From Deliberative Democracy to Utopophobia: Themes from Estlund's Democratic Authority

The four articles in this special issue – by Colin Farrelly, Ben Saunders, Patrick Tomlin and Alan Hamlin and Zofia Stemplowska – pick up on themes stemming from David Estlund's recent book Democratic Authority: A Philosophical Framework (Estlund, 2008), and address various aspects of the view he expounds there. Democratic Authority is widely accredited with being one of the most important and original interventions within the field of deliberative democracy in recent decades, and is already spawning an array of critical responses.

Estlund attempts to steer a path between purely procedural justifications of the authority of democracy on the one hand, and epistemic justifications on the other. Proceduralism locates the value of democracy (if only aspirationally) as residing in the fairness of its framework. By giving everyone such things as equal voting rights, and equal rights to express their political views, democracy provides every person with the equal chance to influence the outcome of a decision. What could be fairer? Yet notice how this justification of the value and authority of democracy is intrinsic to the process; it places importance on one of the conditions of the democratic framework (equal participation) and not on what democracy actually does, or is supposed to do (its outcomes). Thus it says little, if anything, about the quality of the outcomes of democratic decisions. Really bad decisions might be perfectly legitimate. Or if fairness is the main foundation of democracy's worth, then what reason is there to prefer it to flipping a coin when faced with a political choice? Coin flipping is as fair as, and in many ways cheaper and easier than, elaborate democratic procedures.

The alternative – epistocracy in Estlund's terminology – puts value in the ability of the democratic mechanism to produce the right, or better, decision (outcome). As such, it is instrumentally valuable. The analogy – albeit a very imperfect one according to Estlund (2008, pp. 10–2) – is that of the jury system. Given the independently valuable outcome – to convict the guilty – the jury system achieves this better than other systems. If it did not do this, but randomly decided who should go free, then it is difficult to see what authority it would have. As Estlund writes: ‘[I]ts epistemic value is a crucial part of the story. Owing partly to its epistemic value, its decisions are (within limits) morally binding even when they are incorrect’ (Estlund, 2008, p. 8). The epistemic authority of democracy has a precedent, of course, in Rousseau's Social Contract (Rousseau, 1997 [1762], 4.2), where democracy better achieves the (independently valuable) outcome of realising the general will. The problem with epistocracy, however, is that it seems to privilege some rather than others, potentially mandating rule by an elite of political decision makers or ‘knowers’. Political decisions are, it is said, unlike medical decisions, for example, where specialist knowledge is better relied upon than democracy, but even if a class of knowers did exist, privileging them would be unsavoury because it seemingly leads in an undemocratic direction.

Against this background, Estlund steers a course that seeks to argue that democracy has some (modest) epistemic value – and is therefore better than flipping a coin – and that its outcomes have authority or legitimacy because of its proceduralist dimension. Hence, the ‘goal is to show how a concern for the quality of political decisions, properly constrained by other principles, supports democratic political arrangements’ (Estlund, 2008, p. 1). This ‘epistemic proceduralism’ seeks to:

The tactic, therefore, is to establish that epistemic proceduralism is superior to other epistemic theories of democracy because it establishes the very modest threshold of being slightly better than random in arriving at the correct outcome, but without sliding into a reliance upon the outcome always (or nearly so) being correct (by some independent standard).

Colin Farrelly's essay in this volume, ‘Virtue Epistemology and the “Epistemic Fitness” of Democracy’, takes this concern for the quality of the outcomes of the democratic process as its point of departure, and argues for an alternative account of establishing democracy's epistemic value. Farrelly questions on what grounds democracy might be thought to be the best epistemic device available, and proposes and defends an account he terms ‘virtue epistemology’. This alternative route places virtue ethics at its foundation. Virtue ethics is a moral theory that is concerned with questions of how the agent should live, or what kind of person he or she should be, and not – as per deontic or telic theories – how one should act. To the usual list of moral virtues (courage, generosity, etc.) Farrelly – drawing on the wider literature in this area – adds the ‘intellectual virtues’ of open-mindedness, fairness, intellectual humility, insight into persons, and several others, all associated with the pursuit of knowledge. In democratic terms this has a certain Deweyan flavour to it regarding the search for scientific knowledge and its relation to democracy. Cultivating a citizenry that displays the higher-order virtue of phronesis (practical wisdom) is a boon to democracy because it provides valuable resources for making the types of decision that democracy is concerned with. The judgements made within democracies are social, in that they are arrived at by collectivities that contribute diverse amounts of information. They are provisional, in that they are open to revision when new moral or empirical discoveries are made. Democracy, on Farrelly's view, is much more than a simple voting function akin to the analogy of Condorcet's Jury Theorem. Virtue epistemology, rather, places emphasis on key democratic characteristics such as public discussion, a free press, and mutual influence prior to voting. Accordingly, it has three major benefits. First, virtue epistemology, as an account of democratic deliberation, takes ‘realism’ seriously. That is, it incorporates considerations of cognitive limitations and the biases displayed by humans into its framework, rather than eschewing such considerations in favour of ideal circumstances. Second, it builds certain normative criteria into its assessment of the ability of different decision-making arrangements to achieve better outcomes. Third, the Deweyan conception of democracy that it leads to is considered to be more robust than simpler aggregative models because of its closer fit with a full understanding of democracy.

Ben Saunders' essay, ‘Democratic Politics between the Market and the Forum’, takes up this contrast between aggregative models of democracy (often associated with public choice theory) and deliberative democracy. Aggregative democracy is often characterised as analogous to the economic marketplace, where, like consumers on the high street, voters vote in order to promote their own interests, and where other considerations such as the common good are absent. By contrast, deliberative models of democracy are often characterised as requiring voters to exclude considerations of their own interests and vote in accordance with what they take to be the common good. Saunders argues that this contrast is overblown; that even in deliberative models there are occasions when self-interest is permissible (even if they are less frequent than in the market), and that there are occasions on which consumers in the market must look to the common good rather than their own interests.

The contrast, as it is elaborated in the work of both John Rawls (1999) and Jon Elster (1997), provides the framework for the argument. Rawls suggests several contrasts between the market and legislative politics: the market aims at efficiency, whereas the forum aims at justice. Second, the market is a perfect procedure, whereas politics is imperfect. Third, individuals in the market pursue self-interest, whereas the political process requires impartiality. And fourth, the market accommodates intensity of preferences, whereas the legislative process does not. Saunders seeks to show that these contrasts are overdrawn by Rawls, and that on Rawls’ own account, justice sets the boundaries of permissibility for laws, but, within those bounds, the pursuit of self-interest can be permissible in certain indeterminate cases, so long as it is consistent with justice and not concerned with questions over constitutional essentials. In a similar manner Saunders elaborates and explores Elster's contrasts between the market and the forum, arguing that, once again, the difference is one of degree. Justice will constrain political behaviour more than market behaviour, but motivations of self-interest need not be absent.

The motivation of individuals within the deliberative framework raises an important issue about the ‘realistic’ nature of the assumptions embodied in a theory. This is something that Estlund is sensitive to about his own conception, and raises in the chapter ‘Utopophobia’ in Democratic Authority. Epistemic proceduralism relies on the assumption that individual voters must go beyond the casual pursuit of self-interest if democracy is to have the authority that Estlund claims. The challenge here, of course, is that this is not a standard of behaviour that individuals are capable of meeting. The familiar arguments are about the usual ignorance and selfishness of voters, such that they will not act in a deliberative fashion out of a concern for the common good in the way required by the theory. The question is: what if this is correct? Does the fact that people will not behave in the way the theory requires provide a devastating objection to epistemic proceduralism itself? For Estlund, the fact that voters will never act in such a way is no objection to the theory itself (only would that, coupled with the fact that it is not something they could do, be an objection), and political theorists generally should not be gripped by ‘utopophobia’ – the fear of normative political principles that are unlikely ever to be met.

Some of these themes are explored by Patrick Tomlin in ‘Should We be Utopophobes about Democracy in Particular?’ Tomlin contests Estlund's view along two dimensions. First, he argues that concerns about realism and utopia in political theory are unlikely to admit of some blanket solution or answer. Political concepts exist, according to Tomlin, at different levels of generality or abstraction, and how much regard for real circumstances we need to pay them will, in part, depend upon the concept in question. The usual debate revolves around the concept of justice, but we need not think that all concepts exist at the same level of abstraction as justice; democracy might be very different. Second, Tomlin contests Estlund's reasons for thinking that the utopian nature of a theory of democracy is no objection to its worth. He does so by asking what type of concept democracy is, and argues that it is a problem-solving concept. That is, democracy helps us solve the problem of what justice requires. Justice provides the common rules by which we should live, but the problem is often that we do not know what justice requires, and here democracy provides a way of determining that. But such problem-solving concepts cannot, Tomlin claims, ignore the circumstances of the problem itself. They should not, that is, be utopian.

The utopophobia question is one of several debates about the place or importance of the realities of the political world to political theory. In addition to questions about utopophobia, there are issues concerning the distinction between ideal and non-ideal theory, the value of ideal theory, and the so-called ‘realist’ criticisms of ideal political theory. Alan Hamlin and Zofia Stemplowska's essay ‘Theory, Ideal Theory and the Theory of Ideals’ represents an important intervention in these debates by seeking to clarify the terrain on which the ideal/non-ideal debate takes place, and to show that this territory is only a subset of the wider territory of normative political theory. Consequently, the ideal/non-ideal debate misses important aspects of what is at stake in normative theorising.

For Hamlin and Stemplowska, the ideal/non-ideal distinction is better re-described as a continuum that is multidimensional. Comparison of theories across dimensions such as full versus non-compliance, idealisation versus abstraction, sensitivity versus insensitivity to facts, or other ways of characterising the ideal/non-ideal distinction, is possible, but unlikely to produce any clear-cut distinctions between theories. The purpose of the ideal/non-ideal continuum is to identify the social arrangements that will best promote or deliver the relevant ideals. Yet this is not the only purpose of theorising, according to Hamlin and Stemplowska, and to focus purely on this aspect misses the wider purpose of the normative political enterprise. This other aspect is the identification, elucidation and clarification of the ideal or ideals themselves – what is termed the ‘theory of ideals’. Distinguishing in this manner means that non-ideal theory can draw on ideals (without being considered as applied ideal theory) via the theory of ideals, and ensure that theorists do not omit the analysis of values because of the mistaken belief that a feasibility constraint normally associated with non-ideal theory must be adopted for the clarification of the values at stake as well.