1.1 – State of the art and approach of the study

5With the exception of few works such as James Coleman’s (1990), formal decision and social choice theory have overlooked decision-making by tacit consent or taken it for unanimity. Certainly, “it has not attracted the attention its significance as a way of making collective decisions deserves” (Olsen 1972, p. 273). The reasons lie partly in the nature of the rule itself: First, it is often used informally, and then primarily enforced via social norms; since it, then, isn’t codified, there is no written document from which the application of the rule can be deduced unambiguously. Second, it is precisely the absence of an observable act – the avoidance of any ordered and active expression of group preferences – that constitutes its voice concentration mechanism; therefore, its logic is often not grasped and implications of this rule of preference collection are neglected. [5] Besides, this ambiguity of individual preferences frustrates formal modelling, which is essential to social choice theory. And finally, in our cultures “we learn that political choices are made mainly through collective voting procedures” (Olsen 1972, p. 267), and this underlying assumption shapes our view on decision-making in all kinds of settings.

6Yet, if we fail to realise the prevalence and often institutionalised nature of this way of collective decision-making in certain contexts, we may take its outcomes for unanimous decisions (e.g. Tsebelis 2013, p. 1084) and fail to grasp why critical voices did not use their alleged veto right in the respective decision-making contexts, as Thomas König and Dirk Junge (2009, p. 508) do. Or we will assume that all participants of a ‘tacit consent’ decision-making process fully support the apparently unanimous decision and be surprised when the decision turns out to be less binding than we would expect. Yet, collective decisions without voting are very often not defective outcomes of the unanimity rule, but outcomes of a decision-making rule in its own right, which follows certain patterns and differs in important respects from unanimous voting. Only when we acknowledge the distinctness of the mode and its dynamics are we able to assess the degree of commitment, acceptance and convergence behind those joint consensual decisions, and judge their democratic legitimacy as well as their empirical suitability adequately.

7The distinctness of this phenomenon has been recognised by a range of case studies that investigate contexts where this decision rule is in force (cf. della Porta 2009; Ehlermann and Ehring 2005; El Hakim 1978; Krick 2013, 2015; Lewis 2010; Nail 2013; Novak 2013; Steiner and Dorff 1980; Dorff and Steiner 1981; Urfalino 2012; Wiredu 1998; Vignes 1975). Yet, theoretical considerations of this decision rule that go beyond the single case and try to describe general features or the logic of this technique of decision-making have been relatively rare (cf. however Coleman 1990; Krick 2015; Olsen 1972; Thiele 2008; Urfalino 2014). These bodies of work will be built upon prominently to describe the logics of tacit consent decision-making and to point to typical empirical phenomena and variations. Yet, most of these contributions explicitly attribute veto power to each member of a consensual decision round, and in none of these is the idea of an individual veto option seriously questioned. This position will be challenged here by scrutinising the meaning of ‘veto’ in collective decision-making contexts that avoid the use of voting techniques and where the expression of critique and disagreement is part of the preceding deliberation process.

8For clarifying the logics of (consensual) joint decision-making and discussing the legitimacy of tacit consent outcomes, this analysis also draws on studies on decision rules and deliberation (Dryzek and Niemeyer 2006; Manin 1987; Moore and O’Doherty 2014; Novak 2014; Offe 1983; Rae 1975; Romme 2004; Sadurski 2008) as well as works on conflict resolution, collective decision-making and communication (cf. e.g. Buchanan and Tullock 1962; Hopmann 1996; Kuran and Sunstein 1999; Nullmeier and Pritzlaff 2009; Olsen 1972; t’Hart et al. 1997). For describing how (particularly informal) rules are applied, institutionalist studies will be consulted additionally (cf. Azari and Smith 2012; Helmke and Levitzky 2004; March and Olsen 1989).

1.2 – Concentrating preferences without voting

9The above-mentioned studies on decision rules point out two interdependent core characteristics of the decision-making technique in focus here – vote avoidance and the absence of open dissent. These aspects can be considered two sides of the same phenomenon and together they establish a social mechanism: since no voting technique is used, decisions are not taken by active consent, but passively, by tacit consent or non-opposition (cf. Coleman 1990, p. 857). Turned around: since the absence of any open opposition suffices, voting can be avoided. [6]

10Mutual bindingness, group coherence and “collective belief formation” (Kuran and Sunstein 1999, p. 683) in settings that do not vote can be generated via ‘acceptance acts’ of the participants, which are interlinked, e.g., through the technique of “proposal re-narration”, whereby participants “integrate the proposal into their own, differing background assumptions and evaluative standards, into their individual ‘web of meaning’” (Nullmeier and Pritzlaff 2009, p. 368). These acceptance acts are usually rounded up by a “confirmation act” in the name of the collective as a whole, typically performed by the head of the group (ibid.) who states, for instance: “I take it, we have reached consensus then” or “if there is no further objection, I consider the decision to be taken”. So, under consensus rule, a decision is understood to have been taken when there is no obvious act of disapproval that repeals the final “confirmation act”. The respective threshold of such disapproval or open opposition is of course open to interpretation and depends on the specific situation. While some groups and committee chairs may not accept mumbling and frowning as below the threshold of open opposition, for others only well-argued contestation will count as dissent.

11In any case, there is no systematic cast and count of votes under consensus rule, no systematic expression of preferences (cf. also Urfalino 2012, p. 186), and thus there is uncertainty about the degree of consent behind the outcome (Steiner and Dorff 1980, p. 7). Indeed, outcomes can cover various degrees of actual agreement, ranging from full informed consent of a group to decisions that are actually backed only by a majority over to extreme cases of unilateral domination (Hopmann 1996, p. 248). When voting is avoided, there is no need for full, active unanimity; “partial unanimity” or “joint acceptance” to not argue further but “let something stand as the position of the group” can suffice (Moore and O’Doherty 2014, p. 302). “The support for the chosen alternative may (…) vary from enthusiasm to abstaining from obstruction” (Olsen 1972, p. 273), for reasons as various as inferiority, “keeping face”, identification with the group, lack of interest in the particular decision and, indeed, full group consent.

2.1 – Institutionalisation of formal and informal rules

13In general, it is by sanctioning mechanisms, the “punishment of non-conformers” (Azari and Smith 2012, p. 40) that behavioural regularities are institutionalised. Of course, “not every unwritten (or, indeed, written) rule is enforced vigorously, but the consistent absence of sanctions applied to observed violators would suffice to demonstrate the nonexistence of a rule” (ibid.). While particularly formal rules may be imposed and enforced by direct coercion through official sanctioning channels, informal rules are chiefly governed by social norms and expectations. They are often “part of a code of appropriate behaviour that is learned and internalized through socialization or education” (March and Olsen 1989, p. 22). Informal sanctioning mechanisms can be summarised as practices of withdrawing recognition of the other.

14While the application of a formal rule is usually less contested, the institutionalisation of an informal use is easily overlooked. Yet, we can speak of an informal rule or institution to be in place, if a certain behavioural regularity is usually shown in certain contexts and if we can be sure that this is neither arbitrary nor an uncoordinated reaction to a common stimulus (Helmke and Levitzky 2004, p. 727) – like the (collective) act of leaving the office building in case of fire – but a social practice whereby people relate to each other. Since the informal rule is usually applied, while deviant behaviour is likely to be considered a violation and be sanctioned, it shapes social behaviour, even though it is not written down.

2.2 – Social norms under consensual rule

15Like all decision rules, the consensus rule can be applied formally or informally, i.e. it can be a written or an unwritten rule. [8] It can be accompanied by other decision rules that constitute fall-back options in cases of severe obstruction, for instance. We can speak of this rule to be in force if a decision-making entity usually (or concerning certain decisions) decides collectively without voting and by absence of open dissent, and participants thus expect such a mode of decision-making in general (or with regard to particular issues). Surely, this does not mean that expectations are always met or that the rule is always respected. Yet, it means that violation usually has consequences. An example of violating the consensus rule – or “inappropriate behaviour” – would in most cases be to press for a vote in situations where arguments are still being put forward, where delegations show commitment to compromise, and where group coherence is growing (Lewis 2010, pp. 171f.). In tacit consent arenas, rule-breaking behaviour such as pressing for a vote during the process of deliberation can be collectively rejected or else be sanctioned via withdrawal of recognition or by confrontational orientations in the future – a serious sanction in permanent joint decision-making groups that rely on reciprocity and trust (Scharpf 1997, p. 144; cf. also Lewis 2010, p. 180; Olsen 1972). Since sanctions can affect status, project success, and other valued outcomes (Azari and Smith 2012, p. 40) and rule-violation is very costly for the violator, it will happen rarely and usually only be ventured by very inexperienced or socially independent, very audacious or very powerful actors (Ehrenberg and Ehrling 2005, p. 66; Novak 2013, p. 1101; Olsen 1972, p. 276).

2.3 – Voting as a fall-back option

16Social interaction in collective decision-making settings is of course governed by uncountable rules, regarding everything from appropriate clothing over to the adequate tone or form of statements and to the quorum of a decision-making entity, which will often be codified. Whatever their status of formalisation, there can be different rules governing the same issue under different conditions, which can complement, accommodate, substitute or compete with each other (Helmke and Levitzky 2004, pp. 728ff.; cf. also Azari and Smith 2012; March and Olsen 1989, pp. 24ff.)

17The consensus rule very often co-occurs with the decision rule of majority voting (Thiele 2008, p. 297). The WTO Agreement states, for instance, that “where a decision cannot be arrived at by consensus, the matter at issue shall be decided by voting” (UN 1995, p. 13). While there are several cases like this where this particular combination of decision rules is codified, it is much more often the case that an informal consensus rule replaces an either formalised or informal majority rule. [9]

18In cases where the consensus rule accommodates or substitutes majority voting, consensual decisions without voting are usually the preferred outcome, while majority voting constitutes the fall-back option. If a group is unable to agree on a joint decision tacitly despite intensive debate, a (usually majoritarian) vote constitutes the last resort. The “shadow of the vote” (Lewis 2010, p. 181; cf. also Ehlermann and Ehring 2005, p. 52) works as “a threat, an inducement to achieve a consensus” (Vignes 1975, p. 120), it disciplines participants to play by the rules and show cooperative behaviour and it contributes to stabilising the consensus rule.

19Thus, when voting occurs in arenas that usually decide by tacit consent, it is neither generally a rule violation nor does it means that the arena as a whole is no longer shaped by the consensus rule. Rather, because time is up or because of ongoing conflicts, the group may have replaced the tacit consent rule by a voting rule for this particular decision. Besides, even hard-boiled consensus arenas can be accustomed to regularly use voting mechanisms for certain decisions, such as “technical” or “procedural” issues. In fact, most “real-life” arenas will use both informal and formal mechanisms of voice collection, depending on the situation and the issue in question (Olsen 1972). And indeed, voting can also be used strategically when unanimity exists to underline the unity of the organisation (Olsen 1972, p. 280).

3.1 – What constitutes a veto?

20For discussing the general assumption of individual veto power under consensus rule (cf. El Hakim 1978, p. 56; Moore and O’Doherty 2014, p. 305; Nail 2013, p. 30; Thiele 2008, p. 299; Urfalino 2014, p. 187), it needs to be clarified, first of all, what is meant by ‘veto power’. The meaning of ‘veto’ is not straightforward and warrants a much closer look in contexts where the explicit stating of preferences at the end of the process is avoided while the decision process is characterised by deliberation and debate – of which disagreement and objection are integral parts.

21The term veto refers to a unilateral refusal of assent to a proposal that stops a collective decision, while the absoluteness of these stopping powers varies. In constitutional law, one differentiates minimally between suspensive vetoes that postpone and absolute vetoes that ultimately end decision-making. Most successful attempts to block a proposed decision de facto have only suspensive effects because those issues tend to re-enter the agenda (Romme 2004, p. 706). Yet, while absolute vetoes may be rare in most political systems (McCarty 2000, p. 509), suspensive vetoes, which postpone or stop a particular decision until conditions change, are also a powerful, a prevalent and potentially disruptive tool in decision-making contexts.

22Within the informal, debate-based settings in focus here, it is not easy to draw a clear line between a (suspensive) veto and “regular” objections and disagreements. Expressions of reservation, disagreement and counter-arguments should characterise the course of every efficient deliberative process and they will usually have a suspensive effect – they will rule out or depreciate certain options, they will continue and extend the process of reason-giving and deliberation, and they will sometimes put off decision-making to the next meeting. One surely wouldn’t like to call any such critical remark, any argument or anything that is not plainly affirmative, a veto. The expression “veto” should be reserved to behaviour that stops a certain proposal from becoming a decision. This is also what authors seem to have in mind when they mention vetoes under consensus rule that are able to effectively block a certain decision (Nail 2013, p. 30; Thiele 2008, p. 299), prevent a proposal from becoming a decision (Urfalino 2012, p. 187), and paralyse the decision process (Thiele 2008, p. 352).

23“Veto” also needs to refer to expressions of dissent at the end of a decision process. What constitutes the end of a process is of course more difficult to ascertain in consensus procedures where decisions are not taken at one particular point in time, by way of collecting and fusing preferences formally, but where the decision develops slowly and where the moment of decision-making is indistinct. With regard to consensual decision processes, a veto would be an expression of disagreement at a point where a joint decision already asserts itself, i.e. when no further arguments are put forward and acceptance acts confirm each other.

24The assumption that this study argues against, is the idea of individual, effective veto power under the rule of tacit consent, sometimes described as a veto right. [10] Of course, I do not claim that there are never vetoes under consensus rule or that there are no consensus arenas where individual veto powers exist. The power to veto differs from arena to arena, and depends on the decision and communication culture and on the power structure of the collective, as well as on the relationship of the consensus rule with other decision rules. Yet, there are similarities between all consensus arenas that follow from their technique of preference collection and that undermine the use of vetoes very effectively, and we can elaborate on the preconditions that invalidate the veto option in many empirical cases.

3.2 – How the “logic of appropriateness” restricts veto power

25Under consensus rule, vetoes are rarely seen even when disagreement prevails. One reason for this is that an expression of disagreement at the end of a consensual decision process constitutes a severe disruption and is therefore usually avoided. Consensus arenas share certain behavioural patterns that facilitate the achievement of collective decisions without voting and, in effect, undermine the appropriateness of vetoes: a group will have to debate excessively in order to develop jointly acceptable policy solutions, i.e. decisions that are (tacitly) approved by the whole group. Participants of consensus procedures will typically show cooperative attitudes and a certain commitment towards a joint group decision. For this matter, they will engage in intensive communication and largely inclusive deliberation (Coleman 1990, p. 381), and ensure that all serious concerns have been allayed (Urfalino 2014). [11] Possible objections become apparent during the process and can be dealt with in communication. Vital interests are expected to be voiced during the process. Yet, a systematic expression of preferences is not intended in consensual settings. By way of ‘internal’ and ‘external’ deliberation, agents will develop their initially incomplete and incoherent set of preferences (Manin 1987, pp. 350-352), adapt their own individual position as well as the group’s positioning, and a joint decision can assert itself. The final confirmation act will usually only be performed when – after intensive discourse – no new and incontestable counter-arguments are put forward and group coherence has reached a stable level. Trying to block a decision that is already emerging by showing sudden disagreement at the end of a decision process, without putting forward further convincing arguments and showing commitment to find a solution, is likely to be met with contempt by the group. Research on collective communication and decision-making has pointed to social dynamics that make it more and more difficult to utter rejection when group coherence about a proposal grows. Once started, “acceptance loops” (Nullmeier and Pritzlaff 2009, p. 368) and “spirals of silence” (Moore and O’Doherty 2014, p. 306) constitute “self-reinforcing process[es] of collective belief-formation” (Kuran and Sunstein 1999, p. 683) that can be de facto impossible to breach.

26Apart from these general reasons, vetoes will also be avoided due to reasons that are specific to different types of consensus institutions:

27In arenas that are characterised by a strong collective identity, such as social movement organisations, families, colleges or epistemic communities, an expected commitment to the common goal and a “unity of purpose” (Della Porta 2009, p. 79) can undermine the use of vetoes. It can be expected to “let something stand as the position of the group” (Moore and O’Doherty 2014, p. 302) and restrain from selfish objections and obstructionism (Urfalino 2014, p. 17) for the sake of a joint decision. In consensual arenas that are particularly small and close-knit, “dense relationships” (Coleman 1990, p. 381) of trust and even friendship (Olsen 1972, p. 275) and a culture of co-operation, social cohesion and deliberation tend to develop (Scharpf 1997) that will prevent voting (Olsen 1972, p. 276). Conflict potentially destroys these relations and will be perceived as costly, which makes disruptive behaviour more unlikely (ibid.). A “logic of appropriateness” (March and Olsen 1989, p. 22) that lets one choose the behaviour which is most appropriate for a given role and circumstance, shapes social interaction and the open pursuit of personal interests can be deemed illegitimate. If such arenas are additionally stable entities that exist over time, reciprocity norms can help to reinforce a culture of tacit consent (Lewis 2010, p. 180; Scharpf 1997, p. 144) because participants can be rewarded for tacit consent by package deals that extend into the future. In addition to that, vetoes can be prevented by a “logic of consequentiality” (March and Olsen 1989, p. 23f.), which makes individuals choose the course of action with the best consequences for themselves. In the Council of the EU, for instance, participants often expect that open opposition may cause criticism from their constituents, which triggers a “strategy of blame avoidance” by silently consenting to adopted measures, as Stéphanie Novak (2013, p. 1100) points out. Siding with the winning team in consensus procedures prevents players in power politics to lose face externally. Not vetoing a decision in these cases is then not so much attributed to internal social norms of cooperation and to the common good but to the personal costs linked to such behaviour. International organisations seem to be shaped particularly by this logic of power and consequences, where participants are concerned with protecting their state’s sovereignty within a relatively large group of members that cannot develop pronounced personal ties.

28In all cases and particularly when decision rules are not formalised, the articulation of a veto needs to be recognised as valid by the collective to gain any effect. For instance, in expert committees, which are chiefly shaped by epistemic norms, you have to give reasons to justifiably oppose a position. Here, “the value of contestation, its ‘veto’ power, is conditioned by whether or not the arguments used against the proposal are recognised as valid” (Urfalino 2014, p. 18).

3.3 – The unequal distribution of veto power

29Part of the assumption regarding vetoes under consensus rule is that each member of a decision round has the power to veto decisions (El Hakim, 1978, p. 56; McCarty 2000, p. 509; Moore and O’Doherty 2014, p. 305; Urfalino 2012, p. 187). But can we really speak of effective individual veto power under consensus rule?

30Since vetoing collective decisions tends to breach the social norms of collectives that follow the rule of consensus, not all actors will dare to use this option alike – rather, it will be exercised more easily by the more powerful actors or by “social hermits” (Olsen 1972, p. 276). Philippe Urfalino (2014, p. 16) notes that members of consensus arenas only have veto power vis-à-vis equally or less powerful agents. This is confirmed by Claus-Dieter Ehlermann and Lothar Ehring (2005) with reference to the WTO, who state that “consensus does not provide for equality (in terms of decision and influence) because not every member has the same ability to maintain vetoes” (p. 51). In clans, but also in International Organisations, more powerful agents will make sure to underline the importance they attach to certain decisions so that ‘normal’ members will be aware of the high costs of vetoes and not show disruptive behaviour (Urfalino 2014, p. 16). “Where a member is alone in opposing a decision, it can find itself in quite some isolation and exposed to quite some pressure which arguably only robust, big members can sustain for an extended period of time” (Ehlermann and Ehring 2005, p. 66). In the WTO, the veto capacity of members tends to be linked to size and importance in international trade (Ehlermann and Ehring 2005, p. 66). In expert committees, power is not built on military and economic strength but on resources like habitus, argumentative wit and astuteness. In clan and family structures, the different weight of junior and senior members is often explicit, and the asymmetry of veto powers is apparent and often recognized by the group. Since the veto option is generally linked to power structures, and since power imbalances are a fact of social life, it can be called an asymmetrical power (Rae 1975; McCarty 2000). Certainly, the assumption of individual veto power of every participant of consensus arenas therefore needs to be reconsidered.

31Some of these arguments generally apply to arenas that depend on deliberation and accommodation. The unequal distribution of veto options has also been mentioned for unanimous voting (Romme 2004, p. 707), where the equivalent to remaining silent when you are too weak to speak up would be to adapt your preferences to the more powerful agents and vote with them. Yet, there are differences between the logics of these decision rules and the contexts of their application that restrict veto options far more under consensus rule: The consensus rule tends to be informally applied; since the question of validity of a veto plays a larger role when rules are not formalised, a successful attempt to block a consensual decision is less likely here. Under unanimous voting, by contrast, one would not expect the individual right to veto, aggravating and disruptive as it may be, to be questioned as such. Besides, and decisively, since preference conversion under consensus rule happens incrementally and since a systematic voicing of preferences is explicitly not intended, an expression of disagreement, particularly a sudden one at the end of a process, is likely to be much more inappropriate than when voting rules are used. Most notably, it is specific to the consensus rule that it is usually accompanied by a majority voting rule as a fall-back option. This peculiar arrangement invalidates the veto option under consensus rule completely, as shown in the next section.

3.4 – De-activated veto options in the shadow of majority voting

32While the preceding arguments call attention to the asymmetrical distribution of the alleged veto power under consensus rule, there are – quite common – empirical circumstances where the veto option is de-activated altogether. As mentioned before, the consensus rule often replaces majority voting as a fall-back option. If this fall-back option is not defunct, there is no effective veto power for any member: If you oppose a decision ultimately under these circumstances, you will not stop the decision itself from emerging, because the majority rule will be applied instead of the tacit consent rule. In effect, your disagreement is then not a trump, but a majority decision will be taken on the issue without your approval. Veto power thus does not exist in the shadow of the vote, since expressions of disagreement do not stop the respective decision but lead to exclusion of the opponent. [12]

33Although an attempt at vetoing a consensual decision in the shadow of majority voting will not stop the respective decision from being taken, and therefore shouldn’t be called an effective veto, it can nonetheless be politically rational because it reduces group commitment, amongst other things. Compare the interesting case of the British Prime Minister David Cameron who in June 2014 threatened to call for a vote and veto the European Council’s decision on the Commission Presidency (Corbett 2014; Watt and Traynor 2014). Since “negative votes and abstentions can be useful tools for ministers to communicate with lobbies or their electorate or to escape responsibility for an unpopular decision” (Novak 2013, p. 1100), this was obviously a strategic move. Calling alternately for a vote and threatening to veto the respective decision is of course contradictory, because a vote on the Commission presidency is formally subject to majority rule, which does not give room for vetoes. This led to an application of the fall-back option of qualified majority voting and resulted in David Cameron and the far right Hungarian Prime Minister Viktor Orbán voting together against the candidate. It did not give Cameron veto powers, however, since the decision went through without his approval.

4.1 – The unanimity and the majority principle

36Following the principle of unanimity, there has to be full agreement or there will be no valid decision. When applied as a strict voting rule, it attributes an effective and equal right to consent and to veto to every member of the group (Novak 2014, p. 682; Rae 1975, pp. 1277f., Sadurski 2008, p. 48). This “most inclusive decision rule” (Buchanan and Tullock 1962, p. 69) is the only rule of preference concentration that grants pareto optimality (ibid.; Dougherty et al. 2014, p. 360). There is a long tradition of regarding the unanimity rule as ethically superior (Buchanan and Tullock 1962; Novak 2014; Romme 2004, p. 707), as the ultimate means of authorization in democracies (Beatty/Moore 2010, p. 198), “the gold standard of political justification” (Dryzek and Niemeyer 2006, p. 634) and on principle the only legitimate basis for collective rule and political power under conditions of individual liberty and equality (Manin 1987, p. 341). [13] The downside of its theoretically highly inclusive nature are its impracticabilities, since it is generally slow, even utterly infeasible in large groups and fraught with high transaction costs and a bias towards the status quo, amongst normative dangers such as the tyranny of a minority (Beatty/Moore 2010, p. 200, p. 204; Kelsen 1955, p. 25; Romme 2004, pp. 705f.). Both its strengths and its weaknesses are attributed to its main characteristic, the requirement of absolute consent and its flipside, the individual veto option. The majority rule, which clearly subordinates the minority to the majority (Novak 2014), is often described as a second best (Sadurski 2008; Novak 2014, p. 684), but much more workable solution for everyday collective decision-making that “offers a maximum of certainty that decisions will be made at all and with relatively small costs“ (Offe 1983, p. 713). Apart from these important procedural qualities, it has also been attributed intrinsic normative value, because of its respect for individual freedoms (Kelsen 1955, p. 25) and social diversity (Beatty/Moore 2010, p. 199; Novak 2014, p. 684) and because of the equality of weight it attributes to every participant during the process (Dougherty et al. 2014; Sadurski 2008). It has therefore been described as the best synthesis of freedom and equality (Kelsen 1955, p. 25) and thus seems to lend itself particularly well to mass decision-making in liberal democracies.

37These two voice collection logics will now be used as benchmarks against which the consensus rule is assessed.

4.2 – Input-legitimacy: democratic self-determination under consensus rule

38Comparing the equality of participants under consensus rule with voting rules, where the rule of one-man-one-vote applies, [14] Coleman states that “the explicit allocation of rights given by the unanimity rule over the events in question makes more egalitarian, at least in principle, the power distribution among members of the decision-making body” (1990, p. 861). By contrast, imbalances of power can take much more hold under consensus rule: Powerful agents, i.e. those with access to informational, constitutional or monetary resources, can influence the decision more than others (Coleman 1990, p. 381, p. 858; Ehlermann and Ehring 2005, p. 67). Since individual veto options are essentially invalidated, the consensus rule does not protect individuals from the coercion of others as does the unanimity rule, where no-one has to comply with unwanted directives (Manin 1987, p. 338; Sadurski 2008, p. 48).

39In its own way, the consensus rule can protect minorities or, more precisely, those that have not shaped the joint decision decisively (Ehlermann and Erhing 2005, p. 66): The soft mode of voice accumulation and conversion and the invisibility of disagreement that may lie behind a consensus decision avoid inflicting an explicit loss on those that do not particularly agree with the joint decision (Ehlermann and Erhing 2005, p. 66; Urfalino 2014, p. 333). Under consensus rule, remaining silent – and not being able to explicitly veto a decision – can save these individuals from losing face in competitive voting and facilitate blame avoidance (Novak 2013, p. 1100). Yet, for one thing, this feature underlines how the consensus rule obscures and accommodates the fact of inequality (Urfalino 2014, p. 339). Due to the extended discussions under consensus rule, intensities of preferences additionally determine an agent’s weight and thus increase inequalities (Coleman 1990, p. 858). [15]

40In comparison to majority voting, the consensus rule is surely not as open towards social diversity, since its processes are strongly shaped by a strive towards a joint, undisputed decision; when disagreement remains, there will be strong pressure not to speak up and claim a veto right (Ehlermann and Ehring 2005, p. 66; Urfalino 2014), but to support a proposal for the sake of the group, for instance.

41From a perspective of transparency and accountability, the consensus rule suffers from opacity in contrast to voting rules. Some opacity is typical for negotiations in general, even those that end with majority decisions (Ehlermann and Ehring 2005, p. 66; Scharpf 1997). Yet, “a formal vote, usually taken by hand, creates the highest possible degree of transparency of opinions in a context of a face-to-face meeting, simply by its clarity of signs” (Nullmeier and Pritzlaff 2009, p. 366). Tacit consent decisions, by contrast, are distinctly characterised by the ambiguity of the group’s preferences. If the decision-making unit is made up of delegates, this feature undermines the accountability relationship of decision-makers towards the represented severely.

4.3 – Output-legitimacy: Efficiency and effectiveness of the consensus rule

42While Thiele (2008, p. 298) sees no difference in terms of bindingness and group commitment between the unanimity and majority rule and the often informal consensus rule, Jürg Steiner and Robert Dorff (1980, p. 8) argue that consensual decisions are likely to lack persistence. In comparison to voting rules, consensus rules do not produce clear and procedurally undisputable choices (Novak 2014, p. 684) and this larger room for interpretation about the content of a decision will reduce their bindingness partly. In comparison to majority decisions, the consensus mode usually assembles a greater share of the group behind the decision, and can thus be expected to generate a higher degree of acceptance and commitment. In contrast to unanimous voting, however, individual participants can hide behind joint decisions that were made without voting and feel less committed individually, since they have not actively and clearly expressed their support, nor their disagreement by using veto powers. Yet, one has to acknowledge that consensus decisions are often backed by alternative sources of authority and legitimacy that lend them validity. For instance, tacit consent procedures are typically characterised by processes of deliberation and negotiation, i.e. ongoing processes of reason-giving and arguing for an outcome, whereby agents create a normative foundation for the authoritative character of a decision, which render it effectively binding in practice (Nullmeier and Pritzlaff 2009, p. 358, p. 368).

43For the collective bindingness and the implementation of decisions, not only overall group commitment but also substantive backing by the most powerful can be important. The consensus rule is striking in this respect, as it advantages the stronger agents, or, in other words, reflects the stakes that members have in the issue as well as their influence (Ehlermann and Ehring, 2005, p. 66); “no decisions are likely to be taken against the opposition of the big and mighty, who generally need to implement the decision for it to have practical value” (ibid, 67).

44The tacit consent rule is characterised by relatively high overall efficiency: in comparison to the unanimity rule, it generates relatively low transaction costs, i.e. costs that the individual participant of a collective has to incur, in the form of time or other resources (Buchanan and Tullock 1962, p. 68). They are the highest under the most inclusive voice concentration rule, unanimity (Romme 2004, p. 705), since everyone’s consent is essential for the final decision, and they have a potentially paralysing effect (Thiele 2008, p. 277; pp. 352 ff.). In comparison to majority voting, however, tacit consent decisions are clearly more costly in terms of time (Thiele 2008, p. 299) and are thus usually not envisaged when time is a scarce resource or when a decision is not considered important enough for such a costly process (Olsen 1972, p. 274). At the same time, when compared to majority voting, the rule has been argued to come with relatively low ‘external costs’ – i.e. “costs that the individual expects to endure as a result of the actions of others over which he has no direct control” (Buchanan and Tullock 1962, p. 45) – because of the general orientation towards consensus and the deliberative nature of the process, which potentially gives a voice to all participants during the process. Thus, the rule of tacit consent strikes a balance between the two types of costs of collective decision-making: the group concedes a certain degree of deviance (external costs) for the sake of collective problem-solving within reasonable time (transaction costs).

45Finally, there is a chance that orientation towards consensus and deliberation promotes better, more adequate policy solutions to problems in comparison to the competitive majoritarian mode, because problems are likely to be looked at from different angles in deliberative interaction (Young 2000, p. 6, p. 29, p. 76) and because of the ‘double dimension’ of deliberation that promotes processes of internal deliberation or learning alongside external collective exchanges of reasons and perspectives (Manin 1987, pp. 350-352). Yet, the powerful silencing effect of growing group coherence that characterises small, face-to-face interaction can also undermine the quality of a consensus decision (Nullmeier and Pritzlaff 2009, p. 396; t’Hart et al. 1997, pp. 10-11) and run counter to an exhaustive pondering of reasons and solutions. During deliberative interaction, particularly when the social dynamic of ‘acceptance loops’ has started, voicing counter-arguments quickly becomes virtually impossible and the epistemically perilous “group think” and “spirals of silence” may prevail particularly within homogeneous groups (Romme 2004, p. 706; Moore and O’Doherty 2014, p. 306). These social mechanisms and the transaction costs of deliberated decision-making can lead to neglect of crucial information concerning risks of certain policies (t’Hart et al. 1997). These aspects surely apply to all deliberating arenas and are thus not unique to, yet probably typical for the consensus rule.

46Due to the restricted veto options, the consensus rule is not as biased towards change as is the unanimity rule, which tends towards extreme conservatism, because “even a single person can block change altogether” (Romme 2004, p. 706), and which “seems to be based on the unstated presumption that the costs of inaction are nil” (Sadurski 2008, p. 49). Yet, in comparison to majority voting, the consensus rule surely “unfairly privileges the status quo” (Beatty and Moore 2010, p. 200).