Reasonable Disagreement and Political Deadlock

Of course, before clarifying what I mean by reasonable disagreement, it helps to say something about disagreement first. The notion of disagreement suggests that actors hold different opinions, ideas, or claims about an issue at hand.² It is distinct from the idea of difference, which merely implies that actors do not have the same characteristics. Rather, disagreement implies that there is an absence of consensus on an issue and that each party thinks that the other’s position is wrong. Because this chapter is ultimately concerned with disagreement in the UNFCCC, I want to focus on a specific area of disagreement, namely: disagreement over the fair terms of cooperative action. That is, this chapter considers situations where there is a need for cooperation on matters of common concern, and there is a common understanding that the terms of this cooperation should be fair. At the same time, there is disagreement about what these terms should be.

This reflects the situation in the UNFCCC itself, where climate change requires collective action, and all parties to the convention have agreed to participate in the cooperative project. At the same time, there is disagreement about the exact terms of cooperation. There are different types of disagreement over this issue, and the likelihood of moving towards agreement depends in part on what type of disagreement actors are faced with. For this reason it is worth distinguishing between these different types of disagreement that might arise in relation to the terms of a cooperative arrangement.

Epistemic disagreement can be disaggregated into two types:

This typology is not exhaustive; there are many other kinds of disagreement about collective action. The reason for highlighting these specific types of disagreement is that they are the most relevant forms of disagreement that arise over the fair distribution of emission rights.

The important point to draw from this typology is that some types of disagreement are likely to be more intractable than others. For instance, some epistemic disagreement is just a matter of fact; disagreement can be resolved by providing the correct information about empirical facts, or by ensuring that actors cooperate in good faith. Other types of disagreement are much more intractable.³ Matters of interpretative or weighting disagreement are likely to rest on differences that cannot be resolved by providing correct information, or by cooperating in a fair way. The full implications of this point are to follow. For the moment, it is only necessary to point out that different forms of disagreement are easier, or harder to resolve.

This typology does not explain how disagreement can come about. There are many reasons why actors disagree about collective decisions. As Amy Gutmann and Dennis Thompson point out, actors might disagree simply because they adopt self-interested positions in an argument, which are likely to be different for different actors (Gutmann and Thompson 1996, p. 18). This explains why actors disagree in certain situations. But here I’m primarily concerned with the notion of reasonable disagreement, rather than disagreement more generally. As I alluded to earlier, the fact that there is reasonable disagreement in the UNFCCC necessitates attention to procedural fairness. For this reason, I move on to discuss the concept of reasonableness in the following section, before moving to a discussion of how reasonable disagreement can arise in the section following that.

Reasonableness is a property that governs how actors behave, and subsequently the types of claims that they are willing to make and accept in a fair procedure.⁴ A reasonable proposal is not simply a matter of self-interest, nor is it wholly altruistic. Rather, a reasonable actor is one that conforms to certain requirements, which fall into two categories: sound reasoning, and reciprocity.

First, reasonable actors are those who act under certain conditions of reasoning. That is, reasonable actors act in accordance with principles that govern the competency of the actor’s reasoning and judgement. These conditions require, for example: that an actor reasons in a consistent way, using methodologies that are commonly understood. This is the first requirement of reasonableness.

The first requirement of reasonableness implies that actors act with sound reasoning, in the sense that they do not employ unsound, or incapable reasoning, and there are no obvious logical errors or flaws. But reasonableness concerns more than just sound reasoning. As Thomas Scanlon argues, one can reason competently, whilst also behaving very unreasonably (Scanlon 1998, p. 192–3).⁵ In a bargaining situation, a rational actor may adopt hard bargaining tactics, mislead others, or exploit those who are in weaker positions. Whilst such behaviour may meet the first requirement of reasonableness defined above, it is clearly not reasonable in a broader sense of the term. Rather, reasonableness also requires some normative constraints on the behaviour of an actor.

In his discussion of reasonableness, John Rawls provides a useful starting point for thinking about this additional requirement. In Political Liberalism, Rawls provides two conditions for reasonableness that incorporate this normative concern. The second of Rawls’ conditions relates to the ‘burdens of judgement’, which I discuss in the following section. For the moment, I am concerned with the first of Rawls’ conditions, which states that:

‘Persons are reasonable in one basic aspect when, among equals say, they are ready to propose principles and standards as fair terms of cooperation and to abide by them willingly, given the assurance that others will likewise do so’. (Rawls 1993, p. 49)

Reasonableness not only implies that people act under certain conditions of competent reasoning. It also implies that people constrain the types of claims that they make according to some normative requirement. Rawls provides this necessary requirement by providing a normative commitment in his first condition of reasonableness. Rawls’ first condition therefore provides the basis for our own second requirement of reasonableness:

Reasonable actors do not pursue merely self-interested ends. Reasonable actors make claims that appeal to a sense of fair cooperation.⁶ Now, the important upshot of all this is that other reasonable actors cannot dismiss claims that are made under these conditions as unreasonable. That is, if a reasonable actor puts forward a claim then other actors cannot reject this claim as a reasonable proposal for the fair terms of cooperation. To be sure, other reasonable actors can reject this claim as an incorrect proposal in their own opinion. But the fact that the claim is made under conditions of reasonableness means that it should be respected as a reasonable proposal for the fair terms of cooperation. This is because reasonable actors are willing to engage with each other in a fair way for the purposes of identifying fair terms of cooperation. They are therefore willing to recognise the reasonableness of another actor’s claims and treat them appropriately. The requirements of reasonableness put the necessary constraints on the types of claims that one can make so that these claims are justifiable to other actors who are also seeking fair terms of cooperation. Reasonableness therefore implies a willingness to accept the views of others as well as to temper one’s own claims.

These two requirements form the basis of our understanding of reasonableness. This represents a starting point for considering what a reasonable claim is, and what people should expect from reasonable actors.

Having specified a typology of disagreement and outlined some properties of reasonableness I now turn my attention to the concept of reasonable disagreement. Put simply, reasonable disagreement is disagreement that exists among reasonable actors. It represents a failure on the part of actors to agree on an issue despite the fact that they are acting under the conditions of reasonableness outlined above. Following the discussion so far, reasonable disagreement means two things: (i) reasonable actors will not converge on any one answer, and (ii) reasonable actors should accept that other reasonable actors hold different views and that those views should not be dismissed as unreasonable.⁷ The important point to note here is that, because each view can be reasonably upheld, such views cannot be dismissed as unreasonable. Whilst each actor can reasonably deny that the answers that others endorse are correct, these actors cannot dismiss these claims as unreasonable. Therefore, someone can reasonably reject a particular proposal as the right answer, whilst also accepting that it is not an unreasonable position to adopt. This means that no single view is likely to enjoy the full support of reasonable people. This is obviously a problem in situations that require the consensual cooperation of an entire group of actors.

But there’s a further reason that reasonable disagreement is important. If we are merely faced with a situation of disagreement among actors, then it should still be possible to specify some substantive outcome that a given procedure should aim to achieve. Whilst different actors may disagree because they hold different views that are based on mistaken reasoning, or incorrect evidence, this doesn’t mean that it isn’t possible to determine what the correct outcome actually is and then convince the disagreeing parties that they should endorse this outcome. Of course, if it’s necessary to gain each actor’s voluntary cooperation on an issue then it is still necessary to persuade the disagreeing parties to accept this procedure. If each actor’s interests persistently clash then there might be persistent disagreement here. But whilst achieving agreement may be difficult in such cases, it is not a wholly impossible thing to hope for. What is more problematic is the fact that it isn’t possible to specify a correct outcome in cases of reasonable disagreement. This is because reasonable disagreement reflects difference that should be respected, rather than dismissed as erroneous.

The upshot of this is that, if there is reasonable disagreement, then agents are unlikely to reach agreement on the substantive merits of particular proposals for the fair terms of cooperation. The fact that disagreement is reasonable means that actors are unlikely to reach an agreement even when they argue in good faith and for the public good.

Of course, there are other important implications of reasonable disagreement that relate to how people should think about fair procedures. There are also unresolved questions concerning why actors should act reasonably in the first place and what should be done about those who behave unreasonably. There have been many instances in COP negotiations where parties have acted unreasonably, and there are questions about the extent to which concessions should be made to these actors. But this is to get ahead of my current argument. I discuss the full implications of reasonable disagreement in the latter parts of this section. Here, the most pressing question to ask is why actors should treat reasonable claims with respect. I answer this question by considering how reasonable actors come to hold different views in the first place.

If reasonableness requires that certain conditions of reasoning are met, then how can reasonable actors come to disagree in the first place? After all, reasonable actors employ sound reasoning and engage in cooperative behaviour, but reasonable disagreement presumes that reasonable actors will disagree regardless. So how can reasonable actors fail to reach agreement on an issue?

As I pointed out earlier, actors can disagree for many reasons. But I’m concerned with why reasonable actors disagree. Here I draw on John Rawls’ discussion of the ‘burdens of judgement’, which he uses to explain how reasonable people can come to have different opinions on a matter (Rawls 1993, p. 54–8). Rawls argues that these are ‘the hazards involved in the correct (and conscientious) exercise of our powers of reason and judgement in the ordinary course of political life’ (Rawls 1993, p. 55–6). Whilst Rawls uses the burdens of judgement to explain disagreement about the comprehensive philosophical doctrines, these can also be used to explain disagreement about political ends (Waldron 1999, p. 112). The burdens of judgement are:

According to Rawls, this means that actors are likely to disagree even when they act in a reasonable way.⁸ The fact that different positions can be reasonably held means that, if there is a large number of heterogeneous parties, there is unlikely to be a single option that all parties immediately agree is correct. Whilst unreasonable disagreement can be resolved by exposing self-interested or erroneous views, reasonable disagreement can persist despite continued deliberation. This suggests that reasonable disagreement is more intractable than other forms of disagreement, which would be more easily resolvable under deliberative discussion.

But this is not the main issue here. The important point to take from Rawls’ discussion is that some issues require a judgement or interpretation on the part of the actor. Other issues require judgements about the values that people hold and how to prioritise them. The way that each actor judges, or interprets an issue depends on his or her previous experiences and worldviews. Given that each actor’s experiences and worldviews are different, each actor judges or interprets issues differently. This means that reasonable actors can come to hold different opinions of the same thing. But this difference is not just a question of mistake or error on the part of an actor. Nor is it a case of differential self-interest. Rather, it is the fact that actors are different that leads them to disagree, even if they are acting cooperatively. This brings us to Rawls’ second requirement of reasonableness, which states that not only do the burdens of judgement explain how reasonable disagreement can arise, but also that, under his account of reasonableness, reasonable people will recognise the burdens of judgement as reasons for accepting the reasonableness of others’ views (Rawls 1993, p. 54). The burdens of judgement therefore play a secondary role, not only explaining how reasonable disagreement can arise, but also providing reasons for why people should respect and acknowledge the reasonableness of other people’s views, even if they disagree.

Drawing on what has been argued above, it is worth clarifying the various points made so far. Here, I define reasonableness and its associated concepts as follows:

The result of this is that if a claim is reasonable, then another reasonable actor cannot reject it as unreasonable. This is not to say that a reasonable actor cannot reject a reasonable claim as wrong, or incorrect. A reasonable actor may hold his or her own views about what is right in a certain context, and reject other reasonable proposals as incorrect. But reasonable actors recognise the status of reasonable claims and treat them with due respect accordingly. This means accepting that other reasonable people hold different views and not dismissing these views as unreasonable. Following this, the next section shows that there is reasonable disagreement over the distribution of emission rights by showing that there are different reasonable proposals for the fair distribution of this good. Because each of these proposals is reasonable, reasonable actors cannot dismiss any of them as unreasonable.

Many allocation rules have been proposed in the UNFCCC for the fair distribution of emission rights, both indirectly from academic circles and political activists and directly from state delegations.⁹ The Fourth Assessment Report of the IPCC defines 14 such allocation rules that are frequently proposed in the literature on this subject.¹⁰ In this section I analyse some of the allocation rules that are proposed for the fair distribution of greenhouse gas emission rights, rather than for any other end. The most prominent of these rules are:

This is just a sample of all of the proposals that are made. But these allocation rules represent the most prominent allocation rules that are advocated in both theory and practice. Whilst similar issues would arise with other allocation rules, this chapter illustrates the existence of reasonable disagreement using just these four rules.

The remainder of this section considers each of these in turn. In doing so, it is possible to show four things. First, each allocation rule is a reasonable proposal for the fair distribution of emission rights. Following the argument in Sect. 2.2, the fact that each allocation rule is a reasonable proposal means that none of these proposals can be dismissed as an unreasonable interpretation of what fairness requires. Second, even if there is a consensus that one particular allocation rule is appropriate for the fair distribution of emission rights, the application of each rule involves choosing between variables that are unspecified by the allocation rule itself, and there are different reasonable interpretations of what these variables are. Third, even if there is agreement on these first two matters, there are different reasonable weights for the relative importance of each allocation rule in question. Fourth, I show that this is a matter of disagreement in both theory and practice, by highlighting where different principles are actually proposed in climate change negotiations. Taken together, these points show that the fair allocation of emission rights is subject to reasonable disagreement in a number of distinct ways and that there is no reason to expect that there will be agreement between reasonable persons on this matter. The fact that there are a number of different proposals that cannot be reasonably rejected means that there is unlikely to be one solution that will command the reasonable assent of all of the participants of the UNFCCC. Given this, I go on to argue that it is essential to have a fair way of dealing with disagreement that can help generate a consensus in Chap. 3. With this in mind, I now consider the main allocation principles that are suggested.

The Basic Emissions Rights proposal stipulates that there is a threshold level of well-being below which individuals are not obligated to limit their emissions.¹¹ This proposal is based on the premise that everyone should be allowed to achieve a basic minimum level of well-being. Given that achieving this basic level of well-being requires undertaking activities that generate greenhouse gas emissions, so the argument goes, there should be no limitations on the emissions that are brought about by these activities. According to Henry Shue, this means that each ought to be allowed at least the minimum amount of emissions necessary for a decent life.’ (Shue 1995, p. 252)

The Basic Emission Rights proposal receives strong support in the literature, and almost every proposal for the fair distribution of emission rights assumes that mitigation policies should only apply to those who already enjoy a threshold level of welfare. There is a moral threshold here and people shouldn’t enact policies that prevent people from meeting that threshold.

Despite the appeal of this proposal, there is at least one way in which its suitability as a principle of fairness might be reasonably disputed. The Basic Emission Rights proposal is premised on the assumption that there is a sufficient emissions budget to provide everyone with the necessary emissions to meet his or her basic needs. But because climate change is already occurring, one might reasonably argue that there is insufficient atmospheric space to exempt those emissions that are necessary for people to meet these needs. Whilst people frequently argue that the climate can tolerate a 2 °C rise in preindustrial temperatures without causing dangerous climate change, some hold that a much more stringent target is needed, even if this means preventing some from meeting their basic needs (Allen 2009). A great deal of harm is already being caused to the planet and there are many uncertainties about potential positive feedback loops, threshold limits, and tipping points in the climate system that may bring about severe and catastrophic changes to the environment. It is therefore reasonable to propose that there is an insufficient carbon budget to implement the Basic Emission Rights proposal.¹² At the same time, it is also reasonable to hold that the atmosphere can tolerate at least some limited amount of further emissions before people risk causing dangerous climate change, and that there is sufficient atmospheric space for providing basic emission rights so that people can meet their basic needs. That is, because the full effects of greenhouse gas emissions are unknown, there may be a sufficient carbon budget to implement the basic emission rights proposal.

Both of these positions represent different but reasonable positions about whether the Basic Emissions Rights proposal is suitable as a fair allocation rule. Since these claims are both reasonable, neither can be reasonably rejected as interpretations of what constitutes dangerous interference with the atmosphere. Disagreement arises here for two separate reasons. On the one hand, people may disagree about what atmospheric concentration of greenhouse gases constitutes dangerous climate change, which is a form of empirical disagreement (following the typology in Sect. 2.2). On the other hand, people may agree that it is desirable both to achieve a minimum level of well-being and to avoid dangerous climate change, whilst disagreeing about what should take priority here. Different reasonable positions cannot be rejected on this issue. Because this reflects the relative importance that an actor places on an issue, this represents a form of weighting disagreement.¹³

Even if there is agreement that there is sufficient atmospheric space for the Basic Emissions Rights proposal, it is not clear what the threshold level of welfare for exemption should be. The proposal for Basic Emission Rights stipulates that people should be exempt from those emissions that are necessary for achieving a basic minimum level of well-being for an individual. But determining what this level actually is involves a judgement about the minimum level of welfare that an individual should achieve. This is a matter of ongoing philosophical debate,¹⁴ and there are different reasonable positions about what this should be. One might reasonably contend that a basic level of welfare involves having access to a wide range of life opportunities, including the right to cultural practices. Alternatively one might reasonably argue that a basic level of welfare should be as minimal as possible, incorporating only the most absolute and basic needs of an individual. The point is that each of these positions is a reasonable interpretation of what a basic level of welfare should be. Because each of these positions is reasonable, neither of them can be reasonably rejected. This means that there is reasonable disagreement about how to apply the Basic Emission Rights proposal, because there is reasonable disagreement about what level of welfare it should apply to. Since this represents a disagreement about how to interpret this allocation rule, this is a form of interpretative disagreement.

There is one further point to make regarding this proposal. Thus far I have suggested that there are different reasonable positions regarding the interpretation of this principle, owing to disagreement over what a basic level of well-being should be. But in addition to any notion of reasonable disagreement, the Basic Emission Rights proposal is an incomplete account of what a fair distribution of emission rights is because it does not state how emissions should be distributed above the minimum level needed to achieve a basic level of well-being (if indeed there is any remainder to distribute). Whilst this point does not represent an area of reasonable disagreement itself, it does indicate that this allocation rule is insufficient in determining what a fair distribution of emission rights is and that it needs to be supplemented by a further principle. There may, of course, be reasonable disagreement about what supplementary principle, or principles these should be, which is something I explore when I discuss Multi-criteria rules.¹⁵