The concept of democracy is central to our contemporary political vocabularies, yet agreement on how to conceptualise democracy is far from widespread.1 As Adam Przeworski has recently remarked: ‘Perusing innumerable definitions, one discovers that democracy has become an altar on which everyone hangs his or her favorite ex voto.’2 Certainly we can say that democracy is a form of government that appeals to an idea of popular sovereignty and, hence, an answer to the question ‘who rules?’ – but to flesh out this answer will very quickly mire us in controversy. This point is of more than merely academic interest for two reasons. First, how we understand the concept of democracy guides our practical reflections on how to design or reform democratic institutions, it generates criteria governing what we can reasonably expect from democratic government and it animates our debates concerning political legitimacy. Second, in so far as reasonable disagreement is an abiding circumstance of politics and democratic rule is a condition of political legitimacy, it follows that disagreement concerning the nature of democratic rule, and hence of the criteria of political legitimacy, is itself liable to be a persistent feature of democratic politics. For this reason, this chapter will begin by considering a recent minimalist view of democracy, before going on to consider two important contemporary models of democracy: the interest-aggregating model and the deliberative model. It will then briefly consider a supplement to each of these models in the form of ‘contestatory’ democratic mechanisms. The chapter will conclude by indicating what is arguably the main contemporary challenge to democratic theory and practice in the era of globalisation.
1 A minimal view of democracy
Perhaps the best known minimal view of democracy is that advanced by Schumpeter as ‘a system in which rulers are selected by competitive elections’,3 where such elections are held on a regular basis and under conditions of universal suffrage. The main question that arises here is this: why should we value democracy understood in this minimalist way?
Przeworski has argued that we do have good reasons to value democracy on a minimal understanding of it; even if it is the case, as he also argues, ‘that choosing rulers by election does not assure either rationality [of decision-making], or representation [of the interests or will of the people], or equality [of citizens]’.4 He advances this argument on two main grounds: first, ‘the mere possibility of being able to change governments can avoid violence’ and, second, ‘being able to do it by voting has consequences of its own’.5
With respect to the first point, Przeworski puts his argument thus:
assume that governments are selected by a toss of a, not necessarily fair, coin . . . the very prospect that governments would alternate may induce the conflicting political forces to comply with the rules rather than engage in violence, for the following reason. Although the losers would be better off in the short run rebelling rather than accepting the outcome of the current round, if they have a sufficient chance to win and a sufficiently large payoff in the future rounds they are better off continuing to comply with the verdict for the coin toss rather than fighting for power. Similarly, while the winners would be better off in the short run not tossing the coin again, they may be better off in the long run peaceably leaving office rather than provoking violence to their usurpation of power.6
Notice that this argument suggests that the chances of maintaining democratic rule are increased in having at least two relatively matched political parties and where political loyalty is primarily to political parties rather than individuals. In the absence of these conditions, as the current state of affairs in Zimbabwe illustrates, democratic rule may become very fragile.
Przeworski’s second point turns on the fact that we do not actually toss coins but vote:
Voting is an imposition of a will over a will. When a decision is reached by voting, some people must submit to a decision different from theirs or to a decision contrary to their interests . . . Voting generates winners and losers, and it authorises the winners to impose their will, even if within constraints, on the losers. This is what ‘ruling’ is.7
This fact has consequences not because voting imposes an obligation to respect the results of voting (although it may, Przeworski is more sceptical of this claim that I am) but because ‘voting does reveal information about passions, values and interests’:
If elections are a peaceful substitute for rebellion, it is because they inform everyone who would mutiny and against what. They inform the losers – ‘Here is the distribution of force: if you disobey the instructions conveyed by the results of the election, I will be more likely to beat you than you will be able to beat me in a violent confrontation’ – and the winners – ‘If you do not hold elections again or if you grab too much, I will be able to put up a forbidding resistance.’8
In other words, voting not only provides a mechanism, like coin-tossing, that avoids violence but also provides current rulers and possible future rulers with information concerning the political constitution of those subject to their rule, information which (given their interest in re-election and election, respectively) is likely to inform the character of their rule.
These are, I think, compelling arguments for the value of democracy even on a minimal view. But that we should value democracy on these grounds does not imply that we cannot reasonably and plausibly expect more – at least from relatively mature and relatively wealthy democracies – than simply the avoidance of internal violence and the provision of informational constraints on rational rulers. What more can and should we expect? The reflections offered in this chapter will begin from the delineation of a formal concept of democracy as: a mode of government in which the members of the unit of rule are equal consociates and have collectively an effective capacity to govern, either directly or via intermediaries, matters of common interest (or concerning the common good) qua membership of this unit of rule.
This formal concept highlights two features which are typically taken to be basic to any substantive account of democracy: the political equality of citizens and the idea of collective self-rule. Hence on any more than minimal view of democracy, it is suggested, we can and should expect that democratic institutions will, at least to a significant extent, be shaped by commitments to ensuring the political equality of citizens (in terms of, for example, public and private rights) and to facilitating ‘collective self-rule’, where this phrase implies not simply a right to the periodic selection of one’s rulers by way of competitive elections but also that ‘important decisions on questions of law and policy depend . . . upon public opinion formally expressed by citizens’.9 How we understand these commitments (and the obligations that they impose), however, will hang to a large extent on the way in which we conceptualise democracy. For this reason, the next two sections of this chapter examine two different models – or, more strictly, regulative ideals – of democracy, tracing the distinct ways in which these commitments are cashed out.
2 The interest-aggregating model
This first model begins with the intuitively simple and appealing thought that the basic substance of political reflection and action is interests, where these interests are expressed by political actors as preferences. This basic thought orients the justification and conceptualisation of democracy as a mode of political government. The argument for democracy on this understanding runs as follows:
If the good or interests of everyone should be weighed equally, and if each adult person is in general the best judge of his or her good or interests, then every adult member of an association is sufficiently well qualified, taken all around, to participate in making collectively binding decisions that affect his or her good or interests, that is, to be a full citizen of the demos. More specifically, when binding decisions are made, the claims of each citizen as to the laws, rules, policies, etc. to be adopted must be counted as equal and equally valid. Moreover, no adult members are so definitively better qualified than the others that they should be entrusted with making binding collective decisions. More specifically, when binding decisions are made, no citizen’s claims as to the laws, rules, and policies to be adopted are to be counted as superior to the claims of any other citizen.10
To argue against democracy on this view requires that one reject either or both of the conditional statements at the beginning of Dahl’s remarks. Hence, the anti-democrat must argue that the interests of everyone should not be weighed equally and/or that each adult person is not in general the best judge of his or her interests. Although historically both of these anti-democratic arguments have been made, it is not clear that they can easily be sustained.
Given this interest-oriented argument for democracy, what ideal standards are appropriate to a democracy? Dahl suggests the following five ideal standards as the core normative commitments of democracy:
- Effective participation
- Equality in voting
- Gaining enlightened understanding
- Exercising final control over the agenda
- Inclusion of adults.11
Why these criteria? Dahl’s response is that, given the assumption that democracy is to be understood in terms of interests, ‘each is necessary if the members . . . are to be politically equal in determining the policies of the association’.12 This establishes (1), since to deny any citizen the opportunity to express their preferences, place questions on the agenda or give reasons for endorsing or rejecting a given proposal is to mute their political voice; it justifies (2), since to weigh votes unequally is to weigh interests unequally; it implies (4), since otherwise the agenda may not represent the full range of interests of the citizens; and it establishes (5), since otherwise the interests of some competent persons are not counted. But what of (3)? Dahl’s point here is that citizens must have an adequate and equal opportunity ‘for discovering and validating (within the time permitted by the need for a decision) the choice on the matter to be decided that would best serve the citizen’s interests’,13 since otherwise some citizens may be disadvantaged relative to others in terms of being able to accurately determine the choice which best expresses their fundamental interests with respect to a given decision. Essentially the democratic idea expressed here involves two elements. First, that what count as matters of collective interest (that is, matters on which political decisions may be appropriate) should be determined by equal citizens as a collectivity within the constraints imposed by the conditions of democratic rule. Second, that what is held to be in the collective interest (that is, what is the best course of action for the polity to adopt) in relation to a given issue should be decided by equal citizens collectively in accordance with the principles of democratic rule. This is why Dahl stresses the importance of both the collective ability of citizens as political equals to set the agenda and to decide what to do. Any form of government that fails to meet both of these conditions subjects citizens, in one way or another, to the rule of guardians.
But how are decisions concerning what is in the collective interest to be determined? The aim of democracy, as David Miller noted, ‘is to aggregate individual preferences into a collective choice in as fair and efficient a way as possible’.14 Now in so far as these collective choices take the form of a single-issue choice between option A and option B, such aggregation is straightforward and, for issues which are not constitutionally basic, it seems reasonable to adopt the principle of majority rule, that is, that the option with the majority of votes wins.15 But what of collective choices where there are more than two options and no option receives an absolute majority of the votes cast? Here determining the collective interest may be harder than it immediately appears to be.
Consider that there are numerous ways in which an aggregation of these votes might be accomplished. On the one hand, there are majoritarian decision-making rules such as the plurality rule (whichever option gets the most votes wins) and the Condorcet rule (the winning outcome is the option which defeats each of the others in a vote on every pair of alternatives). The problem here is that the plurality rule implies that a choice is simply that of the largest minority, while the Condorcet rule is incomplete in that for a given distribution of votes there may not be a Condorcet winner. On the other hand, there are positional decision-making rules such as the Borda count (the winning outcome is determined by scoring each option according to its place in a voter’s ranking, thus the top option gets n points, the second n–1 points, etc., and then aggregating the point for each option across all votes cast). The problem here is that ‘it may make the decision among quite popular options depend on the way that some voters rank way-out or eccentric options if these are on the ballot paper’16 and, for just the same reasons, this decision rule is highly vulnerable to strategic voting. Things get worse! As Arrow has demonstrated in his famous Impossibility Theorem, given certain reasonable conditions which we might wish a decision rule to satisfy,17 ‘if there are more than two alternatives, any method for making social decisions that ensures transitivity in the decisions must be necessarily either dictated by one person or imposed against the preferences of every individual’.18 The implication drawn from this and related social choice theorems is that no rule for collective decisions can be discovered that does not produce arbitrary or meaningless outcomes and hence both (a) there is no decision rule for aggregating preferences which is clearly fair and rational and hence superior to other possible rules and (b) different rules may produce different outcomes. In this context, Riker has argued that it makes no sense to speak of discovering the ‘popular will’ at all.19
This is clearly in principle a very serious challenge to this model of democracy and it is one that might encourage us to give up trying to make sense of the notion of majority rule. We might, then, instead adopt a principle of unanimity in which each citizen has a right of veto over laws, policies, etc.20 The main objection to this proposal by Buchanan and Tullock is that it is unfeasible, making government (to all intents and purposes) impossible.21 This conclusion is a little quick since, under conditions in which each of us knows that any of us can veto a given proposal, it seems likely that we will rapidly develop a political culture of bargaining, compromise and trade-offs such as characterises existing political contexts in which decisions are subject to this rule (for example, treaty negotiations). Still although the conclusion drawn by Dahl is hasty, it is the right one – under contemporary conditions, it makes little sense to propose that a community’s capacity to govern itself be subject to the dogmatic convictions of every member. However, since unanimity is the only way of avoiding the problems concerning majority rule raised by social choice theory, we are thrown once more back on to the task of making sense of this notion. One way to avoid this problem would be to adapt the unanimity principle so that what requires unanimity is not a proposal but a choice between two proposals – and then decide between these proposals on the basis of a majority vote. This would avoid the most obvious difficulty with the unanimity principle while ensuring that all decisions can be unproblematic expressions of majority rule. However, although such an adapted unanimity principle would avoid the problems raised by social choice theory, it is difficult to see how such a demanding principle would be practicable.
But does the arbitrariness of decision rules really matter that much? Recall that the problem is this:
The problem of arbitrariness arises because it is not clear which of the many possible rules best matches our intuitive sense of ‘finding the option which the voters most prefer’, or to put the point another way, for any given rule it is possible to give examples where using the rule produces an outcome that seems repugnant to our sense of what a democratic decision should be.22
It was for this reason that Riker insisted that the notion of ‘popular will’ is just an empty phrase. But to reach this conclusion Riker would need to assume that the fact that different decision rules produce different results or expressions of the ‘popular will’ vitiates the notion of the ‘popular will’ and this does not follow: the popular will (or collective interest) is just whatever is the result of the decision rule that we take to be authoritative for a given decision, that other rules may produce other results is neither here nor there. The real question is how, knowing that different rules may produce different results, we decide which rule to employ for a given decision. Now it might seem that this simply pushes the whole problem back one stage but it does not. From a democratic point of view, what matters is not that a given decision rule may produce results which conflict with our intuitions concerning what a democratic decision would be but that the democratic community has determined what the authoritative rule is for a given class of decisions – and this is the point at which the principle of unanimity has a role to play. In other words, what matters democratically is not that the rule may produce counter-intuitive outcomes but that the choice of the rule is not arbitrary – and the only way to secure this is to adopt the principle of unanimity. It is, in other words, to treat the choice of decision rule (or rules, where different rules may be used for different decisions) as if it were analogous to a treaty – or contract! – in which all parties are required to agree. Indeed, this is, I think, why Rousseau insisted that the contract to form a political community must be unanimous. (Moreover, if the members of a community are particularly troubled by the thought of strategic voting, they might choose to introduce a meta-rule to the effect that the decision rule will be chosen randomly once the votes are cast. This would have the advantage of blocking attempts at the strategic manipulation of the outcome.) But does not this proposal suffer from the same disadvantage as the proposal of unanimity as a decision rule? No. The problem there was that where unanimity is used as a decision rule for specific issues, any member who had particularly strong or dogmatic convictions on that issue could veto any proposal that did not conform with their convictions. But in this case unanimity is being used as a decision rule for deciding between decision rules and not for deciding between proposals on concrete topics – and this means that no member of the community will be in a position to know (in more than very general terms) whether or not the choice of decision rule x for treating decisions of type y is likely to work out in ways that tend to favour them and hence members will have no interest-based motive to veto the proposal of particular decision-rules. (Note that if this turned out to be false, it would provide a general incentive for adopting the randomising meta-rule suggested above.) In this respect, it seems to me that we can talk intelligibly of the popular will or collective interest, and hence that the apparent threat posed by social choice theory to this model of democracy is dissolved.
To conclude this section, let us return to the question of what more we might expect from democracy in terms of cashing out the principles of political equality and ensuring an effective capacity for collective self-rule. In terms of this interest-aggregation model, it is clear that what we can and should expect is a commitment to realizing as fully as practicable the five ideal standards which Dahl sketches and thus, most importantly, to measures which seek to ensure our equal freedom to form and identify our own interests, our equal freedom to express these interests at all stages of the democratic process from agenda-setting to final decision-making and the equal weighting of our interests in determining our collective decision. Our democratic institutions are to be evaluated in terms of their design and performance against their satisfaction of, and commitment to, such measures.
3 The deliberative model
Whereas the first model takes interests as the basic currency of politics, the second model takes public reasons as occupying this position. Joshua Cohen has helpfully summarized the distinction thus:
According to an aggregative conception of democracy . . . decisions are collective just in case they arise from arrangements of binding collective choice that give equal consideration to . . . the interests of each person bound by the decisions. According to a deliberative conception, a decision is collective just in case it emerges from arrangements of binding collective choice that establish conditions of free public reasoning among equals who are governed by the decisions. In the deliberative conception, then, citizens treat one another not by giving equal consideration to interests . . . but by offering them justifications for the exercise of collective power framed in terms of considerations that can, roughly speaking, be acknowledged by all as reasons.23
The deliberative argument for democracy, thus, emerges from a claim about the political equality of citizens as grounded on their equal moral status as autonomous individuals capable of giving and exchanging reasons – and, hence, on what Rainer Forst has called their basic moral right of justification, that is, the basic right to have exercises of collective power over their free activity as citizens justified by reasons that are acceptable to them as citizens.24
Given this argument for democracy, what ideal standards does it invoke? On Cohen’s account, these standards can be given by presenting ‘an idealized procedure of political deliberation, constructed to capture the notions of free, equal and reason that figure in the deliberative ideal’.25 We can summarise this ideal procedure thus:
- All citizens acknowledge the freedom of each citizen to participate.
- Citizens are formally equal in that each has the same rights to propose issues and solutions, to offer reasons for or against proposals, and to have an equal voice in deciding the outcome.
- Citizens are substantively equal in that each has an equal opportunity to exercise their rights of participation.
- Citizens are reasonable ‘in that they aim to defend and criticize institutions and programs in terms of considerations that others, as free and equal, have reason to accept, given the fact of reasonable pluralism and on the assumption that those others are themselves concerned to provide suitable justifications’.26
The standards invoked by this idealised procedure are interestingly analogous to those proposed by Dahl in that this procedure ensures the opportunity for effective participation (proposing issues, solutions and offering reasons), equality of voting (each has an equal say and each counts as one in a vote), gaining enlightened understanding (where this now implies the opportunity to consider and determine, within the time available, the reasons that one considers best concerning the issue at hand), exercising final control over the agenda (equal rights to propose issues) and inclusion of adults (all persons with the deliberative capacities). In addition, this model also invokes the standard of reasonableness, namely, that citizens acknowledge the fact of reasonable pluralism and seek to offer reasons that other reasonable citizens could not reasonably reject.27 In this respect, the deliberative model is more demanding than the interest-aggregation model since it requires that citizens exercise a form of democratic self-restraint, namely that they reflect as citizens and not in terms of their private interests.
At this stage, we can turn to the issue of what it means to talk of ‘collective self-rule’ in this context. Ideally, of course, collective self-rule here refers to the generation of a consensus such that for any given law or policy, it can be justified to all citizens in terms that they could not reasonably reject. (This does not require that the justification is the same for all citizens; an overlapping consensus will do as well as a common consensus.) However, since it is unlikely that such a consensus will emerge on most, if not all, issues, it is clear that most decision-making will require the use of some decision-rule in this same way as required for the interest-aggregration model.28 The deliberative model is, however, arguably better placed with respect to the issues raised by the need to select a decision-rule. Thus David Miller has argued that the interest-aggregation model is exposed to social choice dilemmas in part because it posits choosing a decision rule independently of consideration of the content of the citizens views, whereas because the deliberative model sees the content of citizens’ judgements concerning the collective interest as emerging in the course of reasonable deliberation, it is less vulnerable to these problems in three ways.29 First, we may plausibly expect that the deliberative process may produce voters’ rank orderings that are ‘single-peaked’, that is, ‘the alternatives can be arranged on a continuum such that if, say, a voter ranks the alternative on the left highest, he does not rank the alternative on the left above that in the centre’.30 This matters because in cases where such single-peaking occurs, there is always a Condorcet winner. We can expect this because in many cases the policy options represent a choice between two values and voters’ ranking reflect their weighing of these values. Second, where such single-peaking does not occur, it is likely to be the result of more than one dimension of disagreement emerging in the deliberative process – but precisely because the dimensions of disagreement become apparent through deliberation, it may be possible to disaggregate the original choice scenario into components such that for each component there is single-peaking. Third, if we consider majoritarian (for example, Condorcet) and positional (e.g., Borda count) types of decision-rule, we can note that whereas the former aim to satisfy as many people as possible, the latter aims to satisfy everyone collectively as much as possible and, consequently, which of these aims is best may hang on the nature of the issue over which we are seeking to come to a decision. Miller’s point is that deliberative democracy will be well placed to choose decision-rules that are appropriate to the issue at hand since the nature of this issue will be revealed in deliberation, whereas the interest-aggregating model of democracy will not be so well placed. Whether or not it is actually the case that the deliberative model is better placed, however, depends on whether or not citizens do in fact exhibit the commitment to reasonableness that this model calls for, and that, in turn, hangs to a significant extent on whether the claim that involvement in deliberation with free and equal others has transformative effects on individual citizens, that is, cultivates a disposition to reasonableness.
In considering the question of what more we might expect from democracy in terms of cashing out the principles of political equality and ensuring an effective capacity for collective self-rule, we can note that the deliberative model involves a commitment to realising as fully as practicable a framework of social and institutional arrangements that ‘facilitate free reasoning among equal citizens . . . while ensuring that citizens are treated as free and equal in that discussion; and tie the authorization to exercise public power – and the exercise itself – to such public reasoning’.31 It is against such criteria that our democratic polity is to be judged.
4 Contestatory democracy
Given the above account of the two models of democracy, we might still reasonably be concerned by the potential problem posed by the tyranny of the majority. While both models provide effective arguments for liberal freedoms such as freedom of speech and freedom of association as intrinsic to democratic rule, it remains plausible that, under non-ideal conditions, the interests or reasons expressed by minority groups may be ignored or, at least, not granted equal status within the decision-making process. In the practical context of democratic rule by way of representative government, addressing this problem ‘would require not just that the majority are heard in determining what common, perceived interests ought to be pursued by government, but also that the relevant minorities get a hearing, ‘So the question is whether there is any way of subjecting government to a mode of distributive or minority control in order to balance the electorally established mode of collective or majority control.’32
Pettit’s suggestion is to introduce contestatory mechanisms. These are mechanisms through which a minority group who hold that the decision reached has not adequately acknowledged their political voice can contest the decision through ‘a procedure that would enable people, not to veto public decisions on the basis of their avowable, perceived interests, but to call them into question on such a basis and trigger a review; in particular, to trigger a review in a forum that they and others can all endorse as an impartial court of appeal’.33
This supplement to both the major models is important not least because it contributes to maintaining an effective sense of political belonging among minority groups. After all, as Pettit notes:
There is an enormous gulf between being subject to a will that may interfere in your affairs without taking your perceived interests into account and being subject to a process such that, while it takes your interests and those of others equally into account, it may deliver a result – for reasons you can understand – that favours others more than you.34
Consequently, under non-ideal conditions, a concern with promoting democratic stability in the sense of a strong identification of citizens with their democratic institutions entails that we have good reasons to adopt a contestatory supplement to our basic democratic fora and that this is the case whether our democratic understanding is primarily informed by either of the two main models considered.
5 Globalisation and democracy
In this final section, I want to mention a significant contemporary issue for democratic theory, namely, the problem of the embeddedness of democratic polities within networks of transnational governance.
It is increasingly becoming a commonplace that democratic polities in the form of the sovereign nation-state are situated within, and subject to, forms of regional and/or global governance by way either of regional organisations such as the European Union (EU) and the North Atlantic Free Trade Area (NAFTA) (which may or may not be polities themselves: the EU is a polity but NAFTA is not) or the multilateral institutions of global governance such as the World Trade Organisation (WTO), the International Monetary Fund (IMF) and the World Bank. In this context, recall our basic formal definition of democracy as a mode of government in which the members of the unit of rule are equal consociates and have collectively an effective capacity to govern, either directly or via intermediaries, matters of common interest (or concerning the common good) qua membership of this unit of rule. Up to this point we have been considering the issue of citizens possessing ‘ an effective capacity to govern’ in terms of the internal arrangements and conditions of a polity, but it is equally clear that such an effective capacity can be undermined by virtue of being subject to external forms of governance such that, for example, policies concerning the subsidisation of manufacturing industry within the polity are politically disconnected from public opinion on this topic and subject to the authority of the WTO. The dilemma, therefore, is that a polity which has the form of a democratic polity may only exhibit the appearance of democratic rule rather than the reality of such rule – or, more strictly, may only exhibit the reality of such rule in externally limited and constrained contexts. In this real world contemporary context, maintaining the claim that a polity is democratic requires either a highly implausible and impracticable reassertion of its autonomy or a democratisation of the forms of transnational governance to which it is subject. In this respect, as David Held has powerfully argued, a concern with democratic rule cannot be restricted to the level of the sovereign state but must track the levels of governance to which we, as peoples, are subject.35 This raises a plethora of theoretical and practical issues for democracy but the challenges posed by these difficulties need to be met if we are to maintain our practical identities as democratic citizens.
This chapter began with reference to a minimalist view of democracy and it did so because reflecting on this view should remind us of a point which needs to be borne in mind when considering the concept of democracy. This is that the more we demand from democracy, the more we are likely to be disappointed by our democratic institutions and, in our disappointment, to lose sight of the very real benefits which democracy delivers even on a minimalist view. With this warning in mind, however, it has been argued that we can, at least in some circumstances, expect more from democracy than the minimalist view admits and I have tried to sketch what more we might reasonably expect by reference to the two major contemporary models – or regulative ideals – of democracy as well as suggesting that contestatory mechanisms can play an important supplementary role with respect to these models. Finally, the chapter has indicated the central issue for the future of democratic theory and practice, namely, our subjection to forms of transnational governance over which, at present, we can exercise little or no effective democratic control.