"Unlocking the Iron Cage: Public Administration in the Deliberative Democratic Theory of Jürgen Habermas"

(published in Administration and Society, March 2004)

Dr. Terrence Kelly

Department of Philosophy

California State University, Hayward

Public administration is rarely viewed as fertile ground for deliberative democracy—and for good reason. As administrative, bureaucratic power functions according to an instrumental, hierarchical logic, it seems the least likely place for deliberation in general, let alone for the complex and delicate forums needed to bring diverse groups of citizens together in conjoint discursive problem solving.

However, there is a growing body of literature within public administration studies that suggests that public administrators can effectively turn to deliberative democracy as way to increase civic participation, reduce citizen and administrator alienation, and create consensus-based policy/administrative decisions. Administrators are using deliberative democracy for a wide range of issues: delivery of health care (Folley 1998), neighborhood renewal (Grey and Chapin 1998), energy policies (Timney 1998), the writing and regulating of EPA regulations (Sabel, et al 1999) and so on. This literature suggests that it may be fruitful to re-examine deliberative democracy from the perspective of the administrative state.

In this paper, I take up such a re-examination through an analysis Jürgen Habermas’s democratic theory. Habermas’ work provides one of the most comprehensive accounts of deliberative democracy. More importantly, for our purposes, Habermas’ work is motivated precisely by the question of how a modern administrative state can remain robustly democratic. On this issue, I argue that a shift has occurred in the way Habermas conceptualizes public administration, a shift that makes it possible to see both the plausibility and necessity of deliberative democratic forms of state administration. I argue that, for Habermas, public administration now stands as the unique site of direct citizen participation in democratic governance. This development in Habermas’ work is important because it provides a link between the theoretical design of deliberative democracy and the growing empirical literature on the practical implementation of deliberative democracy in public administration.

The Theory of Communicative Action (TCA), Habermas’ major work of the 1980s, represents a significant departure from the early Frankfurt School’s reliance on the social theory of Max Weber. From the standpoint of public administration, however, TCA’s reliance on systems theory continues the Weberian legacy in Frankfurt School social theory.

Weber had argued that as the modern state grew in size and complexity, efficient organization and administration would become a foundational social/political principle. Thus, in the name of efficiency, the modern state becomes increasingly bureaucratic in structure, that is, it becomes more specialized, rule bound, hierarchical, and closed off to feedback. Most importantly, the modern state is guided by zweckrational, or instrumental, means-ends oriented rationality. Bureaucratized and instrumental, the modern state can efficiently administer complex state functions.

However, the turn to instrumental rationality comes at a high price. As increasingly bureaucratic and instrumental, the state must treat its citizens as objects to be processed and controlled. Thus modern society becomes locked in an "iron cage" of instrumental rationality. Modern life—for all its tremendous complexity-- is purchased with human freedom.

Dramatically, Weber uses modern democracy itself as an example of the pathological forms of modern rationality. National campaigns and elections, open political parties, and efficient news delivery brings more citizens into the political process. However, politics at such a scale requires efficient organization. Political parties, for instance, become increasingly bureaucratic if they are to maintain order within their ever-expanding ranks. In becoming bureaucratic organization, parties close off avenues of feedback response, thus isolating them from the de-stabilizing effects of public sentiment. The result: mass democracy is made possible by the very same strategies that make popular sovereignty impossible.

While Habermas agrees with Weber that strategic rationality is a distinguishing feature of modern life, he rejects the claim that modern rationality, as such, is instrumental. In his "iron cage" argument, Weber presupposes instrumentality as the fundamental norm of social action in modern societies. In contrast, Habermas argues that there are forms of rationality that do not exhibit the means/ends efficiency that are characteristic instrumental rationality. For instance, when agents seek to communicate with one another, they use a form of rationality that is based on arriving as a reasoned consensus rather than mere efficient ends fullfillment. This is particularly true when agents make a good faith effort to come to understand one another. It is also true when agents face problems that require a consensual solution. Habermas calls this second form of rationality "communicative rationality."

Communicative rationality is the underpinning of a range of practices that Habermas categorizes as "communicative action." Habermas uses the everyday quality of this form of rationality to launch an empirical criticism against Weber. The iron cage thesis is too idealistic in that it ignores the communicative rationality that permeates the everyday practices of agents. As enthnomedological analyses indicate (Garfinkel 1969), everyday communication is shot through with normative, non-instrumental concerns whereby speakers attempt to genuinely communicate with one another in order to reach an agreement.

Instead of an iron cage of instrumental social structures, Habermas characterizes modern life as an ongoing tension between instrumental and communicative rationality—between forms of life that are consensual and communicative, and forms of life that are instrumental and systematic.

Despite his rejection of Weber’s account of modernity, from the perspective of public administration, Habermas remains in fundamental agreement with Weber. The administrative state is more or less Weberian in structure and can lead to pathological forms of social development that are manipulative and undemocratic. For Habermas, these insights needed to be cobbled into in a new social theory that could also take seriously the communicative dimension of everyday practical rationality. In The Theory of Communicative Action Habermas does this by distinguishing "systems" from the "lifeworld."

In the concept of "system," Habermas retains a Weberian component in his social theory. A system is a sphere of human interaction centered around non-communicative steering mechanisms (i.e. money and power) through which agents generally act instrumentally. In the system, the instrumental micro-macro link theorized by Weber is preserved—but only for unique spheres of social interaction. In these spheres, such as the economy and the administrative state, agents reason instrumentally. The result is that their actions create forms of unintended macro developments. Indeed, Habermas agrees with systems theorists, such as Nicolas Luhman, that at the systems level, social analysis requires little reference to agents.

Counterbalanced with "system" is the "lifeworld." The lifeworld, for Habermas, is the sphere of interaction centered on culture, society and personality that is communicatively structured and reproduced through justificatory (argument) and expressive (narrative) communicative acts. Unlike systems, actions in the lifeworld are by and large not predictable, and agency takes a central role. Here, the macro analysis of systems theory gives way to the fine-grained microanalysis of ethnography, which aims not at prediction, but at understanding.

Whereas Weber and the early Frankfurt School tell a story of modern life as systems, Habermas tells a more nuanced story of the tension between systems and lifeworld. Because the lifeworld is the sphere of the full expression of the human person, it is important that it remain independent of the leveling effects of social systems. However, Weber, Horkheimer and Adorno are correct in their assessment that in techno-industrial societies, the allure of instrumental rationality is powerful, and maintaining a firewall between lifeworld and system is difficult. When this firewall breaks down, the lifeworld become, in Habermas’ terms, "colonized." This occurs when "the media controlled subsystems of the economy and the state intervene with money and bureaucratic means in the symbolic reproduction of the lifeworld." (356) One of Habermas’ central examples here is the "juridification" of the lifeworld through the administrative welfare state.

Habermas sometimes describes the welfare state as "halfway successful." On the one hand, the welfare state gives institutional flesh to those social rights possessed by citizens which aim at securing their well being. However, the further the state becomes involved in the everyday well being of its citizens, the more it tends to bring systematic power into the lifeworld. The resulting juridification of the lifeworld replaces the spontaneous, communicative organization of everyday life with legal and administrative relations.

The encroachment of bureaucratic power on the communicative forms of interaction in the lifeworld naturally gives rise to alienation. However, the welfare state finds itself in something of a dilemma. To secure the well being of citizens, it must intrude into the personal and intimate spheres of social life. At the same time, in order to be efficient, it seemingly must be bureaucratic. As a result, clients of the welfare state are subjected to a "violent abstraction" where, ironically, their personal well being becomes depersonalized through bureaucratic means. (TCA 363) This "dilemmatic structure" of the welfare state alienates citizens from the distant, unresponsive, and yet intrusive structure of administrative power. The lifeworld become both separated from, and yet controlled by, the instruments of state power. In such a condition, citizens rightly feel separated and isolated from the state and often from one another.

A related problem is justification. As the administrative state penetrates the lifeworld, it finds a different logic of justification. The lifeworld is organized around "consensual mechanisms" of communication where thick ethical values and moral norms play a critical role in the maintenance of everyday interaction (Garfinkel 1969). Such norms are critical components of justification and the sense of appropriateness in everyday communicative contexts. Administrative power, on the other hand, does not function according to this logic of justification. Because it is instrumental and systematic, it typically appeals to legality and efficiency as sources of justification. But the legality or efficiency of administrative actions, while sufficient for justification within the system, are insufficient as administrators act within the lifeworld. In the lifeworld, ethical and moral justifications will be demanded if administrative actions are to be seen as justified. Again the dilemmatic structure of the welfare state is unable to cope with this demand. Because of its systematic and instrumental nature, the administrative state is often unable to provide non-instrumental justifications for its actions. What’s worse, it often lacks a proper communicative forum where any kind of justification can be offered to citizens. For instance, "one shot" public hearings often provide citizens with little opportunity to engage in dialogue with planners and administrators.

The problems of alienation and justification are, of course, interlinked. As citizens become depersonalized they demand richer forms of justification from the administrative state. As the state cannot meet those justificatory demands, citizens become even more alienated. Likewise, alienation itself can be caused by the lack of normative justification for administrative actions.

The early critical theory of Habermas makes clear the task of a satisfactory theory of modern democracy. Democratic theory must balance the competing demands of system and lifeworld in a way that preserves citizen autonomy and makes possible robust forms of administrative justification. Habermas’ theory of deliberative democracy takes these problems as two sides of the same coin and proposes a structural/discursive solution to them.

In his 1988 essay "Popular Sovereignty as Procedure," Habermas takes up the issue of justification and administration at some length. Here, Habermas does not rely on the system/lifeworld distinction so central to TCA. Instead, he relies on a distinction between communicatively generated and administratively employed power. Habermas’ critique, despite the new distinction, remains similar to that from TCA. Still operating under a Weberian account of administrative power, Habermas again characterizes the administrative state as necessarily instrumental:

The administration obeys its own rationality criteria as it operates

according to law…what counts is not the practical reason involved

in applying norms, but the effectiveness of a given program. Thus the

administrative system primarily deals with the law instrumentally. (483)

As in TCA, Habermas argues that as the state penetrates ever more deeply into the lifeworld, instrumental rationality may not serve as an appropriate or acceptable justification for state action. In the lifeworld, and in the public sphere, citizens use "communicatively generated power" to justify actions and form communicatively structured joint intentions. Communicative power aims at motivating citizens into action through collective agreement and will formation. While citizens may use instrumental arguments in this process ("we should do A because it is more cost efficient than B"), it is by no means the exclusive form of communicatively generated power. For instance, citizens also rely on moral, ethical and appropriateness argumentation in communicative action. Moral arguments aim at determining those principles that citizens believe to have a binding force on all (i.e. the requirements of justice). Ethical argumentation determines the principles or codes that one should follow to be authentic to one’s personal identity/community (i.e. racial, gender, ethnic and religious). Finally appropriateness argumentation determines the best way in which moral, ethical and instrumental considerations should be applied in the particular cases at hand.

The fullness of rationality used in the lifeworld comes into conflict with the one-dimensional rationality of the administrative state when the law is applied administratively into the lifeworld. These cases are so wide spread in the administrative welfare state that they can hardly be avoided. For instance, administrators and citizens often conflict over infrastructure projects. Where should the new highway be built? Where should the sewage line run? Should we cut down trees in a neighborhood because of repeated forest fire problem? Answering these questions often leads to intractable conflict because citizens and administrators use different forms of rationality (Forester 1989: 60). Whereas administrators operate instrumentally (i.e. cutting down the trees is the cheapest and quickest way to reduce fire risk), citizens use more robust forms of practical rationality (i.e. the trees are an essential element to the character of the neighborhood). As these two ships pass in the night (and they often do), the result is alienation for citizens and administrators. Administrators grow frustrated with citizens and begin to systematically exclude them from public discussion (i.e. public hearings held at times and locations that make it difficult for citizens to attend). For their part, citizens reject the legitimacy of administrative action and either withdraw from the political process, or resist administrative action through litigation, protest, or violence (King & Stivers 1998: 7-12).

This situation again highlights the dilemmatic structure of the administrative welfare state. The state increasingly takes up issues of the welfare of its citizens, only to find itself unable to do the job in a way that its own citizens can find legitimate. The tension between administrative employed power and communicatively generated power exposes a fundamental problem in modern democracies. If a state is to be democratic, then the law must be said to be the product of popular sovereignty. That is, to put it loosely, the law is somehow the product of the decision making power of the people themselves. However, this seems impossible when law is created and applied administratively—often in direct conflict with the moral/ethical reasoning used by citizens themselves. Under administrative power, the public no longer recognizes itself in the power under which it is ruled.

The task of "Popular Sovereignty as Procedure" is to construct an account of popular sovereignty that takes seriously the existence of systematic forms of administrative power. Its central question: "How can administratively employed power be linked to communicatively generated power?" One common position in classical liberal theory is that administrative power is linked to the public sphere through the elected representatives of the people (Locke). In this model of governance, which Habermas elsewhere calls the "self programming" (Habermas 1996: 480) society, the public appoints representatives through elections. Representatives, in turn, create laws in the name of the people. Administrators then execute or manage the dictates of the legislature in the name of the public good. On this model, if the public is alienated by administrative actions, then they can demand changes at the legislative level (i.e. elect different representatives). The advantage of this model is that it creates a link between popular sovereignty in a way that demands very little from citizens aside from voting. Furthermore, it explains why instrumental administrative power is consistent with democracy. Because administrators are merely executing the public will (that is law that is already validated by the public through their representatives), there is good reason for doing so efficiently. In fact, the idea of democracy seems to demand nothing less than the execution of the public will in the most efficient manner possible. As Habermas puts it, the hope here is to create a society by which "the people program the laws, and these in turn program the implementation and application of law…" (Habermas 1996: 481).

However, the self-programming society suffers from deep descriptive and normative shortcomings. Descriptively, it misconstrues the activity of the administrative state as one of mere management or execution. While this has been an ideal for traditional administrative theory (Wilson 1887), one cannot separate law making and law managing in such a clean way.

It is a Wittgensteinian insight that rules cannot determine the range of their own applications. As many administrators painfully discover, the same is true of law. Laws cannot be written so tightly that they can determine in an a priori fashion the appropriate mode of their execution (Gunther 1993). This means that administrators must use a great deal of interpretive practical judgement in deciding when and how law should be applied. This ranges from mundane decisions (what laws will be prioritized under budgetary decisions, what kind of staff is hired) to substantive judgement calls on how a law will be applied in a specific community (Forester 1989: 74). From the standpoint of the citizen, the law making/managing distinction fails because for them, law is the force that they face in the lifeworld itself. When an administrator executes the law in a particular manner, that decision is part of the law to citizens. Through their interpretation and execution of the law, administrators are never merely managers—they are constantly adding something to the law itself. In this sense, as the legal realists (and CLS) argued, those who execute the law are also lawmakers themselves.

If this is correct, then the liberal model of the administrative state requires modification. Popular election of legislative representatives provides no guarantee of popular sovereignty because those representatives are not the only lawmakers. As administrators also make law (not in the formal or positivist sense, but in the realist sense) they stand outside of popular sovereignty. Mere elections do not provide a link between the public and the fine-grained details of the law. Here the devil is in the details. Once a gap exists between the public will and the law, execution of the law in the name of efficiency often flies in the face of the public’s values and priorities.

A second common proposal to link popular sovereignty with the administrative state calls for a more active citizenry to take part in the process of the administrative state itself. For republican theorists, the only way to close the gap between citizens and the administrative state is to include citizen participation in the administrative state itself. Indeed, at its most radical, this position calls for the replacement of the administrative state with citizen activism and volunteerism. As a result, republicans often emphasize the role of civil society where organized citizens engage in the process of public life and relieve the administrative state of its duties. On republican grounds, citizens become self-legislators in the fullest sense only when they live out their public lives in such a way as to concretize their political will through their direct action within the state.

Habermas rejects the republican approach for both empirical and normative reasons. Empirically, he argues that "[a]ll the radical varieties of Rousseauianism labor under this moral overburdening of the virtuous citizen." (473). In modern societies, it is unreasonable to expect that citizens can consistently live out the kind of robust public life that republicanism calls for. This empirical problem also underscores a normative point. Republicans typically argue that such a burden is necessary for citizens because the "good life" of citizens is found in their public, not private, lives. This, Habermas argues, is consistent with the republican subordination of civil rights and individual liberties to political rights and popular sovereignty. However, the reduction of individual liberties to an outgrowth of popular sovereignty creates a dangerous environment in which individual freedoms can be curtailed in the name of a robust "public life." This is particularly problematic because modern societies exhibit a degree of pluralism that seeming makes the common ethos envisioned by Rousseau an unlikely social achievement.

Having rejected both the liberal and republican alternatives, Habermas proposes an account of deliberative democracy that fashions a connection between the public and the administrative state in way that respects individual rights, but also creates procedures by which communicatively generated power can steer (or "countersteer") administrative power.

Still following a Weberian conception of administration, Habermas argues that administrative power is predominantly instrumental in nature—decisions are made from the standpoint of efficiency. The norms and justifications that legitimate law "are regarded in the language of administrative power as rationalization appended to decisions that were previously induced." (484) Reducing norms to mere rationalizations structurally allows administrative power to exclude the influence of communicatively generated power. It is precisely this exclusion that Habermas seeks to undermine. Even if administrators view ethical and moral reasoning as mere rationalization, if administrative power is limited to those rationalizations that have been generated by free and open communication in the public sphere, then the communicatively generated power of citizens can exert influence on the administrative state. As Habermas puts it, the idea is that the public sphere assumes responsibility for the "pool of reasons from which administrative decisions must draw their rationalizations." (484) This places limits on the range of administrative action. It is not that "anything goes" (484) provided it is justified in the name of efficiency. Instead, when popular sovereignty is realized, the political system is structured in such a way as to create procedures by which the free and open public sphere can generate and communicate ideas that mark the range of appropriate administrative justification. If particular administrative decisions depend exclusively on publicly invalidated norms or arguments, then those decisions themselves are invalid.

One project of constitutional democracy, therefore, is the construction of accountable, democratic procedures by which unstructured public opinion can countersteer administrative power (489). Effective procedures are critical because, on the one hand, the public sphere, as Habermas construes it, is relatively unorganized and spontaneous. Within the public sphere, movements, activist groups, communities and interested individuals come to form "subjectless flows of communication" (486) which carry ideas, themes, norms, reasons and arguments. Democratic procedures aim precisely at creating forums by which the "spontaneous, unsubverted circuits of communication in a public sphere" can be linked to "institutionally structured political will-formation." (485)

When democratic procedures successfully make this linkage, then a form of popular sovereignty emerges by which the communicatively generated power of citizens can control the administrative power of the state. However, unlike republican conceptions of popular sovereignty, Habermas’ account produces only an "indirect steering" influence on the part of citizens (484). The communicative power of citizens does not take over the state—a prospect that is unrealistic in Habermas’ eye given the efficiency demands that complex modern forms of life place on the state. Instead, communicative power acts in "the manner of a siege" (486) as it provides a "normative filter" which limits the range of legitimate decision making in the administrative state.

When successfully implemented, the deliberative democratic state provides for popular sovereignty, if not in the Rousseau’s sense of self-legislating sense, then at least in the weaker sense in that citizens recognize their own rationality in the procedures that create the laws by which they must abide. Because only norms and arguments that deliberating citizens have validated in the public sphere can justify legal and administrative decisions, the gap between communicatively generated and administratively employed power is closed. Or is it?

For many, Habermas’ procedural account of popular sovereignty has been less than satisfactory. Republicans and radical democrats question a theory of popular sovereignty that requires no direct citizen participation (Bohman 1993). While Habermas is perhaps correct in his estimation that republican theories of democracy place too great a burden on citizens, his own solution requires too little from them to solve the problems of justification and alienation that his earlier work identified.

As I noted above, in TCA, the problem of justification vis a vis administrative systems begins when the welfare state penetrates the lifeworld. Agents in the lifeworld demand robust justifications from administrators, who can only reply in instrumental terms. This leads clients of the welfare state to undergo a "violent abstraction" and become alienated from the state. Habermas’ procedural account hopes to overcome this problem by appealing to rationally acceptable procedures through which administrators can justify their actions with ideas and opinions that have been produced by the public sphere. However, as I argued above, once citizens are depersonalized into anonymous chains of communication, the robust justifications necessary to overcome alienation are impossible. This is because lifeworld demands on justification have a performative dimension that the countersteering model cannot satisfy.

In lifeworld interactions, when agents justify their actions they rely on much more than the illocutionary content of their arguments. They also take up the practice of justification by which they engage their interlocuters with reasons that are debatable and redeemable within that particular context. As Habermas himself rightly argues, the performative aspect of engagement is a critical feature of lifeworld justification. It requires that agents take their interlocuters seriously as rational agents who must be persuaded through a dialogical process of argumentation. This means that the process of lifeworld justification must have space for the contentious parrying between parties as they offer arguments and counterarguments. Indeed, one of the lessons of Garfinkel’s breeching experiments is that agents will not validate justifications if they do not believe they are being taken seriously as interlocuters—even if they agree with the illocutionary content of the justication.

It is precisely this performative dimension of justification that Habermas’ countersteering model lacks. Because citizens have been reduced to anonymous chains of communication, there is no one for administrators to engage in the practice of justification. Thus the crafting of justifications becomes the exclusive task of the administrator, with citizens only supplying the raw material of opinions and ideas. Even if administrators craft justifications that have agreeable illocutionary content, there is no reason to expect that citizens will be any less alienated given their lifeworld expectations regarding the dialogical practice of justification.

The countersteering model also runs afoul of the lifeworld expectation of sincerity in justification. As Habermas argues, in communicative situations, agents rely on the truth, rightness and sincerity of their interlocuters. Among other things, "sincerity" would include the idea that interlocuters actually hold or believe in the justifications they offer to others. When agents fail to meet this expectation, they are often criticized for being cynical or deceptive—regardless of the logical status of their justification itself.

However, given the way Habermas has set up the problem, administrators cannot possibly be sincere in their justifications. Because administrators reason exclusively in instrumental terms, they can only append normative justifications as "rationalizations." The administrator actually bases her decision on instrumental reasoning, using public opinion formation as a boundary marker for the range of legitimate decisions. When the administrator then offers a justification drawn from public discourse, they fundamentally offer a justification they themselves do not hold and which was not constituative of the decision at hand.

Of course, for Habermas these objections miss the point that given the systemic nature of the administrative state, the robust demands of justication made in the lifeworld cannot possibly be satisfied at the administrative level. However, the problems of justification and alienation result from the friction between lifeworld and system forms of justification. Because Habermas construes administrative decisionmaking in purely Weberian or systems theory terms, the gap between administratively employed and communicatively generated power cannot be closed. The best one can hope for is to transform public discourse into digestable inputs for the system and aim for a congruence between public opinion and administrative decisions. If the administrative state is fundamentally instrumental, then anything like direct democracy is impossible and political alienation is simply a fact of modern life. But this claim only goes as far the empirical account of public administration that supports it. If this account is not accurate, then the tension between lifeworld and systemic forms of justification might be resovled after all.

Habermas’ Weberian tendencies, expressed through systems theory, leads him to a description of public administration as essentially systematic and instrumental. However, such a description of public administration ignores the role of communicative rationality with the practice of administration itself. Indeed, starting with the depersonalizing assumptions of systems theory, Habermas commits the same mistake for which he rightly criticized Weber—an idealistic account of social practices that reduces the rich use of practical rationality to a single dimension of it.

Theorists of public administration, as well as administrators themselves, have been dissatisfied for some time with bureaucratic models of administrative practice. Some have noted that bureaucratic modes of organization turn out, over time, to be rather ineffective due to the increased alienation they create within the organization itself (Jun). Others, such as John Forester (1989, 1993), focus on the ways in which, even in bureaucratic settings, administrators use communicative rationality in their planning and policy practices. Through questions, information, and collaboration, administrators focus attention in differing ways. They also can spread responsibility for public planning and management in different ways as well. For instance, contra Weberian assumptions about efficient administration, there is growing literature on the uses of collaboration between administrators and citizens in creating political viable public policy and management plans. When focusing attention on a problem, providing information to citizens and spreading responsibility, administrators routinely turn to the skills of communicative action and rationality (Foester 1989: 27). At the very least administrators, from the very start, must appeal to the richer forms of communication (and justification) when they use "organizational tact" (Forester 1993: 29) to present and justify their decisions within the administrative state itself. This is critical, for instance, when administrative agencies interact with one another. Appeals to technocratic expertise, for instance, are unimpressive when dealing with other "experts." Without justification through a process of communicative action, administrators often find their work undermined by other agencies (Flyvberg 1998: 85) or simply ignored by politicians (Forester). The role of attention focusing, responsibility sharing, and organizational tact are indicative of the ways in which administrative practice is shot through with communicative action. Administrators routinely expect and utilize the richer forms of justification that only communicative rationality can provide.

Communicative rationality also enters into administrative practice because appeals to efficiency are often indecisive in decision making. As Rawls noted in "Distributive Justice" (1999) and A Theory of Justice (1973), in many situations multiple outcomes can be identified as optimally efficient. In such cases, appeals to efficiency, while providing for a range of choices, cannot provide justification for any one particular choice. This situation is all the more tenuous because efficient outcomes can only be determined by identifiable preferences. However, administrators often face situations of value pluralism in which different accounts of the good lead to different efficiency judgments. As communitarians argue, when administrators make decisions about efficiency, they cannot help but invoke (often implicitly) an account of the collective good—which itself cannot be justified by appeals to formal instrumentality.

On closer examination of the everyday practice of administration, the Weberian and systems theory assumptions appear suspect. However, if administrative practice is always already instrumental and communicative, then the systems/lifeworld, administratively employed/communicatively generated power dichotomies seem overly idealistic. If administrative power is always (at least partially) communicatively generated, then Habermas’ account of the administrative state is too one sided.

This problem is something that Habermas himself seemed to recognize after "Popular Sovereignty as Procedure." In Between Facts and Norms, Habermas further develops, but also departs from, the model of popular soveriegnty e outlined in "Popular Sovereignty as Procedure."

The central problematic of Habermas’ theory of democracy in BFN is linking political and administrative power with public deliberation given the constraints of modern complexity. Increasingly in complex and pluralistic societies, the public sphere becomes more "abstract" and less localizable within concrete dialogical encounters. In such publics, dubbed "weak" publics by Nancy Frasier, far flung participants are linked through mass media in an ongoing public discussion. However, this discussion is by and large virtual and decentralized as the discussion is "uncoupled from the thick contexts of simple interactions, from specific persons and practical obligations." Given the nature of weak publics, how can public deliberation (if we can even call it that) be linked to political power.

Habermas construes of the problem as one involving both input and output considertations. On the input side, links must be developed between the abstract public sphere and legislative power—between public deliberation and the creation of law. On the output side, the challenge is to link public deliberation to the application or administration of law.

As in POP Habermas proposes a reflexive form of law that takes its own legitimacy into questions by developing procedures by which the inchoate discussions of the weak public forms the basis for legislative deliberations. Once again, Habermas uses the metaphor of a normative filter. However, this is taken more as procedures that allow well formed public opinion to enter legislative deliberations with an eye towards identifying those elements of public opinion that can form the basis of legislative justifications with universal (at least within the legal community) acceptability. Legislative deliberation aims at translating the particular opinions emerging from the weak public sphere into arguments that can generate the obligatory force of law.

However, coupling the abstract or weak public sphere with the legislative process cannot satisfactorily account for the "output" side of state power. As throughout his writings, Habermas holds that modern, complex states cannot help but rely on the discretionary power of state administrations in managing welfare and security programs. As a result, administrations gain a certain autonomy from the public, elected officials and positively enacted law.

The administration predominantly programs itself by

steering the legislative process through its bill proposals

by extracting mass loyalty from the citizenry through parties

that have become arms of the state and by making direct

contacts with its clients.

However, his theory of administration has become much more nuanced. There he argues that the practice of administrative decision making can never be strictly instrumental because administrators, insofar as they make law and policy (in the realist sense described above), must justify the norms used in creating such policy. But, Habermas notes, "these normative questions cannot be answered from the standpoint of effectiveness, but demand that normative reasons be dealt with rationally." (436) As normative, administrative practice cannot be accurately described as simply instrumental. Indeed, Habermas argues that a techno-efficiency description of administrative practice "in no way leads to a matter of fact treatment of administrative problems. Rather, it results in opportunistic and unreflective ways of reconciling value complexes without the guidance of reasonable criteria." (436)

This new account of public administration contrasts sharply with the account offered in TCA and "Popular Sovereignty as Procedure." There Habermas argued that "what counts is not the practical reason involved in applying norms but the effectiveness of a given program. Thus the administrative system primarily deals with the law instrumentally." (483) Now, in BFN, administrative practice is described as normative from the very beginning. Administrators cannot help but appeal to norms in crafting policy and planning decisions. Given this new account of public administration, Habermas argues:

To this extent, my image of a democratically besieged fortress

By "democratization" Habermas still insists on a countersteering model by which rationally acceptable procedures create a filter by which public deliberation takes responsibility for the range of justifications available to administrators. However, Habermas’ new account of administrative practice no longer supports the necessity of a countersteering model. The argument for the countersteering model depends on a strong communicatively generated/administratively employed power distinction. Once this distinction breaks down, a countersteering model of popular sovereignty is no longer conceptually necessary. While administrative and lifeworld forms of justification often conflict, they do not necessarily do so. If citizens and administrators use communicative reasoning, then direct deliberative decision making is at least possible. This means that deliberative democratic theory can aim for much more than the a countersteering role for citizens. It can aim, at least ideally, for a more direct form of democratic deliberation between citizens and administrators—a form of governance that has called "collaborative government."

Certainly, collaborative government would better meet the normative expectations of lifeworld justification than Habermas’ countersteering model. In direct deliberations, administrators are in a better position to engage citizens and satisfy both the performative and illocutionary demands of justification. It also provides a forum by which administrators can present their sincere arguments to citizens. Ideally, the outcomes of such deliberation would be decisions based on justifications that are sincerely acceptable to all.

But regardless of its normative value, is a more direct form of citizen/administrator deliberation pragmatically viable? Even if collaborative government best meets the normative demands of popular sovereignty, it may come at a steep price in terms of efficient governance. Might not such a method of decision-making drastically reduce the efficiency of administrative decisions both in terms of timeliness and effectiveness?

I argue that there good reason to believe that collaborative government not only meets the demands of a robust account of popular sovereignty, but does so in a way that is acceptable in terms of efficiency. Indeed, in at least in some contexts, collaborative government is more efficient than the traditional models of administrative decision-making.

As I argued above, administrators already appeal to norms in their decisions and must justify those norms—at least to their staff, superiors and other agencies when using organizational tact. Despite claims to the contrary by technocratic accounts of administration, the give and take of time consuming deliberation is not new to administrators (Flyvbjerg 1998, Forester). Collaborative government simply requires that administrators extend the deliberative process to include citizens. While this requires a drastic change in the way that most administrators view citizens, it simply expands (and makes more rational) the deliberative and communicative content of administrative practice.

In expanding deliberations to include citizens, administrators can take advantage of their unique structural position. Administrators, more than any other part of the state, deal with specific citizens in a face to face manner as they connect everyday forms of life with the law. Recall that in TCA Habermas noted that administrators interact with citizens, but subject them to a "violent abstraction" by use of depersonalizing bureaucratic means. However, within this intimacy lies the potential for something beyond colonization of the lifeworld. Instead of subjecting citizens to "violent abstractions" and depersonalization, administrators could build direct communicative links between state and lifeworld. By including citizens in administrative decisions, these links can be built on trust (if not solidarity) and citizen empowerment rather than objectification and manipulation. The following case study also suggest that this also creates a forum by which the administrative welfare state can become a framework for communal solidarity and cooperative action.

Joseph Chapin and Linda Grey’s study (Chapin & Grey 1998) of the South Apoke Project illustrates the virtues of collaborative government at work. The town of South Apopke lies along the rural edge of Orange County, Florida. Far from established lines of county services, the town has traditionally been a site of extreme poverty and underdevelopment. In 1991, in response to increases in juvenile crime throughout the county, the Targeted Community Initiative (TCI) was established to help revitalize the county’s poorest communities; particularly South Apopke, which accounted for 25% of the county’s juvenile arrests but less than 1% of its total population. The TCI had broad public support and was seen by many as a long overdue public attention to the neglected areas of the county, or at least as a necessary measure to address the sources of juvenile crime.

Despite increased funding and administrative commitment, citizen apathy and time-consuming pressure on administrators hampered improvements in the community. Beyond the physical improvements to the community, both citizens and administrators were dissatisfied with the slow progress of the TCI. Citizens felt that administrators "bossed them around," while administrative staff came to dread assignment to South Apopke and the "24/7" work commitment it entailed. As a result, the TCI simply failed as an effective program to improve the quality of the town’s social condition as a whole.

This initial struggle of the TCI in South Apopke suggests that even when laws are legitimated by norms produced by the "normative filter" of public discourse, the alienation created by welfare state administration need not subside at all. In this case, the level of trust with citizens of this extremely marginalized community was so low, that any administrative intervention provoked alienation. Indeed, in a town meeting at South Apopke, Linda Chapman, the county’s new executive was "caught offguard" by the "torrent of hostility and resentment" that her effort to address the town’s problems created. "Citizens felt that they had been neglected for so long by local government that they were unwilling to believe that anyone could be sincere about changing that relationship." (175)

To salvage the TCI, collaborative government replaced the techno-expertise model of administration. Residents were brought into sessions aimed at deliberately identifying the actual interests of the community. In one session, for example, citizens were given paper and crayons and asked to draw their ideal neighborhood. In comparing their drawings, citizens began a process of self-examination and articulation of common goals and values. This hermeneutics of self-discovery was followed by "citizen-friendly" planning meetings. Administrators were surprised to find that citizen apathy dissipated when key issues were put in terms they could understand. Grey noted that citizens were confused and put off by bureaucratic terms such as "code enforcement and land use" (189). By making the agenda more accessible, Grey was able to dramatically increase citizen participation.

As stronger participants, citizens were able to become co-policy makers with administrators. Grey and Chapin note:

the sharing of budget information was probably one of

the most important things public administrators could have done to

build public trust. It also relieved staff from the burden of explaining

to citizens why not all of their needs and expectation were realistic,

given fiscal constraints (187).

By working with teams of citizens, administrators were surprised to learn that when empowered as co-policy makers, citizens were adept at understanding "the complex issues (187)." As they developed community programs, citizens were able to work within the instrumental constraints of the budgetary process.

The results of this form of "collaborative government" were dramatic. Legislative bodies approved the plans drawn up by citizens-administrator teams with little modification. Furthermore, implementation of those plans was made easier and more effective because citizens took an active role in the management and staffing of new community programs. This made the administration of these programs less expensive without sacrificing efficiency.

Public administration can be an ally, however unlikely, of deliberative democracy. Such governance, I believe, is the best way for the administrative state to live up to the ideals of legitimacy that Habermas defends. At a time when most Americans believe that "government is the problem, not the solution," redefining the role of the administrator is more important than ever before. When it functions as a system, the administrative state is slowly drained of credibility and legitimacy. Alienated citizens increasingly become cynical, indifferent, or hostile towards their government, while frustrated administrators attempt to insulate themselves from the public. Collaborative government reconnects citizens with administrators in a way that takes advantage of the structure of the welfare state in order to create real deliberative contexts in which citizens exercise decision making power and assume joint responsibility for administrative actions.

I have not addressed the question of collaborative government and the legislative branch of government. Certainly, collaborative government at the administrative level alters the relationship between the public and legislature. Public policy is often initially generated at the administrative level and subsequently made law. As that policy is written in part by active citizens engaging in collaborative government, citizens come to have a more direct relationship with the legislature that addresses that policy. This is seen in the South Apopke example. Citizens working with administrators draw up budgets and public policies regarding their neighborhood. These policies then become the basis for the legal construction of neighborhood programs. There is research that also indicates that collaborative government can have this effect at the state (Timney) and even the federal (Sabel et. al.) of governance.

However, these interesting questions are beyond the scope of this paper. Here I have tried to show that a Habermasian theory of public administration should move beyond the Weberian and systems theoretic assumptions of administrative practice. By reconceptualizing administrative practice from one that is necessary instrumental to one can be (and often is) communicative, the iron cage of administrative rationality can be unlocked, allowing for the possibility for a true form of radical democracy that can live up to the ideal of popular sovereignty.

Flyvbjerg, Bent. 1998. Rationality and Power. trans. Steven Sampson. Chicago: University of Chicago Press.

Forester, John. 1999. The Deliberative Practioner. Cambridge, Mass.: MIT Press.

_______. 1993. Critical Theory, Public Policy, and Planning Practice. Albany: State University of New York Press.

_______. 1989. Planning in the Face of Power. Berkeley: University of California Press.

Garfinkel, Harold. 1969. Studies in Ethnomethodology. Englewood Cliffs: Prentice Hall.

Grey, Joseph and Linda Chapin. 1998. "Targeted Community Initiative: "Putting Citizens First!"" In King & Stivers. 1998.

Gunther, Klaus. 1993. The Sense of Appropriateness. Trans. John Farrell. Albany: SUNY Press.

Habermas, Jürgen. 1998a. Between Facts and Norms. trans. William Regh. Cambridge, Mass.: The MIT Press.

_______. 1984. The Theory of Communicative Action Volume One. Trans. Thomas McCarthy. Boston: Beacon Press.

_______. 1987. The Theory of Communicative Action Volume Two. Trans. Thomas McCarthy. Boston: Beacon Press.

_______. 1998b. "Popular Sovereignty as Procedure." Appendix to Between Facts and Norms.

Honneth, Axel. 1993. The Critique of Power. trans. Kenneth Baynes. Cambridge, Mass.: The MIT Press.

Horkheimer, Max and Theodore Adorno. 1997. The Dialectic of Enlightenment. trans. John Cummings. New York: Continuum Press.

King, Cheryl Simrell and Camilla Stivers. eds. 1998. Government is Us. Thousand Oaks: Sage Publication.

Rawls, John. 1971. A Theory of Justice. Cambridge: Harvard UP.

________. 1999. "Justice as Fairness." In Collected Papers. Ed. Samuel Freeman. Cambridge: Harvard UP. 47-73.

Timney, Mary. 1998. "Overcoming Administrative Barriers to Citizen Participation." In King & Stivers. 88-102.

Walzer, Michael. 1995. "The Civil Society Argument." Theorizing Citizenship. Ed. Ronald Beiner. Albany: SUNY Press. 153-175.

Young, Iris. 1997. "Difference as a Resource for Democratic Communication." In Deliberative Democracy. Ed. James Bohman & William Regh. Cambridge: MIT Press. 383-407.