Justification by Reflective Equilibrium in Rawls's More Recent Work
1. Famously, John Rawls uses the method of reflective equilibrium (RE) to justify his principles of justice. (1) But the point of justification by RE in Rawls's more recent work is not that easy to establish, since he regards his own work still as contractarian. Accordingly, it is peoples', citizens', or rational deciders' acceptance of the basic notions, methods, and results of Rawls's framework at its different stages (2) that is to establish his Justice as Fairness. Since every single one of us supposedly has already accepted a moral view of the world, though not the same one, it is in the end with regard to that moral view of the world, (3) or in Rawls's terms, that comprehensive theory of the good, (4) that the principles of justice have to be justified. (5) From the point of view of every one of us who reads Rawls's work or from the point of view of any particular citizen engaged in a discussion of a just society nothing else matters (6) - or so it seems. What, then, can be the point of justification provided for by RE that is not provided for by our comprehensive theories of the good? (7) Imagine we are discussing what kind of society is just. Here the plurality of views is likely to foreclose easy agreements. I take RE in Rawls's more recent works to widen the range of possible agreements in that situation, and, hence, facilitate agreements otherwise impossible to arrive at.
The request of RE may facilitate aggrements in two ways: first, by asking us to reconsider initial refutations (in this case) of Rawls's proposed principles, and, second, on settling what may be seen as the second best solution of what a just society is like from the point of view of the single comprehensive views. I take Rawls to have both of those requests in mind when he asks us to go back and forth between a set of principles of justice and our existing considered judgments (8) - or so I want to argue.
2. Let's begin the argument with the concept of RE. Clearly, what Daniels expounded in general as "narrow reflective equilibrium" (9) could only be a method for the first of these two requests: the proposal of principles of justice may lead us to consider the RE between these newly proposed principles of justice and our theory of the good, and reconsider the RE between our own principles of justice and our own theory of the good.
Yet the likely results of such narrow equilibria should be sobering: normally, we will stick to our own principles of justice, which are part of our theory of the good, and we will accept the new proposal only insofar as this new propsal repeats, specifies, or explicates our own principles of justice. So, narrow RE will be no more than a plea for consideration and (re)consideration, in short a plea for (re)consideration.- What then about wide RE?
Following Daniels's interpretation of Rawls, (10) a wide RE consists basically of three narrow reflective equilibria (REa): one narrow RE between an intuition about a single case and a proposed rule or principle, another narrow RE between this rule or principle (11) and a cluster of non-contradicting and often more general principles which might form a theory, and a third narrow RE between some (other) principles of this cluster and some intuitions we hold concerning other single cases. These narrow REa are of different kinds - in that respect Daniels's account is insufficient:
We are in the (permanent) procedure of establishing a RE between the intuitions we have concerning a certain case and proposed rules or principles. RE here describes a constructive procedure of choice, where both principles/rules and descriptions for our intuitions are to be chosen in a mutually constructive way. (12) Let's call this RE as constructive procedure of choice.
It is uncertain whether or not in Rawls's view there already exists a RE between these principles/rules and the background cluster of principles or whether we are to create that RE in due process. In any case, the RE involved in this process is of an already familiar kind: it is the plea for (re)consideration of some rather firmly held or some newly proposed principles.
Finally, there is a RE between parts of the background cluster and some other intuitions. In this sense we already have proven background principles, that is principles we have proved to fit our intuitions about some cases (though perhaps not others). As a result there is a RE between those background principles and some relevant intuitions. Here, clearly, is another kind of RE involved; let us call this third kind of RE the safe-ground RE.
These different kinds of narrow equilibria are prone to different problems. (13) Here, one of them is of special interest: The safe-ground RE is only safe once chosen, but its choice may be in need of further justification. According to Rawls this justification can only be provided for by the ongoing process of reasoning and reflection. (14) But how can we expect this kind of reasoning to facilitate agreement given the plurality of reasonable theories of the good?
3. In my view Rawls uses the wide RE not only to make us reconsider our intuitions against the background of our respective theories of the good, but - I propose - to introduce another point of view and with it another set of background principles against which his principles of justice are to be seen. He introduces a new safe ground RE which is to become part of the wide RE. This new point of view is the point of view of being-a-citizen in society, and the background cluster comprises principles regarding the functions of society and its citizens: that a society consists of citizens with certain basic capacities, that they are equal and free as citizens, that society is a venture not only of coordination but of cooperation, and so on. It is this last point of view of a citizen with its background principles which in Rawls's theory gives the method of wide RE its point: only this wide RE leads us both to reconsider the proposed principles of justice, and to revise our own ones. And it is a position in which we as democrates supposedly all share the same background cluster of the function of a society and its citizens, and thus a starting point on which we can all agree.
As a result two wide reflective equilibria emerge, to be distinguished by two different safe-ground equilibria: in one of them the safe ground is our moral view or comprehensive theory of the good (the moral ground RE), in the other it is the theory of being-a-citizen in society (the citizen RE). Being-a-citizen in society is supposedly (15) not in apparent contrast with the most firmly held parts of most comprehensive theories of the good. (16) But for those of us who hold other moral views Rawls in order to reach a result asks us again to bring into RE our moral ground RE and our citizen RE. That is: Rawls asks us to form a RE between two wide reflective equilibria. This RE - though certainly being a wide RE - has all the other features of a constructive choice RE. No new kind, no new conception of RE seems necessary here. The constructive choice in question is, of course, the choice of the all things considered best principles of justice. (17) Wide RE in Rawls's political philosophy thus is a highly complex procedure of justification, which in the end may ask us both to reconsider our principles of justice and - at least in some cases - to settle for the second best solution as seen from the single comprehensive theories of the good. (18)
4) Now, I couldn't give the single elements of this RE their due, but some final remarks seem appropriate:
First, the conception of wide RE in the newer works of John Rawls as I have reconstructed it bears some resemblence to theories which explicate what it means to value something by reference to higher order nonkognitive attitudes (desires, preferences, emotions), esp. the normexpressive theory of Allan Gibbard, who thinks that we can justify our value judgments by reference to rational standards and the rationalization of standards has to be done with reference to acceptable normsystems. (19)
Second, this reconstruction of wide RE interprets Rawls as leaving open the question of the universalizability of his theory (at least as far as the RE for the Political Liberalism part of it is concerned). This universalizability depends on the kind of justification for the conceptions of society, citizen, cooperation and the like: if that is done by theories of democracy which imply universalizability, Rawls theory remains universalizable, otherwise it becomes perhaps universaly true, but nevertheless partizan.
Third, I didn't say anything about how the different narrow REa are to be combined into one wide RE, given the rationale to facilitate agreement by way of revision. How can revision be prompted by RE? What kind of RE is necessary for revision? So far, the only plausible candidates for explaining revision are RE understood as a plea for (re)consideration and RE understood as safe ground. It looks as if elements of both conceptions of RE are involved when we say someone has a rationale for revision: he has no more safe ground for the justification of something, and there is a plea or even a need for reconsideration of the justification of it. This kind of unity, however, seems to be more intimate than the kind of unity otherwise necessary in wide RE. At least part of the more intimate relation is that - to capture revision - (RE as a plea for) (re)consideration has to be caused by the missing of a safe ground RE. The additional relation, then, is a causal relation between the two kinds of RE. Since other wide RE may be imaginable for other purposes, (20) the wide RE Rawls needs for Political Liberalism is of a special kind - and it is still a necessary part of Rawls's theory of justice.
(1) This method resembles Brandt's "qualified attitude method" in: Brandt (1959), pp. 244 ff.
(2) Anderheiden (1996); Rawls restricts assessment of his "Justice as Fairness" by RE to the point of view of you and me, see Rawls (1993). p. 28.
(3) This point is emphazised by Kersting (1996), pp. 234 - 236.
(4) Rawls (1993), pp. 19 f., 74.
(5) I assume that the comprehensive theories of the good also include the terms for moral justification.
(6) And it does not matter why we accept or refute certain proposed principles of justice. Thus, we might either accept or refute them since they accord with our highest principles taken to be fixed (Kantian method). Or we might do so since they capture what is common in our judgments taken to be unshakeable (Aristotelian method). Both of these approaches are much more complex and allow for considerable variations. But the role they play in justification amounts to the same.
(7) To show this additional justification is a reason why Daniels introduced wide reflective equilibrium. With regard to Rawls, however, he conceded that wide RE is fixed at the (comprehensive) theories of the person and of society Rawls advances, see Daniels (1979), p. 260 f. Yet in Rawls (1993) those comprehensive theories are replaced by normative conceptions, see Rawls (1993), p. 18 note 20.
(8) Rawls (1971), p. 20.
(9) Daniels (1979), pp. 256 ff.
(10) Daniels (1979).
(11) For the difference between "rule" and "principle" see Dworkin (1976), pp. 22-28, 71 ff.
(12) Rawls explicetely rejects an interpretation of RE as an intuitionist procedure, see Rawls (1993), pp. 95 ff. For a similar discussion see Dworkin (1976), pp. 158 ff.
(13) Some other problems: The constructive choice RE between principle and intuition is prone to the instability of the intuition-description caused by the numerous possibilities of situation-descriptions and the possible normative-descriptive terms used thereby (which in turn might cause different psychological reactions and hence new intuitions). The plea for (re)consideration equilibrium between a principle and a background cluster of principles is heavily influenced by the problem of whether there can be different possible background clusters and if so, which of them is the right one for the rationale of differently structured life-worlds.
(14) Rawls (1993), p. 96.
(15) This explains the necessity to restrict the theory of justice as an enterprise to convince democrats only. However, this restriction has nothing to do with a restriction to specific western values and experiences, and I take those verdicts to be premature which take Rawls to have become a communitarian.
(16) Rawls goes on to show that his proposed principles are in line with different comprehensive theories of the good; see Rawls (1993), pp. 145 ff.
(17) So Rawls in his Political Liberalism on that account can only succeed if citizens can achieve a highly complex RE between the principles of justice, their moral views, their view of how a society functions, their intuitions about a functioning society, and their intuitions about just decisions regarding the basic structure of society.The principles might even be justified if some people do actually reject some of the principles of justice (or think they are not the best principles all things considered) if there is reasonable hope that they might change their minds in future. Understood this way, Rawls's plea for RE concures with Scanlon's contractarian formula of "reasons nobody can reasonably reject". See Scanlon (1982), p. 110 - Obviously, it would require a seperate paper to establish the relationship between Scanlon's and Rawls's contractarian theories. On this see Barry (1995), pp. 67 ff.
(18) Insofar Rawls's method of RE resembles the method of reflective endosement which Christine Korsgaard ascribes to Hume, Mill, and Williams, see Korsgaard (1996), chap. 2, esp. p.78.
(19) Vgl. Gibbard (1992), pp. 55 ff.; and Holmer Steinfath's (1997) talk at the Conference of the Society for Analytical Philosophy: Rationality, Realism, Revision. Munich.
(20) Obviously, much more work has to be done to establish how exactely (the lack of) RE causes revision.
Anderheiden, Michael (1996): "John Rawls", Information Philosophie Heft 4, S. 36 - 39.
Apel, Karl-Otto/Matthias Kettner (1996): Die eine Vernunft und die vielen Rationalitäten. Frankfurt a. M.: Suhrkamp.
Barry, Brian (1995): A Treatise on Social Justice, Vol. II: Justice as Impartiality. Oxford: Clarendon Press.
Brandt, Richard (1959): Ethical Theory. Englewood Cliffs NJ: Prentice-Hall.
Brandt, Richard (1979): A Theory of the Good and the Right. Oxford: Oxford University Press.
Daniels, Norman (1979): "Wide Reflective Equilibrium and Theory Acceptance in Ethics", Journal of Philosophy 76, pp. 256 - 282.
Dworkin, Ronald (1976): Taking Rights Seriously. Cambridge MA: Harvard University Press.
Dworkin, Ronald (1986): Law's Empire. Cambridge MA: Harvard University Press.
Gibbard, Allan (1992): Wise Choices, Apt Feelings. Cambridge MA: Harvard University Press.
Kersting, Wolfgang (1996): "Spannungsvolle Rationalitätsbegriffe in der Philosophie von John Rawls", in: Apel/Kettner (1996), pp. 227 - 265.
Korsgaard, Christine (1996): The Sources of Normativity. Cambridge: Cambridge University Press.
Rawls, John (1971): A Theory of Justice. Cambridge MA: Harvard University Press.
Rawls, John (1993): Political Liberalism. New York NY: Columbia University Press.
Scanlon, Thomas (1982): "Contractualism and Utilitarianism", in: Sen/Williams, pp. 103 - 128.
Sidgwick, Henry (1907): The Methods of Ethics. 7th ed. London: Macmillan and Co.; reprint: Indianapolis: Hackett, 1981.
Steinfath, Holmer (1997): "Wertungen und Wertungsrevisionen: Zur Rolle höherstufiger nonkognitiver Einstellungen in Wertungen" (Ms.).