Gastil and Wright envision a sortition chamber within a bicameral system, and in this chapter, I challenge their proposed retention of an elected chamber. I contend that in a democracy, elections are neither a fundamental nor an effective way to achieve accountability and political equality, and hence legitimacy.
This view strikes most people as absurd, or at least counter to the received wisdom that elections are the cornerstone of—or synonymous with—democracy. From a historical perspective, this response is unfounded. Since the so-called birth of democracy in ancient Athens until the publication of Alexis de Tocqueville’s influential Democracy in America (1835–40), elections were synonymous with aristocracy, meritocracy, or—at best—“republican” government, whereas “pure democracy” entailed more direct methods of self-government, including rotation and sortition (that is, using random selection to fill public offices).1
Historical meanings aside, choosing representatives by way of elections leaves much to be desired, when compared to the sortition alternative. Below it is argued that democratic accountability under sortition would be as (or even more) robust as it is with elections. It is then argued that the superior political equality and deliberative quality of a sortition legislature would help it secure greater public legitimacy. In the final section, I argue that a unicameral sortition assembly would better secure democratic values than would Gastil and Wright’s proposed bicameral legislature as the latter retains of an elected chamber.2
Accountability
Perhaps the most obvious justification for elections is their purported ability to hold rulers to account. In this view, the prospective fear of sanction—or the retrospective act of being sanctioned—results in politicians that are responsive to the electorate. Before addressing this “folk theory of democracy,” however, it is instructive to take a step back to consider the general concept of political accountability.3
Avoiding Horizontal and Vertical Hazards
Political theorist John Dunn holds that accountability is, at heart, an attempt to ensure that rulers (and ruling groups or factions) do not abuse their position of power. Dunn recognizes two principal, potentially avoidable hazards of living in societies: vertical hazards from rulers and horizontal hazards from other members and groups in society.4 Both of these hazards are real—and all too familiar.
The first two ways societies have attempted to limit these hazards, according to Dunn, is by instituting legal and constitutional constraints on governments. The programs of ancient and modern liberty (such as legal equality; civil liberties; and freedom from torture, arbitrary violence, and imprisonment, as encapsulated by constitutions and the “rights of man [sic]”) are realizations of this ideal. In theory, the state retains a monopoly on legitimate violence, tightly constrained by rights, obligations, constitutions, and laws. These constraints are independent of the system of rule. Proponents of classical liberalism were concerned with ensuring rights in monarchies and aristocratic regimes; few foresaw the coming democratic struggles of the nineteenth and twentieth centuries.
Even if rulers face legal constraints, however, the risk remains that governments will translate their society’s horizontal hazards directly into vertical hazards as antagonistic social groups compete for control of these institutions. The winning group, or coalition of groups, can then use the resources of the state to oppress or exploit other groups, or to simply strengthen their own social and financial position through preferential treatment and laws.
Political accountability attempts to address this breach. Dunn posits that the risk of horizontal hazards being transformed into vertical hazards can be minimized by “holding the key agents and implementers of public choice effectively responsible for the manner in which they make and implement these choices … The locus of democratic accountability is this second approach to limiting the inherent hazards of political subjection.”5
It is here that elections are used as the crutch on which political theorists prop up theories of accountability in representative democracy. Dunn acknowledges that the idea “has had a very good run for its money in modern political thinking” but “is an astonishingly optimistic way of envisaging political relations.”6 Electoral accountability fails due to the informational asymmetry between ruled and rulers and the impossibility of assigning a clear causal link between intention and consequence in any complex societal system.7 To meaningfully sanction politicians would require a clear thread, involving “highly determinate conceptions of action,” joining political intention, action, and outcome. Such a thread cannot, under usual circumstances, and by ordinary uninformed, uninterested citizens, be untangled—if it exists at all.8
What Motivates Voting Behavior
Dunn’s more theoretical objections are confirmed by a wealth of research highlighted in a new critique of American elections by political scientists Christopher Achen and Larry Bartels. In Democracy for Realists: Why Elections Do Not Produce Responsive Government, they show that people often draw fictitious links between political cause and effect when assigning political responsibility. Even when voters make choices based on real shifts in their own financial positions, they do so myopically—forgetting anything other than the most recent changes.9 If this amounts to “the consent of the governed,” it is an ill-informed, shortsighted consent at best. As Achen and Bartels put it, retrospective sanctioning “simply will not bear the normative weight that its proponents want to place on it.”10
Achen and Bartels are equally skeptical of the myth that politicians outline a program of ideas and an informed, attentive electorate votes for the set of ideas they most admire, and punishes those who fail to deliver on their promises.11 Here again, the evidence reviewed by Achen and Bartels does not support this view. The crux of their argument is that “group and partisan loyalties, not policy preferences or ideologies, are fundamental in democratic politics.”12
Who Gets Represented
Notwithstanding the above, politicians do react to events and attempt to win elections by convincing voters that they are responsive—at least when they are not out wooing financial backers for their next campaign.13 Bartels’s earlier book, Unequal Democracy, along with Martin Gilens’s Affluence and Influence, shows that elected representatives in the US respond overwhelmingly to one particular social group—the affluent.14 Bartels demonstrates that elites listen predominately to their own, with US senators being “vastly more responsive to affluent constituents than to constituents of modest means … [while] the views of constituents in the bottom third of the income distribution received no weight at all in the voting decisions of their senators.”15
Gilens finds a similar correlation between expressed preferences and policy outcomes, which “suggest that the political system is tilted very strongly in favor of those at the top of the income distribution.” Analyzing mountains of national survey data and federal policy choices, Gilens concludes that “when preferences between the well-off and the poor diverge, government policy bears absolutely no relationship to the degree of support or opposition among the poor.”16
One benign explanation for this pattern could be that middle- and upper-income preferences are similar. Thus, disregard for the poor could indicate a responsive majoritarian democracy, albeit a clear case of horizontal hazards being translated into vertical ones. Closer inspection of the data, however, disproved this hypothesis.17 Low- and middle-income citizens do get what they want sometimes: in those instances when their preferences align with the rich. This fact (that middle-income citizens sometimes get what they want) is what contributes to the illusion of direct responsiveness.18
Political Agency and Accountability Beyond the Electoral System
However bleak this picture may seem, democratic governments are not free to do as they please. In The Life and Death of Democracy, political theorist John Keane locates the nexus of accountability in the myriad institutions of civil society, the media, the judiciary, and other institutions that can keep power in check—all part of a “monitory” democratic system.19 The few minutes it takes to cross or number a ballot cannot be the primary mode of holding rulers accountable for their actions. A second-term US president, who will never face the electorate again, is still held accountable in many important ways.20
This systemic view of accountability foregrounds the institutions and social practices that encourage—or even demand—that rulers justify their decisions and actions. A diverse, independent, and free press plays a key role in questioning representatives. An independent judiciary constrains the kinds of laws enacted and should ensure that rulers who break laws are prosecuted. An active civil society can scrutinize legislation and mobilize political activity to promote or oppose specific laws. Civil liberties guarantee that if dissenters find official justifications unconvincing (or unethical), they have legal avenues to express their outrage and organize resistance; even beyond lawful protest, nonviolent civil disobedience has a proud history in democracy.
More broadly, rulers are held to account through well-established (if too often flouted) norms requiring that they give reasons for their decisions and actions. Obviously, elected politicians can and do dissemble, with equivocations, convoluted strategic rationales, and explanations that please partisans but contradict the factual record. That reasons are given at all, however, shows the persistence of this norm and underpins the abiding faith in the possibility of a more deliberative politics.21
For the argument here, the important point is that elections are not a necessary component of this systemic accountability process. Ironically, elections may undermine effective accountability by shifting the media and the public’s focus from substance to spectacle.22
Even if elections are the key moment of political agency for most people in modern democracies, agency should also be understood in much broader terms. Political activity has many facets, such as writing to a representative, signing a petition, participating in a town-hall meeting, donating to a political party or cause, or attending a demonstration. Democracies must include a plethora of avenues for citizen participation beyond elections.
If instituted, sortition would require continued attention to maintaining these possibilities for mass political participation. Support for sortition and support for stronger forms of participatory democracy go hand in hand. For example, a sortition legislature might be complemented effectively by a periodic political event or series of citizens’ assemblies, such as a modified version of the Deliberation Day proposal, where a list of legislative priorities is collectively developed and delivered to legislators.23
More generally, a sortition system could work well alongside participatory budgeting, open consultative meetings, and strong social movements. Eliminating the expense of elections could free up human and financial capital for such civic activities. Whereas an elected representative must prioritize the interests of economic elites because of fundraising imperatives, a representative selected by lot might choose to meet, learn from, and respond to a far more diverse array of individuals and social groups.24
Equality, Legitimacy, and the Simulation Claim
A legislature populated using sortition would be accountable if the liberal freedoms and democratic norms of public justification of decision-making were continued—or hopefully even enhanced. The loss of elections would represent neither a lethal nor a particularly serious blow to democratic accountability, and as argued elsewhere, sortition would result in a descriptively representative assembly—unlike the gender-, racial-, and age-imbalanced assemblies of today.25
What about political equality? In an electoral system, this democratic ideal is fulfilled by giving every adult one vote—although it is relatively easy to show that this does not lead to political equality in practice.26 With a unicameral sortition legislature, political equality manifests most directly in the equal probability of being selected to political office.
In a more concrete sense, however, sortition should produce an equality (or equivalence) of outcomes: the randomly selected sample should make the same decision that the entire adult population would have made, if every citizen had an equal opportunity to deliberate together. In more practical terms, the sample of people who happen to be chosen for a sortition legislature should produce the same laws that any such random sample would enact. It is this equivalence that makes a sortition legislature normatively desirable and from which it derives the most significant aspect of its legitimacy.
We want the randomly selected microcosm to simulate how any (or all) of us would decide policy matters under circumstances as close to ideal as possible.27 As political theorist John Dryzek explains, this “simulation claim” requires a minipublic to give “a simulation of what the population as a whole would decide if everyone were allowed to deliberate.”28 Since having every adult participate in a process of informed, respectful deliberation is, however, infeasible, the best alternative is to build institutions that simulate this ideal. Sortition does precisely this.
This simulation claim comes in two versions: strong and weak. The stronger version requires that a representative microcosm of society should, under deliberative conditions, replicate exactly the decisions of any comparable microcosm. This is probably impossible to achieve, although it is amenable to empirical test by running parallel sortition assemblies on the same topic.29
The weak—and more practical—version of the simulation claim holds that, within a given community, groups of descriptively representative, term-limited, and randomly selected people who deliberate (under conditions where good deliberation can occur) will approximate what the entire community would have decided under similar conditions, over time and across the broad array of issues. This weaker claim allows that on certain topics, at certain times, and under less than ideal conditions, a sortition assembly or minipublic’s decisions might deviate from what a similar group (or all of society) would have decided. Over time, however, and as the participants in an assembly are replaced and issues are revisited, such bodies should fulfill this weaker simulation criterion.
This weak version of the simulation claim is, to me, persuasive, but those who remain skeptical might appreciate an even weaker version. In this minimalist view, a descriptively representative microcosm of people need only get closer to the normative deliberative ideal than an elected body of politicians (even under the best possible electoral system) to be more legitimate and therefore desirable. For the reasons outlined earlier, there would likely be a large gulf between many important legislative decisions of politicians and those of a sortition legislature. This weakest simulation claim merely posits that the gap between what is normatively desirable and what we have at the moment would be narrowed significantly through sortition.
The Sortition Alternative
If our democratic ideal is popular sovereignty, and we recognize the impracticality of continual deliberative democracy on a mass scale, then sortition should outperform elections as a means of securing this ideal. We should select a representative sample of people to establish laws, while maintaining the larger accountability system secured through civil liberties, judicial oversight, social movements, and civil society more generally. Giving a sortition body legislative power (with rotation, time-limited terms of office, and a tried and tested deliberative decision-making process) would provide a robust test of at least the weaker versions of the simulation claim and thereby achieve—and increase—democratic equality and public legitimacy.
Though Gastil and Wright’s proposal for a sortition chamber heads admirably in this direction, it falls short by preserving a parallel elected chamber. Political parties are the dinosaurs of our times. Elected politicians are not trusted, and party membership has suffered a long-term collapse—whereas membership in civil-society organizations has flourished of late.30 Social movements and the mobilization of social forces do and can happen outside of political parties.
Gastil and Wright also defend the elected chamber as an important site of political bargaining. While it is true that bargaining may, in some contexts, be necessary, and can be a source of strength when otherwise disenfranchised social groups become organized, it is difficult to defend bargaining as an ideal mode of legislative decision-making, where the normative ideal of informed deliberation should predominate.31 Preserving an elected chamber preserves horse-trading, pork-barrel politics, and illusory zero-sum battles.
The capture and distortion of the political process by vested interests and the influence of money also undermines Gastil and Wright’s assertion that political parties are necessary for the development of political leadership. Electoral imperatives taint all party leaders, who compare poorly to leaders who emerge from civil society. Moreover, if levels of trust in our political representatives are indicative of satisfaction with the electoral system, the broader public has stated plainly its dissatisfaction with the status quo.
Fortunately, there is an alternative to electoral democracy, and recent examples of sortition-based policy making abound. Minipublics, in various forms and sizes, have a strong track record, at least in the primarily advisory charge given to them most often.32
The larger and more significant step from occasional minipublics to institutionalized sortition could take many forms. Sortition assemblies could be used as a tool for constitutional reform (see Fishkin’s and Arnold’s chapters in this volume), or they could become the local democratic norm if the G1000 democratic experiments continue their spread from Belgium to the Netherlands, Spain, the UK, and beyond. Or local governments experimenting with sortition, such as Toronto’s Planning Review Panel or some of Rotterdam’s area commissions, may inspire cities to take this method even farther.33 Sortition will probably continue to be used to inform contentious policy decisions, as in the Nuclear Fuel Cycle Engagement policy jury in South Australia, or the deliberations on same-sex marriage and the constitutional ban on abortion at the Irish Constitutional Convention and the Irish Citizens’ Assembly (see Arnold et al. in this volume).34
Or, as proposed by Gastil and Wright and others, sortition may be used to populate a second chamber in parliament or congress. Strategically, the first place to promote this could be at the subnational level, where a unicameral legislature exists amid strong public demand for a people’s chamber. One example is Scotland, where a group of organizations has recently proposed such an assembly to complement the Scottish Parliament.35 This could also happen in Australia, where the newDemocracy Foundation has a well-established record of policy decision-making using sortition and a mission to institutionalize sortition in government.36 In France a campaign has been launched to replace the French Senate with a sortition chamber, and French president Emmanuel Macron has proposed that presidents should be held to account by an annual sortition body.37 These examples show that sortition is a political idea with a promising future. Its modern rediscovery and resurgence is impressive.
Though a standalone sortition body should be our aim, for the reasons outlined above, the bicameral proposal of Gastil and Wright could be an important stepping stone toward unicameral sortition. An empowered sortition chamber would enable a direct comparison of the decision-making (and antics) of an elected chamber to that of a sortition chamber. If current research into sortition assemblies generalizes to legislative chambers, then the broader populace will have a higher level of trust and confidence in the outcomes of the sortition chamber.38 A move to then eliminate the elected chamber would initiate the demise of the electoral oligopoly experienced in many nations, the elimination of the aristocratic device of election, and the end of politicians as we now know them. Though the idea of ending elections may be controversial now, I am confident that after people witness a sortition chamber in action, few would be sad to see elected politicians go.