Introduction

One important democratic quality that we seek in our political system is accountability. We would like to have accountable institutions, politicians, policymakers and so on so forth. For this aim, political institutions are equipped with different checks and balances, and here we often find principal-agent type of relationships, which mostly involve authorization and accountability. However, it is a quality that is not discussed much in the context of minipublics. Warren (2008) is one of the few works that discuss accountability in relation to a minipublic. In addition, although deliberation and accountability are discussed in the context of electoral politics, as Bächtiger et al. (2018: 8–9) put it ‘Deliberation in other forums would require other forms of accountability that still remain to some degree untheorized.’

One reason for the lack of accountability discussions in the literature can be that there seems to be a tension between citizen deliberation and accountability. In the context of minipublics, ‘legitimating bonds of authorization and accountability between participants and non-participants’ are missing (Parkinson 2006: 33). There is no principal-agent relationship between the participants and the represented. Hence, electoral accountability as we will expect in conventional politics is not applicable in the context of minipublics (Brown 2006; Mansbridge 2004; Warren 2008). For some, the lack of electoral accountability is good for deliberation because, without the pressures of principal-agent bonds, participants are thought to be freer in their deliberations (Mansbridge 2004; Warren 2008).

This article will enable us to move beyond the accountability of participants. As will be seen in the following pages, the accountability of organizers is important. Especially their deliberative accountability will be highlighted, and it will be argued that it is a part of organizing minipublics. Here, the term organizers involves all the actors who take part in the organization of minipublics. It is clear that organizing a minipublic involves different tasks and responsibilities. Hence, organizers are not one homogeneous group. Rather they are a group of actors with different roles in minipublics. We can follow Gül (2014, 2019) to divide organizers functionally into three: initiators, project managers and field staff. Initiators are the actors who decide to organize a minipublic. Project managers are responsible for the operational tasks before, during and after the organization of minipublics. These may include the chair, the secretariat and civil society groups. Finally, field staff is responsible for various tasks such as logistics, catering and very importantly facilitation.

The plan of the article is as follows. The first section will discuss deliberative accountability as it is usually mentioned in the discussions of deliberative democracy. It will be argued that since deliberative accountability is part of deliberation, which is understood as reason-giving, it does not contribute much to our understanding of accountability of participants. While the accountability of deliberators is obvious, the deliberative accountability of organizers is not discussed much, yet it appears to be a crucial aspect of organizing a minipublic. The second section will present the basic questions and considerations found in the study of accountability. The relational core of the concept, descriptive and normative studies of accountability and the sanctioning as a way to categorize accountability as weak and strong would be presented. This section will establish the background for the third section that will provide us with a snapshot of the accountability relations in the context of minipublics by asking three questions used in the accountability literature: who is accountable?, to whom? and for what? The answers will enable us to see the accountability relations among participants, between participants and organizers, among organizers, and between organizers and external actors. This will highlight the importance of accountability for organizers who are in various accountability relations in and out of minipublics. In the final section, three points will be raised to enhance our understanding of accountability in the context of minipublics. First, it will be argued that trust-based selection model of principal-agent accountability (Mansbridge 2009, 2014) might offer us a different perspective on the weak accountability of participants and it directs our attention to the selection done by organizers. Second, the empowerment of minipublics is important when we think about whether we want stronger accountability mechanisms in minipublics and finally, it will be argued that organizers might be held accountable for the decisions made by an empowered minipublic.

Deliberative Accountability

Deliberative theorists generally consider accountability as giving an account of the reasons for one’s opinions or decisions (Escobar & Elstub 2017; Gutmann & Thompson 1996; Mansbridge 2014). For instance, Gutmann and Thompson (1996: 129) argue for a deliberative version of accountability for elected representatives, who are concerned with more than re-election or abiding by the constitutional rules: ‘In a deliberative democracy representatives are expected to justify their actions in moral terms.’ This justification process asks more than giving reasons to their voters. Gutmann and Thompson (1996: 144–145) argue that representatives have moral constituencies that involve people beyond their constituency (e.g. people in different areas or countries or future generations). Thus, reason-giving should involve this moral constituency as well.

In addition, Borowiak (2011) argues that deliberative accountability involves publicly articulating, explaining, and justifying public policy. In deliberative accountability, an actor can be held accountable by being obliged to justify. The justification is not only about their acts but also about the rationale behind them. Representatives need to provide reasons for their acts ‘on the terms others could accept’ (Borowiak 2011: 105). These might be available in electoral contexts too. However, deliberative accountability ‘spans the political process’ (Borowiak 2011: 106). In other words, accountability is more like a process that extends during election times. Thus, deliberative accountability as reason-giving or being obliged to justify one’s actions and rationale is a moral act that goes beyond one’s constituency and election times. Similarly, Staszewski (2009: 1284–1285) argues that deliberative accountability is more dynamic than electoral accountability, and it transcends electoral boundaries. While political accountability is understood as electoral accountability that is based on sanctioning, ‘deliberative accountability is premised on the need for public officials and citizens to persuade one another of the merits of their positions’ (Staszewski 2009: 1286).

Goodin (2008) offers a similar accountability understanding and he names it as discursive accountability. Discursive accountability is non-electoral, hence, it is broader than electoral accountability, and it involves ‘accountability to disbursed networks rather than merely accountability to democratically empowered electors’ (Goodin 2008: 155). This is the network-based accountability that is different to electoral accountability of state-based accountability and financial accountability of market-based accountability (Goodin 2008: ch. 8). Although he discusses network-based accountability in the context of the third sector, he argues that it can be extended to civil society in general. In such an accountability model, actors justify their acts to their peers, and only this way actors can give a good account. He argues that this sort of network accountability is the discursive accountability that deliberative democrats need to aspire for (Goodin 2008: 185). In addition, ‘Discursive accountability to one’s peers within networks, nationally and transnationally, can be a powerful supplement to electoral accountability, …’ (Goodin 2008: 149). Gutmann and Thompson (1996: 144) make a similar point by arguing that ‘Reiterated deliberation, punctuated by periodic elections, is the best hope for the principle of accountability.’

We can see that deliberative accountability is mainly perceived to be a supplement to the existing accountability mechanisms. It acts as an extra accountability measure and can be used to boost accountability in that system. If elected representatives were required to give an account to the public for their decisions, we would have a more accountable system. The above discussions consider the actors with a claim to represent or to act for others. Seen in this way, the use of deliberative accountability in the context of minipublics is dubious as we lack both already existing mechanisms of accountability and the participants claiming to represent (Gül 2019).

In a different discussion, O’Neill et al. (2008: 13) argue that deliberative accountability includes: the examination of the framing of the question; the examination of experts and their relevant expertise; the consideration of other options; and the assessment of the credibility of the information. They argue that such features of deliberative accountability bring larger questions that are not about the composition of the practice (i.e. descriptive representation), but about the organizing bodies. Although O’Neill et al. (2008) raise an important point about organizing bodies, it is doubtful whether the above-mentioned features are of deliberative accountability or deliberation. Having a critical eye on the questions, experts or information seems to be part of deliberation. We can have the same doubt for deliberative accountability understood as reason-giving since reason-giving is also part of deliberation. The participants give accounts of their arguments during the process of minipublic deliberations and this is where we find deliberative accountability. However, giving an account also denotes deliberation as a process of reason-giving. Then, we can argue that deliberation comes with deliberative accountability. In other words, deliberative accountability understood as reason-giving is embedded into the act of deliberation.

If we accept the last point, we can see that the deliberative accountability of deliberators is obvious. Hence, deliberative accountability understood as reason-giving does not contribute much to our discussion of accountability of participants in minipublics as it will always be there. In addition, discussing deliberative accountability like above limits the discussion to the deliberators, and limiting the discussions as such would miss the larger web of relations found in minipublics. As will be seen in the third section, organizers are deliberatively accountable for their organizational choices and the deliberative accountability of organizers can be seen as a normative criterion for organizing minipublics. Before this, we need to go back to the basics of studying accountability as a concept. This will provide us with the background to look at the accountability relations in minipublics.

Key Dimensions of Studying Accountability

In this section, I would like to introduce general ways to examine accountability and relate them to our examination of minipublics. Accountability is a relational concept that can be studied normatively and descriptively (Bovens 2010; Bovens et al. 2014; Philp 2009). Its relational character is named as ‘the minimal conceptual consensus’ by leading scholars of accountability (Bovens et al. 2014: 3). Studying accountability normatively treats it as a virtue that professionals or organizations should possess. Normative accountability studies are active (focusing on actors’ behaviour), evaluative, concerned with legitimacy and prescriptions for good governance. In its descriptive sense, accountability can be seen as a social mechanism (Bovens 2010; Bovens et al. 2014). It is a narrower sense of the term than the normative sense. It looks at the arrangements and institutional designs in which individuals are held accountable (Bovens et al. 2014). Descriptive studies are passive, look at the relations between actors and forums and seek to identify accountability there (Bovens 2010: 961–962). Another way to put the difference between the two is like this: If we ask whether accountability relations and mechanisms exist or not, it is descriptive; but if we ask how much of it or what kinds of mechanisms we would want in our political system, it is normative (Philp 2009: 32). However, it is important to highlight that descriptive and normative studies of accountability are not mutually exclusive as they are closely related. They feed each other. Social mechanisms cannot be thought without normative judgements because without some standards, we cannot assess one’s conduct. Thus, the point is that they are distinct yet ‘mutually reinforcing’ (Bovens 2010: 962).

In any discussion of accountability, the value of sanctioning comes to the fore. However, it seems to be an unresolved issue. For instance, for Bovens (2010: 952), the possibility of sanctions is a constitutive element of accountability yet the word ‘sanction, has a rather legal and formal connotation,’ so it is a strong word and often associated with negative situations. Hence, he suggests using the expression ‘face consequences’ instead of sanctions (Bovens 2010: 952, emphasis in original). In addition, Mulgan (2000: 556) argues that the involvement of sanctions in the definition of accountability is contestable because it may be beyond the scope of giving an account. On the other hand, he continues, calling someone into account is incomplete without sanctions.

A fruitful way of discussing sanctions is to see the problem as a matter of degree rather than as an all-or-nothing situation. For instance, Rubenstein (2007) argues that sanctions are part of accountability but should not be too mild as it is the sanctions that give ‘teeth’ to accountability and ‘distinguishe[s] it from responsibility, responsiveness and deliberation,’ yet they should not be too severe either (Rubenstein 2007: 619). Then, we can think of accountability as lying on a continuum the ends of which denote the weight of sanctioning. At one end, we have strong accountability, and at the other, we have weak accountability. The strong accountability suggests that to have accountability, there needs to be some sort of sanctioning, some teeth, so to speak. At the other end of the continuum, we have weak accountability, which does not focus on sanctioning or punishing. Instead, the possibility of giving an account serves as a deterrent. Reporting and justification replace sanctions. It is not the possibility of sanctions that serves as a check on one’s conduct, but the possibility of being required to justify one’s conduct. (Mulgan 2000: 555–558). As seen, accountability is more about the potential of holding someone accountable and so, it is ‘account-ability’ (Mulgan 2000: 560, emphasis in original). We can even argue that it is the institutionalized potentiality of holding someone accountable that makes accountability. This way of understanding the place of sanctions is helpful for us to examine minipublics in which there are different accountability relations with varying degrees of sanctions, or consequences as Bovens would like to name it.

Finally, in the literature, we find several questions to ask about accountability relations (Bovens 2010; Bovens et al. 2014; Mulgan 2003). Although there are slight differences among scholars, some questions are common, and they are helpful for this article. We will mainly consider three questions: who is accountable?, to whom?, for what?. Who? question seeks the account-givers (i.e. accountability holdees). To whom? question seeks the forum/actors to which an account is given (i.e. accountability holders). Finally, for what? question seeks the subject of accountability. These are the matters that actors are accountable for. These might be their actions, results of their actions and their intentions in general (Goodin 2008); finance, fairness and performance in particular (Behn 2001: 6).

Accountability Relations in Minipublics

On the background provided in the previous section, we can state that our examination of accountability in minipublics is a descriptive study of accountability relations yet with normative implications for organizers, whose accountability is rather on the stronger side of the accountability continuum compared to the accountability of participants. Below is an overview of accountability relations that we might find in minipublics (Table 1). The discussion does not claim to be exhaustive, but it aims to show what kinds of accountability relations might take place between different actors.

Participant Selection, Stronger Accountability and Accountability for Decisions

This section will discuss three points to further our understanding of accountability in minipublics. The first point discusses the possible usefulness of trust-based selection model of principal-agent accountability in the context of minipublics as it might give us a different perspective on the weak accountability of participants (Mansbridge 2009, 2014). Mansbridge (2014: 58) argues that trust-based accountability requires putting ‘in considerable effort ex ante, to select the right agent, whose interests are aligned with the principal’s interests […] rather than putting in all the effort ex post, to monitor and sanction the agent.’ In other words, the efforts to select good agents replace the efforts for sanctions afterwards. Such characterization fits well with minipublics and their organization. In minipublics, organizers select their participants with care. Although the rigorous selection phase is to achieve descriptive representation, it also creates the conditions for trust-based selection model of principal-agent accountability. In fact, the selection is one of the most important aspects of organizing minipublics and it is one of the distinguishing characteristics of them (Smith 2009: 72). In many cases, the selection is done using stratified random sampling. Even for some, mere random sampling is not enough, we need data that compares participants and non-participants drawn from the sample (Fishkin 2009: 112). Every organizer needs to decide whom to sample, choose the relevant characteristics for stratification, decide if the result is sufficiently representative and add more participants if necessary. At the end of this process, we have a descriptively representative citizen body whose members take their tasks seriously that require less monitoring.

Mansbridge (2014) continues that for the selection understanding to work, it is important to find agents having a similar interest with the principals. ‘If the agent’s and the principal’s interests are aligned well, the principal can afford to engage in less monitoring and sanctioning after the selection, …’ (Mansbridge 2014: 58). In minipublics, we see that instantly. Although in strict terms we do not have principals and agents, we can safely argue that organizers authorize participants to become deliberators in minipublics and participants are willing to act as deliberators. So, we can say that the interests of organizers and participants are aligned. However, it should be noted that we cannot say there is no monitoring in minipublics. Facilitation can be seen as monitoring. Facilitators are important actors to keep deliberations up to the deliberative standards. Clearly, there might be cases where facilitators are thought to be involved in the deliberations of participants more than they should (Landwehr 2014). At least in principle, we can think facilitation as a limited monitoring mechanism, but it is also needed for non-monitoring purposes (Escobar 2019). In any case, the point made here is that the rigorous selection phase can be seen as preparing the conditions for less sanctioning and less monitoring in minipublics.

Secondly, although minipublics lack accountability and democratic legitimacy relations (Escobar & Elstub 2017; Lafont 2015; Parkinson 2006), their use in policymaking is on the rise. Then, one important question is: Is it desirable to have a citizen body with possible influence on the decisions made without strong(er) accountability mechanisms? Here, the empowerment of minipublics seems to be the key. The more empowered a body, the more accountability we would expect from it. For instance, if there is direct decision-making power, then we need to think more about the accountability of minipublics. Most minipublics do not have such empowerment, though. To date, only very few cases (e.g. deliberative polls in Greece and China) had direct decision-making power (Fishkin et al. 2008; Fishkin et al. 2010). We have a middle ground like the Canadian citizens’ assemblies, which had a higher level of empowerment than the most cases, as their organizers committed themselves to the process and promised to take the assembly’s recommendations to a binding referendum (Fournier et al. 2011).

As said, many minipublics lack higher levels of empowerment but we might still think they have the potential of influencing the public opinion. Although minipublics are usually not widely known by the general public, let’s assume that a minipublic is known by the majority of the public. Even in that case of high awareness, its influence might not always necessitate more accountability because whether the awareness can change the voting preferences of the public does not have a straightforward answer. For instance, empirical research on British Columbia Citizens’ Assembly showed that awareness of it made a difference. ‘The more voters knew, the more likely they were to say Yes, to vote for a change in BC’s electoral system,’ which was the recommendation of the assembly (Cutler et al. 2008: 186). On the other hand, while Oregon Citizen Initiative Review (CIR) is trusted and known by some members of the population, this does not necessarily change the voting preferences (Gastil et al. 2016). In brief, although every minipublic would be different in their specificities, we can argue that the conditions for and the necessity of stronger accountability are linked to their empowerment and that in principle at least, more empowerment would require stronger accountability.

Having said that, we should also be aware of the possible implications of stronger accountability measures for participants. While strong accountability in any context might bring the problem of innovation, which suggests that too much focus on accountability curbs innovative thinking because actors will be too afraid of possible consequences (Behn 2001; Rubenstein 2007), it poses a serious threat to the fundamentals of deliberation. As Warren (2008) and Mansbridge (2004) argue, strict accountability measures are not good for the genuine deliberation of the participants that requires the free flow of ideas and arguments. If participants feel that they will be accountable for their decisions or if they feel that there is a strict principal-agent relationship in which they are seen as the agents, deliberation will be hampered.

Finally, even if we want stronger accountability in the case of an empowered minipublic, we need to decide who the accountable actor is. Imagine a situation in which an empowered minipublic recommends a policy that leads to grave consequences. Who is to blame? Who needs to give an account of that policy decision? Can we hold participants individually accountable for the decision? Probably not. Can we think of collective accountability of minipublics? Maybe, but still, it would be difficult, and not very desirable to impose sanctions or bring consequences as discussed above. Here, I would argue that the collective accountability of minipublics falls on the shoulders of organizers. As said above, organizers need to justify their decisions about organizing minipublics. If they empower a minipublic, they need to give an account of that decision, too. However, the accountability of organizers does not stop there. If there is a failure related to the decision made by an empowered minipublic, it seems that the accountability for the decision shifts to the organizers even though they have no official contribution to the recommendation made. They might be held accountable by other political actors for delegating decision-making power to a citizen body.

Conclusion

This article has located itself in the descriptive studies of accountability and understands accountability as a relational concept that lies on a continuum depending on the severity of sanctions/consequences. A look at deliberative democratic literature and how it has engaged with accountability have shown that deliberation and accountability have been discussed mostly with regards to deliberators and mostly in the context of electoral politics. There seems to be a consensus on the role of deliberation as an external accountability booster. That’s to say, elected representatives should give reasons for their actions as part of their accountability, in addition to their electoral and democratic accountability.

However, this article has argued that it is not clear what differentiates deliberation from deliberative accountability, if both of them are understood as reason-giving. In addition, focusing only on the deliberative accountability of deliberators fails to see the variety of accountability relations in the context of minipublics that this article has identified. Moreover, the deliberative accountability of organizers has been highlighted and it has been argued that this rather descriptive account corresponds to our normative demands from the organizers. We would like to have deliberatively accountable organizers. We would like them to give justifications and explanations for their organizational choices.

Finally, three further points have been discussed. First, it has been argued that trust-based selection model of principal-agent accountability conception gives us a different perspective and contributes to make sense of how weak accountability of participants results in no irresponsible behaviour. Here, rigorous selection done by organizers is the key factor. Second, empowerment influences the desired strength of accountability mechanisms in minipublics. When we move from merely advisory minipublics to more institutionalized and embedded minipublics with certain legislative powers or potential to influence public opinion and formal decision-making, accountability appears to be an important concern to discuss. In principle, more empowerment seems to necessitate more accountability of a stronger kind. Third, the accountability of a “bad” decision made by an empowered minipublic shifts to its organizers even though they do not contribute to the decision-making process. Seen in this way, it seems that accountability works as a check on the conduct of organizers. It might not be formal and might not have serious consequences every time, but we can see that it is real.

Competing Interests

The author has no competing interests to declare.