Voting

Unanimity effectively gives each decision-maker an equal vote in the decision-making process, since any one member can either accept the decision or veto it by abstaining from the vote.⁵ This seems important for procedural justice for a number of reasons.

Because it depends on finding an outcome that is mutually acceptable to all, unanimity ensures that no actor has to accept a policy that it does not endorse as correct. This means that unanimity does not infringe independence or autonomy by forcing a person to go against his or her will. In this respect, unanimity means that decision-makers act as authors of the decisions that they are subject to, thereby allowing them to act as autonomous agents in decisions.

Unanimity encourages actors to deliberate and justify their reasons to one another. If each actor has a veto right over a decision, then everyone’s position has to be recognised and accommodated in the decision-making process. This means that decision-makers listen to each other, exchange information, and acknowledge each other’s opinions. It also helps actors to form opinions that acknowledge the views and positions of others. In this respect, unanimity serves an important function in encouraging actors to justify their decisions to one another.

But giving each an equal power to veto a decision is also detrimental to equality in a different sense. By giving each actor the right to veto a decision it, unanimity privileges those who favour the status quo.⁶ Each member of a group has the power to prevent a new option from being collectively adopted. Under unanimity, one actor can veto a decision, even if there is consensus from the rest of the group. Contrary to giving equal consideration to each actor’s interests, which was one of the requirements of procedural justice that I introduced in Chap. 5, this gives greater say to those who favour the status quo. This is the primary reason for thinking that unanimity is inappropriate for fairness.

Unanimity also rewards those who hold extreme views. Because it’s necessary to get everyone on board in order to make a decision, actors who do not value an agreement as much as others can hold out for what they want rather than compromising on an issue. In doing so, unanimity can discourage people from deliberating with one another and encourage them to adopt hard bargaining strategies with the promise of getting what they want. Jorgen Wettestad suggests that this is something that happened in the early stages of the negotiations for the UNFCCC, where major oil producers advocated unanimity so that they could later block emission reduction commitments (Wettestad 1999, p. 216).

Aside from concerns about fairness, there are also other problems with unanimity. For one thing, voting procedures should provide a way of reaching agreement even when people are unable to arrive at a unanimous agreement. As I argued in Chap. 2, there is reasonable disagreement over some issues in the UNFCCC. If this is the case, unanimity either leads to stalemate or brings about outcomes that represent the lowest common denominator (Sohn 1974, p. 445; Zamora 1980, p. 571). Several authors suggest that the use of unanimity in the UNFCCC has held up decisions and stifled ambition (Harstad 2009, p. 3; Dimitrov 2010; Payton 2010, p. 1).

If unanimity is problematic for fair decision-making, then an alternative voting rule to consider is consensus. Recent discussions show that the meaning of consensus is open to interpretation and still subject to much academic debate.⁷ Whilst consensus is traditionally understood as the absence of expressed objection, Lavanya Rajamani suggests that this definition should be expanded to include cases where there is some opposition to a decision (Rajamani 2011a, p. 515). This is something that’s become evident in the COP itself. For example, Jacob Werksman notes that in several other cases the COP has adopted decisions even though some parties have objected to the agreement (Werksman 1999, p. 12). This was also evident at the UNFCCC COP16 in Cancun (2010) where the Chair accepted the negotiation text despite express opposition to the decision by Bolivia. These represent more lenient standards for accepting an agreement than unanimity, since they can accommodate some limited disagreement.

By appealing to a less demanding standard for agreement, consensus avoids at least some of the deadlock and political stalemate associated with unanimity. A decision can be adopted without the endorsement of every actor, so long as there isn’t any expressed objection or abstention. Consensus therefore represents a compromise between the need to push ahead with a decision and the need to respect each actor’s position on an issue. This is a significant benefit in situations where there is a premium on making decisions quickly.

Despite these benefits, consensus fails to address the problems that led us to reject unanimity. Like unanimity, consensus still faces the problem of stalemate when there is reasonable disagreement on something. It also fails to meet our requirement for the equal consideration of interests, since it gives those who favour the status quo the ability to prevent agreement. A further concern is that consensus masks disagreement rather than resolving it (Brunnée 2002, p. 10). If a group presses ahead with a decision even though some parties harbor disagreement, then this might just postpone conflict until some later time. For example, some saw the COP20 negotiations in Lima as simply delaying discussions around the most contentious issues until COP21 in Paris the following year.⁸ Whilst consensus can provide a more efficient way of making decisions, it might mean that some actors later renege on the outcome of the decision because they feel that their interests weren’t sufficiently taken into account during the decision.

In practice, the search for consensus can be just as elusive as unanimity and may lead to political stagnation. It’s also unclear why consensus is an improvement in terms of fairness. Given that the UNFCCC is renowned for its lack of progress, and given the need for urgent action on climate change, it is worth considering the merits of majority rule procedures as a possible remedy.

Some now see majority rule as a desirable alternative to consensus and unanimity in multilateral institutions.⁹ Some might think that majority rule doesn’t fare well in terms of autonomy because it forces people to accept decisions that they do not endorse. It is true that some element of independence is lost in a majority rule process. But majority rule doesn’t mean that people give up all of their independence. For one thing, people still express their interests and views in the deliberative processes. People also maintain an element of autonomy through the voting process itself, expressing their interests or judgements about decisions. Whilst unanimity may score better on the grounds of autonomy in some sense, it’s also important to consider how to make decisions when people disagree.

Some might also criticise majority rule in terms of deliberation. The purpose of deliberation is to arrive at a mutual consensus. In this respect, one might argue that majority rule is inappropriate because it can bypass deliberative discussion, when decision-makers should really be trying to reach mutual agreement (Sohn 1974, p. 441). But this is an incorrect view of the role of majority rule in deliberative discussion. Majority rule doesn’t replace deliberation, nor is it a procedure that people should turn to once decision-makers have failed to reach a consensus. Rather, majority rule is a way of expressing our judgement about something after deliberative discussion. Jeremy Waldron convincingly argues that aiming for consensus in deliberation doesn’t mean that it should be supported as the appropriate political outcome (Waldron 1999, Chap. 5). Rather, people should first deliberate and then make a decision by a fair voting rule. In this sense majority rule is fully compatible with deliberation.

Whilst the desirability of majority rule is questionable in terms of autonomy or deliberation, there are at least two ways in which it is important for fairness. Importantly, majority rule is more favourable than either unanimity or consensus on some accounts of equality. Earlier, I suggested that there is some sense in which unanimity is an unequal voting rule, because it gives undue power to a minority. Majority rule, on the other hand, is fairer in this respect. Under majority rule, it is possible to give decision-makers different numbers of votes and in some situations, this is important for equality. Assuming that actors are roughly equal, in the sense that they are roughly affected by a decision to the same extent, then giving each actor an equal vote in majority rule voting procedure is fair because it gives each an equal say in that decision (Christiano 1996, p. 88). But actors are unlikely to be equal in this respect for many decisions. As I argue later, there are many cases where it is important to give more votes to certain actors, if people have different stakes in a decision, or if a decision-maker represents more actors. This isn’t possible under unanimity or consensus, since under these rules, each decision-maker has the same power to either support or reject an agreement. For this reason, majority rule can be more advantageous for the equal consideration of interests, because, unlike consensus, it is possible to give actors different amounts of votes, which is important for fairness in some cases.

A second way that majority rule is important for fairness concerns respect. For Waldron, majority rule respects individuals by accommodating some difference over different conceptions of justice and the common good (Waldron 1999, p. 111). Striving for consensus can play down difference, encouraging us to think that someone is wrong, or incorrect if they disagree with us. But majority rule allows individuals to maintain their difference. It accommodates the fact that people might simply disagree on some matters and in this sense it respects individuals where unanimity and consensus do not.

These two arguments give us strong reasons for thinking that majority rule is more desirable than unanimity and consensus when it comes to fairness. There’re also good reasons to think it’s advantageous in terms of procedural efficiency. But there’s still a mixed picture here and there are at least two potential objections to majority rule that need to be addressed.

One might object that majority rule is not a fair way of making decisions since it allows a minority to rule over a decision. Buchanan and Tullock argue that those who hold more orthodox views support this claim (Buchanan and Tullock 1962, p. 243). The orthodox argument is that if people have a choice between two options, A and B, and more than a simple majority is needed to reach a decision (i.e. more than 51 % of the vote) then it’s possible to end up with a situation in which the minority controls the final decision. That is, a majority of 75 % is needed for a decision and 74 % support A whilst 26 % support B, then the minority can prevent the adoption of option A. According to this argument, the minority controls whether or not a decision is made. Buchanan and Tullock rightly criticise this view, arguing that it equates positive decisions to authorise group action with negative decisions that block actions proposed by others. Whilst the minority can prevent the adoption of option A, it cannot bring about the adoption of option B. To this extent, neither group controls the outcome, but rather blocks an action from coming about.

A second concern is that majority rule marginalises those who are in the persistent minority (Bodansky 1999, p. 607). If a minority has preferences that are consistently different to the majority, then that group is always overruled in the decision-making process. This might mean that an entire group is marginalised and disenfranchised from collective decisions. This is a serious concern, but there are ways of resolving this problem and it doesn’t mean that majority rule should be rejected altogether. Individuals should be required to listen to one another and take into account each other’s interests during deliberation, which should encourage the majority to take the interests of the minority into account when making decisions. In extreme situations, where a minority is significantly affected, it might be necessary to grant veto rights to a minority group for certain decisions. This might involve appointing independent actors to observe decisions and determine whether a certain group is persistently marginalised by decisions to a significant degree. These are measures that should be taken in light of specific decision-making contexts, and the point here is to give some preliminary thoughts about how these problems can be avoided.

So majority rule is not just a fair way of making decisions; it also scores better in terms of other important values. Further, I’ve shown that it stands up to two potential objections. But this is only one side of the coin, and the following section takes up the second challenge of how to weight votes in a fair way.

Let’s consider (i) first – the magnitude of someone’s stake in a decision. It is often claimed that fair, or democratic voting processes are those that give each voter an equal vote (Kelsen 1955; Saunders 2010). The idea is that each member of a group should have an equal say over collective decisions. But some authors suggest that fairness requires that those who have a greater stake in a decision receive more votes. This is known as the proportionality principle, and it implies that those who are affected more by a decision should have a greater say in how it is made.¹¹ In what follows, I introduce the argument for proportionality and discuss its merit as a principle of fairness. I argue that it is a fair way of making decisions when actors are voting to advance their interests in a decision. I then go on to argue that proportionality is not a fair way of making decisions when voters are voting to express their judgement or opinion in a decision.

I start by looking at Brighouse and Fleurbaey’s comprehensive review of the proportionality principle, in which they give four arguments for favouring it over equal suffrage (Brighouse and Fleurbaey 2010, p. 138). Their first argument is based on premise that fair decisions are those that treat individuals with equal respect. For the authors, equal respect means giving equal consideration to each individual’s interests in a decision. In this respect, Brighouse and Fleurbaey share Thomas Christiano’s justification for democracy (Christiano 1996, 2008, p. 78). But the problem with Christiano’s approach is that giving each actor an equal say in a decision doesn’t give equal consideration to each actor’s interests when some are affected by a decision more than others. For Christiano, this means that a necessary condition for democracy is that every member of a group is affected by its decisions to roughly the same extent, which puts a strong restriction on the sorts of decisions people can make. For Brighouse and Fleurbaey, however, if some are affected by a decision more than others, then these actors should be given more votes for the sake of achieving an equal consideration of interests.

This is the main argument for proportionality. But there are other arguments in support of this principle. The proportionality principle is also relevant for fairness because it is important for autonomy. If individuals have unequal stakes in a decision, then giving each an equal vote means that those who are more affected by a decision are subject to the will of those who are less concerned. By giving each actor a say according to the stake that they have, the proportionality principle gives individuals greater control over these decisions, thereby enhancing autonomy and independence.

Proportionality also appeals to epistemic concerns. Those who are more affected by a decision are more likely to be better placed to make decisions about it. This means that giving more votes to the more affected should lead to better outcomes, in the sense that these outcomes have higher epistemic quality. Although this isn’t an argument for fairness, it does give additional support to proportionality.