童世骏《“Overlapping Consensus” on “Overlapping Consensus”》

All these three propositions, in my view, convey the idea that we should respect diversity as well as unity, but none of them makes it clear enough how these two apparent extremes are to be reconciled and integrated. This reminds us of the idea of “overlapping consensus,” which is used by its main proponent John Rawls, as well as many others, to address the issue of political stability in pluralistic domestic societies and the issue of multiple cultural bases of universal human rights at the global level. In order to see more clearly whether this concept is helpful in answering the question how diversity and unity can be taken care of at the same time, we will have an overview of the ideas proposed by thinkers in various countries and of different schools of thought regarding this concept. I want to argue that these different understandings of the idea of “overlapping consensus” can be interpreted as characterizations of different levels of overlapping consensus to be reached in our efforts to deal with pluralism with the aim of “social stability for the right reasons” (in Rawls’s words). This also means that some deficiencies found in each of these understandings, especially in the best-known version, i.e. the Rawlsian version, of the idea can be overcome by a mutual complementation between them.

1.

The concept of “overlapping consensus” first appeared in Rawls’s A Theory of Justice published in 1971, in which he observes that many disagreements among citizens in their understanding of justice can nevertheless lead to similar political judgments. These similar political judgments, he says, are “overlapping rather than strict consensus.” (Rawls 1971, pp.387-388) The logical implication of the existence of this kind of consensus is very simple: “different premises can yield the same conclusion.” (Rawls 1971, p.387)

Since mid-1980s, the idea of overlapping consensus has turned itself from an idea, mentioned only in passing in Rawls’s earlier book on justice to a major conceptual instrument in his later works on political liberalism for dealing with the phenomenon that “the political conception of justice that regulates its basic institutions is endorsed by each of the main religious, philosophical, and moral doctrines likely to endure in that society from one generation to the next.” (Rawls 1999, p. 473) Rawls considers this idea as the first of the three major ideas of his “political liberalism” (the other two being “the idea of the priority of right” and “the idea of public reason”). (Rawls 1996, p. 212)

       What is called by Rawls “political liberalism” is different both from the Hobbes’s version of liberalism and the Kantian or Mill’s version of liberalism. To Hobbes, liberalism is a modus vivendi, or a temporary compromise between conflicting interests of individuals and groups coordinated and balanced by some well-designed institutions. To Kant or Mill, liberalism is based on certain metaphysical doctrines or “comprehensive” moral doctrines. Neither version, according to Rawls, can solve the problem of social stability under pluralistic conditions: in the case of Hobbes, as soon as the balance of power changes, the existing stability based on it ceases to exist; in the case of Kant and Mill, Kantianism and liberal utilitarianism as comprehensive doctrines with “autonomy” and “individuality” as their respective key concepts are both held by particular parties, and cannot support the basic institutions of a society in which there are many other comprehensive doctrines as well, secular or otherwise.

The key to this position is “the fact of reasonable pluralism” (Rawls 1996, p. 36) in Rawls’s mind. Here the term “pluralism” refers to the existence of various “comprehensive doctrines,” and the term “reasonable” refers first of all to the attitude and mentality of those who, in inter-subjective relations, are ready to engage in fair cooperation and to follow public rules in cooperation as long as other participants follow them as well, which is different from “rational” in that the latter refers to the careful choice of efficient means to specific goals, or the prudent ordering of different goals within a whole life-plan. In order for a society to have a social fact called “reasonable pluralism,” its citizens should be reasonable in this sense. Rawls thinks that the liberal societies in the contemporary West can be characterized in this way; therefore the fact of “reasonable pluralism” can be regarded, in Rawls’s view, as the actual basis of, and political cultural support for, his idea of “justice as fairness,” among other political conceptions of justice.

In Rawls’s position presented above there is an interesting switch between the role of the philosopher and that of the citizen: while the political philosopher should not justify a political conception within a particular philosophical system, the citizen should behave like a philosopher, i.e. he or she should understand and support the conception of justice from his or her own world view and value system in addition to justifying it by means of public reason. Only then, Rawls holds, can there be “stability for the right reasons” (Rawls 1996, pp. 388-389) under the pluralistic condition. For only then can his idea of “justice as fairness,” for example, be truly accepted by the people on the basis of reasons, rather than as a result of external pressure, or their ignorance, or their uncritical acceptance. These “reasons” are different from person to person, and the consensus on a certain idea reached by different people for these reasons is what is called “overlapping consensus.”

After it was advocated in late 1980s, especially since early 1990s, the idea of overlapping consensus has attracted wide attention among political and social theorists. In the following discussion I will review some interpretations or elaborations of the idea that in my view are especially suggestive and helpful in our efforts to explore its full potential when addressing the issue of political stability in pluralistic societies and the issue of multiple cultural bases of universal human rights at the global level.

2

The idea of “overlapping consensus” can at the first stage be understood as the situation in which people, though different in their positions on a certain relevant issue, are similar in their attitudes to each other: they all treat each other in a reasonable way.

The major feature of this understanding of the idea of “overlapping consensus” is to loosen the bond between “reasonable” and “consensus” and to stress the connection between “reasonable” and “disagreement.” From C. S. Peirce to J. Habermas, all those who argue for the so-called “consensus theory of truth” have paid special attention to the connection between “reasonableness” and “consensus,” regarding “reasonable consensus” or “rational consensus” as the equivalent to or guarantee for truth, or in Habermas’s words: “The truth of a proposition means the promise to reach a rational consensus on what is said.”[2] Habermas does not only apply the consensus theory to the problem of truth, but also applies it to the problem of normative rightness, and what makes and fulfils the promise to reach a rational consensus is the practical rather than the theoretical discourse. In practical discourses, “just those action norms are valid to which all possibly affected persons could agree as participants in rational discourses.” (Habermas 1996, p. 107)

Although Habermas is clear that real situations of discourse are far from being perfect, and the goal of reaching consensus is far from being certain, he does give us the impression that, of various ideas respecting the same problem, as long as they differ from each other, at least one is not rational or reasonable. In other words, Habermas seems to admit the possibility of “reasonable consensus” alone and reject the possibility of “reasonable disagreement.” But the latter is just one of the main ideas in Rawls’s political liberalism. Many disagreements, according to Rawls, quite possibly do not result from our prejudices, ignorance, selfishness and wishful thinking. In normal political life and in the process of exercising our faculty of reasoning and judging, the complexities involved in the relation between ends and means, in valuing each person’s claims, in applying our theoretical ability and weighing our evidences and so on, there are many difficulties that cannot be avoided altogether. Rawls calls these difficulties “the burdens of judgment;” and as a result of these burdens, even very reasonable people can make different judgments on the same problem: “many of our most important judgments are made under conditions where it is not to be expected that conscientious persons with full powers of reason, even after free discussion, will all arrive at the same conclusion.” (Rawls, 1996, p. 58)

Rawls’s idea of “reasonable disagreement” is highly regarded by H. Grimen, a Norwegian philosopher, who thinks that Rawls has convincingly shown that “an ideal argumentation situation in Habermas’s sense is subject to the burdens of judgment,” (Grimen 1997, p. 276) and we should never expect that disagreements will totally disappear in the long run. A conclusion from this is that “reasonable political actors must learn to live with reasonable disagreements.” (Grimen 1997, p. 276) Because of these “reasonable disagreements” resulting from “burdens of judgment” we very often have to be satisfied with “overlapping consensus” rather than “qualified consensus” (Grimen 1997, p. 289) or what Habermas calls “begruendete Konsensus,” (Habermas 1995, s. 135) which means the consensus on the basis of reasons accepted by all concerned. Grimen’s emphasis is not on the indispensability of “overlapping consensus” as a result of the improbability of “qualified consensus,” but on the inappropriateness of raising problems on which we cannot possibly expect to reach a qualified consensus for public discussion as a result of the reasonableness of “reasonable disagreement.” To be more exact, the point Grimen emphasizes is what Rawls calls “the method of avoidance”: “In following the method of avoidance, as we may call it, we try, so far as we can, neither to assert nor to deny any religious, philosophical, or moral views, or their associated philosophical accounts of truth and the status of values.” (Grimen 1997, pp. 279-280).

Appropriating the idea of “reasonable disagreement” implied in the idea of “overlapping consensus” is really very important to political unity and social stability under pluralistic circumstances, because in many cases we do have to avoid seeking any qualified consensus or any consensus based on shared reasons that is evidently beyond our reach. This understanding of “overlapping consensus,” however, has the weakness of being too passive in that it focuses more on “overlapping” than on “consensus,” or more on “disagreement” than on “agreement.” In the idea of “reasonable disagreement,” actually, both “disagreement” and “agreement” play crucial roles: “reasonable disagreement” is “reasonable” because people share the same feature of reasonableness in their attitude to each other, and, as Rawls sometimes also argues, in the doctrines they hold. To say that people are “reasonable,” according to Rawls, is to say that 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 1996, p. 49) To say that a doctrine is reasonable, moreover, is to say that this doctrine meets the theoretical demands for a certain degree of consistency and coherence, and meets the practical demands for a certain way of ordering and balancing various values, and it is stable over time and not subject to sudden and unexplained changes. (Rawls 1996, p. 59)

These two senses of “reasonableness,” especially the first one, are, to my view, rooted in our common experiences in the life-world. In some sense it is like what is called li xing by Liang Shuming (1893-1988), one of the most eminent Chinese philosophers in the 20^th century. The Chinese term li xing is usually used as the Chinese equivalent of the English terms reason, rationality, reasonablness, and in some cases, intellect. Liang Shuming draws a distinction between ”li xing” and ”li zhi,” similar to Rawls’s distinction between ”the reasonable” and ”the rational.” Both li xing and li zhi are concerned with speaking, reasoning and thinking, but li zhi is more of calculation of means with regards to ends, while li xing is more of co-living and co-experiencing with fellow human beings in a way that takes life as a whole and as the end in itself. ” If you observe other people or engage in introspection,” Liang Shuming says,“whenever people are seen to be calm, easy-minded, without anything in their mind, most receptive to others, and most capable of reaching understanding with their partners by means of speaking with each other, they are people of li xin. The so-called li xing is nothing but the mental state of being calm and understanding.” (Liang 1990, p. 123) At this level, that is to say, a monological reflection is sufficient to show what is reasonable, and whether an overlapping consensus at this basic level has been reached or not.

3.

“Overlapping consensus” can, at the second stage, be understood as the situation in which people disagree in their values but agree on norms: people holding different values accept and follow a shared set of norms.

The Canadian philosopher Charles Taylor can be regarded as a representative of those who devote efforts to making a conceptual distinction between “norm” and “value” and then applying this distinction to the understanding of the idea of “overlapping consensus.” In discussing the problem what is the unforced consensus on human rights Taylor said: “I suppose it would be something like what Rawls describes in his Political Liberalism as an “overlapping consensus.” (Taylor 1999, p. 124) In all cultures, Taylor says, we can find condemnation of genocide, murder, torture and slavery, and what these condemnations express are the action norms on which there is a universal consensus. Below these shared action norms there are “deep underlying values” which usually “belong to the alternative, mutually incompatible justifications.” (Taylor 1999, p. 125)

Here Taylor relies on the conceptual distinction between       norm and value upon which he himself did not elaborate. In Juergen Habermas’s theory of communicative action this distinction plays an important role. The development of this distinction, according to Habermas, is an important achievement both in the process of development of individual moral consciousness and in the process of rationalization of a community’s life-world. When he is criticizing Rawls for treating rights as goods in the latter’s ideal experiment of “original position,” Habermas elaborates upon the differences between “norm” (to which rights belong) and “values” (to which goods belong) which can be summarized into the following four points: “...norms differ from values, first, in their relation to rule-governed as opposed to purposive action; second, in a binary as opposed to a gradual coding of the respective validity claims; third, in their absolute as opposed to relative bindingness; and, last, in the criteria that systems of norms as opposed to systems of values must satisfy.” (Habermas 1995, p. 115) These abstract distinctions between norms and values manifest themselves in the ways of their applications in everyday life: to the same problem “what I should do,” a reply on the basis of norms is categorically different from a reply on the basis of values. A norm “commands” me to do something, while a value “advises” me to do something. The former tells us what is good to all, or what is in everybody’s interest, while the latter only tells us what is good to me or to us.

Although both accepting the conceptual distinction between norms and values, and agreeing that people with different values can accept and follow the same norms, Habermas and Taylor consider the connection between norms and values differently. While Taylor, like Rawls, thinks that different values can be alternative bases of the same norms, Habermas pays more attention to justifying norms in moral discourse (concerning the question “what is good to all parties concerned”) instead of ethical discourse (concerning the question “what is good to me or to us”). Habermas admits that in pluralistic societies political consensuses are reached mainly among people holding different “values” (and with different interests) on common “norms” (including “principles” as “higher-level norms”[3]), but he argues that we should make a distinction between “consensus” as a social event and “consensus” as an epistemic achievement，or a distinction between acceptance and acceptability. (See Habermas 1995, p. 122)  Individual subjects’ agreements on a certain norm on the basis of their respective value systems can at most converge into an event that is the acceptance of the norm; but the persons who claim the acceptability of the norm should provide reasons in the process of debating with others and manage to convince them to accept these reasons as valid. Participants of the debate are neither observers of objects, nor members of the same particular cultural community who discuss their shared values, but participants in a “moral discourse” which demands each to have the same “moral point of view,” but not the same “ethical way of life.” That is to say, participants in moral discourse usually need to reach consensus with people of other cultures and value systems; and in order to do so, they should have the competence to evaluate each other’s perspective, or even consider an idealized perspective, in order to decide whether a certain norm is in the interest of all parties concerned.

Rawls does not agree with Habermas’s criticism of him, but in his reply to Habermas, Rawls seems to be both clearing up Habermas’s “misunderstanding” of him and moving closer to Habermas’s position. The “overlapping consensus” on political conceptions of justice in his sense, he argues, is not the agreement of interests sought by the politicians in everyday politics, but the “reasonable overlapping consensus;” and it is reasonable because the political conception is not only given a pro tanto justification on the basis of public reason at the first stage, and then given a full justification by members of civil society as individual citizens at the second stage, but also given a public justification by the political society in the form of the overlapping consensus among all the reasonable members of the political society as a result of their integration of the political conception in their several reasonable comprehensive views. (See Rawls 1996, pp. 386-387) Rawls argues that the stability he seeks is thus “stability for the right reasons,” and the overlapping consensus that underlies this kind of stability actually demands something that Habermas would call a “moral perspective.” In explaining what he calls the “wide and general reflective equilibrium” which is the kernel of this public justification, Rawls makes it clear that “this equilibrium is fully intersubjective: that is, each citizen has taken into account the reasoning and arguments of every other citizen.” (Rawls 1996, p. 385)

Here one point deserves special attention, that is, Habermas argues for a distinction between “consensus” as a social event and “consensus” as an epistemic achievement, or a distinction between the acceptance of a norm and the acceptability of a norm, not only in order to tell “stability for the right reasons” from stability without right reasons, but also in order to avoid a particularistic or contextualistic understanding of the principles of justice. Rawls emphasizes the role of political culture in his theory, regarding the political culture of the democratic society in his mind as the realistic basis of political liberalism. “The political culture of a democratic society,” Rawls says, “is always marked by a diversity of opposing and irreconcilable religious, philosophical, and moral doctrines. Some of these are perfectly reasonable, and this diversity among reasonable doctrines political liberalism sees as the inevitable long-run result of the powers of human reason at work within the background of enduring free institutions.” (Rawls 1996, pp. 3-4) Rawls’s concepts such as “reasonable pluralism,” “the public use of reason,” and the “overlapping consensus” on the conception of justice, etc., are predicated upon this political culture and at the same time provide conceptual instruments for the latter’s self-understanding. Political culture as a “fact,” however, is a particular thing; if at the public level the principles of justice are only predicated on the political culture of a particular region or a particular tradition, there is a hidden danger of giving up the demand for a universalistic justification for these principles. This particularistic position can lead to two consequences. On the one hand, it would make it difficult to find out elements in the existing political culture as the basis for the immanent critique and immanent transcendence of this culture and the institutions based on it. On the other hand, it would make it difficult to talk about the universality or universal legitimacy of a certain institution outside a particular society and cultural tradition. Habermas does not agree with Richard Rorty, who regards Rawls as a fellow particularist with “a thoroughly historicist and antiuniversalistic’ attitude.” (Quoted in Habermas 1996, p. 62) But Habermas insists that, in order to keep a clear distance from this tendency, a conception of justice achieved through a hermeneutical clarification of a contingent tradition should be put to test in a moral discourse to see whether it is not only accepted, but also acceptable. (Habermas 1995, p. 122)

When we agree with Habermas on emphasizing the imp