common Law" Buffalo Law Review 29:600- .
Jaffe, Louis (1937) "Law Making by Private Groups", Harvard Law Review, 51:201-___.
Jasanoff, Sheila (1986) Risk Management and Political Culture. New York: Russell Sage Foundation.
Jencks, Christopher, et al. (1972) Inequality: A Reassessment of the Effect of Family and Schooling in America. New York: Basic.
Katz, Jack (1982) Poor People's Lawyers in Transition. New Brunswick: Rutgers.
Keniston, Kenneth (1968) The Young Radicals. New York:Harcourt, Brace, Jovanovich.
Kramer, Daniel C. (1972) Participatory Democracy: Developing Ideals of the Political Left. Cambridge, Mass.:Schenkman Publishing Co.
Kolko, Gabriel (1965) Railroads and Regulation, 1877-1916. Princeton: Princeton University Press.
Konrad, George (1984) Antipolitics. New York: Harcourt, Brace, Jovanovich.
Krieger, Martin H. (1986) "Big Decisions and a Culture of Decisionmaking", Journal of Policy Analysis and Management 5:779-797.
Krober, A.L. and Clyde Kluckhohn (1963) Culture: A Critical Review of Concepts and Definitions. New York: Vintage.
Landis, James (1938) The Administrative Process. New Haven: Yale University Press.
Latin, Howard (1985) "Ideal versus Real Regulatory Efficiency: Implementation of Uniform Standards and 'Fine-Tuning' Regulatory Reform", Stanford Law Review37:1267-1332.
Lee, Dorothy (1959) Freedom and Culture. Prentice Hall.
Leff, Arthur (1978) "Law and", Yale Law Journal 87:989-1011.
Lempert, Richard and Joseph Sanders (1986) An Invitation to Law and Social Science: Desert, Disputes and Distribution. New York: Longman.
Levine, Adeline G. (1982) Love Canal: Science, Politics, and People. Lexington, Mass: Lexington Books.
Levins, Richard (1966) "The Strategy of Model Building in Population Biology",American Scientist 54:421-431.
Lindblom, Charles E. (1959) "The Science of Muddling Through". Public Administration Review 19:79-88.
Llewellyn, Karl (1960) The Common Law Tradition: Deciding Appeals. Boston:Little Brown.
Lowi, Theodore (1969) The End of Liberalism: Ideology, Policy and the Crisis of Public Authority. New York: W.W. Norton.
Lowi, Theodore (1970) "Decision Making versus Policy Making: Toward an Antidote for Technocracy", Public Administration Review 30:314-325.
Lowi, Theodore (1987) "Two Roads to Serfdom: Liberalism, Conservatism and Administrative Power" American University Law Review 36:295- 322.
Macaulay, Stewart (1963) "Non-Contractual Relations in Business: A Preliminary Study", American Sociological Review 93:28-55.
MacNeil, Ian R. (1983) "Values in Contract: Internal and External", Northwestern University Law Review 78:340- .
Macpherson, C.B. (1979) The Political Theory of Possessive Individualism: Hobbes to Locke
Manning, Peter K. (1977) Police Work: The Social Organization of Policing. Cambridge, MA: MIT Press.
Mansbridge, Jane (1980) Beyond Adversary Democracy. Chicago: University of Chicago Press.
__________ (1991) "A Deliberative Theory of Interest Representation" in Mark P. Petracca, ed., The Politics of Interests: Interest Groups Transformed. Boulder, CO: Westview Press.
Marx, Karl (1965) Capital London:
Mashaw, Jerry (1985) "Prodelegation: Why Administrators Should Make Political Decisions", Journal of Law, Economics, and Organization 1:81-100.
Mashaw, Jerry (1983) Bureaucratic Justice New Haven: Yale.
Matheny, Albert R. and Bruce A. Williams (1986) "Rethinking Participation: Democratic Theory and the Siting of Hazardous Waste Disposal Facilities. Paper prepared for the annual meetings of the Law and Society Association, May 29- June 1, 1986, Chicago, IL.
McConnell, Grant (1966) Private Power and American Democracy. New York: Knopf.
McGarity, Thomas O. (1986) "Regulatory Reform in the Reagan Era", Maryland Law Review 45:253-273.
McGarity, Thomas O. (1991) Reinventing Rationality: The Role of Regulatory Analysis in the Federal Bureaucracy. Cambridge and New York: Cambridge University Press.
Mead, Margaret (1928) Coming of Age in Samoa: A Psychological Study of Primitive Youth for Western Civilization. New York: William Morrow.
Meidinger, Errol (1985) "On Explaining the Development of 'Emissions Trading' in U.S. Air Pollution Regulation", Law and Policy 7:447-449.
Meidinger, Errol (1987) "Regulatory Culture: a Theoretical Outline", Law and Policy 9:355-386.
Meidinger, Errol, Barry Boyer and John M. Thomas (1986) "Proposal for a Working Group on Sociology of Regulation", Baldy Center for Law and Social Policy, State University of New York at Buffalo.
Mensch, Elizabeth (1981) "Freedom of Contract as Ideology", Stanford Law Review, 33:753-___.
Mintz, Sidney W. (1982) "Culture: An Anthropological View", Yale Review (n.s.) 71:499-512.
Moore, Sally Falk (1973) "Law and Social Change: The Semi-Autonomous Field as an Appropriate Subject of Study", Law and Society Review 7:719.
Nedelsky, Jennifer (1990) Private Property and the Limits of American Constitutionalism.
Nelkin, Dorothy and Michael S. Brown (1984) Workers at Risk: Voices from the Workplace. Chicago: University of Chicago Press.
Offe, Claus (1976) "The Theory of the Capitalist State and the Problem of Policy Formation", in Lindberg, et al. eds., Stress and Contradiction in Modern Capitalism. Lexington, Mass.:Lexington Books.
Parsons, Talcott (1951) The Social System. New York: Free Press.
Perrow, Charles (1984) Normal Accidents: Living With High Risk Technologies
Pierce, Richard J. Jr., "Political Accountability and Delegated Power: A Response to Professor Lowi", American University Law Review 36:391-418.
Polsby, Nelson W. (1980) Community Power and Political Theory (2d Ed.) New Haven, CT: Yale University Press.
Piaget, Jean (1966) Structuralism.
Pocock, J.G.A. (1975) The Machiavellian Moment. Princeton, N.J.: Princeton University Press.
Posner, Richard A. (1986) Economic Analysis of Law (3rd Ed.) Boston: Little, Brown, and Co.
Quine, W.V. (1952) The Methods of Logic. London:
Rabin, Robert L. (1986) "Federal Regulation in Historical Perspective", Stanford Law Review 38:1189-1321.
Rappaport, Roy A. (1986) "Desecrating the Holy Woman: Derek Freeman's Attack on Margaret Mead", The American Scholar 55:313-347.
Rawls, John (1971) A Theory of Justice. Cambridge: Harvard.
Rayner, Steve (1984) "Disagreeing About Risk: The Institutional Cultures of Risk Management and Planning for Future Generations", in Susan G. Hadden, ed.,Risk Analysis, Institutions, and Public Policy, Port Washington, NY: Assoc. Fac. Press.
Reich, Robert B. (1985) "Public Administration and Public Deliberation: An Interpretive Essay", Yale Law Journal 94:1617-1641.
Robinson, Glen (1990) American Bureaucracy.
Sabatier, Paul (1988) "An Advocacy Coalition Framework of Policy Change and the Role of Policy-Oriented Learning Therein", Policy Sciences 21:129-168.
Sagoff, Mark (1981) "At the Shrine of Our Lady of Fatima or Why Political Questions Are Not All Economic", Arizona Law Review 23:1283-1298.
Saperstein, Saundra and George Lardner Jr. (1986) "Rehnquist: The Accidental Jurist,The Washington Post Weekly Edition __:6-8 (July 28, 1986).
Scalia, Antonin (1978) "Vermont Yankee: The APA, the D.C. Circuit, and the Supreme Court", Supreme Court Review 1978:315-372.
Schattschneider, E.E. (1960) The Semi-Sovereign People: A Realist's View of Democracy in America. NY: Holt, Rinehart and Winston.
Schnaiberg, Allan (1980) The Environment: From Surplus to Scarcity. New York and London: Oxford.
Schoenbrod, David (1983) "Goal Statutes of Rule Statutes: The Case of the Clean Air Act", U.C.L.A. Law Review 30:740.
Scholz, John T. "Voluntary Compliance and Regulatory Enforcement", Law and Policy4:385-404.
Sen, Amartya (1970) Collective Choice and Social Welfare. San Francisco: Holden-Day.
Shannon, Margaret (1985) Assessing Communication Effectiveness in Developing Forest Plans and EIS Documents. Report to the U.S. Forest Service (August).
Shannon, Margaret A. (1989) Managing Public Resources: Public Deliberation as Organizational Learning. Doctoral Dissertation, University of California at Berkeley.
Silbey, Susan S. and Sally E. Merry (1986) "Mediator Settlement Strategies", Law and Policy 8:7-32.
Steinfeld, Robert (1991) The Invention of Free Labor: The Employment Relation in English and American Law and Culture, 1350-1870. Chappel Hill and London: University of North Carolina Press.
Stewart, Richard B. (1985) "The Discontents of Legalism: Interest Group Relations in Administrative Regulation", Wisconsin Law Review 1985:655-686.
Stewart, Richard B. (1983) "Regulation in a Liberal State: The Role of Non-Commodity Values", Yale Law Journal 92:1537- 1590.
Stewart, Richard B. (1976) "The Reformation of American Administrative Law," Harvard Law Review 88:1169-1813.
Stewart, Richard B. (1987) "Beyond Delegation Doctrine", American University Law Review 36:323-343.
Sunstein, Cass R. (1983) "Deregulation and the Hard Look Doctrine", Supreme Court Review __:177-213.
Sunstein, Cass R. (1985) Interest Groups in American Public Law, Stanford Law Review, 38:29-87.
Sunstein, Cass R. (1986) "Factions, Self-Interest and the APA: Four Lessons Since 1946", Virginia Law Review, 72:271-296.
Suskind, Lawrence and J. McMahon (1985) Documentation of EPA's Non-Conformance Penalties (NCP) Negotiated Rulemaking Demonstration. Washington, D.C.: United States Environmental Protection Agency.
Thompson, E.P. (1963) The Making of the English Working Class. New York:Vintage.
Tucker, John (197 ) [environmentalism as an elite, purely self-interested movement]
Turnbul, Colin M. (1972) The Mountain People. New York:Simon and Schuster.
Van Maanen, John and Stephen R. Barley (1985) "Cultural Organization: Fragments of a Theory". In Peter J. Frost, et al, eds., Organizational Culture. Beverly Hills: Sage.
Vining, Joseph (1978) Legal Identity: The Coming of Age of Public Law. New Haven:Yale.
Warren, Roland L. (1972) The Community in America. 2nd Ed. Chicago: Rand McNally.
Weber, Max (1968) Economy and Society: an Outline of Interpretive Sociology (Ed. G. Roth and C. Wittich). Berkeley, CA: Univ. of Cal. Press.
Williston, Samuel (1920) The Law of Contracts. New York:Baker, Voorhis.
Wright, Erik Olin (1979) Class Structure and Income Inequality. New York: Academic Press.
Wyman, Bruce (1903) Principles of Administrative Law. St. Paul: Keefe-Davidson.
Zablocki, Benjamin D. (1971) The Joyful Community. Baltimore: Penguin.
1. This is a standard, albeit minimalist definition of the term. See, e.g., Aristotle, Chap. VI, Section I ("In democratic states, for example, the people [or demos] is sovereign..."); Oxford English Dictionary, Compact Edition ("Government by the people...").
2. I have devoted considerable space to working out the concept of regulatory culture in a previous paper (Meidinger 1987). For the sake of clarity here, I recapitulate some of the key arguments of that paper. By "culture" I mean a set of shared understandings which makes it possible for a group of people to act in concert with each other (see e.g., Becker, 1982; Van Maanen and Barley, 1985). Such a set of shared understandings is necessary to social life, because without it people would not be able to act with meaningful reference to each other. Any act of one would be uninterpretable by another. Any expectation of one would be uncomprehended and unmet by the other.
Culture is neither monolithic nor purely rational, however. Rather, shared understandings can vary greatly from one particular social arena to another. One subgroup of actors within a larger collectivity can interact in terms of numerous understandings (or "local culture") quite different from those of another subgroup, although the subgroups will ordinarily share many other understandings. Thus while most types of administrative regulation share many assumptions, they also tend to have many assumptions peculiar to their own practices.
Second, the internal terms of any given set of understandings need be neither logically deducible from nor even consistent with each other. More likely, some assumptions will be seriously at odds with others. Thus we live in a society that espouses equality and authority, democracy and property, principle and expediency, and so on. Our culture is "partially incoherent". On the other hand, for a group to function and persist, its organizing culture must have at least a certain amount of coherence (e.g., Giddens, 1984; Parsons, 1951). Thus its lived patterns of interaction must entail practical ways of constraining the conflicts among inconsistent principles.
Third, culture is not simply a composite or aggregation of individual values or understandings. Quite the opposite. Culture is the primary source and often the measure of those values. It defines the typical forms of individual character, and thus of self-interest. Conversely, the conflicting principles in a culture create considerable space for individual difference and change.
The preceding attributes suggest, fourthly, that culture is likely to be an important factor in any form of political action. It provides the images and standards used to formulate, define and evaluate social behavior or relationships. Conversely, changing behavior or relationships will ordinarily require modifying cultural assumptions. Accordingly, some of the most important political action is aimed at creating and legitimating new images of social behavior (or at destroying and delegitimating old ones). Culture is thus both a major source and a central object of political action. Accordingly, it is also a source of both continuity and change.
Fifth, culture is very detailed, or "dense". It is best conceived not as a set of abstract principles from whichappropriate behaviors can be deduced, but as an extremely elaborate set of images and assumptions that infuse the world with meaning. Many are "latent" rather than conscious (Becker and Geer, 1960). Although it is not language, culture is more like language than like mathematics. Despite its density, however, culture is often indeterminate regarding appropriate behaviors in any given situation. This is so both because cultures ordinarily contain inconsistent principles and because no conceptual code can cover every empirical situation that might possibly arise. Culture may eliminate some behaviors as inappropriate, but it will usually include more than a single acceptable alternative. Often, moreover, actors engage in concerted efforts to change the classification of given behaviors from inappropriate to appropriate, or vice versa.
At the next higher level of abstraction, then, culture can be described as having four salient features. First, it effectively "governs" social behavior. It does so in both a traditionally descriptive (or "positive") sense, since it provides the repertoire of images from which behavior is constructed in any situation, and in a normative sense, since the images have moral components. Second, culture creates the basis for social structure; it is the carrier of the shared expecations which make possible and are reinforced by the enactment and reenactment of patterned social relationships. Third, culture is inherently political, since it defines social behavior and relationships as acceptable or unacceptable, desirable or undesirable, etc. Culture is a product of people (whether a tribe or an administrative agency) which simultaneously defines and structures who they are and how they live. Changing social life in any systematic way requires changing culture. Fourth, while culture gives pattern and structure to social life, it is also inherently dynamic. Its conflicting principles, partial indeterminacy, and continual subjugation to political action give it an open ended and future oriented nature.
3. In focusing on democracy I am of course ignoring a number of other possible criteria, including fairness, rationality, precision, and efficiency. Because efficiency has lately become the criterion of choice for many scholars seeking an ostensibly objective standard to apply to legal institutions (e.g., Posner, 1986), it may be useful to consider the efficiency criterion in terms of the framework developed here.
The fundamental goal of the efficiency criterion is tosatisfy the maximum social wants at the minimum social cost. While it is not therefore inherently objectionable, it has no claim to a status as an exclusive or even predominant standard for two reasons. First, the efficiency criterion ignores the processes by which wants (or interests) are shaped and reshaped. The cultural perspective posits that interests are in significant part socially constructed and reconstructed; it is always possible that people's desires could be different than they are in any given context. Efficiently satisfying existing wants, therefore, is neither inherently nor exclusively a good thing. Second, efficiently satisfying existing individual wants may not be the primary goal of a given group of people. They might, for example, prefer to concentrate on building a social system emphasizing non-economic relationships or one emphasizing religious commitment to a deity. It is true that people might collectively decide to pursue such a system because they believed they would be better off in the end. And the ideal of efficiency might retain a role in their evaluation processes. But the standard of reference would not necessarily be the individual consumer who is the primary unit of efficiency theories. Even if it were, as Sagoff (1981) has so trenchantly argued, those advocating efficiency as a predominant ideal might not be able to pay those advocating (or desiring) other ideals enough to get them to accept efficiency.
The critical utility of efficiency analysis for evaluating regulation is problematic, in sum, because it is indeterminate in application, does not account for the social construction of interests, and cannot conceptualize the affirmative political activities by which interests and norms are selected, shaped, and changed. It has no standing as an evaluative criterion independent of the process of collective choice. However, I do not propose to substitute a standard which has independent standing. Quite the contrary; such a possiblity is literally outside the terms of this argument.
4. See generally, Pocock (1975), MacPherson (1979); Ackerman (1984).
5. I am of course ignoring the problem of defining the exact nature of the right not to live near a repository. Clearly, significant cultural discussion would also necessarily go into deciding whether it should take the form of a liability right, a property right, or an inalienable right (Calabresi and Melamed, 1972), as well as whether it should be held by individuals or collectivities.
I might also note that such a decision to vitiate the operation of democracy could be made democratically. This is no more paradoxical than the idea of constitutionalism generally. I do not find the possibility particularly paradoxical in light of my general position that rights are socially constructed and that they can be constructed, used, and vitiated in many different ways. It does of course raise a political challenge, since I have no a prioritrump to Justice Rehnquist's assertion that "in the long run it is the majority who will determine what the constitutional rights of the minority are." (Saperstein and Lardner, 1986). It would be nice if matters were otherwise; alas, I have no choice but to engage in political interaction and citizenship.
For a fascinating example of the ability of democracy to radicalize conservatives and conservatize liberals, see City of Eastlake v. Forest City Enterprises, Inc., 426 U.S. 668 (1976), in which Chief Justice Berger's majority opinion upholds a city charter provision requiring voter ratification of proposed land use changes and Justice Stevens' dissent (Brennan concurring) attacks it.
6. See the Nuclear Waste Policy Act of 1982. Similar tendencies are evident in the regulation of toxic waste disposal sites. See e.g., Rollins Environmental Services (F.S.), Inc. v. Parish of St. James, ___ F.2d ___(Cir. 5, 1986), holding local regulation of waste disposal sites involving PCBs preempted under TSCA. But the courts show considerable ambivalence on questions such as these. For example, with regard to Sharon Stell v. City of Fairmont, ___ W.Va. __ (1985?), upholding a local toxic waste disposal ban, the Supreme Court refused to hear an appeal (___ U.S. ___, 1986).
7. Formal equality ("equal protection of interests") as exemplified by majority vote and secret ballot procedures is not necessary in unitary democracies because of the basic commonality of members' interests and the possibility of coming to agreements on best policies through extended deliberation. (Mansbridge, 1980: 5)
8. The indeterminacy of rules refers to the general proposition that rules cannot be specified to govern every contingency for cases involving more than a single decision criterion, and that application of rules to particular cases therefore will generally entail exercising discretion. For a more complete explication of this problem, see Meidinger (1986a).
9. The other main challenge to substantial delegations is based more centrally on rule-of-law predictability grounds (Davis, 1969). I will not deal with it beyond the argument suggested in the first section, which is that local regulatory cultures are likely to provide considerably more predictability than any system of externally imposed rules.
10. Indeed, the courts of this country may be in the process of resecuring most social organizations in their recently questioned private status. For example, the Supreme Court's 1968 ruling that union picketing on the grounds of a private mall was protected by the First Amendment (Amalgamated Food Employees Union v. Logan Valley Plaza, 391 U.S. 308) was implicitly overruled four years later, when the Court held that anti-war protesters could not distribute leaflets in a shopping center without the owner's consent (Lloyd Corp. v. Tanner, 407 U.S. 529, 1972), and explicitly overruled four years after that, when that the Court asserted that a shopping center must be public for all purposes or none (Hudgens v. NLRB, 424 U.S. 507, 1976). (However, the court later did uphold a California state constitutional provision guaranteeing free speech in privately owned, publicly used shopping centers. Pruneyard Shopping Center v. Robbins, 447 U.S. 74, 1980).
Similarly, after invalidating a whites-only pre-primary election used by a Texas group to make meaningless a racially open primary (Terry v. Adams, 345 U.S. 461, 1953), the Supreme Court held that a private club's racial discrimination in serving liquor is not subject to the equal protection clause despite the requirement of a government liquor license (Moose Lodge No. 107 v. Irvis, 407 U.S. 163 (1972), that an electric company with a state monopoly is not subject to procedural due process requirements in terminating service (Jackson v. Metropolitan Edison Company, 419 U.S. 345 (1974), that a warehouseman using a state statute to allow him to sell an indigent's furniture to pay an undue bill was not subject to procedural due process requirements (Flagg Brothers v. Brooks, 436 U.S. 149, 1978) and so on.
11. [e.g., Saturn Plant location (NYT cites); Poletown case.]
12. [e.g., Sixty Minutes segment on executive dislocations in suburban communities]
13. [NYT cites on drug testing]
14. The real pattern of political authority in society is thus an exceedingly complex, overlapping, and continuously changing crazy quilt of relationships. If authority could be quantified and measured, most of it would probably be found to exist outside state agencies. (The relationship between political authority and power is addressed in subsection E of this section.) For a systematic analysis of the law of business associations as an exercise in the delegation of sovereignty, see Jacobson (1980).
15. There are, however, good reasons to prefer regulatory delegations. See, e.g., Jaffe (1937).
16. The most common contemporary justification is the argument that private delegations are inherently more efficient than public ones because of incentives to structure conduct and use resources in income maximizing ways. See, e.g., Posner's justification of property rights (1986:29-77). A related, but potentially more radical argument for private delegations is premised on a libertarian argument that they act as a necessary curb on state power. (Friedman, 1962:7-36) Neither position, however, addresses the argument, sketched most powerfully in the modern era by Cohen (1923), that property is actually a form of delegated sovereignty. I suspect the reasons for the lack are both theoretical (society is conceived entirely as an aggregation of individuals who start with a large set of baseline rights) and strategic (it may be most advantageous simply not to address such a difficult problem).
17. Arrow's "paradox of voting" holds essentially that there is no rational way of choosing among more than two alternatives through voting (1951:3). Note that these decisional constraints, particularly the paradox of voting, may also serve to justify what I have described as the oligarchical structure of most regulatory communities. Without a more intensive inquiry, however, I am not prepared to accept that they do. More importantly, as I argue below, the actual decisions of internal oligarchies may in fact meet general democratic criteria.
18. This is consistent with the prescription of K.C. Davis (1966), for example. Experience to date, however, provides ample examples of agencies that have failed to achieve this subtle role. See, e.g., Boyer (1983). How it can be done is a subject worthy of serious study.
19. Even a confirmed pragmatist like Robinson (1990) finds this argument persuasive.
20. Bardach and Kagan (1982) describe the possibilities of this process as creating a "trusteeship" stratum, a group of people who are entrusted with the social responsibility of making sure a certain function is properly performed. Although they seem to think primarily in terms of operatives within regulated corporations, it may make as much sense to think of regulatory communities as trusteeship strata. Thus public interest groups, responsible regulatory officials, and regulatee staffers may operate as delegated social trustees.
21. For an affirmative defense of the intelligence of legislative decisions not to decide particular issues, see Mashaw (1985:92).
22. Even in writing this paper, as I have intimated above, I am both pursuing my self-interest and my view of the good, in this case good scholarship and good politics. My choice of how to develop a descriptive image of regulation, moreover, is based in part on my view about the normative implications of that description, and where it might lead if integrated into regulatory discourse.
23. Although my terminology here is similar to that of Hirshman (1977), I am using the words in their more general forms, rather than with the specific meanings they have in his historical analysis.
24. Examples of such differences abound. Indeed, they define much of the intellectual and political terrain of modern anthropology. For particularly lucid examples, see e.g., Lee, 1959; Arensbberg and Kimball, 1965.
25. Trying to change self-interested behavior by changing structures can be analogized to the efforts of parents to change children's behavior. One can try to discourage fighting and encourage sharing by telling kids that if they don't work out a cooperative arrangement they will lose whatever they are fighting over. This changes the expected consequence of the objectionable behavior. Sometimes it works, sometimes it doesn't. More often than not, such a gambit will probably need to be accompanied by other (cultural) socialization processes -- i.e., persuading kids what their interests are and how they will best achieve them. As discussed in the next section, being a parent also involves having power, and that creates its own problems.
26. For a recent review of the voluminous literature in political science and political sociology on the nature of power, see, e.g., Debnam, 1984.
27. Marx's analysis of the social transformation of textile production in Britain still provides one of the most cogent and illuminating descriptions that I know of the process by which laborers are made dependent on and subordinate to capital (1965: ).
28. I discuss the basis of social structure in culture in 'Regulatory Culture' (1987). In essence my argument is that social structure (defined as stable, routine patterns of social relationships) is the product of the complex of common expectations that a substantial group of people shares and enacts as a result of common culture.
For a careful analysis of the cultural construction of structural dependency, see the recent paper by Rob Steinfeld (1989).
I am not taking a position here on whether the existence of social structure inherently implies power. In practice, however, it seems clear that primary purpose of virtually every existing form of social structure is the allocation and institutionalization of social power.
29. For an analysis of the culture organizing labor law, see Atleson, 1983.
30. Of course power based on action and power based on structure can and often do converge, as when individuals and groups act assiduously to get themselves into powerful structural positions. Moreover, I am not arguing here that social actors are capable of creating power to achieve any objective under any circumstances. Structure and culture confer much more power on certain types of actors than on others. In any social epoch, certain types of outcomes are no doubt effectively precluded. My point is simply that culture and structure do not decide outcomes, and that strategic social action can and often does have a decisive impact on social policy -- an impact which is not merely a reflection of culture or structure.
31. Conventional examples of the later kind in which power was drawn from the culture of the powerful include, e.g., the American civil rights movement which depended heavily on the values of mainstream culture (and was perhaps in the end limited by them) and the invocation by slaves of customary practices to claim privileges from their masters (Genovese, 197 ). Conversely power can also be created through local cultural activity, as was the case in the English peasant revolt of 1381 (Hiller, 198 ) as well as the rise of the Polish Solidarity movement in this decade.
32. See e.g. the progression of positions represented in the movement from Williston (1920) to Hale (1923) to Dawson (194 ) to Llewellyn (1960). Mensch (1981) provides an excellent discussion of some of the dynamics and implications of the process.
33. This project typically translated into documenting systematic inequality and explaining its persistence. Much like Rawls (discussed below), sociologists generally presumed that individuals would not voluntarily consent to bargains in which they regularly benefited less than other parties unless they had little effective choice (e.g., Jencks, 1972; Wright, 1979).
34. [short footnote on the legal process movement]
35. Compare, e.g. Habermas 1975 (legitimation crisis based on failure of capitalism to fulfill socially constructed needs) to 1983 (legitimation crisis based on failure to maintain a realm of pure, expressive -- as opposed to instrumental -- action). The two views are, of course, consistent to the degree that the failure to fulfill socially constructed needs results from a failure of pure communication.
36. Note that language becomes the empirical manifestation of culture, and that truth is both descriptive and normative for Habermas.
37. For example, through inequality of expressive skill (power?) between intellectuals and manufacturing workers.
38. Since the outcome of neutral dialogue is to be based on the quality of arguments put forward, each speaker is not given formally equal influence (such as might be done by requiring either majority voting or consensus). Rather, conversational equality is secured by two limited constraints: (1) no speaker may claim to possess moral insights intrinsically superior to those of other speakers; (2) no speaker may claim to be an intrinsically superior person (1984:99). Thus speakers may have differential power, but it will be based entirely on the quality of their arguments. I will suggest below that this gambit both retains culture as a source of social power, since there is no other fundamental basis for determining the quality of arguments, and fails to address the problem that existing culture was created in significant part by power.
39. For a critical discussion of the relationship of the consensus assumption to liberal and radical strains in modern legal thought, see Binder, 1988.
40. See the limited discussions of indeterminacy and Arrow's theorem above, and the more thorough ones in 'Regulatory Culture' (1987).
41. In this sense, the imaginative enterprise as engaged in by Rawls and Ackerman is inherent in the politics of social life. The reason their essentially deductive approaches are unpersuasive, however, is that they are not capable of providing for the open-ended, creative dimension in the social construction of culture.
42. A persuasive demonstration of this thesis will have to await another paper.
43. By "legalistic" I mean simply that efforts to constrain agency actions have been pursued largely in courts and that the courts tend to frame the issues mostly in terms of the legal rights of regulatees, particularly under the Due Process clause and the Administrative Procedure Act and the legal prerogatives of the agency. The issue in the first case is not whether the administrative action reflects the democratic will, but whether it gives due consideration and deference to the protected rights of the regulatee. The issue in the second case makes assumptions about democratic legitimacy, but focuses primarily on fidelity to a statute.
44. I refer here to the approximately century-long period between the time local regulatory bodies generally were conceptualized as legislatures (rather than corporations) through the classic cases of 1935 invalidating New Deal delegations. Although the decisions on the non-delegation doctrine at the state level are checkered and, so far as I know, little documented by scholars, it seems clear that the doctrine played a major role in the policy thought of both legislatures and courts.
45. See Industrial Union Department v. American Petroleum Institute, 448 U.S. 607 at 687 (1980) (Rehnquist, J., concurring) and American Textile Manufacturers' Institute v. Donovan, 452 U.S.490 (1981) (Burger, C.J., dissenting).
46. See, e.g., Portland Cement Ass'n v. Ruckelshaus, 486 F.2d 375 (D.C. Cir, 1973)
47. Chevron, U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837 (1984).
48. [Scenic Hudson to recent NEPA cases]
49. Not only is there considerable disagreement on the actual degree to which agencies have been captured, there is little concurrence on whether capture operates through industry domination of appointment processes, "revolving door" career patterns, excessive empathy due to working together too long, agency desires to maximize their budgets, or agency dependence on industry for information. Of course failure of observers to agree on exactly how capture occurs is not to be taken as evidence that it does not occur. Quite the contrary, capture could presumably operate through any combination of these avenues and others. Moreover, the processes could differ significantly in different regulatory arenas.
50. Why those expanded powers were obtained is not an issue I will address here, although, as I indicated in Meidinger (1986a), I think it is still poorly understood and worthy of further discussion.
51. It also depends on the dubious proposition that regulatory decision-making would achieve more democratic results in the absence of beneficiary power individuals and groups involved.