Introduction

There is more to politics than bargaining and the mere aggregation of preferences.¹ Political decision making is also a deliberative process where the force of the better argument is often decisive (Habermas, 1984, 132; Manin, 1987; Benhabib, 1996). However, the deliberative approach is often criticized for being a normative ideal rather than an empirically based interpretation of real-life politics in modern democracies. By implication, the approach is largely written off as uninteresting and irrelevant. In this article, I argue that the deliberative approach is both a viable alternative and an important supplement to the traditional conception of politics as a perpetual struggle for power and influence. More specifically, I will try to demonstrate how it can enrich and deepen our understanding of political decision making in local government.

Post-modernists, in particular, claim that the conditions for a rational discourse and consensus are poor today. There is no public sphere where the force of the better argument has replaced power as the ultimate political 'currency'. The deliberative model does not acknowledge the complexity of representative democracies and the plurality of values in modern politics (Mouffe, 1999). Some critics also claim that deliberative processes will favour certain (strong) groups (Young, 2003). However, the criticism most frequently and forcefully levelled against the deliberative approach is that it represents a highly utopian model of democratic politics with little, if any, explanatory value.

The argument can only be settled by identifying — and agreeing on — the conditions for deliberation and rational discourse in democratic politics, and by examining whether these conditions can realistically be met in modern democracies. Otherwise, this debate will remain an academic exercise with little impact on real-life politics.²

Parliamentary institutionalized discussions will be characterized by procedural constraints because of their primary aim to make decisions and priorities of more immediate demands (Bächtiger and Steiner, 2005, 155). When we talk about deliberation in modern competitive democracies, we have to adjust the expectation of deliberation to the contexts of ordinary politics. Power politics does not always facilitate deliberative practices. It is important to be aware of the distinction between weak and strong public spheres. Weak public spheres signify deliberations outside the political system, while the latter concept alludes to formally organized institutions, for example, elected assemblies (Fraser, 1992; Habermas 1996, 313). According to Sharp, Shapiro and other pessimistic authors, modern politics in parliamentary institutions (strong public spheres) is about power and bargaining rather than the creation of consensus by the force of the better argument (Bächtiger and Steiner, 2005, 155).

In order to identify the crucial conditions for deliberation in local politics, we must ask what characterizes local politics as such. How decisions are made, and when is it relevant to anticipate deliberation? According to Steiner et al. (2004, 4), deliberation or talk culture depends on institutional settings. The main question, then, is how the policy space provided by the institutionalization of local government is used. Is local politics only about economic issues and sheer practicalities that contenders can bargain over or is the agenda more diverse? Does it contain moral or ethical issues that by their very nature call for deliberation to be properly addressed? Nordic municipalities are generally small, but deliver a wide array of public services, and local governmental institutions are designed to ensure consensus and cooperation — which is also the constitutive standard for legitimacy.

In order to identify those aspects and to say something significant about what constitutes and legitimates local democracy, we must study how debates are conducted, and how decisions are made and justified. This, however, poses theoretical and methodical challenges. Those challenges will be discussed in order to present an analytical framework for analysing local political decision making. Further, this article will present two empirical cases to underpin the relevance of this analytical framework.

Action and System

A starting point for researching the conduct and content of political debates is to take into consideration that political arguments can reflect both normative and empirical options. A deliberative theory of action assumes that participants in any debate will appeal to inter-subjective standards of legitimacy in order to be rational. Participants are expected to give reasons that are meaningful and defend their actions and preferences by referring to commonly accepted values and standards. As a researcher, in order to be able to give an adequate interpretation of a specific course of action, you must understand the language, values and norms of those you study. In other words, we can only have a complete understanding of actions if we can understand the actor's situation and their evaluations of the situation (Habermas, 1984, I, 374).

An empirical study of political deliberation in modern representative democracies must take into consideration that politics is also a system activity where efficiency is an important source of legitimacy. From a system point of view then, results rather than arguments or intentions are the key to political support. However, by understanding politics only in systemic terms, we may overlook the significance of ideals, norms and values as sources of support for certain courses of action and as justification for individual choice. Since democracy is about strategic choice as well as about normative legitimacy, it is necessary to choose a research design that combines these two conceptions of politics.

Political Autonomy

Modern society is characterized by value pluralism and heterogeneity of interests and therefore the prospects for political consensus are poor. Many defenders of the deliberative approach have stressed this fact and underlined the importance of moderating the idealism associated with it.³ The main challenge, however, for political institutions at all levels is to maintain respect and legitimacy for the priorities that have to be made. The main challenge to democratic rationality is to arrive at acceptable formulations of the common good despite this inevitable value pluralism. (Benhabib, 1994, 34)

This means that political decision makers must be willing to deliberate on the practical consequences of their decisions in order to obtain a certain amount of consensus (Knight and Johnson, 1994). In other words, from a deliberative point of view we will assume that politicians, in some way or other, will engage in normative debates about political solutions.

Further on in this article, I will discuss models for analysing political debates in local democracies. Model pluralism is required in order to capture the often complex nature of local politics. Besides, alternative models are useful in that they provide different criteria by which the quality of political debates can be assessed. Models are also helpful in accurately 'diagnosing' the specific type of local democracy under scrutiny. A key question here is whether political discussions are open to deliberation or are simply characterized by pragmatism, bargaining and compromises.

Political Discourses

Studies of the type alluded to above require an analytic framework that combines classical political theory with knowledge of the basic institutional features of modern political systems. Politicians may have the best of intentions and motives, but may realize rather quickly that the government system leads a life of its own. For example, bureaucratic procedures, overloaded agendas and demands for efficiency will certainly limit the time and resources available for political dialogue or (rational) discourse.

A deliberative perspective is based on Habermas' communicative theory (1984). In Habermas' classical model, actors must be willing and prepared to be persuaded by the 'non-coercive coercion of the better argument' (Habermas, 1984, 131). The ultimate goal is to find a reasoned consensus on the validity of competing claims (Habermas, 1984, 1996; Manin, 1987; Bächtiger and Steiner, 2005, 153). From a deliberative point of view, a reflective and reasoned dialogue is necessary in order to arrive at legitimate decisions in modern pluralistic societies. In a modern democracy, deliberation — based on reason and rationality — replaces religion, tradition or a comprehensive moral outlook as the basis for authoritative decisions (Benhabib, 1996; Habermas, 1996; Bächtiger and Steiner, 2005). Deliberative theorists will claim that a deliberative process is a necessary part of democratic decision making, and that it represents an improvement of democratic practice (Jenssen, 1996).⁴

Returning to real-world politics, the question is how this can enrich empirical analyses of local politics. To begin with, we need to agree on the standards by which the quality of political debates should be judged. The different dimensions of politics involve different standards of evaluation, and according to Habermas' concept of deliberation three questions must be addressed in order to paint a complete picture of a political discussion: (a)the pragmatic question of how we can harmonize competing preferences, (b) the ethical-political question of who we are and seriously want to be as well as (c) the moral-practical question of how we should act (Habermas, 1989, 152).

These three questions capture the complexity of political deliberation in modern democracies. Furthermore, they predict different logics of communication and arguing. Different questions call for different discourses that again relate to different forms of rationality. Empirical and 'trivial' questions are addressed in pragmatic discourses, as are issues where political 'clout' rather than the force of the better argument will be decisive for outcomes.

Value questions call for ethical discourses aiming at defining individual and collective identities. When there are competing interests and identities, there will be a need for moral discourses to decide whether standpoints, in case of conflicts, can be generalized.⁵

According to Habermas (1996), democratic procedures contain several discourses and the empirical relevance for local politics will be discussed further in the next paragraph.

Pragmatic Questions

Normative questions concerning what ought to be done will always relate to one or more aspects of the practical reason (Habermas, 1996, 100 ff). This reason involves the different norms of action according to whether the questions to be answered are of an empirical/pragmatic, ethical or moral kind.

Pragmatic questions are questions of fact and require empirical knowledge to be settled in a satisfactory manner. Pragmatic questions that may be raised in, say, a debate on the municipal budget can be of the following type: what can we expect with regard to government grants? Where can we get the best loan terms? How can we improve the financial conditions next year? Who should have the next building contract? These are all empirical questions, and you do not need participation from elected representatives or drawn-out debates to find the 'correct' answers to such questions. The administrative staff will be more competent to deal with such questions. This dimension of the practical reason predicts given ends and the central question is how to choose the most efficient means to reach those ends.

Bargaining relates to questions where recourses are more important than arguments. In a democracy there are issues and situations where the conflicts of interests are of a kind that can be solved by applying a fairly simple set of rules for aggregating preferences. This will be the decision-procedures where the conflicts about ends and values do not disturb those values and preferences that are agreed upon. It is 'weak evaluations' that do not need to involve important value discussions (Taylor, 1989).

In local democracy it is common to make strategic alliances that are temporary and based on cost/benefit analyses. The local political actors will all the time be looking for strategic partners in order to obtain benefits for their own community, party or other interests they may favour.

Pragmatic discourses then, presuppose a certain degree of consensus when it comes to basic values and they are appropriate for settling empirical questions pertaining to efficiency. They are also appropriate for addressing questions concerning the aggregation of preferences.

Ethical–Political Questions

The ethical dimension of practical reason pertains to what is valued — to the question of what constitutes the good life for a given community (Eriksen and Weigård, 2003, 163):'To answer such questions is at the same time to decide who we are and who we want to be, as a collective, i.e. to which values, ideals, and form of life we subscribe, and which goals we want to pursue'.

When local politicians are promoting issues that pertain to local culture, values and norms, they are concerned with who and what they want to be as part of a community of citizens. Ethical norms are different from the pragmatic ones and as such a 'source' of a different set of questions. They will have a collective dimension since there will be a focus on shared values. How should we take care of the elderly? What do we mean by welfare and well-being in our municipality? Who are we and how are we different from the guys in the neighbouring community? This ethical dimension of practical rationality pertains to questions of what is natural, valuable and right for us to do in our community. The only way to answer such questions is to engage in processes of collective self-interpretation. Who are we and how do we fit in as a part of the larger community? What constitutes a good life and a successful community? The answers to the questions of what is good for us can be found in a hermeneutic process that can be understood as a form of 'clinical deliberation' (Eriksen and Weigård, 2003, 163). These are value questions that belong together with strong evaluations (Walzer, 1983; Taylor 1989). According to Habermas, this can also be called 'the reality of the foreign will' (Habermas, 1996, 163). The collective will-formation, through discussions, will help to decide which problems should be in focus and how they should be solved.

While the pragmatic dimension of practical reason are concerned with questions that can be answered by means of empirical knowledge and by aggregating preferences, the ethical dimension is about defining and agreeing on shared values and common goals. However, those two dimensions do not give any guidelines for situations where the politicians have to reach across conflicting interests, opinions and values. Here, a moral dimension is needed.

Moral Questions

Modern society is characterized by pluralism in both interests and opinions about the good life. The necessity of moral discourses becomes evident when local politicians have to forge viable compromises by reaching across conflicting interests and differing ethical opinions. The moral dimension implies inter-subjectivity in order to answer questions of what the right thing to do is when actions have consequences for other people's interests. The question of what is fair or just is also essential in relation to ethical questions about the good life because the answers will be relative to the cultural context: 'What is fair is due to everyone in the same way and to the same extent. The decisive arguments are those than can be made universal' (Eriksen and Weigård, 2003, 163).

In order to find fair and acceptable solutions in the face of pluralism and conflicting interests, democratic arenas are required where politicians have to consider the moral duties that are superior.

To each of the three dimensions of politics there are three different standards of evaluation. Those basic forms of practical reason relate to different types of questions that arise in a political deliberation process and they also represents different categories of answers. Governmental decisions have to take into consideration both that the public agenda is very differentiated and the fact that there are more and complex criteria of what is legitimate. In practical political will-formation there will often be disagreement on the chosen norms and the question is how much deliberation is there in the real world when it comes to making decisions that can be fair for everyone in the municipality.

The Political Process and the Practical Reason

The idea of deliberation has been seen as a question of how to find a reasoned consensus about validity. Habermas, for example, earlier (1984, 1996) analysed practical reason far more one-sidedly and in line with Kant's moral theory and the universality of moral rules (Eriksen and Weigård, 2003, 160). In recent years, however, Habermas has taken up the idea that normative questions have more dimensions and that practical reasoning consists of more than one discourse. Reasoned dialogue and deliberative processes represent improvements of the democratic decision-making process. However, we have to take into consideration the fact that the political process in modern democracies contains more than one discourse. This is important when the deliberative model is submitted to empirical investigations.

The three different dimensions of practical reason presuppose different discourses and — by implication — different decision-making procedures (Habermas, 1989, 152). As long as we are dealing with trivial choices and empirical facts, pragmatic discourses are appropriate. Pragmatic questions require a political-administrative system that is capable to register people's preferences and to transform those into practical politics. For this purpose, you need a system of aggregation that can handle the flow of demands and forge winning coalitions. Questions involving economic or technical considerations are also pragmatic discussions and do not require more than scientific expertise.

Ethical questions, however, require, a political process where questions of what constitute a good society can be addressed; one that allows citizens to meet in order to clarify their common aims and commitments. In modern societies, opinions on what constitutes the good life will vary, and discussions will arise whenever important values are at stake, for example, on environmental issues (Eriksen and Weigård, 2003, 164). Ethical-political questions can be settled at lower levels when solutions only affect certain groups or a limited territory. Local issues are usually more effectively addressed on smaller and more homogenous arenas like neighbourhood councils and user-directed units.

However, local ethical discourses may generate conflicts that affect the whole municipality. In case of disagreement on goals and values, a moral discourse is needed in order to solve such conflicts neutrally and impartially. This is the main area of discourse theory and the requirement of universality constitutes the standard of validity (Habermas, 1996; Eriksen and Weigård, 2003, 165). The moral dimension requires decision-making procedures that force the actors to look beyond their own narrow self-interest. It also implies the existence of arenas where it is possible, through discussions, to reconstruct, put aside or synthesize different traditions and local opinions of what to do (Dryzek, 1990, 19). Moral questions demand political processes and institutions where all the affected parties are included. A political debate on issues of right and wrong, good or bad, implies a set of arguments that refer to a common set of social norms and cultural values that can exceed particular horizons. The moral question is about the justness of norms. For this purpose, you need arenas where the quality of the arguments can be put to the test. This can only be achieved by adopting procedures that allow for genuine deliberation based on norms of inclusiveness, symmetrical interaction, personal engagement, shared information and innovation (Dryzek, 1990; Kettner, 1993) and where participants are forced to argue their case. In such a setting, the deliberation itself serves to legitimate the output (Miller, 1993, 75).

However, as pointed out at the beginning of this article, such procedures are both 'labour' intensive and difficult to implement in their ideal form.⁶ Even if consensus is more of an ideal rather than a reality, moral questions will nevertheless stand a better chance in a context that gives priority to careful reasoning and rational discourse rather than to voting, bargaining and fiat. The ultimate aim of the moral discourse is not to make decisions that are tolerable. The logic of moral discourses is to reach fair and legitimate solutions, but they also have an ambition to answer practical questions in an optimal way.

Politics in Real Life

Real-life politics is not a linear process that starts with pragmatic discourses, proceeds with ethical issues and ends up in a moral discourse. According to Habermas (1996) discourse theory has been expanded to include three basic forms of practical reason, which relates to the different types of questions that will arise in deliberative processes. It is, however, an empirical question how and to what extent those forms of reasoning will occur in the daily political debate, but those three types of practical reasons enable us to design analytical and normative models for empirical investigations.

Modern democracies are not primarily designed for self-realization and personal development but are rather steering systems for coordination and aggregation of preferences. The modern state is a system for regulating, distributing and redistributing resources (Lowi, 1969). The governmental system copes with conflicts without necessarily being able to solve them. In this way the system is invented to handle conflicts effectively. Discussions can be ended without any agreement obtained. The formal procedures do not only relieve the system, they also stimulate compromises between the participants.

Bargaining is an obvious alternative to moral discourse in democratic decision making where the prospects of reaching (a final) agreement are poor. Habermas talks about procedurally regulated bargaining governed by resources and the need for viable compromises rather than arguments (Habermas, 1996, 168): 'Whereas a rationally motivated consensus (....) rests on reasons that convince all the parties in the same way, a compromise can be accepted by the different parties each for its own different reasons' (p. 166).

The concept of bargaining is derived from the pragmatic discourse while the conception of social cooperation as strategic interaction has been developed within rational choice theory and game theory (Eriksen and Weigård, 2003, 176). Here, the political debate is not that important for the outcome; the objective is to get support for already established opinions or fixed preferences. The political and administrative system decides what the relevant arguments are and political actions can be analysed as role performance.

Routine politics often appear as pure system activity where the resources available to participants are more important for the content of decisions than the quality of the arguments. The political debate is about utility calculation and collective priorities with reference to economic interests or procedurally regulated bargaining where the primary aim is to be part of the winning team. Rhetoric and political symbols are important tools in this type of political game. Even if this gives a credible picture of the political debate, is it not entirely adequate. The political debate cannot be understood in its entirety if we only use this system perspective and the discourses associated with this concept of politics. The fact is that weak and poor groups have been given priority in budgetary discussions despite their weak position in the political game: Would members of a parliamentary committee or a caucus engage again and again in discussions on normatively loaded issues (such as stem cell research)unless they presupposed that they can win over people by better arguments? (Habermas, 2005, 385).

Health care, environment and nursing are issues that have been given priority in spite of not being represented by strong groups.

Politics as Deliberation

Politics is also a process where common interests are defined and where there is mobilization for collective action and support for solutions that benefit no one in particular. Such cases are not captured by the strategic model as it tends to exclude the possibility of mutual commitments and a strong collective identity — and the fact that people are sometimes willing to act with reference to the benefit of the common good. In order to conceptualize this part of the political reasoning in practice, I suggest with reference to the ethical-political discourses, a local community model.

In order to conduct an adequate analysis of political reasoning in a local context, you must be able to classify statements with reference to local identity as something else than bargaining or irrationality. When and if the local politicians for instance use the budget meetings to talk about the latest news in the village or talk in the capacity of being neighbour, relative or friend this rationality can only be understood in light of the local context and belonging. The actors are participating, not in order to maximize their preferences or utility, but rather in order to clarify collective identity and self-understanding.

Providing rational answers to ethical and moral questions requires democratic institutions that serve a purpose beyond the mere aggregation of preferences. Democracy in this sense is about arguing one's case with reference to political norms rather than relying on resources and strength when decisions are made.

Analysing the Local Democratic Debate

In this paper, we have promoted a concept of political rationality based on both system- and action-oriented premises. These two perspectives are opposite poles on a continuum; a concept of political decision making as system activity will have efficiency as a normative standard. The action perspective, however, underlines the force of the better argument and fairness as the normative standard. A political discussion without an element of communicative rationality will result in loss of legitimacy. This balance or trade-off between legitimacy and efficiency is a critical challenge to any political system (Jenssen, 1995).

The pragmatic model sees politics as activities of technical and economic reasoning related to collective priorities. In the local setting this will be the case where discourses are defined by state regulations and budgetary considerations. The politicians' main contribution is to approve — or reject — the budget proposed by the administration. The political discussion on the budget is merely décor — a symbolic exercise of little significance. When the political debate is reduced to only pragmatic questions, there is less space for both deliberation and bargaining. The administrative logic and competence are the key factors and therefore we can call it, in the study of local politics, the administrative model.

The conflict model understands politics as bargaining; conflicts, strategies and political constellations are the key dimensions of politics. The political parties are central to this concept of (local) politics, and participants often refer to the different political ideologies represented around the table. In contrast to the first model, goals and preferences are not fixed, but matters of discussion. Decision making is about choosing between (often competing) courses of action — usually by casting a vote.

The conflict model implies that participants act strategically in order to be part of the winning coalition. In contrast to the administrative model, this concept of political reasoning increases the number of options and proposals considered — enhancing the significance of the political debate. The actors will argue with reference to certain ideas of what is fair and on the basis of ideological convictions. Participants will appear as role players, acting on behalf of a party, sector or territory.

Procedurally regulated bargaining is governed by resources rather than arguments (Eriksen and Weigård, 2003, 167) and politicians are not expected to change their opinion during the discussion. This discourse relates to situations where compromises between interests that cannot be universalized are dominant and this is the basis for the Conflict model.

The basic premise of the third model, however, is that politics is chiefly about clarifying collective identities where the purpose of political discussion is to define the interests and values that the participants have in common. The frame of reference is far more complex than in the first two discourses. In this context we name it the Community model of politics because the ways in which participants argue will reflect their close relationship with other citizens in the community. A local councillor is not only representing a political party, he will also define himself as a neighbour, a relative or fellow villager — and sometimes act accordingly. The community model predicts that local norms and values will decide the appropriate course of action. Hence, political decisions must be interpreted in light of the social and cultural setting of which participants are a part. The community discourse has a frame of reference that relates to non-systemic aspects of politics and therefore the reasoning is more unpredictable.⁷ The political debate itself and the deliberation based on local values will have an influence on the outcome. Ethical norms reflect the significance of neighbourhood, loyalty and solidarity within the local community that again limits the scope (and legitimacy) of self-interested action. A strong sense of belonging and mutual identification makes it possible to balance competing interests and concerns without resorting to negotiations and compromises (Offe, 1984). Deliberation serves to confirm local identities (Walzer, 1983; Taylor, 1989; Benhabib, 1992).

In order to conduct an empirical study of community-discourses, we have to use the method of observation. As a researcher, you have to observe the political process as closely as possible. The most important data will be the statements of elected politicians and their behaviour at political meetings. You will, of course, have to collect additional information about social, geographic and cultural aspects of the municipality. This is important because it enables you to make better sense of statements that would be defined as unimportant and irrelevant — perhaps even irrational — within the first two discourses (Table 1).

Table 1 - Model characteristics.

Full table

Models are of course, highly stylized characterizations of a more complex political reality. In real life, local democratic institutions will not be of one (pure) type or another, but display elements of all three models in the table. One of the discourses may, however, dominate — with implications for the quality of local political debates. Any distinctive combination of discourses can be named a political unit's working method (Jenssen, 1996).

Politics as Deliberation or Moral Discourse

Political decision making raises normative questions about what to do, or as Habermas puts it, questions about practical reasoning (1996). The moral discourse relates to whether conflicts of interests and values can be decided according to universal standards (Habermas, 1996, 168; Eriksen and Weigård, 2003, 167). The purpose of this aspect of politics is not to call attention to what is right to do or what the correct goals are. It is rather to underline the formal or procedural conditions under that fair decisions can be made.

The normative standard of the discourse-theoretical perspective pertains to the validity of an argument. In this way, the moral discourse sets the limits for the other discourses and comes into force when the political discussion degenerates, for instance when pragmatic aspects have absolute priority or when strong groups use their power to promote a project that will only serve their own interests. In a political debate preferences and actions will have to be justified in relation to the common interests. Paradoxically, the relevance of this discourse can most easily be seen in situations where participants — when arguing their case — do not commit themselves to common goals and values. This aspect of justice and fairness has to be addressed in political discussions in order to get support and reach legitimate decisions. When this normative standard is activated, the problem under discussion will be framed in terms of the common interest, and seen as independent of local values, different constellations of interests and other pragmatic factors.

Good Deliberative Practice

Good deliberative practice in governmental institutions will have to take place within an institutional framework where political games and pragmatic considerations tend to dominate. We can still develop an alternative model of how group deliberation might proceed in a group not unified by a common purpose (Goodin, 2005, 187). According to Goodin, we have to consider both different steps and different deliberative expectations (p. 188).⁸

In Goodin's table, there are three deliberative expectations to a parliamentary debate. First, the parliamentary chamber is a place to 'put one's case' (p. 191). Each party is expected to set out the most coherent and persuasive justification they can, for their own preferences. The justification of assertions and validity claims are the primary deliberative virtue on display in parliamentary debates. Further, Goodin sees a tight connection between premises and conclusions and a coherent justification as an important premise for a good deliberative practice. And finally, participants should respect — and seriously consider — arguments raised by opponents, but nevertheless attempt to refute them.

Goodin's scheme for deliberative expectations to representative democracy (p. 190) includes neither considerations of the common good nor expectations of a rationally motivated consensus. According to Goodin, a minimum requirement of political deliberation is that it produces viable compromises: 'But in parliamentary debate, no one seriously expects to change any other MP's mind' (Goodin, 2005, 191).

If deliberation is about politics infused with reason and arguments (Bächtiger and Steiner, 2005, 153) and political actors prepared to be persuaded by the 'non-coercive coercion of the better argument' (Habermas, 1984, 1996), then Goodin's deliberative expectations have more in common with rational choice and the Schumpeterian concept of democracy. I find that the most basic criteria for labelling parts of a political discussion as deliberative, is that there is some consideration of the common good and that the actors show willingness or interest in reaching a consensus or a compromise acceptable to all.

Empirically, there are also problems with Goodin's definition of deliberation in governmental debates: How can unanimous decisions in parliamentary institutions, composed of different political parties, be explained? According to Habermas (2005, 386), the conception of 'rational discourse for explaining an epistemic notion of deliberative politics should be stronger. The main purpose of applying a deliberative concept of politics in empirical studies is to address the "problem of legitimation" that pluralistic systems represent'. The democratic process has to decide when there is no religious or other substantive world view that is dominating the decision-making process (Habermas, 2005, 386). Therefore, when we look for deliberative standards or elements of moral discourses in parliamentary debates, we look for 'the presuppositions of inclusiveness, equal communicative rights, sincerity and freedom of repression and manipulation (....)' (Habermas, 2005, 385). According to Habermas, these elements are part of the intuitive knowledge of how to argue. When those elements are lacking, the quality of the debate is reduced. These presuppositions represent an integrative part of deliberative reasoning. Empirical studies in light of discourse theory have a reconstructive character and empirical research on local political debates will tell us to what extent these presuppositions are fulfilled in practice. Any democratic regime will deteriorate if these elements are absent (Habermas, 2005, 385). In practical politics, however, deliberative standards can also be settled as some kind of meta-consensus to regulate pluralism and allows for diversity in values, beliefs and preferences (Dryzek and Niemeyer, 2006, 647) Normative meta-consensus involves recognition across difference, and as such facilitates cooperative search for mutually acceptable solutions to joint problems while respecting value differences that can remain deep and irreconcilable, .......

This means that considerations of the common good are not necessarily clearly expressed in the political debate, but appear as meta-consensus.

Social Context

Social context refers to organizational, institutional and normative matters that can have an impact on the political debate (Ulbert and Risse, 2005).⁹ By conceptualizing the political discussion as a social process, it is possible to perceive of debates, not only as an exchange of resources but also as a communicative process (Ulbert and Risse, 2005, 354). Under what conditions should deliberation be expected, and how does this harmonize with local democracy?

Organizational matters

The community model assumes that local politics can be understood as an ethical discourse. The implication is that small political systems can do without political parties and traditional party-oriented bargaining. Dahl and Tufte (1973) have suggested that size matters, and that small systems facilitate the resolution of political conflicts by informal means. The larger the unit, the greater the extent to which political conflicts are articulated and resolved through formal and impersonal organizations, that is, parties. Furthermore, according to Dahl and Tufte, small units tend to be homogeneous to an extent that decreases the need for party formations (pp. 91–97). At least, in spite of the existence of different political parties, the political debate will have a distinctive feature of consensus despite the political parties. In other words, we will expect that small and homogenous democracies, like local communities, will favour reason-giving and arguing processes.

Whether meetings are open (public) or not may also influence the nature of decision making and thus the conditions for deliberation. Open meetings can have a 'civilizing effect' on the participants because they have to defend their choices in front of an audience (Eriksen and Weigård, 2003, 218). The audience constitutes an 'external authority' to which the participants may feel obliged to defer (Ulbert and Risse, 2005, 357). In any case, the actors are forced to defend their interests with reference to the common good (Elster, 1992; Jenssen, 1996). On the other hand, the presence of an audience — or representatives of the media — may tempt participants to act opportunistically, play their hand strategically or engage in ritualistic rhetoric (Ulbert and Risse, 2005, 357). In this way publicity may also be dysfunctional for a serious discussion.

As for making decisions behind close doors, this is sometimes necessary to bring about a rational and unconstrained exchange of views. This is of course a paradox, because we can only know if decisions are 'correct' if they are tested in an open, public debate. 'Private reason fails a democratic accountability test' (Chambers, 2004, 389). Closing the doors can improve the context so that debates can be carried out without the interference of pragmatic or strategic considerations. Do closed meetings in local politics favour deliberation?

So far I have discussed organizational mechanisms that can influence the conditions for deliberation in local politics. Within these organizational structures there are also institutional settings that will have an impact on the political debate.

Institutional context

As mentioned earlier, how and to what extent the different discourses will appear in real life will vary. In order to explain differences in working methods (Jenssen, 1996), we also have to focus on the institutional context. According to Selznick (1957) organizations are institutionalized when they become infused with values and norms that — among other things — define what constitute appropriate procedures and agendas. Such institutionalized settings reflect a 'logic of appropriateness' (March and Olsen, 1998) and define the type of arguments and justifications that are considered acceptable and relevant (Ulbert and Risse, 2005, 354). In order to do empirical studies of such settings, a good start could be the scheme of Ulbert and Risse linking institutional settings and arguing strategies (pp. 355) (Table 2).

Table 2 - Institutional settings and arguing strategies.

Full table

In order to understand local political reasoning, it is also necessary to adjust to those standards or lack of standards. In addition to the local working methods or, may be more correct, a potential important part of those standards' are the presence of knowledge brokers or normative entrepreneurs. This will be the case where actors try to 'actively formulate a consensual problem definition backed by existing norms' (Ulbert and Risse, 2005, 354).¹⁰ This also means that the conditions for deliberation must be related to the participants in political institutions and their credibility in the meeting.

The Political Debate in Local Political Institutions

I have tried to apply this analytical scheme in empirical studies of political debates in two north-Norwegian municipalities, Sortland and Hadsel. This study took place in two phases within a span of 20 years. The two municipalities are adjacent, relatively small and have the same economic basis. The number of inhabitants is roughly the same and has remained fairly constant with 8,000–9,000 in the first observation period (1985, 1986) and 9,000–10,000 in phase 2 (2005, 2006). What follows is based on observations of the annual budget process in our two municipalities (Jenssen, 1996, 2007a, 2007b).

The annual budget process is an important part of municipal politics because it concerns every activity that involves financial allocations and funding. This implies that all parts of the public sector and all activities within the different areas are or can be attended to. Furthermore, this is an annual occurrence and the most significant opportunity for wielding political influence as well as an important source of information. The political discussions take place every autumn and the final decisions about the budget are taken in November and December, first in the executive committee and then in the elected municipal council.

What is the most important question for the politicians to discuss in connection with the annual budgetary process? Is there a different approach to budgetary decision making today compared to, say, 20 years ago? In order to address this question, we have observed the budgetary process in these two municipalities in two turns; the first one in 1986 and 1987, the second in 2005 and 2006. The observation data include all contributions to the debates from the participants in the council and executive meetings. These reports are verbatim representations of all the contributions in those meetings and therefore represent a unique database for a study of political reasoning and the approach of elected officials to the budgetary process. The next paragraph presents the content of those budgetary meetings and debates in light of the discourses introduced above.

Distinctive working methods

In both municipalities the budget proceedings are characterized by a distinctive working method. Common to both Sortland and Hadsel is the unanimous approval — in both periods studied — of the budget proposals of the chief executives. But this is also the only common feature for those two municipalities.

The budget process in Sortland is completed within a relatively short time span, and accordingly there is no need to neither inquire nor discuss. Both periods are characterized by extensive cooperation among the politicians, and also between the politicians and the administration. There is a common understanding of the challenges facing the municipality, and how these should be met. The budget deliberations in Sortland show politicians perceiving the administration's budget proposals as good working documents, in the sense that they enable them to promptly make decisions from practical estimates.

While the parties in Sortland have distinct political profiles, these were activated mainly in the council meetings. The party-related rhetoric, however, addresses mainly national issues, and one rarely sees parties disagreeing on the actual budget matters.

The budget discussions in Sortland are formal and rarely encourage representatives to relate to each other as neighbours or fellow villagers. On the other hand, the extensive consensus among the representatives is an indication of just such a close relationship across party lines or geographical divides.

The pragmatic discourse is evidently present, given that the representatives' main concern is that all political decisions be rooted in assessments of their economic consequences. The administration and the representatives have a common and clear understanding that this is of fundamental importance when choices are made. The main impression is that the politicians have a common understanding of what constitutes the right thing to do to improve the welfare of the inhabitants in the municipality, regardless of the political divides between the parties.

This combination of both a strongly pragmatic approach and a normative meta-consensus to the budgetary process, which the politicians applaud, has been there for at least 20 years: 'This is a culture that has a life of its own' (Interview, The leader of the Conservative party in Jenssen, 2007a)

In the municipality of Hadsel, however, the budget discussions are of a quite a different nature. They tend to go on for a long time and, while being fairly pragmatic in tone, the content of the pragmatic discourse is different to that of Sortland. There are also differences between the observed phases. In the first phase of the observations, there were both pragmatic and ethical discourses on display. The economic situation was difficult and the debates were mainly about how to balance the budget. The politicians, however, did not put forward any proposals to meet this challenge.

Phase one observations indicate the existence of both ideological, territorial and sectoral conflicts, but it is the territorial dimension that dominates. The municipality consists of two competing communities and this conflict is the object of part of the debate. However, the dissatisfaction with public service is strongest among the politicians representing the outskirts of the community — especially in connection with proposals for restructuring the local school system. A discourse analysis based on the assumption of strategic rationality cannot, however, account for the cross-party consensus. This requires a conceptualization of politics as an ethical discourse. The most important issue for the politicians, in this first phase, is verbally to underline their common identity within the municipality and the will to stick together. This solidarity, which is a significant element of the local political culture, transcends party-based cleavages. 'We have had good life in our community'. 'It's a good place to live' were remarks often repeated in the meetings I attended (Jenssen, 1996, 118).

In phase two (2005, 2006) both the conflict and the ethical discourses are absent. The financial situation in Hadsel is very bad and the administration is unable to come up with suggestions for how to solve the difficult situation. The budget meetings are characterized by irresolution. The proceedings go on for hours, but nothing is changed in the budget proposals. There is no real discussion, but the politicians put forward a number of incoherent questions that people from the administration are unable to answer. 'Why did we end up in this situation?' 'What more can we do?' 'What are we doing wrong, other municipalities have just as much loans as we have?' Although there is no real debate, these meetings qualify as pragmatic discourses. However, while the actors in Sortland are concentrating on how to comply with the long-term plan, the decision makers in Hadsel are looking backwards — wondering why this happened to them (Table 3).

Table 3 - Discourses in Sortland and Hadsel.

Full table

Why are the budget proceedings so different in these two municipalities? Let us turn to aspects of the institutional settings in order to explain this (Table 4).

Table 4 - Institutional settings and arguing strategies in Sortland and Hadsel.

Full table

In Sortland the standards for appropriate behaviour are clear and have been in place for at least 20 years. These are guidelines that regulate both the content of the debates and set the rules for when and how the council members should participate during budget proceedings. This is a working method that is tantamount to a pragmatic discourse — but also built on a meta-consensus on goals and identities that transcend party cleavages. In Hadsel, debates are less constrained by formal rules and procedures and therefore allow for longer, more spontaneous and diverse discourses. The drawback, of course, is a less well-structured, almost irrational, process where the politicians do not really come to grips with the budget, especially in phase two. There is neither meta-consensus nor an institutionalized practice that can guide the political discussions.

Conclusions

In this article, I have sought to demonstrate the relevance of a deliberative concept of democracy to the study of local politics and government. The case for a deliberative approach can only be made by 'confronting' it with alternative approaches. The conclusion to this exercise is therefore that we need more than one approach in order to explain what constitutes and legitimates local politics. My main hypothesis, supported by the study reported here, is that we will find different institutionalized working methods where the structure and contents of political debates will vary from one municipality to the next. Students of local politics and government should examine the nature of such variations and ask how they can be explained.

I have also focused on certain aspects of the social and institutional context of political decision making in order to identify the conditions under which one would expect the deliberative aspects of politics to be activated. There is no set answer to how local democracy could improve its capacity for deliberation. However, by recognizing the significance of social and institutional context, it should be possible to improve the conditions for deliberation in local government.

Notes

¹ I thank Knut Mikalsen, Jarle Weigård, Erik Oddvar Eriksen and Hans Kristian Hernes for helpful comments and criticism.

² Bächtiger and Steiner are editors of two issues of Acta Politica where the possibilities for empirical studies in the light of a deliberative perspective are discussed (Acta Politica, 2005, 40).

³ See for instance Benhabib (1994), Knight and Johnson (1994) and Habermas (2005).

⁴ March and Olsen (1989) make a distinction between an aggregating and an integrative concept of democracy.

⁵ The legal discourses, which pertain to the consistency of judicial norms and rules, will not be dealt with here.

⁶ 'Too much democracy can kill off democracy'(Keane, 1989, xii).

⁷ Cf. the two first discourses where the arguing will be derived from characteristics of the governmental institutions.

⁸ Goodin (2005, 189) , influenced by Birch (1964), presents four stages of representative democracy; the 'caucus room', the parliamentary debate, election campaign and post-election arguing and bargaining, and discuss deliberative expectations to each of those stages. In this paper, I will only address the evaluation standards related to the parliamentary debate.

⁹ As Ulbert and Risse point out, there is always a material context embedding a political discussion and those have been given the most attention (2005, 353). Studies also on local politics have overwhelmingly focused on the material aspects in order to explain given priorities.

¹⁰ In their case studies they found three distinct settings, in which norm entrepreneurs were using distinct strategies of arguing. Although those studies are about international politics, we can expect much the same thing at the local level. As the authors underline, the difference individuals can make depends on the credibility and truthfulness of such speakers. There is no success for the speaker unless his arguments are perceived by the audience (2005, p. 359).

References

1. Bächtiger, A. and Steiner, J. (2005) 'Introduction', Acta Politica 40: 153–168. | Article |
2. Benhabib, S. (1992) Situating the Self. Gender, Community and Postmodernism in Contemporary Ethics, New York: Routledge.
3. Benhabib, S. (1994) 'Deliberative rationality and models of democrativ legitimacy', Constellations 1(1): 26–52. | Article |
4. Benhabib, S. (1996) 'Toward a Deliberative Model of Democratic Legitimacy', in S. Benhabib (ed.) Democracy and Difference: Contesting the Boundaries of the Political, Princeton: Princeton University Press, pp. 67–94.
5. Birch, A.H. (1964) Representative and Responsible Government, London: Allen and Unwin.
6. Chambers, S. (2004) 'Behind closed doors: Publicity, secrecy, and the quality of deliberation', The Journal of Political Philosophy 12(4): 389–410. | Article |
7. Dahl, R.A. and Tufte, E. (1973) Size and Democracy, Stanford: Stanford University Oress.
8. Dryzek, J.S. (1990) Discursive Democracy, Cambridge: Cambridge University Press.
9. Dryzek, J.S. and Niemeyer, S. (2006) 'Reconciling pluralism and consensus as political ideals', American Journal of Political Science 50(3): 634–649. | Article |
10. Elster, J. (1992) 'Arguing and Bargaining in the Federeal Convention and the Assemblée Constituante', in R. Malnes and A. Underdal (eds.) Rationality and Institutions, Oslo: Universitetsforlaget.
11. Eriksen, E.O. and Weigård, J. (2003) Understanding HAbermas. Communicative Action and Deliberative Democracy, London, New York: Continuum.
12. Fraser, N. (1992) 'Rethinking the public sphere: A contribution to the critique of actually existing democracy', in C. Calhoun (ed.) Habermas and the Public Sphere, Cambridge: Cambridge University Press.
13. Goodin, R.E. (2005) 'Sequencing deliberative moments', Acta Politica 40: 182–196. | Article |
14. Habermas, J. (1984) The Theory of Communicative Action,Vol. 1, Boston: Beacon Press.
15. Habermas, J. (1989) 'Towards a communication-concept of rational collective will-formation. A thought-experiment', Ratio Juris 2: 144–154. | Article |
16. Habermas, J. (1996) Between Facts and Norms, Cambridge, Mass: MIT Press.
17. Habermas, J. (2005) 'Concluding comments on empirical approaches to deliberative approaches to deliberative politics', Acta Politica 40: 384–392. | Article |
18. Jenssen, S. (1995) 'Hierarki, spill eller deliberasjon?', in E.O. Eriksen (ed.) Deliberativ politikk. Demokrati i teori og praksis, Oslo: Tano.
19. Jenssen, S. (1996) 'Rasjonalitetsdimensjoner i kommunalpolitikken', Teses Dr. Philos, UiTø Tromsø.
20. Jenssen, S. (2007a) 'The Rationality of Local Politics', in S. Jenssen, A. Røiseland and N. Aarsæther (eds.) Practicing Local Governance: Northern Perspectives, New York: Nova Science Publishers.
21. Jenssen, S. (2007b) 'Den politiske debatten i kommunalpolitiske institusjoner', (work in progress).
22. Keane, J. (1989) 'Introduction: Demcracy and the Decline of the Left', in N. Bobbio (ed.) Democracy and Dictationship, Oxford: Polity Press.
23. Kettner, M. (1993) 'Scientific Knowledge, Discourse Ethic, and Consensus Formation in the Public Sphere', in E.R. Winkler and J.R. Coombs (eds.) Appliedthics: A Reader, Oxford: Blackwell.
24. Knight, J. and Johnson, J. (1994) 'Aggregation and deliberation: On the possibility of democratic legitimacy', Political Theory 22(22): 277–296. | Article |
25. Lowi, T. (1969) The End of Liberalism. Ideology, Policy, and the Crisis of Public Authority, New York: W.W. Norton.
26. Manin, B. (1987) 'On legitimacy and political deliberation', Political Theory 15(3): 338–368. | Article |
27. March, J.G. and Olsen, J.P. (1989) Rediscovering Institutions, New York: John Wiley & Sons.
28. March, J.G. and Olsen, J.P. (1998) 'The institutional dynamics of international political orders', In International Organization 52(4): 943–969. | Article |
29. Miller, D. (1993) 'Deliberative Democracy and Social Choice', in D. Held (ed.) Prospects for Democracy, Cambridge: Polity Press University of Notre Dame Press.
30. Mouffe, C. (1999) 'Deliberative democracy or argonistic pluralism?' Social Research 3: 745–758.
31. Offe, C. (1984) The Contradictions of the Welfare State, London: Hutchinson.
32. Selznick, P. (1957) Leadership in Administration, Berkeley: University of California Press.
33. Steiner, J., Bächtiger, A., Spörndli, M. and Steenbergen, M.R. (2004) Deliberative Politics in Action. Analysing Parliamentary Discourse, Cambridge: University Press.
34. Taylor, C. (1989) Sources of the Self, Cambridge: Cambridge University Press.
35. Ulbert, C. and Risse, T. (2005) 'Deliberately changing the discourse: what does make arguing effective?' Acta Politica 40: 351–367. | Article |
36. Walzer, M. (1983) Speheres of Justice, Oxford: Basil Blackwell.
37. Young, I.M. (2003) 'Activist Challenges to Deliberative Democracy', in J.S. Fishkin and P. Laslett (eds.) Debating Deliberative Democracy, Malden, MA: Blackwell, pp. 102–120.

About the author

Synnøve Jenssen (Dr Philos) is Associate Professor in the Department of Political Science, University of Tromsø where she teaches political theory and public policy. She has done research on prefectoral systems and central–local government relations in Scandinavia and is currently engaged in research on local democratic governance.