Democracy without elections?

Amid growing consensus that democracy faces a crisis, the standard response is to plead for “more” democracy, usually in the shape of direct democracy. Yet this approach does not address the root cause of discontent, which is the perceived lack of legitimacy of political parties and their elected representatives. I argue that sortition – randomly chosing citizens as members of the legislative body – could provide the required legitimacy without sacrificing institutional efficiency.

  April 7, 2017

There is growing consensus that democracy faces a crisis,1 a finding corroborated by studies showing that trust in politicians is steadily declining.2 Democracy as an abstract ideal, however, remains virtually unquestioned in the “western” world. It is the only norm the entire political spectrum, including the most extreme fringes, seems to agree upon. Such a universal consensus is suspicious, especially in light of the current crisis. Rather than blindly calling for more democracy, I would plead for investigating alternative forms of democracy as a way forward.

I will first frame the current crisis in terms of democratic legitimacy, before arguing that its origin lies in the practice of elections. Finally, I will propose solving it by purely and simply replacing elections with random selection.

The importance of intentions

The canonical philosophical argument for democratic participation originates in social contract theory. According to this theory, individuals have to implicitly or explicitly agree to form and confer rights to a state, by entering into a social contract.3 Therefore, the very existence of a state depends on the will of the people agreeing to it, the supreme power within the state – its sovereignty – does not only come from its people, but must also always remain with it.4 In this line of thought, it is then natural to give the people a significant role in organising and running the state, by participating in the legislative, executive and adjudicative powers. The legitimacy of a state allowing for such a democratic participation is usually called input legitimacy,5 because it involves the citizens shaping and controlling the state.

This is not the only type of legitimacy that can be derived from social contract theory. Obviously, individuals only enter into such a contract because they expect some sort of (material or immaterial) benefit from creating a state.6 From this perspective, the social contract serves a specific purpose for the contracting parties and the state is only legitimate if it achieves this purpose. This type of legitimacy is often termed output legitimacy.7 It is not a democratic criterion, since it can also apply to political systems without input legitimacy.8

Modern western democracies seem legitimate both from the input and the output perspective. They offer significant democratic participation in the legislative, executive and judiciary branches and have arguably created unprecedented wealth and comfort for a large majority of their citizens.9 Yet popular discontent with politicians is growing.10 There appears to be another type of legitimacy playing a role in this discontent.

The key factor, which is usually overlooked, is intentional legitimacy.11 It deals with the – subjectively perceived – intentions of political decision makers: are they pursuing the common good or rather their own? are they independent or corrupt? Strictly speaking, intentional legitimacy is not a characteristic of the political system but of the concrete actors involved in it. It is therefore possible for a political system to produce intentional legitimacy at one point in time but not at another, while at the same time maintaining a constant output legitimacy. For instance, many western democracies are failing at creating intentional legitimacy, which was not the case a few decades ago.12

Since the degree of intentional legitimacy does not exclusively depend on the underlying institutions, it could be enhanced by acting only on the image of the political actors without altering the political system itself. But such a solution would only be superficial and temporary, at the mercy of a subsequent change of political actors. I will therefore examine the institutional weaknesses having led to waning intentional legitimacy in the next section.

The fragility of partisan systems

Modern democratic systems, developed in the tradition of the United States Constitution of 1787, are based on two important pillars: elections of legislative representatives and political parties. Both of them are now so ubiquitous and intertwined that it is difficult to grasp their peculiarties and to conceive of democracy without them.

Elections have existed in various forms since Antiquity, but until the late 18th century, they were not considered democratic but rather aristocratic13 and therefore not considered for legislative assemblies. The reason is very simple: in an election, some people have higher chances of winning than others. It invariably introduces a distinction between those who are favoured by the electorate and those who are not. The electorate does not want anyone as a representative, which is the aristocratic element. Of course, there is also an equally obvious democratic element: every citizen has the same weight in deciding who will be part of the ruling aristocracy.

In modern representative democracies, the law-making body is elected by the people.14 The modalities of this election differ considerably (proportional representation, single-winner or mixed), so that the criteria for electing the “aristocracy” vary substantially. However, there is a distinctive feature common to every representative system: a form of market, mediating between “political demand” and “political supply”. From this perspective, electoral campaigns are just mundane advertising campaigns, to influence the electorate in their assessment of the suitability of the candidates.

In such a market, information about the “political product” is intrinsically incomplete. There is no possibility for the electorate to know what its representatives will actually do when elected – not only because there is usually no imperative mandate15 but also because circumstances can change drastically during a legislature. In such a market,16 it is natural for brands to develop, which are necessary to build trust within the electorate. Developing and protecting a political brand requires significant concentration of capital (economic, human and political17) on the supply side, leading to high barriers to entry. The elective aristocracy really is an elective oligopoly.18

Therefore, the entities managing the political supply – political parties – virtually always form a restricted elite. This is a direct consequence of the election process. It also explains the ubiquity of political parties, which exist in every type of representative democracy. This is not to say that election modalities cannot influence party structures.19 For instance, a first-past-the-post system will encourage parties to develop “local brands” for each constituency and invest most in contested constituencies. The very existence of political parties, however, cannot be avoided in a representative system.

Parties, being at the core of political decision-making, are the main actors to which intentional legitimacy can be ascribed. For a party system to be legitimate, it has to provide a sufficiently varied offer, so that the majority of the electorate can identify with at least one party. Therefore, political parties must somehow reflect the political divisions, the fault lines of the society. While this requirement might conflict with the oligopolistic concentration of the political spectrum, it does not have to. As long as the number groups with significantly differing interests is limited, political parties might represent them adequately.

However, conferring political power to political parties may have other deleterious consequences. First, parties might exclusively focus on their own electorate (or, worse, on their own re-election), losing perspective of the people as a whole. This clientilism exacerbates the political divisions, as both the parties and the electorate split into antagonistic factions.20 At a certain point, the dialogue between the factions breaks down, which might lead to civil war. In terms of intentional legitimacy, each faction only considers its own representatives as being legitimate, while opposing groups are seen as pursuing a partisan agenda. The Austrian (1934) and Spanish (1936-39) civil wars are only two examples of such party-driven explosions.

Second, the party spectrum might fail to capture the society’s divisions. This can happen for a number of reasons. Established parties with large inertia might be incapable to adapt to changes in their electorate; parties might be reluctant to embrace conflict because they fear to be exceedingly divisive; new parties might not develop because of the entry barriers of the electoral market. This leads to the estrangement of an increasingly large proportion of the electorate and a perceived loss of intentional legitimacy. The political elite – in this case all relevant parties – appears to simply be ignoring the people in favour of its own agenda.

As mentioned in the introduction, this scenario is currently unfolding in many western democracies. Alarmingly, the process can reinforce itself, as those who actually are represented by the established parties start empathizing with those who are not. As a result, “the establishment” as a whole, including the political elite at its core, is rejected. It is then sufficient to appear as being “outside the establishment” to grab political power. Donald Trump’s election is not the only example of this phenomenon.

To summarize, the election of legislative assemblies, a feature common to all modern democracies, inevitably leads to a partisan oligopoly with a differentiated political elite. This system is extremely fragile, since political parties may not adapt sufficiently fast to changes in society and thus lose their intentional legitimacy. In the next section, I will propose a radical, yet reasonable way to address this weakness of the representative system and to design a political system with inherent intentional legitimacy.

Randomness to the rescue

One way to address the issue of intentional legitimacy is to resort to direct democracy. If the elected representatives and the parties they belong to are not trusted to pursue the interest of the people, it is natural to let the citizens decide on sensitive issues themselves. Proposals to strengthen direct democracy are currently quite popular21 and indeed, matters decided by referendum are undeniably endowed with input and intentional legitimacy. There are, however, a number of drawbacks of direct democracy, the most important of which is that it is only a procedure to adopt or reject, not to draft legal texts. Referenda that aim to introduce yet unspecified legislation (for instance the “Brexit” referendum) are ill-conceived, as the task of deciding what the electorate really wanted is, in the end, left to the representative system. The second crucial defect is that direct democracy is extremely time-consuming and expensive, and only makes sense for a very limited number of decisions.

Another democratic element originating in the Solonian Constitution of Athens is usually given little attention:22 sortition, i.e. random selection of citizens. In modern democracies, it is only used in the judiciary branch, where popular juries are meant to be a safeguard against politically motivated trials. Using sortition to also designate legislative assemblies, as it happened in Athens23 would be a simple and elegant way to combine the advantages of direct and representative democracy.

First, sortition guarantees a near-perfect representation of the population, since any citizen has the same probability to be chosen as a member of the assembly. Second, it ensures that there is no incentive for the formation of stable political elites, completely eliminating the issues arising in its interaction with the rest of the people. Third, it is practical, since it can fully rely on existing institutional procedures, merely replacing legislative elections with sortition.

The main advantage of sortition – its impartial randomness – is also its major weakness. Who would want laws to be drafted and decided upon by people without any legal or political experience, without relevant qualifications?24 However, this flaw is also much overestimated, since most of the drafting and legal advising is already carried out by civil servants and technocrats, not by politicians. Professional politicians indeed are experts, but not in law-making, rather in getting elected.25 Without elections, there is simply no need for professional politicians. It is in fact reasonable to contend that legislative assemblies of random citizens are more productive and efficient than assemblies whose members are indirectly campaigning during nearly half the legislature.

There remain a number of technical issues to be addressed regarding sortition.26 Who should be excluded from sortition? What compensation should the selected citizens receive to avoid corruption? Should the executive branch be responsible to the legislative assembly? Yet these issues are far from insoluble, since they require the same type of institutional fine-tuning that has been necessary to make representative democracy work.

Democracy without elections? Silly as it may sound, it is certainly no contradiction and could well be the foundation upon which to build a more stable, more legitimate, more efficient political system, a system without a differentiated political elite and its side effects. Implementing sortition requires surprisingly few institutional changes; it could even easily be rolled back should the experience be inconclusive. After 250 years of representative democracy, sortition might well the long overdue update.

  1. Cf. Armingeon, Klaus and Kai Guthmann (2014) Democracy in Crisis? The Declining Support for National Democracy in European Countries, 2007–2011 and Gould-Wartofsky, Michael (2015) The Crisis of Liberal Democracy.
  2. Cf. OECD (2015) Trust in Government, Pew Research Center (2015) Beyond Distrust: How Americans View Their Government and NatCen Social Research (2015) British Social Attitudes.
  3. Cf. Locke, John (1763) Two Treatises of Government.
  4. Cf. Rousseau (1762) Du Contract Social.
  5. Cf. Scharpf, Fritz (1999) Regieren in Europa: efektiv und demokratisch?, Frankfurt: Campus Verlag.
  6. Cf. Hobbes, Thomas (1651) Leviathan.
  7. Cf. Scharpf, Fritz (1999) Regieren in Europa: efektiv und demokratisch?, Frankfurt: Campus Verlag.
  8. In fact, Hobbes would claim that output legitimacy is incompatible with democratic participation, cf. Hobbes, Thomas: Leviathan.
  9. Cf. The World Bank (2016) World Development Indicators.
  10. Cf. OECD (2015) Trust in Government and Pew Research Center (2015) Beyond Distrust: How Americans View Their Government.
  11. Which is not equivalent to throughput legitimacy, which assesses the quality of the process leading from input to output, cf. Schmidt, Vivien A. (2013) Democracy and Legitimacy in the European Union Revisited: Input, Output and ‘Throughput’.
  12. Cf. OECD (2015) Trust in Government and Armingeon, Klaus and Kai Guthmann (2014) Democracy in Crisis? The Declining Support for National Democracy in European Countries, 2007–2011.
  13. Montesquieu and Rousseau were among the last to share this view (cf. Manin, Bernard (2012) Principes du gouvernement représentatif, Paris: Flammarion)
  14. This can be taken as a definition of “modern democracy”.
  15. Cf. Manin, Bernard (2012) Principes du gouvernement représentatif, Paris: Flammarion.
  16. Cf. Stigler, George J. (1961) The Economics of Information.
  17. Cf. French, Richard D. (2011) Political Capital.
  18. For a similar conclusion, albeit coming from other lines of thought, cf. Michels, Robert (1915) Political Parties; a Sociological Study of the Oligarchical Tendencies of Modern Democracy, New York: Hearst’s International Library Co and Weber, Max (1919) Politik Als Beruf, München: Duncker & Humblot.
  19. Cf. Weber, Max (1919) Politik Als Beruf, München: Duncker & Humblot.
  20. Cf. Rousseau, Jean-Jacques (1762) Du Contract Social. Amsterdam: Marc Michel Rey and Madison, James (1863) Federalist No. 10.
  21. Cf. Bershidsky, Leonid (2016) Bring on the other E U referendums and Caspari, Lisa (2017) Bundestagswahl 2017: Wahlkampf Mit Dem Volksentscheid.
  22. Although recently, it has increasingly received coverage, cf. Manin, Bernard (2012) Principes du gouvernement représentatif, Paris: Flammarion, Sintomer, Yves (2012) Could Random Selection and Deliberative Democracy Revitalize Politics in the 21st Century? and Pluchino, Alessandro et al. (2011) Accidental Politicians: How Randomly Selected Legislators Can Improve Parliament Efficiency.
  23. Cf. Manin, Bernard (2012) Principes du gouvernement représentatif, Paris: Flammarion.
  24. This was one of the reasons why sortition was abandoned after the 18th century, along with the (nowadays obsolete) impracticality of using it for large nation states, cf. Manin, Bernard (2012) Principes du gouvernement représentatif, Paris: Flammarion.
  25. Cf. Weber, Max (1919) Politik Als Beruf, München: Duncker & Humblot.
  26. Cf. Mueller, Dennis C., Robert D. Tollison, and Thomas D. Willett (1972) Representative Democracy via Random Selection.