A FRAMEWORK FOR FACILITATING META-CONSENSUS
The Nature of Meta-Consensus
Consensus first needs to be broken down into its constituents based on the processes at work when individuals arrive at policy preferences. Three kinds of consensus can be identified. The first, normative consensus, refers to agreement regarding values driving the decision process. The second, epistemic consensus, refers to the judgmental aspect of preference formation, agreement about how particular actions map onto values in cause and effect terms. The third type, preference consensus, pertains to the degree of agreement about what should be done.
Each of these three kinds of consensus has a "meta" counterpart, which can help structure the process of deliberation:
| Element of Preference Construction |
Value |
Belief |
Expressed Preference |
| Type of Consensus |
Normative consensus (Agreement on the values that should predominate) |
Epistemic consensus (Agreement on belief about the impact of a policy) |
Preference consensus (Agreement on expressed preference for a policy) |
| Meta-Counterpart |
Recognition of legitimacy of disputed values |
Acceptance of credibility of disputed beliefs |
Agreement on the nature of disputed choices |
The critical distinction is this: meta-consensus does not require that parties agree on which values should predominate, what the facts actually are, or what policy should be adopted. It requires only that they recognize the legitimacy of contested values, accept that contested beliefs are credible, and share an understanding of how the choice is structured.
Mapping the Landscape of Disagreement
Your first task is to identify the groups defined on the basis of their underlying point of view on the issue at hand. Give each group a descriptive name that captures its orientation. Then map each group's position across three dimensions: the normative value that each group holds most dear, the beliefs about what means are most likely to satisfy that end, and the expressed preference that follows from the combination of values and beliefs.
Consider a dispute over the Bloomfield Track, a road constructed through World Heritage Rainforest in tropical Australia, the future of which remains a matter of public contention. Three groups can be identified:
|
Value (Normative Level) |
Belief (Epistemic Level) |
Expressed Preference (Preference Level) |
| Pragmatists |
Needs of community most important |
Road will benefit the community |
Keep road |
| Preservationists |
Needs of environment most important |
Road will negatively impact the environment |
Close road |
| Optimists |
Needs of environment most important (community matters too) |
Road will benefit the environment |
Keep road |
An examination of this mapping shows there is no universal consensus across all three groups, but there is agreement across elements of their positions. There are normative overlaps among the groups: Optimism with Preservation on the dominance of environmental concern, Optimism with Pragmatism on recognition of the community dimension. At the epistemic level there is dissensus across the three groups, which is why it is possible for agreement on values to coincide with disagreement on policy preference. The converse is also true: Pragmatists and Optimists disagree at the normative level but have similar preferences.
Seeking Normative Meta-Consensus
Normative meta-consensus exists to the extent that there is agreement on recognition of the legitimacy of a value, though not extending to agreement on which of two or more values ought to receive priority in a given decision. The recognition that defines normative meta-consensus is facilitated to the degree that the values in question are not positioned in a necessary zero-sum trade-off, though this perception is not crucial, and will not always be available.
In the Bloomfield Track case, there are two normative dimensions, pertaining respectively to community and environmental interests. While environmental issues are often portrayed in terms of a clash between two kinds of values, most individuals are actually likely to endorse both sets of values. That individuals privilege community needs over environment is not to say that they are hostile to environmental concern, and vice versa. It is just as unlikely that individuals given unconstrained choices would support environmental degradation as it would be for them to want the Bloomfield community to be worse off.
One of the more robust findings of the psychology literature on values is that at the abstract level there is a high degree of agreement on the legitimacy of basic values. Individuals differ mainly on the relative priority of values, and how they apply in particular cases. In this light, one of the main tasks of deliberation could be to uncover existing normative meta-consensus obscured by the strategic actions of partisans who try to delegitimate the values held by their opponents.
Your role as facilitator is to surface this latent recognition. Ask each party whether they reject the values held by the other side or merely weight them differently. Help parties articulate their recognition that others have legitimate values, even values they do not share.
Facilitating Across Deep Difference
Some disputes involve values basic to the identity of participants. A case reported by Forester and Hughes concerns a dialogue on HIV-AIDS policy in Colorado designed to produce advice for the state government. Participants included gay activists, including members of ACT-UP, people with AIDS, and fundamentalist Christians active in anti-gay rights campaigns. Conflicting values were in this case basic to the identity of gay activists and fundamentalists in particular. The fact that the fundamentalists were active in anti-gay rights campaigns showed that their identity in particular required validation through denial of gay identity. Thus the initial conditions were deep normative meta-dissensus and an absence of recognition.
The deep difference between the two sides was crystallized in a statement by a gay activist to the forum that there was a need to shift the discussion of AIDS in Colorado from a moral issue to a public health issue. To the fundamentalist Christians the issue was entirely moral. Each side rejected the way the issue was framed by the other.
As the dialogue proceeded, the two sides could, however, realize that they were not going to change the values of the other side. Their written agreement contained an explicit statement of the participants' recognition of a normative meta-consensus: for communities that include members with a range of moral perspectives, HIV prevention methods need to be appropriate to that range of moral perspectives.
The specific measures to which each side consented were not components of an overlapping consensus. These measures included moral education in schools and sexually explicit material targeted at the gay community, which remained objectionable to gay activists and fundamentalist Christians, respectively. At the end of the process the two sides still had fundamental differences in values and identities; the normative meta-consensus was specific to the HIV-AIDS issue under discussion. This example demonstrates that normative meta-consensus can be achieved much more easily than normative consensus.
Seeking Epistemic Meta-Consensus
Epistemic meta-consensus is agreement on the credibility of disputed beliefs, and on their relevance to the norms that define the issue at hand. Credibility here means that it is accepted by others as reasonable to hold the belief in question. Epistemic meta-consensus can therefore accommodate the multiplicity of perspectives required by epistemic arguments for the political rationality of pluralism.
Complexity of a phenomenon and associated uncertainty can preclude definitive choice across competing explanations and their associated theories or perspectives. Competing explanations may then coexist. This coexistence can apply not just in politics, but also in natural science: for example, explanations of the behavior of light as both a wave and a particle in optical physics.
There are situations where competing epistemic claims can both be accepted as valid. For example, local residents' experiences of the ill-effects of toxic pollution could be regarded as just as valid as the epidemiological studies that typically do not confirm these experiences. Any fault may lie in the demanding statistical requirements of epidemiological proof, as much as in residents' perceptions. This toxic pollution example shows that belief systems and not just isolated beliefs can enter an epistemic meta-consensus.
Epistemic meta-consensus is desirable on the grounds of deliberative economy. To the extent a set of beliefs is accepted as credible and relevant, there is an understanding of what the main issues are, and so no need to debate fundamentals each time a claim is made. In effect, epistemic meta-consensus creates a problem-solving public in the sense of pragmatist philosophers such as John Dewey.
Epistemic meta-consensus can also be instrumental to the creation of both normative and preference meta-consensus. If an issue is characterized by competing interests but no deep denials of identity or morality, then normative meta-consensus is facilitated to the degree values are not positioned in a necessary zero-sum relationship. In the Bloomfield Track example, deliberation reduced the credibility of claims about community access benefits of the track, which in turn led the jurors to recognize that community and environmental values were not necessarily diametrically opposed.
Seeking Preference Meta-Consensus
Preference meta-consensus consists of agreement on the nature of disputed choices across alternatives and has two aspects. The first concerns the range of alternatives considered acceptable. Most theorists who have contemplated the issue want the acceptable range to be limited by constitutional or legal means. However, deliberation itself could produce consensus on the range of acceptable alternatives.
The second aspect of preference meta-consensus concerns the validity of different ways that choices across alternatives can be structured. One such type involves agreement on a single important issue-dimension along which preferences are to be structured. If there is agreement that degree of access is the most important dimension, policy preferences should be single peaked along this dimension.
However, single-peakedness is not the only type of structuration of choices. Say, for example, that there is normative meta-consensus on the Bloomfield Track that environmental and community values are the important considerations, and epistemic consensus that the mere presence of the road will continually increase environmental damage by opening up new areas for development. Preferences may then be structured not along degree of access, but simply between closure and all other options where the road remains open. This sort of preference meta-consensus involving option reduction became observable and important at a key point in deliberation on this issue.
Preference meta-consensus is valuable because it makes social choice less vulnerable to arbitrariness, instability, and manipulation by clever strategists. To the extent that the preferences of actors are single peaked along one dimension, the chances that there is a Condorcet winner across policy options—one that beats all others in pairwise comparisons—is increased.
Distinguishing Three Types of Issues
We can distinguish between three types of issues requiring different emphases. First, there are those that involve clashes of identity in a divided society. The key here is production of a normative meta-consensus that remains contestable. Second, there are issues that involve deep moral conflicts. The key here is production of a stable normative meta-consensus. Third, there are issues that feature competing interests but no deep denials of identity or morality across participants. The key here is uncovering an existing normative meta-consensus that is obscured by the strategic machinations of partisans.
Normative meta-consensus is especially urgent in situations featuring deep difference in identities and value commitments. Attention to the bounds of epistemic meta-consensus is especially important in settings where powerful actors invoke questionable empirical claims in support of their material interests. Preference meta-consensus matters most in situations where one or more actors is in a position to manipulate decision processes.
Guarding Against Distortion
Meta-consensus might still be under the sway of ideological constriction. Elites can manipulate public opinion using arguments that invoke symbolic values and beliefs. The effect is to privilege particular norms invoked by symbolic arguments over others, so that normative meta-consensus is manipulated.
Particular factual claims can be deployed strategically in the interests of particular normative positions. So oil companies will finance studies that point to the absence of serious climate change resulting from increased carbon dioxide emissions, while environmentalists will make claims about imminent catastrophe unless greenhouse gas emissions are curbed. Here, credibility judgments can be made based on the source of a claim. If a claim is backed only by science funded or undertaken by partisans, then it should not be accepted into an epistemic meta-consensus.
Meta-consensus produced by symbolic politics in the service of partisans or hegemonic actors is much less defensible than meta-consensus produced by relatively uncoerced dialogue. Whatever the mechanism used for collective choice, meta-consensus makes collective choice more tractable. But to recognize this benefit is not to imply that any meta-consensus should be applauded.
Articulating the Achievement
When meta-consensus emerges, make it explicit. The Colorado HIV-AIDS dialogue produced a written agreement containing an explicit statement of the participants' recognition of normative meta-consensus. Such documentation stabilizes the achievement and provides a foundation for subsequent problem-solving.
The standard for evaluation is this: outcomes are democratically legitimate to the degree they are structured by free and reasoned meta-consensus among individuals subject to them. This does not guarantee agreement on what should be done, but it provides a shared framework within which continuing disagreement can proceed with mutual recognition rather than mutual delegitimation.
