Abstract
In the quest for a workable ideal of democracy, the systems approach has recently shifted its perspective on deliberative democratic theory. Instead of enquiring how institutionalized decision-making might mirror an ‘ideal deliberative procedure’, it asks how democracy might be construed as a ‘deliberative system’. This leads it to recommend de-emphasizing the role of parliament and focusing instead on non-institutionalized actors and communications. Though this increased emphasis is undoubtedly warranted, the importance of parliament must not be downplayed. In the debate about transnational democracy – and about EU democratization in particular – it is widely held that making democracy safe for transnational politics entails finding new, non-parliamentary ways of organizing it. Here, the deliberative systems perspective may be misunderstood as offering an alternative, non-parliamentary route to transnational democracy. This article argues that a deliberative system without a parliamentary legislature is tantamount to deliberation without democracy and that an elected parliamentary legislature is constitutive of democracy as a deliberative system – national or transnational. To substantiate this claim, the article suggests looking at Habermas’ discourse theory in a new light, as a sociological-reconstructive approach that aims to explicate the cognitive dimension of modern democratic decision-making. Acknowledgement of discourse theory’s sociological intent enhances our understanding of democracy as a deliberative system in two important ways. First, it helps elucidate the epistemic meaning of democracy differently from the accounts given in alternative theories. Discourse theory holds that what generates democratic legitimacy in modern democracies is a particular epistemic quality which the democratic process lends to decisions by enabling citizens to view the latter as valid outcomes of a cooperative practice of collective problem-solving among equals. From this pragmatist understanding it follows, second, that the presence of a parliamentary legislature is crucial to the establishment and operation of democracy as a deliberative system.
With its ‘coming of age’, deliberative democratic theory evolved from a ‘theoretical statement’ into a ‘working theory’ that subjected its own hypotheses to empirical testing. 1 Now it is poised to enter a further stage of reflection. A number of scholars have recently proposed adopting a systems approach to deliberative democracy and looking at it from a macro-perspective. 2 They argue that the study of deliberation in current democracies, and the attempts to improve deliberative processes, have so far focused on individual (mostly legislative) sites and that what we should be doing instead is to consider the legitimacy-enhancing role of deliberation from a systems perspective. Viewed from a systems standpoint, no individual site, however deliberative, is of itself sufficient to legitimize political decision-making. What lends legitimacy to policy-making in the deliberative system of democracy is, rather, the interplay and division of labour between political institutions, formal and informal sites of deliberation, and deliberative and non-deliberative communication, both in the wider public sphere and in everyday talk.
In thinking about ways to democratize democracies from a deliberative-systems standpoint, the systems perspective assumes that deliberative democratic theory is about how to apply an ‘ideal deliberative procedure’ 3 to real-world democracies in order to lend them maximum legitimacy. I term this the ‘normative-prescriptive approach’ to deliberative democracy. The systems perspective here brings along with it a revision of some of the central assumptions ascribed to deliberative democratic theory. In a deliberative system, for example, it is not necessary for all communication to be deliberative, or for all decisions to be reached by rational consensus, in order for democratic legitimacy to obtain: it is sufficient that the constituent elements work together as a deliberative system. Furthermore, the systems approach de-emphasizes parliamentary deliberation and focuses instead on non-institutionalized actors and communications. While a reconsideration of the ways in which deliberative democratic theory addresses the phenomena of consent, conflict and communication within a democracy is undoubtedly long overdue, the important role that a parliamentary legislature plays for democracy as a deliberative system should not be downplayed. Regarding the conditions of transnational democracy, for example, it is widely held that making democracy safe for transnational politics entails finding new, non-parliamentary ways of organizing democracy. Thus, in the debate about the European Union – to quote a specific example – the deliberative-systems perspective may be misunderstood as offering an alternative, non-parliamentary means of securing a multi-level EU democracy. But a deliberative system without a parliamentary legislature is tantamount to deliberation without democracy. In other words, as I will argue here, a parliamentary legislature is an enabling condition of democracy as a deliberative system – in transnational politics as in other spheres.
To substantiate this claim, I suggest pushing the systems perspective one step further and applying it to deliberative democracy not only in normative-prescriptive but also in sociological-reconstructive mode. I do not consider these two strands mutually exclusive; they simply differ in scope and focus because they pursue different epistemological aims. However, there is often little appreciation of the fact that the Habermasian discourse theory of democracy is a well-elaborated example of the sociological-reconstructive approach. What this theory aims to do is to explicate the cognitive dimension of the decision-making processes in modern democracies. In other words, it seeks to understand the inner logic and rationale of democracy as a deliberative system.
I argue that re-evaluation of the sociological-reconstructive import of Habermasian discourse theory enhances our understanding of democracy as a deliberative system and does so in two important ways. First, it helps both to elucidate the epistemic meaning of the concept of democratic legitimacy and to highlight the aspects in which this account differs from alternative theories of epistemic democracy. A central premise of the discourse theory of democracy is that what generates democratic legitimacy in modern democracies is the interplay of different (institutional and non-institutional) elements in the decision-making process (however defective this may be) and that this happens because the interplay lends democratic decisions a particular epistemic quality – namely, it enables citizens to view democratic decisions as valid outcomes of a cooperative practice of collective problem-solving among equals. From this pragmatist understanding of democracy’s epistemic meaning, it follows, second, that the presence of a parliament as the highest legislative authority is crucial to the establishment and operation of democracy as a deliberative system.
In what follows here, I develop this argument in three stages. I begin by outlining the debate about deliberative democracy, pointing up two particular strands of it, which I dub ‘normative-prescriptive’ and ‘sociological-reconstructive’ (I). I then suggest revisiting the concept of ideal discourse. Drawing on my observations here, I consider the sociological-reconstructive interpretation of the epistemic dimension of democratic practice offered by discourse theory and the features that distinguish this interpretation from alternative epistemic approaches (II). Finally, I set out the reasons why an elected parliamentary legislature is crucial in generating the epistemic meaning, and thus the legitimacy, of democracy as a deliberative system (III).
I Two strands of deliberative democratic theory
1 The early critique of liberal democracy
The debate on deliberative democracy began in the early 1980s with a ‘theoretical statement’ 4 targeting the then dominant school of thought: liberal democratic theory. 5 Liberal theory views the legitimacy of democracy as deriving from the latter’s instrumental value in realizing equal individual autonomy. The liberal concept of political legitimacy assumes individual will as a basic normative category, and the unimpeded pursuit of happiness as the epitome of the good life. The two pillars on which it rests are thus: the consent of all concerned as the basis of legitimate decisions, and the equal influence of individual interests in political decision-making – in other words, fairness. This is the tenor of the highly influential economic theory of democracy, 6 which saw the democratic process as analogous to the workings of the market. 7 The interplay of general elections and majority decisions is viewed as a means of generating decisions effectively while also upholding the principle of fairness. And elections are seen, not as a collective public practice but as an aggregation of private electoral decisions based on private individual preferences. On this view, majority decisions, though violating the requirement for substantive consensus, are nonetheless legitimate within the framework of a democracy because there is consensus about the moral quality of the procedure – in other words, about the fact that the democratic process gives equal and undistorted influence to all individual interests.
The assumption that equal individual autonomy is crucial to democratic legitimacy is one also held by deliberative critics. However, such critics contend that the procedures involved in general elections and majority decisions are not of themselves sufficient to ensure the realization of that autonomy. They have legitimizing force, so the argument goes, only because they are embedded in processes of political deliberation. In the view of deliberative democrats, there are, broadly speaking, two respects in which the liberal account ignores the role of deliberation – one empirical, the other normative. Empirically, so they argue, it is inappropriate to describe the act of voting in a democracy as private: if democratic elections were merely a summation of private decisions, voters – like consumers – would each form their preferences privately and come to discrete political judgements before voting. But such a scenario does not tally with the reality of complex modern societies, in which individuals only ever have fragmentary information. It is public deliberation itself that is the chief channel through which ‘individuals acquire new perspectives not only with respect to possible solutions, but also with respect to their own preferences’. 8 Hence public deliberation – so runs the deliberative democrats’ first objection – is crucial to the very formation of individual political will.
Normatively speaking, meanwhile, deliberative democrats object that procedural fairness is not in itself enough to realize equal individual autonomy in a democracy. General elections and the aggregation of votes into majorities give effect to the principle of equal admission and fair treatment of all individual wills. However, individual autonomy in collective decision-making requires more than citizens’ acknowledgement of the moral quality of the procedure. Besides aspiring to procedural equality, say the deliberative democrats, democracy is committed to the goal of generating substantively just decisions, that is, outcomes that are in the equal interest of all. 9 While the mechanisms of election and aggregation, taken by themselves, justify the democratic process as procedurally fair, they offer citizens no reason to accept the outcomes in question as mutually justifiable decisions in substantive terms. 10 This predicament points up another area in which political deliberation comes into its own: without a surrounding public exchange of views and arguments about what policy to pursue and why, democratic decisions are simply expressions of the self-interest of the (arbitrarily shifting) majority. If, by contrast, democratic decision-making is viewed in terms of public deliberation, the aggregation of majorities is not just a pragmatic necessity; it also plays an important role in the search for mutually justifiable outcomes. It is a means of ‘determining where the burdens of proof lie in the deliberative process’. 11 Majority decisions are acceptable even from the minority’s point of view because they simply indicate the minority’s failure to provide convincing arguments on that particular occasion and the need to come up with better arguments in future. 12
To sum up: the deliberative critique of liberal democratic theory argues that the realization of equal individual autonomy depends on general elections and majority decisions being embedded in a context of public deliberation. The democratic process does not simply represent an aggregation of ready-made individual wills; it exerts a (partially) transformative effect by mediating between individual perspectives in a process of public deliberation about what common policy to pursue. Similarly, a democratic decision represents, not a majority aggregation of individual wills, but a publicly generated judgement to which the majority subscribes: ‘[A] legitimate decision does not represent the will of all, but is one that results from the deliberation of all. It is the process by which everyone’s will is formed that confers its legitimacy on the outcome, rather than the sum of already formed wills.’ 13
2 Normative-prescriptive and sociological-reconstructive deliberative democratic theory
In line with its two main criticisms of the liberal view, deliberative democratic theory has evolved two new strands. The first is concerned with the philosophical explication and justification of the normative ideal of deliberative democracy. The most influential contribution here is Joshua Cohen’s translation of the Habermasian concept of ideal discourse into an ideal model of deliberative democracy. 14 Cohen famously identifies 4 features of an ‘ideal deliberative procedure’ which a political process must display in order to constitute a perfect deliberative democracy: first, decisions are the result of free deliberation in which the participants are bound only by the results and preconditions of their deliberation; second, the deliberation is reasoned – in other words, it consists of an exchange of credible arguments for and against proposals for political action; third, the participants are equal; and fourth, the deliberation is aimed at producing a rational consensus. In line with this, Cohen argues that political decisions are democratically legitimate ‘if and only if they could be the object of free and reasoned agreement among equals’. 15
The second, somewhat distinct, strand seeks to demonstrate that modern representative democracies can in fact be viewed as deliberative. The point here is to show that the institutionalized decision-making process is configured in such a way that decisions are based on prior political deliberation. What needs to be done, in other words, is to reconstruct the cognitive potential inherent in the institutionalized order of the democratic process. Thus Joseph M. Bessette explains the institutional order underpinning US democracy in terms of the desire of the founding fathers to replace the tyranny of the spontaneous majority with ‘the rule of the deliberative majority’. 16 Drawing on empirical studies of US policy-making, Bessette argues that law-making is not driven by bargaining alone but passes through an effective system of political deliberation. 17 Although the system of inter-institutional checks and balances is an important means of fostering deliberation in US law-making, Bessette, and also Manin, 18 sees the principle of representative government as the chief motor of democratic deliberation. Together, the ambivalent character of democratic representation (in which representatives are supposed to champion the interests of their voters but have no imperative mandate), the holding of regular elections, and the operation of the majority principle trigger a deliberative dynamic. This in turn points to the formation of majorities through ‘trial by discussion’ 19 as the best way of making policy, given that in such a process individual views will neither be neglected nor written in stone. Not neglected, because representatives, keen on re-election, will take care to pay them heed; and not written in stone, because representatives are expected to depart somewhat from them and find solutions that are acceptable to (at least) the majority.
Although more comprehensive theories of deliberative democracy draw on both these strands, their main epistemological interest is determined either by the first or by the second. An example of a ‘first-strand’ theory is the account of deliberative democracy given by Amy Gutmann and Dennis Thompson. 20 Gutmann and Thompson ask how the ideal of deliberative democracy could and should work in actual societal conditions. As well as discussing how the principles inherent in the ideal of deliberative democracy relate to other fundamental principles and values, they consider the ways in which democratic deliberation can help overcome deep moral disagreement. The motive of approaches such as this is ultimately to substantiate deliberative democracy’s status as a normative yardstick with which we can identify ways of further democratizing existing representative democracies.
The most influential example of the ‘second-strand’ approach to deliberative democracy, I will argue, is Habermas’ discourse theory. This sets out to reconstruct those epistemic features of the modern democratic process that generate the legitimacy of political decisions. It does not claim to provide a picture of an ideal deliberative democracy under actual societal conditions. What it does, rather, is develop two further sociological hypotheses. 21 The first of these asserts that the concept of ideal discourse represents not a philosophically constructed utopia but an attempt to delineate what might be called an ‘empirical utopia’. It refers to the ‘idealizations that orient subjects capable of speech and action’ 22 and that are an enabling condition of everyday language-mediated interactive problem-solving (coping). The second hypothesis asserts that the reason why democracy is able to bring about social integration in modern pluralistic societies is that the institutions of democracy organize decision-making along collective problem-solving lines that mirror crucial features of everyday language-based coping.
3 The narrowing of the discourse on deliberative democracy and the correctives proposed by the systems approach
The further discourse on deliberative democracy is marked by a narrowing of focus, in that proponents and critics alike direct their attention almost exclusively to the first theoretical strand. Both the empirical studies and the major critiques in this area presume that the core aim of deliberative democratic theory is to refine the ‘ideal deliberative procedure’ as a model of democratic politics. In this sense we may speak of normative-prescriptive approaches to deliberative democracy as standard accounts.
Three of the most influential critiques of deliberative democratic theory discuss it in these terms. The first objects that in modern democracy deliberation plays only a minor role compared with the power-based strategic action and bargaining that precedes majority decisions: ‘[D]eliberation theorists … wish away the vulgar fact that under democracy deliberations ends in voting’, which means that ‘it is the result of voting, not of discussion, that authorizes governments to govern, to compel’. 23 But this is to ignore the fact that, although majority decisions do not figure (and do not have to) in the explication of the ‘ideal deliberative procedure’, a crucial aim of the second theoretical strand is to understand the impact of public deliberation on the legitimation of majority decisions. In this case, therefore, the complaint that deliberative democratic theory ignores self-interest and power is unjustified. 24
A similar narrowing is observable in a second, pluralist critique of deliberative democracy that focuses on two features of the ‘ideal deliberative procedure’. The first is the idea that deliberation aims at rational consensus. Agonistic accounts of democracy, which stress the conflictual nature of the political, regard political consensus as an impossibility and view any exhortation to it as nothing but an ideological means of suppressing the societal other. 25 It is also claimed that the aspiration to consensus exerts a homogenizing pressure that restricts the range of acceptable input into political deliberation and marginalizes diverse identities. 26 The second feature of the ‘ideal deliberative procedure’ to come under criticism here is the requirement for rational argumentation in political deliberation. Such argumentation, say the critics, is not a culturally neutral practice. Being rather Eurocentric, abstract and male-oriented, it has a number of exclusionary effects. 27
A third influential critique of deliberative democracy – set out most clearly by John Dryzek 28 – homes in on Habermasian discourse theory. Habermas’ theory is interpreted as an attempt to respond to the first two critiques by offering a more realistic account of how deliberative democracy might be realized in current societies. By doing this, argues Dryzek, Habermas surrenders deliberative democracy’s critical potential by ascribing excessive importance to the institutions of the democratic constitutional state. In particular, political communication only figures in relation to its impact on the parliamentary legislature: ‘There is no sense that the administrative state, or economy, should be democratized further. All that matters is that they be steered by law, itself democratically influenced.’ 29
Given that they discuss premises and hypotheses in relation to the model of an ‘ideal deliberative procedure’, all three critiques can be said to be based on the more normative-prescriptive understanding underlying the first strand of deliberative democratic theory. A similarly narrow focus is observable in empirical approaches to deliberative democracy. Although a number of these explore the degree and quality of deliberation in democratic institutions and processes – mostly parliamentary debates 30 – the vast majority set out to test the assumptions underlying the ‘ideal deliberative procedure’ and the extent to which it is practicable. Studies of this latter type focus on the workability and impact of deliberation in smaller groups – either in experimental settings (deliberative polls) or in ‘real’ sites of political deliberation (citizens’ juries, citizen assemblies, etc.). 31 They ask whether deliberation really does bring about changes in individual preferences, fostering a more informed, well-considered perspective on contested political issues, an attitude of greater tolerance and understanding among citizens, and decisions of a more consensual kind.
Because it focuses on the first strand, deliberative democracy is generally thought of in terms of an ideal model of democracy based on the notion of an ‘ideal deliberative procedure’. As a consequence, it cannot easily be reconciled with the realities of pluralistic societies. The systems perspective seeks to answer some of the objections regarding feasibility by viewing deliberative democracy in the context of large-scale societies. What distinguishes the systems perspective from other approaches is that it no longer focuses on single political institutions or sites of deliberation as potential real-life manifestations of the ‘ideal deliberative procedure’. Instead, it shifts to a macro-analytic perspective and regards democratic society as a whole – with its interplay of institutionalized and non-institutionalized shaping of wills and decisions – and asks how, and how far, it could become a real-life manifestation of that procedure: Deliberative systems include, roughly speaking, four main arenas: the binding decisions of the state (both in the law itself and its implementation); activities directly related to preparing for those binding decisions; informal talk related to those binding decisions; and arenas of formal or informal talk related to decisions on issues of common concern that are not intended for binding decisions by the state.
32
What might be considered low quality or undemocratic deliberation in an individual instance might from a systems perspective contribute to an overall healthy deliberation… Judging the quality of the whole system on the basis of the functions and goals one specifies for the system does not require that those functions be fully realized in all the parts.
36
II Ideal deliberation and the epistemic dimension of democracy in reconstructive deliberative democratic theory
Like the first strand of deliberative democratic theory, the systems approach seeks to devise an ideal deliberative democracy that will serve as a critical standard in the quest for democratization. However, it alters the terms of the basic brief: instead of asking how the institutionalized decision-making procedure should be configured in order to produce an ‘ideal deliberative procedure’, it asks how democracy might be construed as a ‘deliberative system’. As argued above, this shift entails the revision of some crucial assumptions. Most importantly, the nature of consent, conflict and communication in a democracy is reconsidered. However, there is also a re-evaluation of institutionalized democratic procedure: rather than championing state institutions, the systems approach stresses the relevance of non-institutionalized actors and communications for democracy as a deliberative system. This emphasis on non-institutionalized elements is undoubtedly warranted, but it is important not to downplay the importance of democratic institutions, and of the parliamentary legislature in particular, for that same system.
Why this should be so becomes clear when we push the systems perspective one step further and apply it in sociological-reconstructive mode to deliberative democracy. A well-elaborated example of such an approach – as I will argue here – is Habermas’ discourse theory. 41 Habermas 42 looks at national democracy from a deliberative-systems point of view and asks to what extent the interplay of state institutions in democratic decision-making, and the interplay of these institutions with an anarchic general public, influences the emergence of democratically legitimate solutions to political problems and societal conflicts. The intention here is not to transpose an ideal model of deliberative democracy onto real societies but to understand how, following the collapse of a previously solid religious world-order, modern democracy manages to ensure social integration in pluralistic societies marked by growing dissent and conflict.
A new look at discourse theory that captures its deeper, sociological-reconstructive thrust may prove helpful in working out ways of responding to some long-standing deficiencies in standard accounts of deliberative democracy. Crucially, this alteration in epistemological interest brings with it a revision in the status of the concept of ideal deliberation in democratic theory. This concept is viewed, not as a philosophically devised blueprint from which the essential features of ideal democratic legitimacy may be deduced (and then contextualized) but as part of an account of the way in which communicative action actually functions – that is to say, as a world-disclosing [welterschließend] collective cooperative problem-solving practice at work in real societies (see II.1 below). It is important to be aware of this if we are to understand the pragmatist roots of the concept of democratic legitimacy in discourse theory. The conditions for democratic legitimacy are derived not from the concept of ideal deliberation but from the latter’s function as a set of idealizing presuppositions operating in the context of communicative action as a cooperative practice of problem-solving. One important consequence of this is that discourse theory’s interpretation of the epistemic dimension of democracy as a deliberative system differs from that given in alternative epistemic accounts (II.2). Equally important in transnational terms, democracy’s particular epistemic dimension, which is crucial to the legitimacy of its decisions, depends on the existence and operation of a parliamentary legislature. This being the case, the presence of a strong parliament is a (necessary but not sufficient) condition of any deliberative system that seeks to be democratic – even in a transnational context such as the European Union (III).
1 The concept of ideal discourse in a reconstructive theory of deliberative democracy
Although the concept of ideal discourse 43 touches upon the ideal conditions for the unconstrained achievement of agreement, its role in the discourse theory of democracy is not (and cannot be) to provide a blueprint for legitimate decision-making. The concept of ideal discourse is, rather, an essential component in a theory of social interaction – Habermas’ theory of communicative action 44 – which is grounded on the pragmatist assumption that human interaction is an ongoing practice of cooperative and creative problem-solving. In broad terms, the theory assumes that human action proceeds in a ‘routine mode’ of ‘pre-reflexive coping’, in which the actors tacitly accept every new situation as ‘known’, based on their knowledge and normative convictions, until such time as they are confronted with a situation that belies their tacit assumptions. They are then faced with the question of how to interact with the material or social world in order to solve the problem in hand. Although the actors are free to turn their backs on cooperation (and respond with strategic action or violence), if they want to resolve the problem in the most effective and legitimate way they eventually have no alternative but to reflect cooperatively on it and try to find a common understanding. In describing the change in attitude that occurs when actors facing a problematic situation turn from pre-reflexive coping to reflexive problem-solving, Habermas talks of a ‘transition from action to discourse’. 45 This turn of phrase, although merely analytical in nature, is unfortunate in that it may give the impression that discourse is decoupled from action. In fact – and this is central to our argument here – discourse always remains part and parcel of the one process of interaction, in the sense that its sole aim is to resolve a perceived problem that requires action and to do so by coming to a common appraisal of the situation and working out ways of dealing with it based on this appraisal. Discourse therefore takes place under pressure of time: a solution has to be found. 46
Habermas develops the concept of ideal discourse as part of his analysis of the reflexive attitude of actors as they turn towards one another and make the transition from action to discourse. The term ‘ideal discourse’ sums up the idealizing assumptions that participants in a discourse inevitably (and perhaps unwittingly) make in a situation (and every situation) in which they want to reach a common understanding without using threat or coercion. In this context, the ability of the participants to engage in argumentation (in other words, in mutual concession, demand, appraisal, justification, objection, and so on) is actually made possible by their counterfactual assumption that they are holding an ideal discourse. 47 The assumption is counterfactual because the participants make it even when they know that the ideal discourse is in fact an unrealizable utopia, or that the conditions that would make it possible have been repeatedly flouted in past deliberations – or may turn out to be unfulfilled in the deliberations in question. In some cases, the outcome of a deliberation may lose its validity because participants realize at a later stage that certain conditions of the ideal discourse have been breached. Despite this, at the very next occurrence of deliberation, and all that follow it, participants will once again presuppose an ideal discourse. According to discourse theory, this happens because participants ultimately have no alternative but to enter into discourse if they want to achieve outcomes based on fact and fairness. Participants’ mutual counterfactual assumption that they are engaged in an ideal discourse has a practical value: it establishes a perspective as open and unconditional as possible on the matter in hand. In this sense, those participating in the process of resolving problems in the material and social world rely on discursive practices as ‘corroborative authorities’. 48
It is crucial to understand that the main differences between the two readings of discourse theory stem from their differing interpretations of the concept of ideal discourse. If the counterfactual presuppositions clustered in the concept of ideal discourse are seen as ideals underpinning a fundamental moral principle – the ‘right to justification’, 49 for example – this suggests a discourse theory with an epistemological aim that is primarily normative-prescriptive in nature. The concept of ideal discourse is then viewed as offering a blueprint (albeit a purely formal one) for equitable social relations. On this view, the discourse theory of democracy constitutes an example of a first-strand theory of deliberative democracy. By contrast, the sociological-reconstructive perspective – the one which I believe Habermas applies – views the normativity established by the counterfactual presuppositions as something more general and basic. 50 Here, the concept of ideal discourse represents, not a moral ideal but a communicative rationality located deeper than moral convictions. It sets the terms for cooperative problem-solving vis-à-vis both the social and the material world and thus (as a ‘weak transcendental necessity’ 51 ) has a low-level but constant structuring impact, shaping the range of what is acceptable in the social generation both of moral convictions and of material knowledge. In other words, the notion of ideal discourse represents an empirical utopia that operates in all language-based social interaction as an enabling condition of world-disclosing practice. In line with this, the function which the concept of ideal discourse fulfils in a sociological-reconstructive discourse theory is a distinct one: it plays a key role in demonstrating that the historical development of societal moral consciousness, rather than being arbitrary in nature or metaphysically pre-determined, can be understood as a process of rationalization or societal learning. Viewed from a sociological-reconstructive standpoint, Habermas’ theory of democracy is one among several building-blocks of an overall discourse-theory of social evolution and, broadly speaking, has three aims: to uncover the core terms of political (i.e. democratic) legitimacy embodied in the self-understanding of contemporary western societies; to show how these terms differ from those that obtained in pre-modern societies; and to explain why this transition occurred in the way that it did – namely, as a learning process. 52
2 The epistemic dimension of democratic practice
Against this background, discourse theory interprets democracy as an institutionalized practice of collective problem-solving. It views the rationality of democratic practice as analogous to the epistemic meaning of discursive problem-solving in the world-disclosing practice of communicative action. This view must be distinguished from two other established interpretations of the epistemic meaning of the democratic process.
According to the first of these views, the legitimacy of the democratic process derives not only from the realization of procedural fairness and political equality through general elections but also – and independently – from the fact that it provides a means of identifying the best available political option. 53 This view of epistemic democracy is based on the assumption that there are objective criteria for selecting the best policy. The epistemic meaning of democracy thus consists in the capacity to detect political truths that exist independently of democratic practice. Democracy can ‘track the truth’. 54 Drawing on the Condorcet Jury Theorem, many proponents of this view explain majority decisions generated by democratic elections as truth-tracking devices. 55 The difficulty with this account is the untenability of its assumption that the standards of quality relevant to the legitimacy of political decisions are independent of the democratic process. One can doubtless argue that all political decision-making involves facts that can be judged objectively – after all, every sound decision depends to some extent on the deciders ‘getting the facts right’. 56 However, it is also the case that an appropriate solution to a particular political issue cannot be worked out solely on the basis of objective facts. The knowledge that must be taken into consideration in determining the legitimacy of political decisions always includes information as to which citizens will be affected by a decision and how. This ‘situated knowledge’, 57 which only affected citizens can possess, can only become the yardstick for evaluating alternative policy-options if it has previously been fed into the process of public will formation. In other words, because citizens of a democracy are ‘self-originating sources of valid claims’, 58 their equal participation is required not only by virtue of the moral principle of fairness but also for epistemic reasons.
This in turn suggests a second interpretation of the epistemic dimension of democracy – one that is often associated with the first strand of deliberative democratic theory. On this view, equal participation is required for democratic legitimacy because democracy’s epistemic capacity to establish a rational consensus among all those affected depends on it. Here, the democratic process is seen as a device not for detecting political truth but for constructing it: the best decision emerges, through argumentation, in the form of a rational consensus in which citizens give up their varying perspectives and meld them into a single general will. As has been pointed out by the major critiques of deliberative democratic theory discussed earlier, this view takes no account of the highly pluralistic character of modern societies. First, it cannot consistently demonstrate the legitimacy of majority decisions – which are an unavoidable feature of pluralistic societies. Application of the majority principle seems necessary on pragmatic grounds but conflicts with the assumed conditions for democratic legitimacy because it contradicts the democratic process’ function of generating consensus-based decisions. Second, it views the common good in terms of unity and suggests that democratic politics is a process aimed at overcoming difference. Here too then, we have a reading of democracy’s epistemic dimension that violates the fundamental principle of individual autonomy, according to which citizens are sources of valid claims that cannot simply be relegated to second place behind what is viewed as the common good.
Discourse theory’s understanding of the epistemic dimension of democratic practice differs from both these accounts. Its concept of democratic legitimacy is a pragmatist one in which the epistemic function of the democratic process is seen as constructive in character but not dependent on consensual political decision-making.
Democratic practice is seen as paralleling everyday practice in dealing with issues that arise in the world. Language-mediated everyday problem-solving is viewed as being driven by a tension between, on the one hand, the counterfactual assumption that an ideal discourse is taking place – which makes possible a common perspective on the subject in hand – and, on the other, the discourse that is actually taking place and which generates the (always imperfect) solutions. Although participants are aware of this tension and of the discrepancy between the ideal discourse and their own, actual, discourse, they are able to accept the outcomes of the latter for two reasons. First, they realize their discourse is a practical one, aimed at resolving a problem right here and now, under pressure of time, even if the information they have is incomplete. However – and this is the second criterion of acceptability – this applies only if the actual discourse does not violate the conditions of the ideal discourse in ways that were realistically avoidable. If it is clear that voices were not listened to that could have been, the outcome will be unacceptable – and this will be true even if all the participants know that, were these voices to be included, there would still be no ideal inclusion of all possible voices (and can be no such inclusion). According to discourse theory, this tension in world-disclosing practice is also a crucial factor in the analysis of democracy because it is the means by which the social world constitutes itself ‘between facts and norms’. 59 ‘[I]deas enter into social reality via unavoidable idealizing presuppositions of everyday practices and unobtrusively acquire the character of stubborn social facts.’ 60 The same is true of political practice. Discourse theory views political institutions as ‘social facts’ that counterfactually establish the social validity of ideals by which citizens’ behaviour is guided – for example, by making certain types of behaviour mutually expected in particular situations or by providing an effective means of countering non-compliant behaviour. In other words, political institutions embody an ideal cognitive content that has a structuring effect on political action.
This parallel between everyday communicative practice and political practice explains why even political decisions that lack overall consensus are still regarded as acceptable in a democracy – though this does not mean, conversely, that majority decisions are legitimate in and of themselves. Discourse theory sees societal consensus as relating not to the content of political decisions but to the epistemic meaning which citizens ascribe to their shared political practice. This means there has to be common consent about the fact that (1) there are problems that can be resolved only through cooperation, (2) this practice of shared problem-solving must be given permanent shape in legally established institutions, and (3) the purpose of these institutions is to get any problems or conflicts that arise resolved in a way that is equally legitimate for all.
The central claim of the discourse theory of democracy is that, in national democracies, the interplay between the political system and the anarchic public leads to the establishment of a deliberative system that meets the conditions just described – and, as a result, is able to ensure social integration in modern pluralistic societies. In this system, it is the cognitive content of the institutionalized democratic process that determines what constitutes legitimacy or, to put it another way, establishes a particular epistemic meaning of democracy to which 5 conditions attach:
The political process must be open, in the sense that any societal issue can be raised as a subject for shared political decision-making. The process of political will formation must take account of all objective aspects and facts relevant to the subject matter in order to meet the required standard of objective appropriateness. The process of political will-formation must take equal account of the perspectives
61
of all those potentially affected, in order to satisfy the requirement for political equality. Although this requirement is a procedural one, it has an epistemic purpose that other, non-deliberative decision-making procedures (such as tossing a coin or mechanically aggregating votes) fail to fulfil. Political equality requires that every citizen be recognized as a ‘self-originating source of valid claims’
62
whose will must not be arbitrarily relegated to second place, behind ‘the common good’, without the reasons for this being publicly specified.
63
This kind of mediation between individual interests necessitates public deliberation.
Considered in isolation, these criteria suggest that the legitimacy of democratic decisions depends on rational consensus. At the same time, it is clear that the ideals aimed at remain, to some extent, unfulfilled in every democratic process. The question, then, is why democratic decisions are not illegitimate per se. Discourse theory’s response to this problem is twofold. First, it assumes that modern citizens have been developing a more and more reflexive consciousness – meaning that they have increasingly acquired a sense both of the plurality and complexity of their societies
64
and of fallibility.
65
Thus they know from their own experience (1) that because of the diversity of modes of thought and living in the world, consensus is a difficult thing to achieve and (2) that even where it is achieved, it is valid only for the present time, because new experiences bring new insights and interpretations that may necessitate a revision of the overall picture. Second, it assumes that citizens realize that, for a given problem to be solved, a decision in favour of collective action needs to be arrived at. The upshot of this, according to discourse theory, is that in modern democracies, the sense of political legitimacy is ambivalent, involving as it does a tension between, on the one hand, the three ‘discursive ideals’ set out above and, on the other, the need for a decision to be taken. In terms of our list of conditions, therefore:
Decisions must be made in time for a collective action to be effective in overcoming a given problem.
However, in a democratic process, the pendulum cannot just be allowed to swing, unchecked, from the norm of inclusive deliberation to the necessity of making a decision. Where a restriction is imposed on deliberation for the sake of bringing about a decision, the following proviso applies:
The process of formation of wills that actually takes place does not violate the ideal of that process in a way that could realistically have been avoided.
The particular challenge which democratic practice faces is that of navigating between these partly conflicting requirements in a way that ensures decisions are sufficiently warrantable. This threshold is not one that can be determined theoretically: it has to be identified on a case-by-case basis as part of actual democratic practice.
On this sociological-reconstructive interpretation, discourse theory is able to account for the epistemic quality of democratic decisions without falling into the trap of assuming either given objective political truths or rational consensus. It opens up a conceptual space in which majority decisions can be seen as publicly endorsed, mediating compromises located somewhere between arbitrary aggregated majority-decisions and a rational consensus that allows no room for dissent. In this vein, Henry S. Richardson 66 has suggested viewing the collective will that emerges from the democratic process not as a rational consensus but as a ‘democratic intention’ – a joint intention to act collectively to overcome a joint problem, even if this action runs counter to what ‘I personally’ see as the best solution for us.
III Democracy as a deliberative system: A non-starter without parliament
One of the intentions behind the systems-turn in deliberative democratic theory has been to temper the high profile accorded to democratic institutions – and in particular to challenge the notion of parliament as a hub towards which all political communication in a democracy is directed. This echoes Pierre Rosanvallon’s 67 caution against ascribing too elevated a role to parliament in democratic thinking. Rosanvallon argues that democratic theory is still captive to a ‘founding myth’ of modern democracy – namely, the myth of the unity of the people whose will is manifested in the majority decisions of the parliamentary system. In fact, contends Rosanvallon, modern democracy has seen a shift of emphasis in the economy of legitimacy, giving a greater role to non-majoritarian institutions (such as courts and regulatory agencies) as a counterbalance to the normative deficits of electoral-majoritarian democracy. The potency of modern democracy’s founding myth as described by Rosanvallon is demonstrated in the response to the project of EU democratization. The debate surrounding this has revealed a widespread fear that the strengthening of the European Parliament will foster ‘essentialism’ and a ‘pull of oneness’ 68 and thus undermine the transnational character of the EU. Underlying this reaction is the assumption that: (1) democratic representation is about mirroring a pre-political community or demos; (2) there is no unitary European demos or nation (nor should there be); and (3) democratic representation through a strong, electorally based parliamentary legislature must therefore be rejected. Viewed against this background, the systems approach to deliberative democracy would appear to offer a route to a viable conception of transnational EU democracy free of the assumed pitfalls of parliamentary democracy. In a deliberative system, democracy beyond the nation-state could forgo a parliamentary legislature and instead take the form, say, of a ‘directly-deliberative polyarchy’. 69 In other words, it could adopt a ‘horizontal’ mode of monitoring and mutual learning based on deliberation in and among dispersed sites of decision-making, as opposed to a ‘hierarchical’ mode of monitoring conducted through the institution of parliament in its capacity as the highest legislative authority.
However, considered from the standpoint of deliberative democratic theory in its sociological-reconstructive permutation, the argument that making democracy safe for transnational politics entails finding non-parliamentary ways of organizing democracy, does not hold up. The systems approach may be right in arguing that parliamentary practice is not of itself sufficient to generate democratic legitimacy, but democracy, as a deliberative system, cannot do without it. An electorally constituted parliament acting as the highest legislative authority embodies an ideal cognitive content, and the structuring effect which such a body has on political action is constitutive for the establishment of democracy in the sense of an institutionalized practice of collective problem-solving among equals. It should be noted, however, that the cognitive content in question does not represent the ideal of a homogeneous body of the people. 70 The discourse theory of democracy suggests a view of democratic representation not as a mirroring of some pre-existent national community but as a dynamic, constructive political relationship. It sees it as an ongoing process of political claims-making in which the terms of political representation are constantly being contested and reconfigured within the context of a collective deliberative system. 71 Although this democratic contestation depends on non-institutionalized actors getting views that are shaped in everyday communication and among the anarchic public onto the political agenda, the existence of a strong parliamentary legislature in fact functions as an enabling condition of the very process of democratic contestation. It does this in three ways.
First: a parliamentary legislature constituted on the basis of a general election is constitutive of a deliberative system of democracy because it is the very means by which politics acquires its status as a cooperative problem-solving practice among equals. General elections are a collective act through which citizens institutionalize the highest legislative authority. By engaging in this act, citizens publicly renew their mutual pledge to regulate their relations through equal and inclusive decision-making. In this sense, elections to a democratic parliament do not simply mirror a pre-existing community but constantly re-create it 72 and at the same time affirm the social validity of the democratic principle as a guiding norm.
Second: the interplay between parliament – as the highest democratic legislative authority – and the electoral process ties administrative decision-making to non-exclusive public deliberation. 73 The fact that parliament is the locus within the deliberative system where the general public regularly exercises its power to decide makes it a focal point of public deliberation. It becomes a channel [Schleuse] through which decentralized and non-institutionalized publics can feed their views, effectively and equally, into the political process. 74 The existence of a parliamentary legislature is thus a precondition for citizens’ belief that their views have actual political impact.
Third: the process of legislating via parliamentary majority decision-making sets up an oppositional logic. This in turn triggers deliberative mediation between conflicting societal views and, crucially, demonstrates that all views are accorded equal respect in the decision-making process. Competition for the support of the majority shapes public political deliberation in two ways – spatially and temporally. 75 Spatially, the ongoing political competition that results from the oppositional logic forces political actors to develop justificatory narratives in order to win over the majority. 76 This creates a public space in which competing political views and options can be ascribed to particular political actors and are opened up to the broad mass of citizens. Temporally, the oppositional logic ensures that political deliberation is kept open, in the form of an ongoing process of contestation. True consensus in political decision-making is not only unlikely but also potentially costly in epistemic terms, 77 since it suggests that all objections have been met and that political deliberation is therefore complete. By contrast, majoritarian decision-making leaves us with a minority position as an indelible reminder of the fact that any decision is open to contest and that political deliberation is therefore never complete. Minority views do not vanish into thin air; they remain part of the collective memory and add to the pool of arguments available for use in future decision-making. The epistemic dimension of parliamentary practice thus helps in ‘overcoming disaffection and sustaining ongoing commitment to the democratic project’. 78
Conclusion
Deliberative democracy is generally thought of in terms of an ideal model of democracy based on the notion of an ‘ideal deliberative procedure’. The systems perspective is an attempt to reconcile this ideal with the reality of pluralistic societies by applying it to large-scale societies. In the course of this process, it revises some of the fundamental assumptions made by deliberative democratic theory.
The revisions proposed by the systems approach undoubtedly improve our understanding of deliberative democracy, but – as I have argued in this article – that approach needs to be applied with the added sociological aim of reconstructing the inner logic and rationale of democracy as a deliberative system. In this connection, I suggested acknowledging the nature of Habermas’ discourse theory as a well-elaborated example of the sociological-reconstructive approach, and, having reconsidered the concept of ideal discourse, I outlined discourse theory’s sociological contention that the rationale underpinning modern democracy is that of the institutionalization of a cooperative practice of collective problem-solving among equals. This has two important consequences in terms of arriving at a better understanding of democracy as a deliberative system. First – and in contrast to explanations given in alternative accounts – although the legitimacy of democratic decisions rests on their epistemic quality, this quality does not require either that decisions tally with some kind of ‘objective political truth’ or that they take the form of a rational consensus. Reconstructing modern democracy as a practice of collective problem-solving among equals creates a conceptual space in which majority decisions can be seen as publicly endorsed, mediating compromises located somewhere between arbitrary aggregated majority decisions and a rational consensus that allows no dissent. Second, in regard to the conditions for transnational democracy, the epistemic dimension of democracy as outlined here points to the need to rethink the role of parliament in democratic practice. If the epistemic purpose of democracy is to establish a problem-solving community of equals, then the existence of a parliament as the highest legislative authority need not presuppose the existence of a nation or imply an essentialism that fosters nation-building. In fact, a democratic parliament would seem to be a necessary condition for the establishment of any democracy as a deliberative system – national or transnational.