Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

The Deliberative Referendum: An Idea Whose Time has Come?

The Deliberative Referendum: An Idea Whose Time has Come? While deliberative citizens’ assemblies and plebiscitary referendums have long been perceived as antithetical, the idea of combining the two democratic instruments for better connecting administration and society has come to the fore in both theory and practice in more recent years. In this article, three ways of linking citizens’ assemblies to the referendum process are distinguished, exemplified, institutionally compared, and reflectively discussed. The three—the referendum-preparing, referendum-scrutinizing, and referendum-elaborating citizens’ assembly—come with their distinctive features, potential merits, scope limits, and related design questions. Fitting the “square peg of deliberative democracy” into the “round hole of direct democracy” and embedding hybrid design in diverging political systems are overarching challenges of institutional design. The article concludes that considering recent developments in theory and practice, the idea of a deliberative referendum linking citizens’ assemblies to direct voting on issues, seems an idea whose time has come, but also comes with challenges and questions that design thinkers and practitioners have only begun to tackle and answer. Keywords democratic innovation, direct democracy, deliberative democracy, referendum, citizens’ assembly, hybrid reform Tilburg University, Department of Public Law and Governance, The Netherlands Corresponding Author: Frank Hendriks, Tilburg University, Warandelaan 2, Tilburg, 5000 LE, The Netherlands. Email: f.hendriks@uvt.nl 2 Administration & Society 00(0) Small-Group Deliberation and Mass Voting: From Antithetical to Synergistic? In discussions about democratic innovations—instruments intended to forge effective connections between administration and society in democratic sys- tems—much attention is paid to two specific types of citizen participation and collective will formation: referendums, in which the whole electorate, the “maxi-public,” can vote directly on a policy issue, and citizens’ assemblies, in which an—ideally representative—sample of the population, a “mini-public” deliberates on a policy issue (e.g., Elstub & Escobar, 2019; Gastil & Richards, 2013; LeDuc, 2015). Neither innovations focused on direct voting in referen- dums nor those focused on extensive deliberations in mini-publics are ideal on their own (Geissel & Gherghina, 2016, p.88: Hendriks, 2019, p.448). By connecting such innovations, we might be able to create synergies between their relative merits, the benefits of which are potentially greater than the benefits of each separate innovation (McKay, 2019; Saward, 2001, p. 363; Tierney, 2013). A referendum proposal may for instance receive more sup- port and policy impact when a citizens’ assembly has been involved in designing, reviewing, or interpreting the proposition (Fishkin et al., 2015; Gastil et al., 2014, 2018; Suiter & Reidy, 2020). Vice versa, the addition of a referendum to a citizens’ assembly process may add focus to deliberations, and may help to prevent the deliberations of the citizens’ assembly’s disap- pear into thin air (e.g., Farrell et al., 2019, 2020; Fournier et al., 2011). The idea of linking citizens’ assemblies and referendums may seem an idea whose time has come, but it also comes with questions and reservations. The referendum and the citizens’ s assembly emanate from two very different strands of democratic theory and practice, and have long been perceived as antithetical (LeDuc, 2015, p. 139; El-Wakil, 2017, p. 59; Landemore, 2018, p. 320; Parkinson, 2020, p. 485); with the mini-public founded on a belief in patient small-scale deliberation, and the referendum geared at swift aggrega- tion of mass votes. The domain of referendums and other plebiscitary for- mats seems one of “thinking fast”—reflexively, quickly translating individual inclinations to a collective signal—whereas the domain of citizen’s assem- blies and other deliberative formats seems one of “thinking slow”—reflec- tively, patiently weaving a rug of shared meaning (Kahneman, 2011). “The twin objectives of voice and votes too often pull in opposite directions,” as LeDuc (2015: 147) explains. And then the question is indeed, as he nicely put it: “Can the square peg of deliberative democratic theory be pounded into the round hole of direct democracy?” In trying to answer this question construc- tively, Leduc is among a slowly but surely growing group of theorists who explore the scope for integrating deliberation in direct voting, either through Hendriks and Wagenaar 3 investments in public reflection in general (e.g., via information brochures for prospective voters), or through more integrative views on the sequencing of direct voting and reflective deliberation (Altman, 2014; Fishkin et al., 2015; Landemore, 2018; LeDuc, 2015; Levy, 2013; McKay, 2018, 2019; Parkinson, 2020; Saward, 2001; Setälä, 2017; Tierney, 2013). In thinking about democratic innovation, two opposite tendencies can be discerned: mixophobia (fear of institutional pollution and a related preference for pure models) on the one hand, and heterophilia (care for the institutionally different and a related interest in hybrid innovations) on the other hand (Hendriks, 2021; Saward, 2021). In this article, we pick up on the latter, explor- ing more integrative views on the linking of referendums and deliberative mini- publics specifically. Other hybrids of small-group deliberation and large-group voting are feasible and emerging in democratic governance, such as hybrids of participatory budgeting that start with deliberations in small-group settings and are completed with digital voting among the wider involved public that did not participate in the deliberative stage (e.g., Laruelle, 2021; Miller et al., 2019). However, these other democratic hybrids are beyond the scope of this article, which focuses on what can be called “deliberative referendums.” The referen- dum component in the hybrid can be of all common varieties: bottom-up, top- down, or mandatory; proactive or reactive; binding or advisory (Altman, 2011; Morel & Qvortrup, 2017). It always involves secret and direct voting of a maxi- public, a general electorate, on issues of public concern (e.g., open voting in town meetings or direct voting on politicians in recalls are not included here). The hybrid’s deliberative component can be a mini-public of all customary sizes (20–350 random participants) and numbers of meetings (1–20 days) (Bächtiger et al., 2018; OECD, 2020). It always concerns randomly selected, demographically stratified, samples of wider populations deliberating on issues of public concern. Denominators for this have proliferated, commonly starting with “deliberative” or ‘citizens,” and then adding a term such as “assembly,” “council,” “jury,” “poll,” “conference,” “dialog.” As a cover-all concept on this side of the hybrid we opt for the citizen’s assembly, which is both recognizable and often used in this context. In this article, we distinguish three broad types of combining and sequenc- ing citizens’ assemblies and referendums. We employ two criteria regarding the link between the citizens’ assembly and the referendum in these models: (1) the deliberative efforts of the assembly relate directly to the ballot proposal(s) in the referendum and (2) the combination of citizens’ assembly and referendum is not merely hypothetical but has actually been applied in practice. This means that purely hypothetical combinations, never taken to practice are excluded; as are citizens’ assemblies which broadly define priori- ties or topics of interest without a direct link to referendum voting. 4 Administration & Society 00(0) Even though interest in hybrid models of democratic innovation is grow- ing, systematic research on them has thus far been limited, focused on single case studies (e.g., Farrell et al., 2020), or other subsets of hybridization (Sørensen & Torfing, 2019). More research is needed, conceptually as well as empirically, to improve our theoretical and practical understanding of how linking elements of deliberation and aggregation could be achieved, what could be expected from this, and which challenges and questions are bound to arise. With this article, we contribute to the research endeavor by elaborat- ing on an empirically-grounded categorization of key variants, in full realiza- tion that we are closer to the start than to the end of it. Following this introduction, in the second section of the article we introduce and discuss three models which directly link referendums to deliberative citizens’ assem- blies, providing a brief example of an actual case for each. In section three, we further reflect on the comparative characteristics, merits, and vulnerabili- ties of the three types, as well as on related institutional design questions. Section four wraps up with closing reflections. Linking Citizens’ Assemblies to Referendums: Three Variants and Sequences As indicated, we focus on purposive deliberative additions to referendum processes of a specific type: deliberative mini-publics, or in common par- lance citizen’s assemblies. The purpose of the deliberative addition can be to prepare a referendum proposal, to scrutinize a referendum proposal or to elaborate a referendum result. In each variant, it is conceivable that the pro- posal (or proposition) is multiple rather than singular (Wagenaar, 2021)— although the latter, singular, is far more common. We distinguish the following three types of citizens’ assemblies combined with referendums: • Preparatory citizens’ assemblies Example: Irish Citizens’ Assembly (2016–2018) • Scrutinizing citizens’ assemblies Example: Oregon Citizens’ Initiative Review (since 2010) • Elaborating citizens’ assemblies Example: British Citizens’ Assembly on Brexit (2017) Each of these deliberative assemblies fulfils a role in a different stage of the referendum process (see Figure 1). Preparatory assemblies prepare a ref- erendum proposal prior to the establishment of the referendum ballot. Ideally, political commitment to put the prepared proposal to voters in a referendum is established prior to the commencement of the assembly’s work. Scrutinizing Hendriks and Wagenaar 5 Figure 1. Three variants of citizens’ assemblies linked to the referendum process. assemblies are established after a referendum question has been decided for the ballot. A scrutinizing assembly’s core task is to dissect the referendum question and ballot alternatives, and to provide voters with objective infor- mation, considerations and, optionally, voting advice. Elaborating assemblies are established after the referendum vote has taken place and the results have been announced. The purpose of an elaborating assembly is to interpret the referendum result and to provide input for follow-up steps or policies in view of the referendum outcome. In the following three sub-sections, we discuss each of these three types in more detail and we look at various empirical manifestations. Our classification builds on and extends the three-stage distinction— Initiation stage, Campaign stage, Implementation stage—that McKay (2019) proposed. We add the systematic comparison of the resulting hybrids in terms of characteristics, merits and related design questions. The classification par- tially builds on and partially deviates from Gastil and Richards (2013), who distinguished five deliberative designs—Priority Conference, Design Panel, Citizen’s Assembly, Citizens’ Initiative Review, Policy Jury—that can con- tribute to successive stages of the direct-democracy process. Their fifth and most downstream model—the autonomous Policy Jury, that develops direct legislation, dispensing with conventional legislative and electoral pro- cesses—is purely hypothetical and thus missing from our categorization of applied models. Our most downstream variant—the elaborating assembly that follows after the referendum vote—is missing in the classification of 6 Administration & Society 00(0) Gastil and Richards, which is more than ours focused on the real and specific problems of the bottom-up referendum and initiative process in parts of the US. Their three upstream models are collapsed into one variant in our clas- sification: the preparatory assembly, which can perform different tasks in mandatory, top-down, and bottom-up referendums (including the specific tasks of the Priority Conference and the Design Panel in bottom-up, peti- tioned processes). The Priority Conference only relates to our first model insofar as it leads to a concrete proposal for mass voting in a referendum; if it leads to another, more specialized deliberative forum taking over the baton, without the prospect of a follow-up referendum, the practice concerned falls outside the scope of this overview of the deliberative referendum. Their fourth model, the Citizens’ Initiative Review, overlaps with our second vari- ant, the Scrutinizing Assembly. The referendum-preparing citizens’ assembly The referendum-preparing assembly brings together a statistically represen- tative group of citizens in a controlled environment to discuss a policy issue, and to prepare a referendum proposal which thereafter can be presented to the wider electorate. In all events, assembly members are informed about the policy issue by experts after which they deliberate on conceivable policy options among themselves. Deliberating as a mini-public, the preparatory assembly paves the way for a referendum vote by the “maxi-public,” result- ing in either an advisory or binding public decision. The eventual formal trigger of the referendum may be governmental or parliamentary, but the proposition in this variant clearly comes from the citi- zens’ assembly. In cases where a referendum is not immediately foreseen, the recommendation to delegate—to “refer”—the outcome of the deliberations to a referendum vote can also form part of the deliberations. Approval by the maxi-public in a referendum may further legitimize the findings of the delib- erative mini-public in the public eye (Farrell & Suiter, 2019; Farrell et al., 2020; Parkinson, 2020; Suiter & Reuchamps, 2016: 9). A well-known example of a preparatory citizens’ assembly is the Irish Citizens’ Assembly, which was instituted in 2016 with the intention to delib- erate on a set of topics (abortion, the challenges and opportunities of an aging population, fixed-term parliaments, the manner in which referenda are held, and how the state can make Ireland a leader in tackling climate change). The Irish process that led to an authoritative decision on abortion—deliberated by the Citizens’ Assembly in 2016, Citizens’ Assembly in 2017, and ultimately approved per referendum in May 2018, is discussed in box 1. An earlier example is the British Columbia Citizens’ Assembly on Electoral Reform in Hendriks and Wagenaar 7 Box 1. Irish Citizens’ Assembly on abortion legislation. In 2016, following an electoral promise by political party Fine Gael, the Irish parliament voted to institute the Irish Citizens’ Assembly. A total of 99 randomly-selected citizens were brought together to discuss issues under the moderation of an independent chair appointed by the government. Selection of the participants was conducted using stratified random sampling taking into account age, gender, residency and social class. A total of 99 reserve candidates were also recruited (Farrell et al., 2019, 2020; Suiter & Reidy, 2020). The citizens’ assembly deliberated on various constitutional issues over a total of 12 weekends, five of which were dedicated to the eighth amendment of the constitution which regulated abortion. For each topic, the process was divided into several process steps. In a first stage (1), the topic was introduced and participants listened to presentations by experts and interest groups. In a second stage (2), participants discussed the issue more in-depth at round- table discussions. Professional moderators lead the discussions and encouraged participants to work toward concrete recommendations. At the end of the discussions, participants voted on the recommendations among themselves as a mini-public. All meetings were broadcasted live via a video stream. The wider public could also submit inputs, 13,000 of which were received on the abortion issue. Discussions on various cases and contingencies resulted in a sequence of internal votes that clearly showed that the participants in the Citizens’ Assembly recommended making legal abortion on request (within 12 weeks) available to a greater extent than previously thought possible in Ireland. The results, including a description of the legislative redrafting most preferred by the assembly, were cumulated into a final report which was sent to the government (Citizens’ Assembly, 2017). Since referendums are compulsory for constitutional changes in Ireland, it was always foreseen that the wider electorate would have to agree with the proposed changes, which it did on 25 May 2018 with a clear majority; 66.4% of the electorate voted in agreement with the Citizen’s Assembly’s proposal. This paved the way for changing the constitution, enabling the Irish parliament to draw up legislation which expanded rights to abortion. Canada, which was instituted in 2004 to debate whether a new electoral sys- tem for the province was desirable and if so, which system would be recom- mendable. The assembly’s proposal was presented to voters in a referendum in May 2005, receiving 57.5% majority support, but failing to meet the 60% support threshold (Warren & Pearse, 2008). Another example is the 2011 Icelandic Constitutional Assembly, which drafted a new constitution. The broader electorate voted in a referendum in October 2012 on six elements of the new constitution, all of which were approved per referendum, although parliament has stalled further decisions on the new constitution ever since (Landemore, 2020). 8 Administration & Society 00(0) The Irish case diverts from the Icelandic and British Columbia cases in its institutional and political embedding. In Iceland the political realm was largely detached from the process by a Constitutional Council consisting of citizens that “refused, in a self-defeating way, to cooperate with parties and other political elites (. . .) losing the good will of the political class” (Landemore, 2020, pp.120, 181). In British Columbia, the Citizens’ Assembly recommendation did not pass the referendum’s supermajority requirement, elevated high by weary politicians (Tierney, 2013; Warren & Pearse, 2008). In the Irish attempt at “systematizing constitutional delibera- tion” (Farrell et al., 2019) a more productive division of responsibilities between citizens’ assembly, representative politics and electorate has come to fruition. It is no coincidence that it is considered a “world leader in the linking of deliberative democracy (mini-publics) and direct democracy (referendums),” a paragon of “how deliberation can be inserted into the referendum process in a meaningful way” (Farrell et al., 2019, pp. 113, 119–120; also Farrell et al., 2020). The Referendum-Scrutinizing Citizens’ Assembly The referendum-scrutinizing citizens’ assembly embarks on its task after a referendum has been triggered and the referendum question has been decided, but before the vote takes place. The function of the scrutinizing assembly is to reflect on the content and consequences of the ballot options and to inform and enlighten the wider public on the matter before voting takes place. The integration and diffusion of arguments as prepared by the assembly can encourage more meaningful societal debate on the issue by the wider public (Jacquet & van der Does, 2021). A statement by the assembly is distributed to all eligible voters, including arguments for both sides. Deliberations need not, but may, culminate in a voting advice, which does not have to be unani- mously supported within the assembly. The distribution of assembly partici- pants backing either side is commonly reported. While scrutinizing citizens’ assemblies could take various forms, the most developed example is the citizens’ initiative review (CIR), originating in the US state of Oregon in 2010 (Gastil et al., 2014, 2018; Knobloch et al., 2014). In the CIR-model, after a referendum on a popular initiative has been announced, a representative group of 24 citizens is elected to take part in the deliberations. The first stage is an orientation stage to get acquainted with the topic of the initiative. In the next stage, participants hear experts on the topic as well as representatives of the pro and con sides. After these informative stages, the deliberation stage starts, in which participants discuss the infor- mation and arguments in moderated discussion groups. The assembly then Hendriks and Wagenaar 9 Box 2. Oregon Citizens’ Initiative Review on Measure 85. In 2012, a citizens’ initiative on Measure 85 qualified for the ballot in Oregon. The initiators intended to reform the corporate tax kicker in order to channel tax surpluses to public education rather than to companies. As stipulated in Oregon state law, a Citizens’ Initiative Review was established to deliberate the issue and present its voting advice. A sample of 10,000 randomly selected citizens was approached, and out of those that were willing to participate, a representative group of 24 citizens was selected to take part. In the first stage (1), the orientation stage, participants attended hearings in which proponents and opponents elaborated on their arguments. In the next stage (2), participants invited various experts to provide background information to the topic. In the final stage (3), participants deliberated their new insights among themselves with two goals: first, to draw up an overview of arguments in favor and against the initiative, and second, to formulate a voting advice. Both were included in the citizens’ statement which was sent to all eligible voters prior to the referendum vote (Knobloch et al., 2012). In the statement on Measure 85, 19 out of 24 participants supported the initiative (State of Oregon, 2012). In the following referendum, on 6 November 2012, a majority of 59.9% of voters also supported the initiative. Research concluded that two-thirds of voters considered the advice of the Citizens’ Initiative Review to be a useful source of information (Knobloch et al., 2013). draws up its citizens’ statement in which it highlights the most important findings and considerations. The statement is distributed to the electorate as a voter information pamphlet. Citizens’ initiative reviews have been used for state-wide citizen initiatives in Oregon since 2010, following a pilot. Since 2011 they are a legally formal- ized element of each citizen initiative trajectory. Box 2 describes an example case of an Oregon CIR in 2012. In the case of multiple citizens’ initiatives subjected to a referendum vote on the same day, parallel assemblies can be organized. Extensive research on the CIR process concludes that participants obtain issue-specific knowledge and are open to revising their options after reasoning with each other. Moreover, the information pamphlet helps the wider public to overcome information deficiencies and to base their vote on the same information and argumentation as established by the assembly par- ticipants (Gastil & Knobloch, 2020). In more recent years, the citizens’ initia- tive review has been experimented with in various other locations, such as the US states of Massachusetts, Colorado, California and Arizona, the Swiss can- tons of Sion and Geneva, and the municipality of Korsholm in Finland (Setälä et al., 2020). The Finnish experiment concluded that voters considered the assembly’s advice to be trustworthy and useful, and that the advice improved their factual knowledge, issue efficacy, and perspective-taking. 10 Administration & Society 00(0) The Referendum-Elaborating Citizens’ Assembly The elaborating citizens’ assembly is instituted after a referendum has taken place and follow-up steps must be decided on. Its goal is neither to prepare and propose a referendum question nor to inform voters, but rather to make sense of the referendum result and its implications in terms of operationaliza- tion prior to the start of the implementation process by policymakers. For reactive (“veto”) referendums, more specifically, a follow-up citizens’ assem- bly could shed further light on the reasons behind a majority rejection of new legislation and on amendments or alterations that might solve resistance. For proactive (“advancing”) referendums, it could help to fine-tune the opera- tionalization of newly accepted legislation from a citizens’ perspective. To some observers, a citizens’ assembly coming after the fact of a referen- dum may seem illogical, as the wider electorate has decided on an issue, which then seems to be delegated back to a smaller subset of the population. Yet, an elaborating assembly could fulfill a relatively modest but useful sup- port function in the stage directly following the referendum vote. Representative politics is commonly and overall responsible here, but might very well use citizens’ support for hammering out the best possible operationalization strat- egy, especially when a referendum result is open to different ways of follow- ing up or when it is prone to politically strategic interpretations of the referendum outcome which may run counter to the preferences of the societal majority. Investing in a citizens’ assembly the task of indicating the most- supported operationalization in line with the referendum outcome could serve to do justice to the underlying intentions of referendum voters. There are few practical examples of referendum-elaborating citizens’ assemblies. The best-known is the Citizens’ Assembly on Brexit, a pilot proj- ect held in the autumn of 2017. This case does not carry the weight of the previous two exemplary cases, but nevertheless provides an empirical opera- tionalization of linking deliberation to referendum voting. Box 3 describes the process of this assembly. Discussion: Distinctive Features, Institutional Implications and Design Questions In the previous section, we outlined three hybrid models that connect a delib- erative citizen’s assembly to a referendum vote. In Table 1, we summarize the main characteristics of the three. In case of the preparatory assembly, work- ing ahead of the referendum, the citizen’s assembly itself can be considered the leading instrument of democratic innovation. Deliberation serves to shape the proposal(s) of the mini-public and the subsequent referendum serves to Hendriks and Wagenaar 11 Box 3. British Citizens’ Assembly on Brexit. The Citizens’ Assembly on Brexit took place during two weekends in the autumn 2017. The 2016 referendum on EU membership had resulted in a majority preference to leave the EU (52% vs. 48% to remain in the EU), but there appeared to be no consensus over the exact shape that the relationship between the United Kingdom and the EU should take after the withdrawal (Offe, 2017; Renwick et al., 2018). A coalition of scientists and societal organizations thus decided to organize a citizens’ assembly on the topic. Its goal was explicitly not to challenge the referendum result, but to get insights into informed public opinion on what the position of the United Kingdom outside of the EU could look like (Renwick & McKee, 2017; Renwick et al., 2018). To select the participants, 5,000 citizens were approached. Of those willing to participate, participants were recruited such that they would be representative of the British population in terms of age, gender, ethnicity, social status, residency and voting behavior at the Brexit referendum. A group of 50 participants attended both weekends of deliberation. Assembly members went through two stages: (1) a learning stage, in which experts and other speakers elaborated on the topic, and (2) a deliberation stage, in which the participants discussed the topic in professionally moderated panels. Various alternative scenarios were distilled, which were voted on by all participants (Renwick et al., 2018). A clear majority of participants preferred an extensive trade deal with the EU over membership of the internal market, and both options were preferred over a no-deal Brexit. These preferences diverged from the preferences of parliamentary fractions at the time (Renwick et al., 2017, 2018). Perhaps because of its informal status, the policy and societal impact of the citizens’ assembly appeared to be limited. However, according to Renwick et al. (2018), the pilot proved that a citizens’ assembly tasked to elaborate a referendum’s general signal can produce a meaningful advice. further legitimize a proposal by submitting it to “market testing” among the maxi-public. For the other two types of hybrids, the referendum is the leading instrument, and the assembly serves to improve specific parts of the referen- dum process. The scrutinizing assembly serves to inform referendum voters prior to making their vote choice. The elaborating assembly is designed to alleviate the sometimes difficult task of understanding and translating the referendum outcome in terms of follow-up courses of action. The three ways of linking deliberation in citizens’ assemblies to voting in popular referendums each have their specific merits and limitations, as sum- marized in Table 1. For all three types, the desired effect is that deliberation improves the alignment of policies with societal preferences, either through designing a suitable referendum proposition, stimulating informed debate or 12 Table 1. Three types of linking deliberative citizens’ assemblies to referendums: central characteristics, potential (dis)advantages and related design questions. Referendum-preparing citizens’ Referendum-scrutinizing citizens’ Referendum-elaborating citizens’ assembly assembly assembly Timing Before referendum triggering After referendum triggering and After referendum vote before referendum vote Main contribution Exploring feasible and justified Informing voters on the content and Interpreting the referendum result of the assembly propositions fit for public choice implications of referendum options and considering ways forward per referendum (possibly adding non-binding voter advice) Main contribution Bringing collective closure, popularly Deciding the issue through maxi- Indicating the general direction, of the referendum supported, on assembly proposal public choice, following mini-public popularly supported, to further scrutiny of choice options elaborate Leading instrument Assembly Referendum Referendum Potential merits Improved alignment of policy decisions with societal preferences as expressed by mini-public and maxi-public in tandem Retaining equality in mass voting while reducing the rupture of equality through sortition Bypass for public issues that Clear division of labor between Interpretation of referendum result representative politics can’t or referendum initiators, assembly drawing on voter intentions rather won’t touch members and the wider electorate than political strategy Deliberation on multiple scenarios Demarcated, tractable task for Implementation sensitive to citizen rather than one specific proposal deliberative assembly perspective Mini-public proposition market- Evenly informed voting at Exploring scope for redress in case tested by maxi-public referendum of popular veto (continued) 13 Table 1. (continued) Referendum-preparing citizens’ Referendum-scrutinizing citizens’ Referendum-elaborating citizens’ assembly assembly assembly Potential Expanded actions and costs related to public consultation vulnerabilities Contradictory motivations and passions under one design Bypassing politics can slide No opportunities to present Counter-intuitive to let mini-public into offloading policymaking alternative ballot options or in when maxi-public has spoken responsibility, accountability, risk further solutions Entertaining multiple policy routes is Voting advice may be perceived and Risk of neglect or scorn from complicated and risks loss of focus resisted as paternalistic toward political realm and clarity electorate Central design Matching assembly design to referendum design questions Embedding hybrid design in the political-cultural context How to channel open, convoluted How to align assemblies’ balanced How to align advisory deliberation deliberation into closed, targeted option dissections with efforts of with political follow-up voting on propositions mass media and public institutions responsibilities How to optimize political bind and How to credibly insert the (small) How to prevent distortion, remain precommitment to the process jury format into democratic aligned with and organize feedback settings unfamiliar with it to the maxi-public How to extend and not undermine How to justify modest How to match elaboration task to already existing citizens’ rights to supplementary functions vis-à-vis type of referendum signal petition for a referendum added costs and efforts 14 Administration & Society 00(0) bringing the follow-up of referendum results in line with public intentions (Farrell et al., 2020; Gastil & Richards, 2013; Renwick et al., 2018). Furthermore, by involving a representative subset of the voting population, particularly when sortition procedures are applied, citizens’ assemblies can theoretically reduce the rupture of equality compared to propositions initiated by citizens with unequal financial and political resources (Fishkin et al., 2015). Nonetheless, the participation of usual suspects can still be an issue in the recruitment of assembly members (Jacquet, 2017). The referendum-preparing citizens’ assembly provides the strongest case for sequencing democratic innovations, as it has proved fruitful for legislat- ing on contentious issues (Farrell & Suiter, 2019; Farrell et al., 2019, 2020). The scrutinizing assembly, however, has also proven to be effective as an addition to the initiative process, particularly in, but not necessarily limited to, the US context (Gastil et al., 2014, 2018). The elaborating assembly fol- lowing referendum voting is the least battle-tested of the three types, although Renwick et al. (2017, 2018) have shown that a meaningful procedure and process result can be achieved. An elaborating assembly could have proven fruitful in other post-referendum situations, like the one resulting from an advisory corrective referendum in the Netherlands in 2018 in which a major- ity of voters vetoed stricter regulation of the secret services, without provid- ing sufficient guidance for necessary follow-up choices (Jacobs, 2018; Wagenaar, 2019). In terms of vulnerabilities, the elaborating assembly risks being seen as too late in the day; the referendum campaign and vote have already passed, and the results are in. Relatedly this variant risks being scorned or outright neglected by the political actors that have a role to play in the aftermath of a referendum—particularly when the assembly can be perceived or framed as a complicating factor rather than a helping hand in the process. Hence, one of the key design questions here is how to align a credible, advisory role of an elaborating assembly with a legitimate, broader responsibility of representa- tive politics to act on referendum results. A second obvious design question here is how to prevent the deliberative mini-public from distorting rather than elaborating the popular signal received per referendum; some form of public reporting in line with and in terms of the maxi-public’s revealed preference seems desirable and also feasible in combination with advisory reporting to parliament. A third design question follows from the fact that different refer- endums result in different public signals—advisory versus binding, proactive versus reactive—which demand diverging elaboration efforts that need to be designed accordingly. The scrutinizing citizens’ assembly is comparatively handicapped in that it can only dissect the given choice options, and not present alternative Hendriks and Wagenaar 15 options or further solutions. In case of supplementary voting advice, this vari- ant is vulnerable to the criticism of, and resistance to, paternalism, with a small group (the Oregon CIR e.g., assembles 24 citizens) presenting guid- ance to a full electorate of citizens which is seemingly not fully trusted. The latter has been parried in the design of the Oregon CIR, which stresses the relatively modest, supplementary, non-committal, service-to-the-voter char- acter of the assembly recommendation. As indicated, it is possible to refrain from giving voter advice and focus solely on dissecting choice options. In this baseline scenario it is imperative to clarify how the role of the citizens’ assembly in providing balanced information is different from and supple- mentary to the role of mass media and public institutions that inform the public. Another design question is if and how the (small) jury format, familiar to the American state context, can be inserted credibly into democratic set- tings not familiar with this format. When a referendum-scrutinizing assembly was discussed in the Netherlands, in the province of Limburg, a small scruti- nizing jury was perceived as lacking legitimacy, and a larger scrutinizing assembly as potentially more trustworthy—but also more costly. The referendum-preparing assembly may be introduced as an effective bypass for political stalemate, but it can also turn into a way of offloading policymaking responsibility, accountability and risk—a political strategy of non-decision-making. This danger has been noted for the Irish case. Another vulnerability follows from the fact that preparatory assemblies not seldom receive relatively open, ill-structured assignments, which are not only hard to grasp in a relatively short timeframe but also difficult to turn into voting propositions with focus and clarity. Hence, a first design question here is how to maintain a level of openness that fits the preparatory stage of assemblies working toward proposals, while also maintaining a degree of focus that fits the need for clearly targeted propositions in the stage of referendum voting. A second design question here is how to organize enough political bind and precommitment to the process, in order to motivate the efforts and costs of participation. Thirdly, a specific design question for the referendum-prepar- ing citizens’ assembly is how to extend—and not undermine—already exist- ing rights of (groups of) citizen’s to self-organize, and petition in a procedure for bottom-up referendums. How to match assembly choices (type of sortition, organization and dura- tion of deliberation) to referendum variants (bottom-up versus top-down and mandatory, proactive versus reactive, binding versus advisory) and how to make this work in a political-cultural context are overarching questions of hybrid design. They pertain to referendum-preparing, -scrutinizing and -elab- orating assemblies, which also have their own design questions. What we have done here is tease out central design questions, but we have not answered 16 Administration & Society 00(0) or solved them. This is not only because of space limitations and focusing choices, but also and more fundamentally because our collective knowledge on this is still very much fledgling—growing for particular cases of hybrid design, but largely lacking when it comes to overarching insights. Further developing comparative, cross-case knowledge about the nuts and bolts of hybrid design is an important next step to take if the general idea of hybrids connecting the best, while preventing the worst, characteristics of delibera- tive assemblies and plebiscitary referendums is indeed an idea whose time has come. From Occasional Nod to Systemic Attention? Closing Reflections Not so long ago, integrative views of voting and deliberation received only occasional nods in theories of deliberative democracy interested in connect- ing administration and society in more reflective ways (Levy, 2013, p. 573). This has changed under the influence of the systemic turn in deliberative democracy, which has encouraged theorizing about deliberative interventions working in tandem with non-deliberative, including plebiscitary, institutions (El-Wakil, 2017; Landemore, 2018; Parkinson, 2020; Parkinson & Mansbridge, 2012; Suiter & Reidy, 2020). Approaching the matter from the side of referendums, theorists such as LeDuc (2015), Setälä (2018), and McKay (2019) have shown a vivid interest in the addition of deliberative reflection to referendum voting. Prominently, Tierney (2013) has argued that a deliberative referendum could further the involvement of citizens in ques- tion formulation, while addressing some of the shortcomings of the referen- dum process, notably the barrier to meaningful deliberation resulting from the focus on aggregating established opinions, but also the issues of elite manipulation and majority tyranny. Slowly but surely, deliberative assem- blies linked to referendums have also sprung up in practice, motivated by the urge to forge more versatile connections between administration and society. In this article, we have further examined the idea of the deliberative referen- dum, its emerging functional variants, their potential merits, vulnerabilities, and related design questions in particular. The result speaks to debates in public administration and related disciplines concerned with democratic instruments and innovations that are meant to forge more effective connec- tions between administration and society in democratic systems. Considering developments in both theory and practice, as we have done here, the deliberative referendum seems an idea whose time has come; it comes, however, also with challenges and questions that theorists and Hendriks and Wagenaar 17 practitioners of democratic innovation have only begun to tackle and answer. The common idea behind the three general, empirically grounded, models distinguished in this article is that combining the use of referendum voting and deliberative citizens’ assemblies can bolster the merit and correct the vulnerability of both instruments of democratic innovation. In a single instru- ment, either deliberative or plebiscitary, the associated vulnerabilities may prevail, while in a hybrid construction the combined forces may lead to an overall more advantageous situation. Variations in empirical reality denote that this is not easily accomplished. A hybrid design that may work compara- tively well in one democratic setting does not automatically fit another con- text with different structural and cultural features. Each hybrid design involving a deliberative assembly and a plebiscitary referendum comes with particular design questions, explored in this article and distinguished as items for more specific design thinking, experimentation, and research-for-design. An intrinsic difficulty of hybrid design harks back to the discussion that we started with. Deliberative mini-publics and plebiscitary referendums are associated with separate discourses and practices with divergent motivational bases (Knight & Johnson, 1994; LeDuc, 2015; Parkinson, 2020). People who are passionate about deliberative mini-publics are not seldom suspicious of plebiscitary referendums, and the other way around; reaching out to the other side does not come naturally in this field. The challenge of hybrid design may thus get more than twice as tough. It is not only that two components need to be fitted; it is also that at least one of two tends to be approached with reserve and suspicion; “Why add an unruly referendum process, if you can have a safely embedded mini-public?”; “Why add a mini-public, can we not just trust the encompassing demos?” From diverging vantage points in, respec- tively, deliberative and plebiscitary democratic thought such questions are comprehensible. Such reservations are bracketed in emerging experiments with hybrid designs like the ones that we explored, and in more recent schol- arly theorizing that approaches the assembly-plus-referendum as at least a promising hypothesis worthy of testing. Further development and testing of the proposition is work to be done. Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. Funding The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research for this article was partly financed by the Dutch National Science Agenda, NWO file number NWA.1292.19.048. 18 Administration & Society 00(0) ORCID iD Frank Hendriks https://orcid.org/0000-0001-7393-4144 Notes 1. For a more general, philosophical argument about the tension between delibera- tive democracy based on mini-publics and mass participation in various guises, among which referendums, see Lafont (2020). 2. Also beyond the scope of this article are hybrid innovations discussed in the wider field of public and private organizations; see a.o. Billis (2010), Secinaro et al. (2019). 3. And under this cover-all concept, we may refer to specific concepts, used in spe- cific settings, such as the Citizen’s Initiative Review as developed in the context of Oregon, US (see our second model). 4. These may be interesting innovations, which empower citizens to set or adjust the policy agenda. However, when a direct link between a citizens’ assembly and a referendum vote is lacking such innovations are excluded here. 5. A notable example is the Ostbelgien Citizens’ Council, a permanent body of citizens selected by lot, tasked with selecting issues for further debate by a sec- ond—temporary and issue-specific—deliberative assembly, also composed of citizens selected by lot. This Citizens’ Assembly deliberates and formulates concrete policy recommendations (Niessen & Reuchamps, 2019). However, as these recommendations are not directly subjected to a referendum vote we do not include this assembly in our analysis based on our selection criteria. The crite- rion of a direct link between citizens’ assembly and referendum proposal means that citizens’ assemblies only tasked with agenda-setting in a broader sense (as meant by e.g., Dahl, 1989), are excluded from our overview. 6. In specific cases, the assembly may consider whether one or several referendum questions are desirable, and which particular alternatives and voting structure would be most suitable for the purpose. A special variant, using the specific instrument of the “Deliberative Poll” as developed by Fishkin and collaborators, was experimented with in 2011, within the specific context of the Californian initiative process. Related to a project called “What’s Next California” a delib- erative-polling process was started to discuss and consider 30 proposals from a broad coalition of reform groups. This ultimately led to a proposition, Proposition 31, which was ultimately not successful at election time (Fishkin et al., 2015). 7. The Citizens’ Assembly, First Report and Recommendations of the Citizens’ Assembly, The 8th Amendment of the Constitution, 29 June 2017. 8. Following input from participants, situations considered were expanded with contingencies such as: real and significant physical risks to the woman’s life; real and substantial risk to a woman’s life from suicide; serious risk to the woman’s physical health; serious risk to the woman’s mental health; pregnancy as a result of rape; fetal abnormality likely to result in death before or shortly after birth; significant fetal abnormality unlikely to result in death before or shortly after birth; socio-economic reasons; no statement of reasons. Hendriks and Wagenaar 19 9. As outlined in the introduction, we focus on deliberative assemblies whose efforts directly relate to the content of the proposal subjected to a referendum vote. Therefore, we limit our focus on the referendum-elaborating citizens’ assembly to the steps of interpreting the referendum outcome and determining the most suitable policy course. Further along the policy trajectory, other types of deliberative assemblies could be plausible, modeled for example after the imple- mentation committees that monitor the implementation of participatory budget- ing outcomes (see e.g., Wampler, 2007). To the best of our knowledge, there have so far not been any empirical examples of citizen committees overseeing the implementation of referendum outcomes; this means they also do not meet our second inclusion criterion. 10. Personal correspondence David Farrell. See also D. Farrell, We may have overdone it on citizens’ assemblies, Irish Times, 16 February 2022 https://www.irishtimes. com/opinion/we-may-have-overdone-it-on-citizens-assemblies-1.4803375 References Altman, D. (2011). Direct democracy worldwide. Cambridge University Press. Altman, D. (2014). Strengthening democratic quality: Reactive deliberation in the context of direct democracy. Helen Kellogg Institute for International Studies. Working Paper No. 400. Bächtiger, A., Dryzek, J. S., Mansbridge, J., & Warren, M. E. (2018). The Oxford Handbook of Deliberative Democracy. Oxford University Press. Billis, D. (2010). Hybrid organizations and the third sector: Challenges for practice, theory and policy. Macmillan International Higher Education. Citizens’ Assembly. (2017). First Report and Recommendations of the Citizens’ Assembly: The eighth amendment of the constitution. Dahl, R. A. (1989). Democracy and its critics. Yale University Press. Elstub, S., & Escobar, O. (Eds.). (2019). Handbook of Democratic Innovation and Governance, Aldershot: Edward Elgar. El-Wakil, A. (2017). The deliberative potential of facultative referendums: Procedure and substance in direct democracy. Democratic Theory, 4(1), 59–78. Farrell, D. M., Suiter, J., & Harris, C. (2019). ‘systematizing’ constitutional delibera- tion: The 2016–18 citizens’ assembly in Ireland. Irish Political Studies, 34(1), 113–123. Farrell, D. M., & Suiter, J. (2019). Reimagining Democracy: Lessons in deliberative democracy from the Irish Front Line. Cornell University Press. Farrell, D. M., Suiter, J., Cunningham, K., & Harris, C. (2020). When mini-publics and Maxi-publics coincide: Ireland’s National Debate on Abortion. Representation. Advance online publication. https://doi.org/10.1080/00344893.2020.1804441 Fishkin, J., Kousser, T., Luskin, R. C., & Siu, A. (2015). Deliberative agenda set- ting: Piloting reform of Direct Democracy in California. Perspectives on Politics, 13(4), 1030–1042. Fournier, P., van der Kolk, H., Carty, R. K., & Blais, A. (2011). When citizens decide: Lessons from citizens assemblies on electoral reform. Oxford University Press. 20 Administration & Society 00(0) Gastil, J., & Knobloch, K. R. (2020). Hope for democracy: How citizens can bring reason back into politics. Oxford University Press. Gastil, J., Knobloch, K. R., Reedy, J., Henkels, M., & Cramer, K. (2018). Assessing the electoral impact of the 2010 Oregon Citizens’ Initiative Review. American Politics Research, 46(3), 534–563. Gastil, J., & Richards, R. (2013). Making Direct Democracy deliberative through Random Assemblies. Politics and Society, 41(2), 253–281. Gastil, J., Richards, R. C., & Knobloch, K. R. (2014). Vicarious deliberation: how the Oregon Citizens’ initiative review influenced deliberation in mass elections. International Journal of Communication, 8, 62–89. Geissel, B., & Gherghina, S. (2016). Constitutional deliberative democracy and Democratic Innovations. In M. Reuchamps & J. Suiter (Eds.), Constitutional deliberative democracy in Europe (pp. 75–92). ECPR Press. Hendriks, F. (2019). Democratic innovation beyond deliberative reflection: The ple- biscitary rebound and the advent of action-oriented democracy. Democratization, 26(3), 444–464. Hendriks, F. (2021). Unravelling the new plebiscitary democracy: Towards a research agenda. Government and Opposition, 56(4), 615–639. Jacobs, K. T. E. (Ed.). (2018). Het Wiv-referendum: Nationaal Referendum Onderzoek 2018. Stichting KiezersOnderzoek Nederland (SKON). Jacquet, V. (2017). Explaining non-participation in deliberative mini-publics. European Journal of Political Research, 56, 640–659. Jacquet, V., & van der Does, R. (2021). Deliberation and Policy-Making: Three ways to think about minipublics’ consequences. Administration & Society, 53(3), 468–487. Kahneman, D. (2011). Thinking Fast, Thinking Slow. Farrar, Straus & Giroux. Knight, J., & Johnson, J. (1994). Aggregation and deliberation: On the possibility of Democratic Legitimacy. Political Theory, 22(2), 277–296. Knobloch, K. R., Gastil, J., Feller, T., & Richards; R.C. (2013). Empowering citizen deliberation in direct democratic elections: A field study of the 2012. Oregon Citizens’ Initiative Review. Field Actions Science Report, 11, 1–10. Knobloch, K. R., Gastil, J., Reedy, J., & Cramer Walsh, K. (2014). Did they delib- erate? Applying an evaluative model of democratic deliberation to the Oregon Citizens’ Initiative Review. Journal of Applied Communication Research, 41(2), 105–125. Knobloch, K. R., Gastil, J., Richards, R., & Feller, T. (2012). Evaluation report on the 2012 Citizens’ initiative reviews for the Oregon CIR Commission. University of Washington. Lafont, C. (2020). Democracy without shortcuts: A participatory conception of delib- erative democracy. Oxford University Press. Landemore, H. (2018). Referendums are never merely referendums: On the need to make popular vote processes more deliberative. Swiss Political Science Review, 24(3), 320–327. Landemore, H. (2020). Open Democracy. Princeton University Press. Laruelle, A. (2021). Voting to select projects in participatory budgeting. European Journal of Operational Research, 288(2), 598–604. Hendriks and Wagenaar 21 LeDuc, L. (2015). Referendums and deliberative democracy. Electoral Studies, 38, 139–148. Levy, R. (2013). Deliberative voting: Realising Constitutional Referendum Democracy. Public Law. Journal Article (ssrn.com), 555–574. McKay, S. (2018). Designing popular vote processes for democratic systems: Counter-Proposals, recurring referendums, and iterated popular votes. Swiss Political Science Review, 24(3), 328–334. McKay, S. (2019). Building a better referendum: Linking mini-publics and Mass Publics in popular votes. Journal of Public Deliberation, 15(1), 1–29. Miller, S. A., Hildreth, R. W., & Stewart, L. M. (2019). The modes of participation: A revised frame for identifying and analyzing participatory budgeting practices. Administration & Society, 51(8), 1254–1281. Morel, L., & Qvortrup, M. (Eds.). (2017). The Routledge Handbook to Referendums and Direct Democracy, Abingdon: Routledge. Niessen, C., & Reuchamps, M. (2019). Designing a permanent deliberative citizens’ assembly. The Ostbelgien Modell in Belgium. Centre for Deliberative Democracy and Global Governance Working Paper Series No. 2019/6. OECD. (2020). Innovative citizen participation and New Democratic Institutions: Catching the deliberative wave. OECD Publishing. Offe, C. (2017). Referendum vs. Institutionalized deliberation: What Democratic the- orists can learn from the 2016 Brexit Decision. Daedalus, 146(3), 14–27. Parkinson, J. (2020). The roles of referendums in deliberative systems. Representation, 56(4), 485–500. Parkinson, J., & Mansbridge, J. (Eds.). (2012). Deliberative systems: Deliberative democracy at the large scale. Cambridge: Cambridge University Press. Renwick, A., Allan, S., Jennings, W., McKee, R., Russell, M., & Smith, G. (2017). The report of the citizens’ assembly on Brexit. University College of London. https://westminsterresearch.westminster.ac.uk/download/1f5fe9ee6777f06d fe343f5732bdec757edb6b4fe994de1ac63a2bfc496e381a/2889394/Citizens- Assembly-on-Brexit-Report.pdf Renwick, A., Allan, S., Jennings, W., Mckee, R., Russell, M., & Smith, G. (2018). What kind of Brexit do voters want? Lessons from the Citizens’ Assembly on Brexit. The Political Quarterly, 89(4), 649–658. Renwick, A., & McKee, R. (2017). The Citizens’ Assembly on Brexit (I): design and purpose. University College London. The Citizens’ Assembly on Brexit: design and purpose | The Constitution Unit Blog (constitution-unit.com) Saward, M. (2001). Making Democratic connections: Political Equality, deliberation and Direct Democracy. Acta Politica, 36, 361–379. Saward, M. (2021). Democratic design. Oxford University Press. Secinaro, S., Corvo, L., Brescia, V., & Iannaci, D. (2019). Hybrid organizations: A systematic review of the current literature. International Business Research, 12(11), 1–21. Setälä, M. (2017). Connecting deliberative mini-publics to representative decision making. European Journal of Political Research, 56, 846–863. 22 Administration & Society 00(0) Setälä, M. (2018). Deliberative mini-publics in democratic systems. In L. Morel & M. Qvortrup (Eds.), The Routledge Handbook to Referendums and Direct Democracy (pp. 464–478). Routledge. Setälä, M., Christensen, H. S., Leino, M., Strandberg, K., Bäck, M., & Jäske, M. (2020). Deliberative mini-publics facilitating voter knowledge and judgement: Experience from a Finnish local referendum. Representation. Advance online publication, 1–19. https://doi.org/10.1080/00344893.2020.1826565 State of Oregon. (2012). Voters’ pamphlet. Office of the Secretary of State. Suiter, J., & Reidy, T. (2020). Does deliberation help deliver informed electorates: Evidence from Irish referendum votes. Representation, 56(4), 539–557. https:// doi.org/10.1080/00344893.2019.1704848 Suiter, J., & Reuchamps, M. (2016). A constitutional turn for deliberative democ- racy in Europe? In M. Reuchamps & J. Suiter (Eds.), Constitutional deliberative democracy in Europe (pp. 1–14). ECPR Press. Sørensen, E., & Torfing, J. (2019). Towards robust hybrid democracy in Scandinavian municipalities? Scandinavian Political Studies, 42(1), 25–49. Tierney, S. (2013). Using electoral law to construct a deliberative referendum: Moving beyond the democratic paradox. Election Law Journal, 12(4), 508–523. Wagenaar, C. C. L. (2019). Beyond for or against? Multi-option alternatives to a cor- rective referendum. Electoral Studies, 62, 102091. Wagenaar, C. C. L. (2021). Voting beyond vetoing. Agenda-setting and balloting pro- cedures for multi-option referendums. Tilburg: Tilburg University (PhD thesis). Wampler, B. (2007). Participatory budgeting in Brazil: Contestation, Cooperation, and Accountability. Penn State Press. Warren, M. E., & Pearse, H. (2008). Designing deliberative democracy: The British Columbia Citizens’ Assembly. Cambridge University Press. Author Biographies Frank Hendriks is a full professor of comparative governance at Tilburg University, the Netherlands. His research focuses on the design and innovation of democratic governance. He is the author of a.o. “Vital Democracy: A Theory of Democracy in Action,” Oxford University Press, 2010, and co-editor of “The Oxford Handbook of Local and Regional Democracy,” Oxford UP, 2011. Presently, he directs the interuni- versity research project redress (on “revitalized democracy for resilient societies”), from which the present paper originates. Charlotte Wagenaar is a postdoctoral researcher on the redress (“revitalized democ- racy for resilient societies”) research project and is affiliated with Tilburg University. She investigates the design features of hybrid democratic innovations with a special focus on the perspective of elected representatives. Her expertise further extends to multi-option referendums, electoral processes and voting methods. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Administration & Society SAGE

The Deliberative Referendum: An Idea Whose Time has Come?

Hendriks, Frank; Wagenaar, Charlotte
Administration & Society , Volume 55 (3): 22 – Mar 1, 2023
Download PDF
22 pages

Loading next page...
 
/lp/sage/the-deliberative-referendum-an-idea-whose-time-has-come-X4GknNVxrk

References (65)

Publisher
SAGE
Copyright
© The Author(s) 2023
ISSN
0095-3997
eISSN
1552-3039
DOI
10.1177/00953997221140898
Publisher site
See Article on Publisher Site

Abstract

While deliberative citizens’ assemblies and plebiscitary referendums have long been perceived as antithetical, the idea of combining the two democratic instruments for better connecting administration and society has come to the fore in both theory and practice in more recent years. In this article, three ways of linking citizens’ assemblies to the referendum process are distinguished, exemplified, institutionally compared, and reflectively discussed. The three—the referendum-preparing, referendum-scrutinizing, and referendum-elaborating citizens’ assembly—come with their distinctive features, potential merits, scope limits, and related design questions. Fitting the “square peg of deliberative democracy” into the “round hole of direct democracy” and embedding hybrid design in diverging political systems are overarching challenges of institutional design. The article concludes that considering recent developments in theory and practice, the idea of a deliberative referendum linking citizens’ assemblies to direct voting on issues, seems an idea whose time has come, but also comes with challenges and questions that design thinkers and practitioners have only begun to tackle and answer. Keywords democratic innovation, direct democracy, deliberative democracy, referendum, citizens’ assembly, hybrid reform Tilburg University, Department of Public Law and Governance, The Netherlands Corresponding Author: Frank Hendriks, Tilburg University, Warandelaan 2, Tilburg, 5000 LE, The Netherlands. Email: f.hendriks@uvt.nl 2 Administration & Society 00(0) Small-Group Deliberation and Mass Voting: From Antithetical to Synergistic? In discussions about democratic innovations—instruments intended to forge effective connections between administration and society in democratic sys- tems—much attention is paid to two specific types of citizen participation and collective will formation: referendums, in which the whole electorate, the “maxi-public,” can vote directly on a policy issue, and citizens’ assemblies, in which an—ideally representative—sample of the population, a “mini-public” deliberates on a policy issue (e.g., Elstub & Escobar, 2019; Gastil & Richards, 2013; LeDuc, 2015). Neither innovations focused on direct voting in referen- dums nor those focused on extensive deliberations in mini-publics are ideal on their own (Geissel & Gherghina, 2016, p.88: Hendriks, 2019, p.448). By connecting such innovations, we might be able to create synergies between their relative merits, the benefits of which are potentially greater than the benefits of each separate innovation (McKay, 2019; Saward, 2001, p. 363; Tierney, 2013). A referendum proposal may for instance receive more sup- port and policy impact when a citizens’ assembly has been involved in designing, reviewing, or interpreting the proposition (Fishkin et al., 2015; Gastil et al., 2014, 2018; Suiter & Reidy, 2020). Vice versa, the addition of a referendum to a citizens’ assembly process may add focus to deliberations, and may help to prevent the deliberations of the citizens’ assembly’s disap- pear into thin air (e.g., Farrell et al., 2019, 2020; Fournier et al., 2011). The idea of linking citizens’ assemblies and referendums may seem an idea whose time has come, but it also comes with questions and reservations. The referendum and the citizens’ s assembly emanate from two very different strands of democratic theory and practice, and have long been perceived as antithetical (LeDuc, 2015, p. 139; El-Wakil, 2017, p. 59; Landemore, 2018, p. 320; Parkinson, 2020, p. 485); with the mini-public founded on a belief in patient small-scale deliberation, and the referendum geared at swift aggrega- tion of mass votes. The domain of referendums and other plebiscitary for- mats seems one of “thinking fast”—reflexively, quickly translating individual inclinations to a collective signal—whereas the domain of citizen’s assem- blies and other deliberative formats seems one of “thinking slow”—reflec- tively, patiently weaving a rug of shared meaning (Kahneman, 2011). “The twin objectives of voice and votes too often pull in opposite directions,” as LeDuc (2015: 147) explains. And then the question is indeed, as he nicely put it: “Can the square peg of deliberative democratic theory be pounded into the round hole of direct democracy?” In trying to answer this question construc- tively, Leduc is among a slowly but surely growing group of theorists who explore the scope for integrating deliberation in direct voting, either through Hendriks and Wagenaar 3 investments in public reflection in general (e.g., via information brochures for prospective voters), or through more integrative views on the sequencing of direct voting and reflective deliberation (Altman, 2014; Fishkin et al., 2015; Landemore, 2018; LeDuc, 2015; Levy, 2013; McKay, 2018, 2019; Parkinson, 2020; Saward, 2001; Setälä, 2017; Tierney, 2013). In thinking about democratic innovation, two opposite tendencies can be discerned: mixophobia (fear of institutional pollution and a related preference for pure models) on the one hand, and heterophilia (care for the institutionally different and a related interest in hybrid innovations) on the other hand (Hendriks, 2021; Saward, 2021). In this article, we pick up on the latter, explor- ing more integrative views on the linking of referendums and deliberative mini- publics specifically. Other hybrids of small-group deliberation and large-group voting are feasible and emerging in democratic governance, such as hybrids of participatory budgeting that start with deliberations in small-group settings and are completed with digital voting among the wider involved public that did not participate in the deliberative stage (e.g., Laruelle, 2021; Miller et al., 2019). However, these other democratic hybrids are beyond the scope of this article, which focuses on what can be called “deliberative referendums.” The referen- dum component in the hybrid can be of all common varieties: bottom-up, top- down, or mandatory; proactive or reactive; binding or advisory (Altman, 2011; Morel & Qvortrup, 2017). It always involves secret and direct voting of a maxi- public, a general electorate, on issues of public concern (e.g., open voting in town meetings or direct voting on politicians in recalls are not included here). The hybrid’s deliberative component can be a mini-public of all customary sizes (20–350 random participants) and numbers of meetings (1–20 days) (Bächtiger et al., 2018; OECD, 2020). It always concerns randomly selected, demographically stratified, samples of wider populations deliberating on issues of public concern. Denominators for this have proliferated, commonly starting with “deliberative” or ‘citizens,” and then adding a term such as “assembly,” “council,” “jury,” “poll,” “conference,” “dialog.” As a cover-all concept on this side of the hybrid we opt for the citizen’s assembly, which is both recognizable and often used in this context. In this article, we distinguish three broad types of combining and sequenc- ing citizens’ assemblies and referendums. We employ two criteria regarding the link between the citizens’ assembly and the referendum in these models: (1) the deliberative efforts of the assembly relate directly to the ballot proposal(s) in the referendum and (2) the combination of citizens’ assembly and referendum is not merely hypothetical but has actually been applied in practice. This means that purely hypothetical combinations, never taken to practice are excluded; as are citizens’ assemblies which broadly define priori- ties or topics of interest without a direct link to referendum voting. 4 Administration & Society 00(0) Even though interest in hybrid models of democratic innovation is grow- ing, systematic research on them has thus far been limited, focused on single case studies (e.g., Farrell et al., 2020), or other subsets of hybridization (Sørensen & Torfing, 2019). More research is needed, conceptually as well as empirically, to improve our theoretical and practical understanding of how linking elements of deliberation and aggregation could be achieved, what could be expected from this, and which challenges and questions are bound to arise. With this article, we contribute to the research endeavor by elaborat- ing on an empirically-grounded categorization of key variants, in full realiza- tion that we are closer to the start than to the end of it. Following this introduction, in the second section of the article we introduce and discuss three models which directly link referendums to deliberative citizens’ assem- blies, providing a brief example of an actual case for each. In section three, we further reflect on the comparative characteristics, merits, and vulnerabili- ties of the three types, as well as on related institutional design questions. Section four wraps up with closing reflections. Linking Citizens’ Assemblies to Referendums: Three Variants and Sequences As indicated, we focus on purposive deliberative additions to referendum processes of a specific type: deliberative mini-publics, or in common par- lance citizen’s assemblies. The purpose of the deliberative addition can be to prepare a referendum proposal, to scrutinize a referendum proposal or to elaborate a referendum result. In each variant, it is conceivable that the pro- posal (or proposition) is multiple rather than singular (Wagenaar, 2021)— although the latter, singular, is far more common. We distinguish the following three types of citizens’ assemblies combined with referendums: • Preparatory citizens’ assemblies Example: Irish Citizens’ Assembly (2016–2018) • Scrutinizing citizens’ assemblies Example: Oregon Citizens’ Initiative Review (since 2010) • Elaborating citizens’ assemblies Example: British Citizens’ Assembly on Brexit (2017) Each of these deliberative assemblies fulfils a role in a different stage of the referendum process (see Figure 1). Preparatory assemblies prepare a ref- erendum proposal prior to the establishment of the referendum ballot. Ideally, political commitment to put the prepared proposal to voters in a referendum is established prior to the commencement of the assembly’s work. Scrutinizing Hendriks and Wagenaar 5 Figure 1. Three variants of citizens’ assemblies linked to the referendum process. assemblies are established after a referendum question has been decided for the ballot. A scrutinizing assembly’s core task is to dissect the referendum question and ballot alternatives, and to provide voters with objective infor- mation, considerations and, optionally, voting advice. Elaborating assemblies are established after the referendum vote has taken place and the results have been announced. The purpose of an elaborating assembly is to interpret the referendum result and to provide input for follow-up steps or policies in view of the referendum outcome. In the following three sub-sections, we discuss each of these three types in more detail and we look at various empirical manifestations. Our classification builds on and extends the three-stage distinction— Initiation stage, Campaign stage, Implementation stage—that McKay (2019) proposed. We add the systematic comparison of the resulting hybrids in terms of characteristics, merits and related design questions. The classification par- tially builds on and partially deviates from Gastil and Richards (2013), who distinguished five deliberative designs—Priority Conference, Design Panel, Citizen’s Assembly, Citizens’ Initiative Review, Policy Jury—that can con- tribute to successive stages of the direct-democracy process. Their fifth and most downstream model—the autonomous Policy Jury, that develops direct legislation, dispensing with conventional legislative and electoral pro- cesses—is purely hypothetical and thus missing from our categorization of applied models. Our most downstream variant—the elaborating assembly that follows after the referendum vote—is missing in the classification of 6 Administration & Society 00(0) Gastil and Richards, which is more than ours focused on the real and specific problems of the bottom-up referendum and initiative process in parts of the US. Their three upstream models are collapsed into one variant in our clas- sification: the preparatory assembly, which can perform different tasks in mandatory, top-down, and bottom-up referendums (including the specific tasks of the Priority Conference and the Design Panel in bottom-up, peti- tioned processes). The Priority Conference only relates to our first model insofar as it leads to a concrete proposal for mass voting in a referendum; if it leads to another, more specialized deliberative forum taking over the baton, without the prospect of a follow-up referendum, the practice concerned falls outside the scope of this overview of the deliberative referendum. Their fourth model, the Citizens’ Initiative Review, overlaps with our second vari- ant, the Scrutinizing Assembly. The referendum-preparing citizens’ assembly The referendum-preparing assembly brings together a statistically represen- tative group of citizens in a controlled environment to discuss a policy issue, and to prepare a referendum proposal which thereafter can be presented to the wider electorate. In all events, assembly members are informed about the policy issue by experts after which they deliberate on conceivable policy options among themselves. Deliberating as a mini-public, the preparatory assembly paves the way for a referendum vote by the “maxi-public,” result- ing in either an advisory or binding public decision. The eventual formal trigger of the referendum may be governmental or parliamentary, but the proposition in this variant clearly comes from the citi- zens’ assembly. In cases where a referendum is not immediately foreseen, the recommendation to delegate—to “refer”—the outcome of the deliberations to a referendum vote can also form part of the deliberations. Approval by the maxi-public in a referendum may further legitimize the findings of the delib- erative mini-public in the public eye (Farrell & Suiter, 2019; Farrell et al., 2020; Parkinson, 2020; Suiter & Reuchamps, 2016: 9). A well-known example of a preparatory citizens’ assembly is the Irish Citizens’ Assembly, which was instituted in 2016 with the intention to delib- erate on a set of topics (abortion, the challenges and opportunities of an aging population, fixed-term parliaments, the manner in which referenda are held, and how the state can make Ireland a leader in tackling climate change). The Irish process that led to an authoritative decision on abortion—deliberated by the Citizens’ Assembly in 2016, Citizens’ Assembly in 2017, and ultimately approved per referendum in May 2018, is discussed in box 1. An earlier example is the British Columbia Citizens’ Assembly on Electoral Reform in Hendriks and Wagenaar 7 Box 1. Irish Citizens’ Assembly on abortion legislation. In 2016, following an electoral promise by political party Fine Gael, the Irish parliament voted to institute the Irish Citizens’ Assembly. A total of 99 randomly-selected citizens were brought together to discuss issues under the moderation of an independent chair appointed by the government. Selection of the participants was conducted using stratified random sampling taking into account age, gender, residency and social class. A total of 99 reserve candidates were also recruited (Farrell et al., 2019, 2020; Suiter & Reidy, 2020). The citizens’ assembly deliberated on various constitutional issues over a total of 12 weekends, five of which were dedicated to the eighth amendment of the constitution which regulated abortion. For each topic, the process was divided into several process steps. In a first stage (1), the topic was introduced and participants listened to presentations by experts and interest groups. In a second stage (2), participants discussed the issue more in-depth at round- table discussions. Professional moderators lead the discussions and encouraged participants to work toward concrete recommendations. At the end of the discussions, participants voted on the recommendations among themselves as a mini-public. All meetings were broadcasted live via a video stream. The wider public could also submit inputs, 13,000 of which were received on the abortion issue. Discussions on various cases and contingencies resulted in a sequence of internal votes that clearly showed that the participants in the Citizens’ Assembly recommended making legal abortion on request (within 12 weeks) available to a greater extent than previously thought possible in Ireland. The results, including a description of the legislative redrafting most preferred by the assembly, were cumulated into a final report which was sent to the government (Citizens’ Assembly, 2017). Since referendums are compulsory for constitutional changes in Ireland, it was always foreseen that the wider electorate would have to agree with the proposed changes, which it did on 25 May 2018 with a clear majority; 66.4% of the electorate voted in agreement with the Citizen’s Assembly’s proposal. This paved the way for changing the constitution, enabling the Irish parliament to draw up legislation which expanded rights to abortion. Canada, which was instituted in 2004 to debate whether a new electoral sys- tem for the province was desirable and if so, which system would be recom- mendable. The assembly’s proposal was presented to voters in a referendum in May 2005, receiving 57.5% majority support, but failing to meet the 60% support threshold (Warren & Pearse, 2008). Another example is the 2011 Icelandic Constitutional Assembly, which drafted a new constitution. The broader electorate voted in a referendum in October 2012 on six elements of the new constitution, all of which were approved per referendum, although parliament has stalled further decisions on the new constitution ever since (Landemore, 2020). 8 Administration & Society 00(0) The Irish case diverts from the Icelandic and British Columbia cases in its institutional and political embedding. In Iceland the political realm was largely detached from the process by a Constitutional Council consisting of citizens that “refused, in a self-defeating way, to cooperate with parties and other political elites (. . .) losing the good will of the political class” (Landemore, 2020, pp.120, 181). In British Columbia, the Citizens’ Assembly recommendation did not pass the referendum’s supermajority requirement, elevated high by weary politicians (Tierney, 2013; Warren & Pearse, 2008). In the Irish attempt at “systematizing constitutional delibera- tion” (Farrell et al., 2019) a more productive division of responsibilities between citizens’ assembly, representative politics and electorate has come to fruition. It is no coincidence that it is considered a “world leader in the linking of deliberative democracy (mini-publics) and direct democracy (referendums),” a paragon of “how deliberation can be inserted into the referendum process in a meaningful way” (Farrell et al., 2019, pp. 113, 119–120; also Farrell et al., 2020). The Referendum-Scrutinizing Citizens’ Assembly The referendum-scrutinizing citizens’ assembly embarks on its task after a referendum has been triggered and the referendum question has been decided, but before the vote takes place. The function of the scrutinizing assembly is to reflect on the content and consequences of the ballot options and to inform and enlighten the wider public on the matter before voting takes place. The integration and diffusion of arguments as prepared by the assembly can encourage more meaningful societal debate on the issue by the wider public (Jacquet & van der Does, 2021). A statement by the assembly is distributed to all eligible voters, including arguments for both sides. Deliberations need not, but may, culminate in a voting advice, which does not have to be unani- mously supported within the assembly. The distribution of assembly partici- pants backing either side is commonly reported. While scrutinizing citizens’ assemblies could take various forms, the most developed example is the citizens’ initiative review (CIR), originating in the US state of Oregon in 2010 (Gastil et al., 2014, 2018; Knobloch et al., 2014). In the CIR-model, after a referendum on a popular initiative has been announced, a representative group of 24 citizens is elected to take part in the deliberations. The first stage is an orientation stage to get acquainted with the topic of the initiative. In the next stage, participants hear experts on the topic as well as representatives of the pro and con sides. After these informative stages, the deliberation stage starts, in which participants discuss the infor- mation and arguments in moderated discussion groups. The assembly then Hendriks and Wagenaar 9 Box 2. Oregon Citizens’ Initiative Review on Measure 85. In 2012, a citizens’ initiative on Measure 85 qualified for the ballot in Oregon. The initiators intended to reform the corporate tax kicker in order to channel tax surpluses to public education rather than to companies. As stipulated in Oregon state law, a Citizens’ Initiative Review was established to deliberate the issue and present its voting advice. A sample of 10,000 randomly selected citizens was approached, and out of those that were willing to participate, a representative group of 24 citizens was selected to take part. In the first stage (1), the orientation stage, participants attended hearings in which proponents and opponents elaborated on their arguments. In the next stage (2), participants invited various experts to provide background information to the topic. In the final stage (3), participants deliberated their new insights among themselves with two goals: first, to draw up an overview of arguments in favor and against the initiative, and second, to formulate a voting advice. Both were included in the citizens’ statement which was sent to all eligible voters prior to the referendum vote (Knobloch et al., 2012). In the statement on Measure 85, 19 out of 24 participants supported the initiative (State of Oregon, 2012). In the following referendum, on 6 November 2012, a majority of 59.9% of voters also supported the initiative. Research concluded that two-thirds of voters considered the advice of the Citizens’ Initiative Review to be a useful source of information (Knobloch et al., 2013). draws up its citizens’ statement in which it highlights the most important findings and considerations. The statement is distributed to the electorate as a voter information pamphlet. Citizens’ initiative reviews have been used for state-wide citizen initiatives in Oregon since 2010, following a pilot. Since 2011 they are a legally formal- ized element of each citizen initiative trajectory. Box 2 describes an example case of an Oregon CIR in 2012. In the case of multiple citizens’ initiatives subjected to a referendum vote on the same day, parallel assemblies can be organized. Extensive research on the CIR process concludes that participants obtain issue-specific knowledge and are open to revising their options after reasoning with each other. Moreover, the information pamphlet helps the wider public to overcome information deficiencies and to base their vote on the same information and argumentation as established by the assembly par- ticipants (Gastil & Knobloch, 2020). In more recent years, the citizens’ initia- tive review has been experimented with in various other locations, such as the US states of Massachusetts, Colorado, California and Arizona, the Swiss can- tons of Sion and Geneva, and the municipality of Korsholm in Finland (Setälä et al., 2020). The Finnish experiment concluded that voters considered the assembly’s advice to be trustworthy and useful, and that the advice improved their factual knowledge, issue efficacy, and perspective-taking. 10 Administration & Society 00(0) The Referendum-Elaborating Citizens’ Assembly The elaborating citizens’ assembly is instituted after a referendum has taken place and follow-up steps must be decided on. Its goal is neither to prepare and propose a referendum question nor to inform voters, but rather to make sense of the referendum result and its implications in terms of operationaliza- tion prior to the start of the implementation process by policymakers. For reactive (“veto”) referendums, more specifically, a follow-up citizens’ assem- bly could shed further light on the reasons behind a majority rejection of new legislation and on amendments or alterations that might solve resistance. For proactive (“advancing”) referendums, it could help to fine-tune the opera- tionalization of newly accepted legislation from a citizens’ perspective. To some observers, a citizens’ assembly coming after the fact of a referen- dum may seem illogical, as the wider electorate has decided on an issue, which then seems to be delegated back to a smaller subset of the population. Yet, an elaborating assembly could fulfill a relatively modest but useful sup- port function in the stage directly following the referendum vote. Representative politics is commonly and overall responsible here, but might very well use citizens’ support for hammering out the best possible operationalization strat- egy, especially when a referendum result is open to different ways of follow- ing up or when it is prone to politically strategic interpretations of the referendum outcome which may run counter to the preferences of the societal majority. Investing in a citizens’ assembly the task of indicating the most- supported operationalization in line with the referendum outcome could serve to do justice to the underlying intentions of referendum voters. There are few practical examples of referendum-elaborating citizens’ assemblies. The best-known is the Citizens’ Assembly on Brexit, a pilot proj- ect held in the autumn of 2017. This case does not carry the weight of the previous two exemplary cases, but nevertheless provides an empirical opera- tionalization of linking deliberation to referendum voting. Box 3 describes the process of this assembly. Discussion: Distinctive Features, Institutional Implications and Design Questions In the previous section, we outlined three hybrid models that connect a delib- erative citizen’s assembly to a referendum vote. In Table 1, we summarize the main characteristics of the three. In case of the preparatory assembly, work- ing ahead of the referendum, the citizen’s assembly itself can be considered the leading instrument of democratic innovation. Deliberation serves to shape the proposal(s) of the mini-public and the subsequent referendum serves to Hendriks and Wagenaar 11 Box 3. British Citizens’ Assembly on Brexit. The Citizens’ Assembly on Brexit took place during two weekends in the autumn 2017. The 2016 referendum on EU membership had resulted in a majority preference to leave the EU (52% vs. 48% to remain in the EU), but there appeared to be no consensus over the exact shape that the relationship between the United Kingdom and the EU should take after the withdrawal (Offe, 2017; Renwick et al., 2018). A coalition of scientists and societal organizations thus decided to organize a citizens’ assembly on the topic. Its goal was explicitly not to challenge the referendum result, but to get insights into informed public opinion on what the position of the United Kingdom outside of the EU could look like (Renwick & McKee, 2017; Renwick et al., 2018). To select the participants, 5,000 citizens were approached. Of those willing to participate, participants were recruited such that they would be representative of the British population in terms of age, gender, ethnicity, social status, residency and voting behavior at the Brexit referendum. A group of 50 participants attended both weekends of deliberation. Assembly members went through two stages: (1) a learning stage, in which experts and other speakers elaborated on the topic, and (2) a deliberation stage, in which the participants discussed the topic in professionally moderated panels. Various alternative scenarios were distilled, which were voted on by all participants (Renwick et al., 2018). A clear majority of participants preferred an extensive trade deal with the EU over membership of the internal market, and both options were preferred over a no-deal Brexit. These preferences diverged from the preferences of parliamentary fractions at the time (Renwick et al., 2017, 2018). Perhaps because of its informal status, the policy and societal impact of the citizens’ assembly appeared to be limited. However, according to Renwick et al. (2018), the pilot proved that a citizens’ assembly tasked to elaborate a referendum’s general signal can produce a meaningful advice. further legitimize a proposal by submitting it to “market testing” among the maxi-public. For the other two types of hybrids, the referendum is the leading instrument, and the assembly serves to improve specific parts of the referen- dum process. The scrutinizing assembly serves to inform referendum voters prior to making their vote choice. The elaborating assembly is designed to alleviate the sometimes difficult task of understanding and translating the referendum outcome in terms of follow-up courses of action. The three ways of linking deliberation in citizens’ assemblies to voting in popular referendums each have their specific merits and limitations, as sum- marized in Table 1. For all three types, the desired effect is that deliberation improves the alignment of policies with societal preferences, either through designing a suitable referendum proposition, stimulating informed debate or 12 Table 1. Three types of linking deliberative citizens’ assemblies to referendums: central characteristics, potential (dis)advantages and related design questions. Referendum-preparing citizens’ Referendum-scrutinizing citizens’ Referendum-elaborating citizens’ assembly assembly assembly Timing Before referendum triggering After referendum triggering and After referendum vote before referendum vote Main contribution Exploring feasible and justified Informing voters on the content and Interpreting the referendum result of the assembly propositions fit for public choice implications of referendum options and considering ways forward per referendum (possibly adding non-binding voter advice) Main contribution Bringing collective closure, popularly Deciding the issue through maxi- Indicating the general direction, of the referendum supported, on assembly proposal public choice, following mini-public popularly supported, to further scrutiny of choice options elaborate Leading instrument Assembly Referendum Referendum Potential merits Improved alignment of policy decisions with societal preferences as expressed by mini-public and maxi-public in tandem Retaining equality in mass voting while reducing the rupture of equality through sortition Bypass for public issues that Clear division of labor between Interpretation of referendum result representative politics can’t or referendum initiators, assembly drawing on voter intentions rather won’t touch members and the wider electorate than political strategy Deliberation on multiple scenarios Demarcated, tractable task for Implementation sensitive to citizen rather than one specific proposal deliberative assembly perspective Mini-public proposition market- Evenly informed voting at Exploring scope for redress in case tested by maxi-public referendum of popular veto (continued) 13 Table 1. (continued) Referendum-preparing citizens’ Referendum-scrutinizing citizens’ Referendum-elaborating citizens’ assembly assembly assembly Potential Expanded actions and costs related to public consultation vulnerabilities Contradictory motivations and passions under one design Bypassing politics can slide No opportunities to present Counter-intuitive to let mini-public into offloading policymaking alternative ballot options or in when maxi-public has spoken responsibility, accountability, risk further solutions Entertaining multiple policy routes is Voting advice may be perceived and Risk of neglect or scorn from complicated and risks loss of focus resisted as paternalistic toward political realm and clarity electorate Central design Matching assembly design to referendum design questions Embedding hybrid design in the political-cultural context How to channel open, convoluted How to align assemblies’ balanced How to align advisory deliberation deliberation into closed, targeted option dissections with efforts of with political follow-up voting on propositions mass media and public institutions responsibilities How to optimize political bind and How to credibly insert the (small) How to prevent distortion, remain precommitment to the process jury format into democratic aligned with and organize feedback settings unfamiliar with it to the maxi-public How to extend and not undermine How to justify modest How to match elaboration task to already existing citizens’ rights to supplementary functions vis-à-vis type of referendum signal petition for a referendum added costs and efforts 14 Administration & Society 00(0) bringing the follow-up of referendum results in line with public intentions (Farrell et al., 2020; Gastil & Richards, 2013; Renwick et al., 2018). Furthermore, by involving a representative subset of the voting population, particularly when sortition procedures are applied, citizens’ assemblies can theoretically reduce the rupture of equality compared to propositions initiated by citizens with unequal financial and political resources (Fishkin et al., 2015). Nonetheless, the participation of usual suspects can still be an issue in the recruitment of assembly members (Jacquet, 2017). The referendum-preparing citizens’ assembly provides the strongest case for sequencing democratic innovations, as it has proved fruitful for legislat- ing on contentious issues (Farrell & Suiter, 2019; Farrell et al., 2019, 2020). The scrutinizing assembly, however, has also proven to be effective as an addition to the initiative process, particularly in, but not necessarily limited to, the US context (Gastil et al., 2014, 2018). The elaborating assembly fol- lowing referendum voting is the least battle-tested of the three types, although Renwick et al. (2017, 2018) have shown that a meaningful procedure and process result can be achieved. An elaborating assembly could have proven fruitful in other post-referendum situations, like the one resulting from an advisory corrective referendum in the Netherlands in 2018 in which a major- ity of voters vetoed stricter regulation of the secret services, without provid- ing sufficient guidance for necessary follow-up choices (Jacobs, 2018; Wagenaar, 2019). In terms of vulnerabilities, the elaborating assembly risks being seen as too late in the day; the referendum campaign and vote have already passed, and the results are in. Relatedly this variant risks being scorned or outright neglected by the political actors that have a role to play in the aftermath of a referendum—particularly when the assembly can be perceived or framed as a complicating factor rather than a helping hand in the process. Hence, one of the key design questions here is how to align a credible, advisory role of an elaborating assembly with a legitimate, broader responsibility of representa- tive politics to act on referendum results. A second obvious design question here is how to prevent the deliberative mini-public from distorting rather than elaborating the popular signal received per referendum; some form of public reporting in line with and in terms of the maxi-public’s revealed preference seems desirable and also feasible in combination with advisory reporting to parliament. A third design question follows from the fact that different refer- endums result in different public signals—advisory versus binding, proactive versus reactive—which demand diverging elaboration efforts that need to be designed accordingly. The scrutinizing citizens’ assembly is comparatively handicapped in that it can only dissect the given choice options, and not present alternative Hendriks and Wagenaar 15 options or further solutions. In case of supplementary voting advice, this vari- ant is vulnerable to the criticism of, and resistance to, paternalism, with a small group (the Oregon CIR e.g., assembles 24 citizens) presenting guid- ance to a full electorate of citizens which is seemingly not fully trusted. The latter has been parried in the design of the Oregon CIR, which stresses the relatively modest, supplementary, non-committal, service-to-the-voter char- acter of the assembly recommendation. As indicated, it is possible to refrain from giving voter advice and focus solely on dissecting choice options. In this baseline scenario it is imperative to clarify how the role of the citizens’ assembly in providing balanced information is different from and supple- mentary to the role of mass media and public institutions that inform the public. Another design question is if and how the (small) jury format, familiar to the American state context, can be inserted credibly into democratic set- tings not familiar with this format. When a referendum-scrutinizing assembly was discussed in the Netherlands, in the province of Limburg, a small scruti- nizing jury was perceived as lacking legitimacy, and a larger scrutinizing assembly as potentially more trustworthy—but also more costly. The referendum-preparing assembly may be introduced as an effective bypass for political stalemate, but it can also turn into a way of offloading policymaking responsibility, accountability and risk—a political strategy of non-decision-making. This danger has been noted for the Irish case. Another vulnerability follows from the fact that preparatory assemblies not seldom receive relatively open, ill-structured assignments, which are not only hard to grasp in a relatively short timeframe but also difficult to turn into voting propositions with focus and clarity. Hence, a first design question here is how to maintain a level of openness that fits the preparatory stage of assemblies working toward proposals, while also maintaining a degree of focus that fits the need for clearly targeted propositions in the stage of referendum voting. A second design question here is how to organize enough political bind and precommitment to the process, in order to motivate the efforts and costs of participation. Thirdly, a specific design question for the referendum-prepar- ing citizens’ assembly is how to extend—and not undermine—already exist- ing rights of (groups of) citizen’s to self-organize, and petition in a procedure for bottom-up referendums. How to match assembly choices (type of sortition, organization and dura- tion of deliberation) to referendum variants (bottom-up versus top-down and mandatory, proactive versus reactive, binding versus advisory) and how to make this work in a political-cultural context are overarching questions of hybrid design. They pertain to referendum-preparing, -scrutinizing and -elab- orating assemblies, which also have their own design questions. What we have done here is tease out central design questions, but we have not answered 16 Administration & Society 00(0) or solved them. This is not only because of space limitations and focusing choices, but also and more fundamentally because our collective knowledge on this is still very much fledgling—growing for particular cases of hybrid design, but largely lacking when it comes to overarching insights. Further developing comparative, cross-case knowledge about the nuts and bolts of hybrid design is an important next step to take if the general idea of hybrids connecting the best, while preventing the worst, characteristics of delibera- tive assemblies and plebiscitary referendums is indeed an idea whose time has come. From Occasional Nod to Systemic Attention? Closing Reflections Not so long ago, integrative views of voting and deliberation received only occasional nods in theories of deliberative democracy interested in connect- ing administration and society in more reflective ways (Levy, 2013, p. 573). This has changed under the influence of the systemic turn in deliberative democracy, which has encouraged theorizing about deliberative interventions working in tandem with non-deliberative, including plebiscitary, institutions (El-Wakil, 2017; Landemore, 2018; Parkinson, 2020; Parkinson & Mansbridge, 2012; Suiter & Reidy, 2020). Approaching the matter from the side of referendums, theorists such as LeDuc (2015), Setälä (2018), and McKay (2019) have shown a vivid interest in the addition of deliberative reflection to referendum voting. Prominently, Tierney (2013) has argued that a deliberative referendum could further the involvement of citizens in ques- tion formulation, while addressing some of the shortcomings of the referen- dum process, notably the barrier to meaningful deliberation resulting from the focus on aggregating established opinions, but also the issues of elite manipulation and majority tyranny. Slowly but surely, deliberative assem- blies linked to referendums have also sprung up in practice, motivated by the urge to forge more versatile connections between administration and society. In this article, we have further examined the idea of the deliberative referen- dum, its emerging functional variants, their potential merits, vulnerabilities, and related design questions in particular. The result speaks to debates in public administration and related disciplines concerned with democratic instruments and innovations that are meant to forge more effective connec- tions between administration and society in democratic systems. Considering developments in both theory and practice, as we have done here, the deliberative referendum seems an idea whose time has come; it comes, however, also with challenges and questions that theorists and Hendriks and Wagenaar 17 practitioners of democratic innovation have only begun to tackle and answer. The common idea behind the three general, empirically grounded, models distinguished in this article is that combining the use of referendum voting and deliberative citizens’ assemblies can bolster the merit and correct the vulnerability of both instruments of democratic innovation. In a single instru- ment, either deliberative or plebiscitary, the associated vulnerabilities may prevail, while in a hybrid construction the combined forces may lead to an overall more advantageous situation. Variations in empirical reality denote that this is not easily accomplished. A hybrid design that may work compara- tively well in one democratic setting does not automatically fit another con- text with different structural and cultural features. Each hybrid design involving a deliberative assembly and a plebiscitary referendum comes with particular design questions, explored in this article and distinguished as items for more specific design thinking, experimentation, and research-for-design. An intrinsic difficulty of hybrid design harks back to the discussion that we started with. Deliberative mini-publics and plebiscitary referendums are associated with separate discourses and practices with divergent motivational bases (Knight & Johnson, 1994; LeDuc, 2015; Parkinson, 2020). People who are passionate about deliberative mini-publics are not seldom suspicious of plebiscitary referendums, and the other way around; reaching out to the other side does not come naturally in this field. The challenge of hybrid design may thus get more than twice as tough. It is not only that two components need to be fitted; it is also that at least one of two tends to be approached with reserve and suspicion; “Why add an unruly referendum process, if you can have a safely embedded mini-public?”; “Why add a mini-public, can we not just trust the encompassing demos?” From diverging vantage points in, respec- tively, deliberative and plebiscitary democratic thought such questions are comprehensible. Such reservations are bracketed in emerging experiments with hybrid designs like the ones that we explored, and in more recent schol- arly theorizing that approaches the assembly-plus-referendum as at least a promising hypothesis worthy of testing. Further development and testing of the proposition is work to be done. Declaration of Conflicting Interests The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article. Funding The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The research for this article was partly financed by the Dutch National Science Agenda, NWO file number NWA.1292.19.048. 18 Administration & Society 00(0) ORCID iD Frank Hendriks https://orcid.org/0000-0001-7393-4144 Notes 1. For a more general, philosophical argument about the tension between delibera- tive democracy based on mini-publics and mass participation in various guises, among which referendums, see Lafont (2020). 2. Also beyond the scope of this article are hybrid innovations discussed in the wider field of public and private organizations; see a.o. Billis (2010), Secinaro et al. (2019). 3. And under this cover-all concept, we may refer to specific concepts, used in spe- cific settings, such as the Citizen’s Initiative Review as developed in the context of Oregon, US (see our second model). 4. These may be interesting innovations, which empower citizens to set or adjust the policy agenda. However, when a direct link between a citizens’ assembly and a referendum vote is lacking such innovations are excluded here. 5. A notable example is the Ostbelgien Citizens’ Council, a permanent body of citizens selected by lot, tasked with selecting issues for further debate by a sec- ond—temporary and issue-specific—deliberative assembly, also composed of citizens selected by lot. This Citizens’ Assembly deliberates and formulates concrete policy recommendations (Niessen & Reuchamps, 2019). However, as these recommendations are not directly subjected to a referendum vote we do not include this assembly in our analysis based on our selection criteria. The crite- rion of a direct link between citizens’ assembly and referendum proposal means that citizens’ assemblies only tasked with agenda-setting in a broader sense (as meant by e.g., Dahl, 1989), are excluded from our overview. 6. In specific cases, the assembly may consider whether one or several referendum questions are desirable, and which particular alternatives and voting structure would be most suitable for the purpose. A special variant, using the specific instrument of the “Deliberative Poll” as developed by Fishkin and collaborators, was experimented with in 2011, within the specific context of the Californian initiative process. Related to a project called “What’s Next California” a delib- erative-polling process was started to discuss and consider 30 proposals from a broad coalition of reform groups. This ultimately led to a proposition, Proposition 31, which was ultimately not successful at election time (Fishkin et al., 2015). 7. The Citizens’ Assembly, First Report and Recommendations of the Citizens’ Assembly, The 8th Amendment of the Constitution, 29 June 2017. 8. Following input from participants, situations considered were expanded with contingencies such as: real and significant physical risks to the woman’s life; real and substantial risk to a woman’s life from suicide; serious risk to the woman’s physical health; serious risk to the woman’s mental health; pregnancy as a result of rape; fetal abnormality likely to result in death before or shortly after birth; significant fetal abnormality unlikely to result in death before or shortly after birth; socio-economic reasons; no statement of reasons. Hendriks and Wagenaar 19 9. As outlined in the introduction, we focus on deliberative assemblies whose efforts directly relate to the content of the proposal subjected to a referendum vote. Therefore, we limit our focus on the referendum-elaborating citizens’ assembly to the steps of interpreting the referendum outcome and determining the most suitable policy course. Further along the policy trajectory, other types of deliberative assemblies could be plausible, modeled for example after the imple- mentation committees that monitor the implementation of participatory budget- ing outcomes (see e.g., Wampler, 2007). To the best of our knowledge, there have so far not been any empirical examples of citizen committees overseeing the implementation of referendum outcomes; this means they also do not meet our second inclusion criterion. 10. Personal correspondence David Farrell. See also D. Farrell, We may have overdone it on citizens’ assemblies, Irish Times, 16 February 2022 https://www.irishtimes. com/opinion/we-may-have-overdone-it-on-citizens-assemblies-1.4803375 References Altman, D. (2011). Direct democracy worldwide. Cambridge University Press. Altman, D. (2014). Strengthening democratic quality: Reactive deliberation in the context of direct democracy. Helen Kellogg Institute for International Studies. Working Paper No. 400. Bächtiger, A., Dryzek, J. S., Mansbridge, J., & Warren, M. E. (2018). The Oxford Handbook of Deliberative Democracy. Oxford University Press. Billis, D. (2010). Hybrid organizations and the third sector: Challenges for practice, theory and policy. Macmillan International Higher Education. Citizens’ Assembly. (2017). First Report and Recommendations of the Citizens’ Assembly: The eighth amendment of the constitution. Dahl, R. A. (1989). Democracy and its critics. Yale University Press. Elstub, S., & Escobar, O. (Eds.). (2019). Handbook of Democratic Innovation and Governance, Aldershot: Edward Elgar. El-Wakil, A. (2017). The deliberative potential of facultative referendums: Procedure and substance in direct democracy. Democratic Theory, 4(1), 59–78. Farrell, D. M., Suiter, J., & Harris, C. (2019). ‘systematizing’ constitutional delibera- tion: The 2016–18 citizens’ assembly in Ireland. Irish Political Studies, 34(1), 113–123. Farrell, D. M., & Suiter, J. (2019). Reimagining Democracy: Lessons in deliberative democracy from the Irish Front Line. Cornell University Press. Farrell, D. M., Suiter, J., Cunningham, K., & Harris, C. (2020). When mini-publics and Maxi-publics coincide: Ireland’s National Debate on Abortion. Representation. Advance online publication. https://doi.org/10.1080/00344893.2020.1804441 Fishkin, J., Kousser, T., Luskin, R. C., & Siu, A. (2015). Deliberative agenda set- ting: Piloting reform of Direct Democracy in California. Perspectives on Politics, 13(4), 1030–1042. Fournier, P., van der Kolk, H., Carty, R. K., & Blais, A. (2011). When citizens decide: Lessons from citizens assemblies on electoral reform. Oxford University Press. 20 Administration & Society 00(0) Gastil, J., & Knobloch, K. R. (2020). Hope for democracy: How citizens can bring reason back into politics. Oxford University Press. Gastil, J., Knobloch, K. R., Reedy, J., Henkels, M., & Cramer, K. (2018). Assessing the electoral impact of the 2010 Oregon Citizens’ Initiative Review. American Politics Research, 46(3), 534–563. Gastil, J., & Richards, R. (2013). Making Direct Democracy deliberative through Random Assemblies. Politics and Society, 41(2), 253–281. Gastil, J., Richards, R. C., & Knobloch, K. R. (2014). Vicarious deliberation: how the Oregon Citizens’ initiative review influenced deliberation in mass elections. International Journal of Communication, 8, 62–89. Geissel, B., & Gherghina, S. (2016). Constitutional deliberative democracy and Democratic Innovations. In M. Reuchamps & J. Suiter (Eds.), Constitutional deliberative democracy in Europe (pp. 75–92). ECPR Press. Hendriks, F. (2019). Democratic innovation beyond deliberative reflection: The ple- biscitary rebound and the advent of action-oriented democracy. Democratization, 26(3), 444–464. Hendriks, F. (2021). Unravelling the new plebiscitary democracy: Towards a research agenda. Government and Opposition, 56(4), 615–639. Jacobs, K. T. E. (Ed.). (2018). Het Wiv-referendum: Nationaal Referendum Onderzoek 2018. Stichting KiezersOnderzoek Nederland (SKON). Jacquet, V. (2017). Explaining non-participation in deliberative mini-publics. European Journal of Political Research, 56, 640–659. Jacquet, V., & van der Does, R. (2021). Deliberation and Policy-Making: Three ways to think about minipublics’ consequences. Administration & Society, 53(3), 468–487. Kahneman, D. (2011). Thinking Fast, Thinking Slow. Farrar, Straus & Giroux. Knight, J., & Johnson, J. (1994). Aggregation and deliberation: On the possibility of Democratic Legitimacy. Political Theory, 22(2), 277–296. Knobloch, K. R., Gastil, J., Feller, T., & Richards; R.C. (2013). Empowering citizen deliberation in direct democratic elections: A field study of the 2012. Oregon Citizens’ Initiative Review. Field Actions Science Report, 11, 1–10. Knobloch, K. R., Gastil, J., Reedy, J., & Cramer Walsh, K. (2014). Did they delib- erate? Applying an evaluative model of democratic deliberation to the Oregon Citizens’ Initiative Review. Journal of Applied Communication Research, 41(2), 105–125. Knobloch, K. R., Gastil, J., Richards, R., & Feller, T. (2012). Evaluation report on the 2012 Citizens’ initiative reviews for the Oregon CIR Commission. University of Washington. Lafont, C. (2020). Democracy without shortcuts: A participatory conception of delib- erative democracy. Oxford University Press. Landemore, H. (2018). Referendums are never merely referendums: On the need to make popular vote processes more deliberative. Swiss Political Science Review, 24(3), 320–327. Landemore, H. (2020). Open Democracy. Princeton University Press. Laruelle, A. (2021). Voting to select projects in participatory budgeting. European Journal of Operational Research, 288(2), 598–604. Hendriks and Wagenaar 21 LeDuc, L. (2015). Referendums and deliberative democracy. Electoral Studies, 38, 139–148. Levy, R. (2013). Deliberative voting: Realising Constitutional Referendum Democracy. Public Law. Journal Article (ssrn.com), 555–574. McKay, S. (2018). Designing popular vote processes for democratic systems: Counter-Proposals, recurring referendums, and iterated popular votes. Swiss Political Science Review, 24(3), 328–334. McKay, S. (2019). Building a better referendum: Linking mini-publics and Mass Publics in popular votes. Journal of Public Deliberation, 15(1), 1–29. Miller, S. A., Hildreth, R. W., & Stewart, L. M. (2019). The modes of participation: A revised frame for identifying and analyzing participatory budgeting practices. Administration & Society, 51(8), 1254–1281. Morel, L., & Qvortrup, M. (Eds.). (2017). The Routledge Handbook to Referendums and Direct Democracy, Abingdon: Routledge. Niessen, C., & Reuchamps, M. (2019). Designing a permanent deliberative citizens’ assembly. The Ostbelgien Modell in Belgium. Centre for Deliberative Democracy and Global Governance Working Paper Series No. 2019/6. OECD. (2020). Innovative citizen participation and New Democratic Institutions: Catching the deliberative wave. OECD Publishing. Offe, C. (2017). Referendum vs. Institutionalized deliberation: What Democratic the- orists can learn from the 2016 Brexit Decision. Daedalus, 146(3), 14–27. Parkinson, J. (2020). The roles of referendums in deliberative systems. Representation, 56(4), 485–500. Parkinson, J., & Mansbridge, J. (Eds.). (2012). Deliberative systems: Deliberative democracy at the large scale. Cambridge: Cambridge University Press. Renwick, A., Allan, S., Jennings, W., McKee, R., Russell, M., & Smith, G. (2017). The report of the citizens’ assembly on Brexit. University College of London. https://westminsterresearch.westminster.ac.uk/download/1f5fe9ee6777f06d fe343f5732bdec757edb6b4fe994de1ac63a2bfc496e381a/2889394/Citizens- Assembly-on-Brexit-Report.pdf Renwick, A., Allan, S., Jennings, W., Mckee, R., Russell, M., & Smith, G. (2018). What kind of Brexit do voters want? Lessons from the Citizens’ Assembly on Brexit. The Political Quarterly, 89(4), 649–658. Renwick, A., & McKee, R. (2017). The Citizens’ Assembly on Brexit (I): design and purpose. University College London. The Citizens’ Assembly on Brexit: design and purpose | The Constitution Unit Blog (constitution-unit.com) Saward, M. (2001). Making Democratic connections: Political Equality, deliberation and Direct Democracy. Acta Politica, 36, 361–379. Saward, M. (2021). Democratic design. Oxford University Press. Secinaro, S., Corvo, L., Brescia, V., & Iannaci, D. (2019). Hybrid organizations: A systematic review of the current literature. International Business Research, 12(11), 1–21. Setälä, M. (2017). Connecting deliberative mini-publics to representative decision making. European Journal of Political Research, 56, 846–863. 22 Administration & Society 00(0) Setälä, M. (2018). Deliberative mini-publics in democratic systems. In L. Morel & M. Qvortrup (Eds.), The Routledge Handbook to Referendums and Direct Democracy (pp. 464–478). Routledge. Setälä, M., Christensen, H. S., Leino, M., Strandberg, K., Bäck, M., & Jäske, M. (2020). Deliberative mini-publics facilitating voter knowledge and judgement: Experience from a Finnish local referendum. Representation. Advance online publication, 1–19. https://doi.org/10.1080/00344893.2020.1826565 State of Oregon. (2012). Voters’ pamphlet. Office of the Secretary of State. Suiter, J., & Reidy, T. (2020). Does deliberation help deliver informed electorates: Evidence from Irish referendum votes. Representation, 56(4), 539–557. https:// doi.org/10.1080/00344893.2019.1704848 Suiter, J., & Reuchamps, M. (2016). A constitutional turn for deliberative democ- racy in Europe? In M. Reuchamps & J. Suiter (Eds.), Constitutional deliberative democracy in Europe (pp. 1–14). ECPR Press. Sørensen, E., & Torfing, J. (2019). Towards robust hybrid democracy in Scandinavian municipalities? Scandinavian Political Studies, 42(1), 25–49. Tierney, S. (2013). Using electoral law to construct a deliberative referendum: Moving beyond the democratic paradox. Election Law Journal, 12(4), 508–523. Wagenaar, C. C. L. (2019). Beyond for or against? Multi-option alternatives to a cor- rective referendum. Electoral Studies, 62, 102091. Wagenaar, C. C. L. (2021). Voting beyond vetoing. Agenda-setting and balloting pro- cedures for multi-option referendums. Tilburg: Tilburg University (PhD thesis). Wampler, B. (2007). Participatory budgeting in Brazil: Contestation, Cooperation, and Accountability. Penn State Press. Warren, M. E., & Pearse, H. (2008). Designing deliberative democracy: The British Columbia Citizens’ Assembly. Cambridge University Press. Author Biographies Frank Hendriks is a full professor of comparative governance at Tilburg University, the Netherlands. His research focuses on the design and innovation of democratic governance. He is the author of a.o. “Vital Democracy: A Theory of Democracy in Action,” Oxford University Press, 2010, and co-editor of “The Oxford Handbook of Local and Regional Democracy,” Oxford UP, 2011. Presently, he directs the interuni- versity research project redress (on “revitalized democracy for resilient societies”), from which the present paper originates. Charlotte Wagenaar is a postdoctoral researcher on the redress (“revitalized democ- racy for resilient societies”) research project and is affiliated with Tilburg University. She investigates the design features of hybrid democratic innovations with a special focus on the perspective of elected representatives. Her expertise further extends to multi-option referendums, electoral processes and voting methods.

Journal

Administration & Society – SAGE

Published: Mar 1, 2023

Keywords: democratic innovation; direct democracy; deliberative democracy; referendum; citizens’ assembly; hybrid reform

There are no references for this article.