“A Serious Problem”: Motivation, Reasonableness, and the Infeasibility of Overlapping Consensus.

1

“A SERIOUS PROBLEM”:MOTIVATION, REASONABLENESS,

AND THE INFEASIBILITY OF OVERLAPPING CONSENSUS.

In Political Liberalism1 (henceforth, PL) John Rawls proposes the idea of an overlapping consensus (OC) as the blueprint for stabilizing pluralistic societies, stipulating pluralism as a corrective for his assumption, in A Theory of Justice2 (TJ), that all citizens are Rawlsian liberals. Here, I intend to show, first, that Rawls’s move merely replaces this assumption with, at best,the assumption that each and every citizen is a liberal of some kind. From thence, I shall argue that this assumption sends the idea of OC into an infinite regress, which Rawls can escape onlyif he accepts that the problem he is addressing either cannot besolved with an OC, or is a pseudo-problem (that, as such, requires no solution). Because, I believe, he can neither accepteither alternative nor offer a third one, I conclude that implementing the idea of an OC is infeasible wherever it is necessary, and unnecessary wherever it is feasible.

Rawls’s arguments in PL do not avoid promoting a particularconception of the good for a politically liberal society. In fact, they repeatedly and clearly either imply or assume that the citizens of such a society must acquire and retain at least one virtue—namely, reasonableness—if justice as fairness is to be a workable public ethic.3 I shall describe the character of reasonableness in the next section, but in order to do that I

2

must first discuss the important preliminary issue of the kind of motivation which makes citizens capable of affirming Rawls’s political conception of justice. To that end, I begin with the problem of political liberalism, the effort to solve which leads Rawls toarticulate the idea of OC.

Rawls states the problem thus:. . . How is it possible that there may exist over

time a stable and just society of free and equal citizens profoundly divided by reasonable though incompatible religious, philosophical, and moral doctrines? Put another way: How is it possible that deeply opposed though reasonable comprehensive doctrines may live together and all affirm the political conception of a constitutional regime? What is the structure and content of a political conception that can gain the support of such an overlappingconsensus? (PL, xviii.)4

Note that Rawls refers not to a mere diversity of doctrines, butto a diversity of reasonable comprehensive doctrines (PL, 36-7). Rawlstells us—in what he calls a “deliberately loose” account—that a reasonable comprehensive doctrine is an exercise of both theoretical and practical reason, which “normally belongs to, ordraws upon, a tradition of thought and doctrine” (PL, 59), and he assumes that each and every citizen of a well-ordered societysubscribes to some reasonable comprehensive doctrine (PL, 12, 140, 150). This existence of a diversity of reasonable

3

comprehensive doctrines he calls the fact of reasonable pluralism, the persistence of which is the first of three “general facts” he posits about democratic regimes. The other two are, first, the fact of oppression, which is the normative fact that democratic regimes may not oppress their citizens in order to enforce particular comprehensive doctrines (and is therefore, perhaps, something of a misnomer) and, second, the need for any such regime to be “willingly and freely supported by at least a substantial majority of its politically active citizens” (see PL, 36-8). So, the challenge for Rawls is to explain how a society can be stable when riven by the divisions and oppositions inherent in reasonable pluralism while being supported by most or all citizens without the government forcingthem to do so. The short form of the answer is that stability isprovided from within the different reasonable comprehensive doctrines whose diversity gives rise to the problem of politicalliberalism in the first place.

The problem of political liberalism is a species of the stability question, which is the general enquiry into the necessary and sufficient conditions for a scheme of social co-operation tobe, as Rawls puts it in TJ, “more or less willingly complied withand its basic rules willingly acted upon” (TJ, 6). Rawls raises this question in TJ, but addresses it in far more detail in PL. Indeed, the treatment of the problem of political liberalism is needed because of what Rawls calls “a serious problem” in TJ:

4

namely, that its “account of the stability of a well-ordered society is unrealistic and must be recast” (PL, xvii). In TJ, Rawls argues that stability results from the existence of institutions that foster in citizens either “a sense of justice or a concern for those who would be disadvantaged by their defection, preferably both” (TJ, 497). Whence, though, do citizens acquire their sense of justice in the first place? In PL Rawls admits that TJ covertly presupposes that it stems from citizens’ common adherence to a comprehensive doctrine of justice as fairness. In other words, TJ assumes the premise that all citizens are Rawlsians, which contradicts the fact of reasonable pluralism, and hence the need to solve the problem ofpolitical liberalism (PL, xv-xvii).

Since the problem of political liberalism is a form of the stability question, it is fairly easily predictable that the issue of motivation must lay at the heart of Rawls’s proposed solution. As Jürgen Habermas points out, Rawlsian citizens must have the “functionally necessary” motivations to abide by the principles of justice, if the question of stability is to be answered5; and Rawls himself says that “[s]tability is secured by sufficient motivation of the appropriate kind acquired under just institutions” (PL, 142-3). For a reasonably pluralistic society, Rawls advocates OC as a suitable mechanism to provide such motivation. An OC exists whenever a “political” conception of justice has legitimated itself to at least two groups, both

5

found within the same society, that subscribe to different reasonable comprehensive doctrines. Those groups’ divisions withregard to questions of basic justice are thus overcome, and their members acquire the motivation to abide by the political conception willingly (or, in Rawls’s own terms, the motivation to affirm it). Acquiring the right kind of motivation is therefore a necessary condition for bringing about an OC, which itself is the necessary and sufficient condition for the stability of a reasonably pluralistic society.

Any significant doubt about the political conception’s capacity to legitimate itself to at least most citizens must therefore cast significant doubt on the claim that the idea of OC solves the problem of political liberalism. I am going to tryto raise just such doubt, by arguing that the political conception can legitimate itself only to those who are already politically liberal—and even, perhaps, who are already Rawlsian liberals. John Dryzek and Simon Niemeyer, following Chantal Mouffe, have already made point that

. . . overlapping consensus is stronger to the degree different doctrines muster similar reasons for subscribing toit. At an extreme, to make a truly secure overlapping consensus, they will be able to muster the same moral reasons. And these reasons ought to be basic ones for each doctrine. This drive to uniformity warrants Mouffe’s skepticism of what Rawls calls a reasonable pluralism. For

6

Rawls defines a “reasonable” individual as someone committed to basic liberal principles (1996, 249). Thus it is no surprise that these individuals endorse a liberal conception of justice in their overlapping consensus. Setting aside any specifically liberal content of the overlapping consensus and focusing on its formal structure,the general point is that any overlapping consensus requires agreement on the priority of some set of substantive values.6

To that extent, I totally agree. But neither Dryzek and Niemeyernor Mouffe explain exactly why “skepticism about reasonable

pluralism” is justified, other than that it will tend to excludenon-liberals. There is, however, no substantive exploration of Rawls’s idea of OC to show why it will have such consequences. (In Mouffe’s case, her objection is really that the idea of OC is incomplete, and therefore unrealistic, because Rawls has simply ignored the “antagonistic” nature of politics.7) In my view, the whole complex of ideas that go to make up Rawls’s theory of OC is logically confused, such that its consequences are really beside the point. The idea of OC fails because the idea of reasonable pluralism with which it is allied makes it impossible for an OC to succeed under any circumstances other than those in which the idea of OC is unneeded. To back up this claim, I shall first examine Rawls’s account of how parties to the problem of political liberalism acquire the motivation to

7

affirm the political conception. Doing this takes us back to thestability question.

In PL, Rawls’s new perspective on his unrealistic assumption complicates things to the extent that he there distinguishes two stability questions from each other. The firstis the same as that posed in TJ, the answer to which is still, sure enough, “setting out the moral psychology” of the sense of justice (PL, 141). Since, in conditions of reasonable pluralism,taking the sense of justice for granted in citizens would reactivate TJ’s unrealistic assumption, Rawls articulates the second stability question, which is “whether in view of . . . the fact of reasonable pluralism, the political conception can be the focus of an overlapping consensus” (PL, 141). Thomas E. Hill observes that this is not the “simple” stability question of how to ensure that a society endures over time, but that of how to “defend justice as fairness against the charge that . . .to use it as a standard for (inevitably) coercive political decisions would violate the liberal principle of legitimacy.”8 In other words, the second stability question is that of how to legitimate the political conception to parties who are not necessarily Rawlsian liberals. I shall therefore follow Hill in calling the first stability question the simple stability question, andthe second the legitimation question. In short, the legitimation question is that of how to create loyalty to the political conception in Rawlsians and non-Rawlsians alike, whereas the

8

simple stability question is that of how to sustain it in them.What I have said about the link between legitimation and

stability, via OC, should make it clear that the legitimation question must be answered both logically and temporally prior tothe simple stability question. Central to any answer that Rawls can give to the legitimation question is the political conception, which Rawls describes “as exemplifying the content of a liberal political conception of justice” (PL, 6; emphasis mine), even though its principles differ unimportantly from those set out in TJ. This at once raises the question of whether there is any difference between TJ’s unrealistic assumption and PL’s assumption of reasonable pluralism. It seems to me that there is, in that the new assumption is rather weaker than the old, and more realistic in the sense that it describes actual liberal societies more accurately. These differences, though, seem so small and so inappropriate to answering the legitimationquestion that PL’s revised assumption cannot solve the problem of political liberalism; for, as I shall go on to show, Rawls cannot explain satisfactorily how motivation can be created in non-liberal parties to the problem.

This shortcoming is clearest in Rawls’s account of how the political conception legitimates itself to such parties. Legitimation occurs when parties detect congruence of the political conception with the metaphysical (i.e., non-political) conceptions of justice that are associated with their own

9

reasonable comprehensive doctrines. While Rawls never says so inexactly this way, that this is his meaning can be inferred from various claims he makes, such as that “a political conception ofjustice must be one that can be endorsed by widely different andopposing though reasonable comprehensive doctrines” (PL, 38), despite the status of justice as fairness as a “freestanding” moral conception, which is to say that

. . . it does not provide a specific religious, metaphysical, or epistemological doctrine beyond what is implied by the political conception itself. As remarked in [PL] I:2.2, the political conception is a module, an essential constituent part, that in different ways fits into and can be supported by various reasonable comprehensive doctrines that endure in the society regulated by it. (PL, 144-5)

Such congruence as I have described, of the political conceptionwith reasonable comprehensive doctrines, must, then, be what makes it possible for adherents of the latter to affirm the former. Without it, and given reasonable pluralism, it would be logically impossible for adherents to differing reasonable comprehensive doctrines to participate in an OC. To effect legitimation, however, such congruence must be great enough to divert such parties’ existing motivation from affirming their own metaphysical conceptions to affirming the political conception. I shall describe as motivable any parties whose

10

metaphysical conceptions are sufficiently congruent with the political conception to achieve this, and those parties who actually go on to affirm the political conception I shall call motivated. Naturally, one must be motivable in order to be motivated; and, once motivable parties become motivated, the stability question with regard to them progresses from answeringthe legitimation question to answering the simple stability question.

The moment of legitimation is not a Kuhnian paradigm-shift,but merely a co-opting of existing motivation to affirm some metaphysical conception, made possible because the parties recognize their familiar metaphysical conceptions in the hitherto alien political conception. As such, it is reminiscent of a Gadamerian fusion of horizons. The problem, though, is that, if the horizons being fused are defined by metaphysical conceptions and if, also, the political conception is an exemplary liberal conception of justice, is it not very likely that all motivable parties must be liberals already, with regardto matters of basic justice—that is, that their metaphysical conceptions of justice are liberal metaphysical conceptions? If so,then one may well wonder what has become of the profound divisions and deep oppositions, caused by reasonable pluralism, that give rise to the problem of political liberalism in the first place. To the contrary, it seems that, for the problem of political liberalism to have any bite, Rawls’s ambition must be

11

to show how those who are metaphysical non-liberals matters of justice can become motivable, by fusing the horizons of their metaphysical conceptions with that of the exemplarily liberal political conception.

To test whether it is logically possible to bring non-liberals to affirm the political conception, I shall explore Rawls’s account of motivation from the viewpoints of two kinds of parties, which I call actual and potential citizens. Actual citizens are already motivated, while potential citizens are not, as yet, even motivable; and this difference between the twoclasses of persons exists because actual citizens detect congruence between the political conception and their metaphysical conceptions, while potential citizens have yet to detect it.9 As it happens, this distinction shows that the simple stability question applies only to actual citizens, the legitimation question only to potential, because Rawls takes it that simple stability arises from a sense of justice that citizens can develop only by abiding by the principles of justice. Consequently, only actual citizens can possess a sense of justice. If, however, one asks how potential citizens who have no sense of justice may become motivable, one sees at once—as Rawls has—that the account given in TJ is not just unrealistic, but circular; for all it tells us is that actual citizens are motivated because they are motivable, and motivablebecause they are actual citizens. TJ’s argument therefore shows

12

only how to sustain actual citizens’ loyalty, which is to say that it answers only the simple stability question. Surely, though, the problem of political liberalism cannot arise under such conditions. In PL, Rawls acknowledges that he must explain the motivability of potential citizens, instead of assuming all parties to be actual citizens, and therefore attempts to constrain his account of motivability by assuming that at least one citizen of a well-ordered society is non-Rawlsian.

At this point, however, one must take into account an important detail in Rawls’s picture of the citizen because it subverts his assumption of reasonable pluralism. In PL, he supposes all and any motivable parties “to want to be, and to [want to] be recognized as” normal, fully co-operating members of society, on the basis of their “reasonable moral psychology”—that is, on the basis of the very attribute I am calling reasonableness (PL, 81-2). This desire makes them “want to realize in their person, and have it recognized that they realize, [the] ideal of citizens” of a well-ordered society (PL,84). In PL, then, only reasonable parties can be motivable and, hence, actual citizens are necessarily reasonable; but, if OC isto be feasible, not only actual but also potential citizens, also,must be reasonable.

What, though, does it mean to be reasonable? Rawls never defines the term reasonableness as such, instead seeking to “specify two of its basic aspects as virtues of persons” (PL,

13

48). These are, first, “the willingness to propose fair terms ofcooperation and to abide by them provided others do so” (PL, 49); and, second, “the willingness to recognize the burdens of judgment and to accept their consequences for the use of public reason in directing the legitimate exercise of political power in a constitutional regime” (PL, 54). The first aspect is self-explanatory, but what is the relevance to this matter of the willingness to recognize the burdens of judgment? Rawls tells usthat the function of the burdens is to direct “the legitimate exercise of political power in a constitutional regime” (ibid). When taken together, the values of the political and of reasonableness make political liberalism possible, because they

. . . express the liberal political ideal that since political power is the coercive power of free and equal citizens as a corporate body, this power should be exercised, when constitutional essentials and basic questions of justice are at stake, only in ways that all citizens can reasonably be expected to endorse in the lightof their common human reason. (PL, 139-40.)

To put it bluntly, then, recognizing the burdens of judgment determines who may coerce whom into what. Rawls assumes that allparties “share a common human reason, similar powers of thought and judgment: they can draw inferences, weigh evidence, and balance competing considerations” (PL, 55). This view of reason does not change the fact that there are many sources of

14

disagreement between intellectual equals, and the burdens are those very sources of disagreement. Where they are recognized, different groups can come to affirm incompatible comprehensive doctrines, and a reasonable pluralism can flourish.

To make the idea of reasonableness somewhat clearer, I turnto Jon Mandle’s concise explication. Mandle distinguishes four different Rawlsian senses of the term. There is the reasonableness of, first, constraints in the original position; second, the principles of justice; third, conceptions of justice; and, fourth, persons.10 “A person,” Mandle tells us, “would be unreasonable if she didn’t recognize [the] diversity of conceptions of the good and did not rely on principles of justice that all could accept.11 Later on, he explores this variety of reasonableness further.

The idea, to put it briefly, is that in matters of fundamental justice, reasonable persons will only act in ways that they can justify to one another. The ability to justify the use of political power in terms that others canaccept is what Rawls calls the duty of civility. And note that Rawls explicitly connects this idea to the model of citizens as free and equal. Reasonable persons view one another as free and equal and are motivated to maintain thebasic structure of society as a fair system of cooperation among them.12

It is also worth noting that Mandle identifies weak and strong

15

senses of the term, the latter of which applies to the reasonableness of persons and “represents a commitment to the ideal of society as a fair system of cooperation among free and equal citizens, recognizing the burdens of judgment, and accepting the duty of civility.”13 Mandle notes that “it seems too optimistic to believe that an overlapping consensus can be achieved by comprehensive doctrines which are not reasonable in the strong sense.”14 While I agree with this claim, such a view of the necessity that comprehensive doctrines be reasonable in the strong sense has further effects for the reasonableness of the persons who adhere to such reasonable comprehensive doctrines.

For a citizen to be reasonable, Mandle says,. . . she must be committed to a comprehensive view

that is reasonable in the strong sense. At least for purposes of evaluating matters of basic justice, the comprehensive doctrine she holds must direct her to view society as a fair system of cooperation among free and equal moral persons, to recognize the burdens of judgment, and to meet the duty of civility.15

These requirements have very substantial normative import for the conception of reasonableness in persons, to say the least. If Mandle is right—and I think he is—then it seems clear that reasonableness is a virtue of a very particular kind of party toan OC, and of a very particular kind of reasonable comprehensive

16

doctrine.In my view, the assumption of reasonable pluralism, taken

together with Rawls’s new account of the citizen, explains the motivability of potential citizens no better than the unrealistic assumption in TJ did. Consequently, I shall now arguethat Rawls fails to solve the legitimation problem, and therefore fails to solve (perhaps even to raise properly) the problem of political liberalism. First, his argument is an infinite regress. Second, resolving the regress requires Rawls to acknowledge that the problem of political liberalism is either a pseudo-problem, or unsolvable by the idea of OC.

One way to see the infinite regress arising from Rawls’s idea of appropriate motivation is to explore the extremely tightconnection he draws between the political conception and reasonableness, when describing the latter as

. . . a moral psychology drawn from the political conception of justice as fairness. It is not a psychology originating in the science of human nature but rather a scheme of concepts and principles for expressing a certain political conception of the person and an ideal of citizenship. Whether it is correct for our purposes dependson whether we can learn and understand it, on whether we can apply and affirm its principles and ideals in politicallife, and on whether we find the political conception of justice to which it belongs acceptable on due reflection.

17

(PL, 86-7.)But if reasonableness “belongs” to the political conception, it is hardly surprising that reasonable persons are motivable. Indeed, saying that reasonableness is drawn from the political conception makes it look like the former is, in some way, supervenient on the latter. In the absence of any explicit explanation from Rawls, I would suggest that this supervenience consists in citizens actually abiding by the principles of justice, which is the very same relationship that Rawls credits,in both TJ and PL, with creating a sense of justice in citizens. If my suggestion is correct, then it is disastrous for Rawls’s account, since only actual citizens can be assumed to abide by the principles of justice; and, should this be the substance of the relationship between reasonableness and the political conception, then only actual citizens could be reasonable. Thus,in the guise of an answer to the legitimation question, Rawls would simply have offered up the same answer to the simple stability question that he found to be inadequate in TJ.

My question being whether anyone but liberals can be motivable, I need to backtrack here. If, as Rawls clearly says, non-Rawlsian potential citizens are motivable only when reasonable, but (as I am suggesting he must say) can become reasonable only by abiding by the political conception, then anyreasonable non-Rawlsian potential citizens must have affirmed the political conception before having become reasonable—and, ex

18

hypothesi, must have affirmed it without having been motivable, either. This would make potential citizens simultaneously motivated and unmotivable. Rawls’s sole alternative to such egregious incoherence, however, is the admission that, where reasonableness and therefore motivability is concerned, potential citizens are indistinguishable from actual citizens. Since there is no reason to think that Rawls would accept that claim, it is not at all clear how he can explain potential citizens’ reasonableness without assuming that they already affirm the political conception instead of some non-liberal metaphysical conception. If one does not beg that very question,however, then reasonableness becomes an allegedly universal human attribute rather than a particular kind of political virtue—another position that Rawls could never permit himself tohold. I am therefore forced to attribute potential citizens’ reasonableness to their motivability, which must in turn be attributed to their reasonableness, and so on ad infinitum.

Thus does infinite regress emerge from making reasonableness a necessary condition for an answer to the legitimation question. The only response that I can see that Rawls might make to my critique is that I am wrong to assume that parties can become motivable only by abiding by the political conception of justice. Potential citizens could, instead, become motivable by the legitimation of the political conception to them, even though they have never lived under the

19

political conception and are therefore not reasonable.16 Such a response would have the boon of explaining how one could be motivable without being reasonable, but it seems to raise a troublesome question. For, if potential citizens are not reasonable, why would their metaphysical conceptions be congruent with the political conception? Were Rawls to reply that metaphysical conceptions can be reasonable while their adherents are not, then Rawls would be faced not only with the question of how this is possible, but also with an apparent disavowal of the close relationship that he asserts to exist between reasonableness and the political conception—after all, he holds that reasonableness “belongs” to the political conception in some way. If metaphysical conceptions can be reasonable, however, one might well wonder what is so special about the political conception itself. Could one of these other conceptions, rather than the exemplarily-liberal political conception, be the object of an OC? It seems hard to believe that Rawls would accept any such statement; but he nonetheless would appear to be caught between the unwelcome alternatives of making the doctrine of reasonableness belonging to the politicalconception false, otiose, or extremely mysterious.

If my argument so far is sound, then my conclusion, that only actual citizens can be motivable, stands; and, if that stands, then it turns out that, in PL, Rawls unwittingly commitsthe same error as he does in TJ: i.e., trying to answer the

20

legitimation question with an answer to the simple stability question. The only difference between the two errors is that, instead of assuming actual citizens to be Rawlsian liberals, PL

takes them—by virtue of their reasonableness—only to be liberalsof some kind. However, at this point it seems that all possible escape-routes are closed to Rawls.

I shall show the second problem raised by assuming reasonableness in potential citizens by means of two complementary arguments. First, then, Rawls assumes that, if potential citizens are reasonable, they are thereby motivable, while the problem of political liberalism acknowledges the fact of reasonable pluralism—i.e., that both potential and actual citizens are “profoundly divided” from each other by their doctrines. The problem here is that, if all citizens—actual and potential—must be reasonable to be motivable, and reasonable dispositions “belong” to the political conception, then it is hard to see how their equally-reasonable doctrines “divide” themfrom one another in the first place. This is the problem of the problem of political liberalism itself: How can it arise among parties who already generally agree on matters of basic justice?If my first argument is correct, no motivable party is distinguishable from actual citizens in any important way. In that case, the problem of political liberalism is a pseudo-problem with regard to all and any motivable parties, because, by virtue of their reasonableness, they are liberals, if not

21

Rawlsians, already.My second argument runs as follows. Potential citizens are

motivable only when they are reasonable (i.e., only if they are liberals); but any who lack such dispositions (i.e., are non-liberals) are unmotivable in principle, and the problem of political liberalism therefore cannot be solved by the idea of an OC with regard to them. This may sound like a natural consequence of the idea of OC, but if, as the previous argument seeks to show, the problem of political liberalism does not really apply to motivable parties, it can therefore apply only to unmotivable ones—i.e., non-liberals who hold non-reasonable comprehensive doctrines. If they are truly unmotivable, however,the idea of OC cannot be expected to answer the legitimation question with regard to them—indeed, it is questionable whether any liberal conception of justice could be legitimated in their eyes. So, finally, the question Rawls must answer is: if the problem does not apply to motivable parties, and cannot be solved with regard to the parties to whom it does apply (i.e., unmotivable parties), what problem does the idea of OC solve?

I conclude, then, that Rawls’s attempt to answer the stability question in PL fails to eradicate TJ’s unrealistic assumption. Indeed, it makes matters worse. The nexus of the exemplarily-liberal political conception and the reasonable disposition belonging to it make the problem of political liberalism both a pseudo-problem with regard to motivable

22

parties, and a genuine but unsolvable one with regard to unmotivable parties, because the requirement that parties possess reasonable dispositions makes the legitimation question unanswerable, which is the case, in turn, because it makes non-liberal parties unmotivable in principle. Since a cogent answer to the legitimation question is a necessary condition for the successful implementation of the idea of OC, Rawls therefore fails to solve the problem of political liberalism, either—unless one counts it as a solution to show that liberals will tend to affirm a liberal conception of justice. The “serious problem” persists, in a just slightly different form.

So, making the assumption of the reasonableness of parties—which is vital to the viability of an OC—simultaneously transforms the very problem that the idea of OC is intended to solve into either a pseudo-problem that requires no solution, ora genuine problem that is not solvable with the idea of OC so long as that idea presupposes the reasonableness of the potential parties. I conclude, then, that, where the potential parties to an OC are reasonable, there is no problem of political liberalism for the idea of OC to solve; and that, where none of the potential parties is reasonable, the problem of political liberalism cannot be solved by OC.

There is, however, at least one resort left Rawls in attempting to escape my charges. It is the notion of OC being approached by way of a constitutional consensus. Such a move, in

23

fact, would in one respect change the character of the debate completely. Instead of working within the ideal theory that Rawls so frequently stresses to be his primary concern, the developmental process that he sets out in the passages in PL on constitutional consensus historicize the issue of OC, although not with any very deep reference to particular historical precedents. Rather, the theory of constitutional consensus is, as it were, a theory of evolution for OC. If successful, this would at least make the idea of OC somewhat testable against historical precedents, which would be useful in light of the fact that, as Banu Kilan, for example, notes, it is likely that,in the real world, “the reflection leading up to such a consensus cannot be effectively conducted within the confines ofone’s own comprehensive view alone.”17

All that said, though, I do not think that this resort would help Rawls at all. Rawls’s notion of a primordial “constitutional consensus” that transmogrifies itself into an OCis confused to the point at which it is hard for me to tell whether it once again sends the idea of an OC into an infinite regress that can never explain whence parties get their reasonableness, or whether it successfully explains their possession of the virtue of reasonableness and thus once more contradicts his claim that a problem of political liberalism exists. I must therefore take both possibilities very seriously.To begin, here is Rawls’s definition of the notion of a

24

constitutional consensus.In a constitutional consensus, a constitution

satisfying certain basic principles establishes democratic electoral procedures for moderating political rivalry within society. This rivalry includes not only that betweenclasses and interests but also between those favoring certain liberal principles over others, for whatever reasons. While there is agreement on certain basic political rights and liberties . . . there is disagreement among those holding liberal principles as to the more exactcontent and boundaries of these rights and liberties, as well as on what further rights and liberties are to be counted as basic and so merit legal if not constitutional protection. The constitutional consensus is not deep and itis also not wide: it is narrow in scope, not including the basic structure but only the political procedures of democratic government. (PL, 159.)

Naturally, one may wonder how such a consensus might come about in the absence of either deep or wide agreement on the principles of justice that are to be applied to the basic structure. The only possibility that Rawls explicitly entertainsis that liberal principles of justice are accepted in a modus vivendi, intended to bring about peace after tumultuous times (ibid.). Hence, curiously enough for one who so vehemently denies that an OC is a modus vivendi, Rawls allows that the

25

constitutional consensus that is logically and temporally prior to an OC is likely to arise out of a modus vivendi. Rawls never explains why this does not mean that OC, is in fact a modus vivendi that happens to have endured long enough to be formalized with electoral procedures, and so on; but I shall take him at his word, and assume that an OC not a modus vivendi but the resultof a successful constitutional consensus that is, in its turn, the result of a successful modus vivendi. How, though, does such amodus vivendi become a constitutional consensus, and a constitutional consensus an OC?

The answer is, in short: the moral psychology of the sense of justice, coupled with the self-interested eradication of embarrassing inconsistencies between the political conception and the comprehensive doctrines whose members currently affirm the political conception. Over the course of time, Rawls argues,citizens, each of whom subscribes to some comprehensive doctrinebut few of whom subscribe to fully comprehensive ones, come to recognize “the good those principles [of justice] accomplish both for themselves and those they care for, as well as for society at large, and then to affirm them on this basis” (PL, 160). Whenever they happen to notice that their own comprehensive doctrines are inconsistent with the political conception (which, one would imagine, at least some of them should have been able to do before entering the primordial modus

vivendi), then, far from rejecting the political conception,

26

“might very well adjust or revise these doctrines rather than reject those principles” (ibid.). In a footnote to this remark, Rawls revealingly states that

. . . we distinguish between the initial allegiance to, or appreciation of, the political conception and the later adjustment or revision of comprehensive doctrines to which that allegiance or appreciation leads when inconsistencies arise. These adjustments or revisions we may suppose to take place slowly over time as the politicalconception shapes comprehensive views to cohere with it. (PL, 160n25.)

Obviously, then, Rawls sees no likelihood that citizens detecting a lack of congruence between the political conception and their doctrines will reject the former in the light of the latter—or, at least, sees no likelihood great enough to reflect on. In any case, he is, for some reason, not interested in pursuing a possibility which would require him to imagine and forestall the possibility of the collapse of the Rawlsian state in the light of the parties to the OC refusing to revise their comprehensive doctrines in the light of the fact that the political conception is inconsistent with them. Again, I shall take him at his word, for I do not believe that even this gets him out of trouble.

What of the assumption of reasonableness in citizens that, I have shown, so vitiates the idea of OC? Does the idea of

27

constitutional consensus give Rawls any more right to take it for granted that the parties to the problem of political liberalism are reasonable? If this halfway house between modus vivendi and OC can be credited with creating reasonableness in theparties to the problem of political liberalism, then perhaps I am quibbling over mere details rather than really calling Rawls’s whole theory into question. Fortunately for me, Rawls wastes neither time nor space in contradicting his own account. In the middle of his exposition of the idea of a constitutional consensus, he describes the fact of reasonable pluralism as “thefact that leads to constitutional government as a modus vivendi in the first place” (PL, 161). Once again, then, we face the question of how those who have never lived under the political conception at all, let alone affirmed it in either a constitutional consensus or an OC, come to be reasonable in sucha way that the pluralism of their doctrines is also reasonable. And now, I think it is apparent, that Rawls can give no answer to this question. For good measure, though, my critique also raises the question of how a modus vivendi whose object is the political conception and which bridges the diversity inherent inreasonable pluralism is different from an OC which shares this object and result. Once again, Rawls cannot give any answer thatI can see. So, he has not only failed to show how an OC is possible without begging questions that make an OC unnecessary; he has also failed even to remain consistent with, never mind

28

bolster, his oft-stated (and oft-quoted) view that an OC is not a modus vivendi.

Rawls’s notion of a constitutional consensus, then, seems to boil down to this. The political conception is initially accepted by non-liberal parties in a modus vivendi. If the arrangement endures, the political conception gradually becomes ingrained into the moral psyches of citizens. Once that is the case, citizens will change their beliefs in order to be able to go on abiding by the political conception, regardless of the price that their own doctrines must pay for them to do so. Here,one can recognize in this passage a very close relative of Rawls’s well-known theory of reflective equilibrium—one which makes it less puzzling why he takes certain moral intuitions to be what he often calls “our” intuitions, but at the same time portrays reflective equilibrium itself as pandering to one’s individual or one’s doctrinal community’s current self-interest,rather than a principled adjustment of either theory or practicein accordance with what we recognize to be good philosophical reasons. And, of course, this also suggests that those who engage in reflective equilibrium of this type must have held liberal comprehensive doctrines in the first place, in order to make it likely that they would prune those doctrines in the light of the good that abiding by the political conception does them. For, who but liberals would recognize those things as good, in the first place?

29

Taken to their logical conclusion, all of the foregoing considerations would mean that, by the time this kind of societyreaches an OC, all of the most important differences between comprehensive doctrines have been erased. Once again, I find quite baffling the idea that such a citizenry would be riven by conflict and division; and therefore cannot see how the problem of political liberalism would ever arise in such circumstances. Indeed, the problem seems likely to arise only when society heads down the road not taken by Rawls, leading to citizens at least threatening to reject the political conception when they recognize that it is inconsistent with their own doctrines. Then, however, I once again cannot see how one could resolve this problem without once again begging the question that these parties are reasonable—a virtue that Rawls describes as arising from acceptance of, and not from doubt about the political conception.

This dissipates, once and for all, the mystery of why any non-liberal should come to affirm the principles of justice, andtherefore why any non-liberal should be reasonable: and it is, Rawls admits (perhaps unwittingly) in the way I have shown, because they have already become liberals or—more likely, I think—they were liberals all along. Hence, it is clearer than ever that Rawls’s idea of OC in fact evades the legitimation question that he sets himself, because he assumes that the senseof justice, developed during the constitutional consensus that

30

the political conception regulates, brings about an OC. Instead,PL answers the simple stability question in the same way as TJ answered it, which is to say with the moral psychology of the sense of justice—and the inadequacy of exactly that earlier answer, recall, is what Rawls takes to raise the problem of political liberalism in the first place. The problem of political liberalism, viewed from the perspective of the constitutional consensus, is therefore a mirage; and, if that isso, then the idea of OC is a wheel that turns nothing.

In short, then, Rawls’s attempt to solve the stability problem is ridden with the same inconsistencies, no matter whether one looks at it from the point of view of the pure idea of OC or from that of the evolution of a modus vivendi, via a constitutional consensus, into an OC. The efficacy of the idea of an OC, as a solution to the problem of political liberalism, depends utterly on the motivability of parties who are non-liberals in matters of basic justice. The process that creates motivability (and thus answers the legitimation question), however, makes it impossible for any party to become motivable whose metaphysical conception of justice does not already at least include the whole of the political conception. The idea ofOC can therefore solve the problem of political liberalism only where all parties are already liberals in matters of basic justice; but, if this is so, then the problem of political liberalism cannot have arisen between them. With regard to

31

parties between whom the problem can arise, the idea of OC cannot solve it; for, if not all their metaphysical conceptions are congruent with each other in any way, they cannot all be congruent with the political conception, either. This implies that, where OC can be implemented, there is no problem of political liberalism for it to solve; and, where there is such aproblem for it to solve, OC cannot be implemented. Hence, the idea of OC is quite infeasible. The serious problem that Rawls admits to spotting in TJ persists in PL, in almost exactly the same form.

32

ENDNOTES

1 John Rawls, Political Liberalism (New York: Columbia University Press, 1993).

2 John Rawls, A Theory of Justice (Oxford: Oxford University Press, 1971).

3 Since possessing the virtue of reasonableness is, besides anything else, a necessary condition for possessing the liberal virtue of tolerance, Rawls is far from alone among contemporary theorists in recognizing the fundamental importance of persons (and, therefore, citizens) being reasonable. Assumptions very similar to his may be found in the work of Thomas Nagel, Brian Barry, Gary Gutting, and Robert E. Goodin, the first two of whom utilize a “reasonable rejectability principle” which explicitly presupposes a virtue of reasonableness that is applicable to the principles of justice that citizens propose to each other as worth of them all living under. In each case, their principles stem from T.M. Scanlon’s “Contractualism and Utilitarianism,” in Utilitarianism and Beyond, ed. Amartya Sen and Bernard Williams (Cambridge: Cambridge University Press, 1982): 103-128. Barry, for instance, states that impartial principles of justiceare ones that “nobody could reasonably reject” (Brian Barry, A Treatise on Social Justice, Volume II: Justice as Impartiality [Oxford: Clarendon Press, 1995], 168), while Nagel describes the reasonable rejectability principle as characteristic of

. . . a new type of liberal theory, exemplified by the workof Rawls and others, whose distinctive feature is that it bases the legitimacy of institutions on their conformity to principleswhich it would be reasonable for disparate individuals to agree on, where the standard of individual reasonableness is not merely a premoral rationality, but rather a form of reasoning that includes moral motives. (Thomas Nagel, “Moral Conflict and Political Legitimacy,” Philosophy and Public Affairs 16 [1987]: 215-240; 220.)

4 I shall cite page-numbers of PL and TJ in parenthetically, and put all other references into endnotes.

5 Jürgen Habermas, “Reconciliation through the Public Use of Reason: Remarks on John Rawls’s Political Liberalism,” Journal of Philosophy 92 (1995): 109-131; 120.

6 John S. Dryzek and Simon Niemeyer, “Reconciling Pluralism and

Consensus as Political Ideals,” American Journal of Political Science 50.3 (July 2006): 634-49; 636. Emphases authors’. The reference to Mouffe is to Chantal Mouffe, “Democracy, Power and ‘The Political’,” in Democracy and Difference: Contesting the Boundaries of the Political, ed. Seyla Benhabib (Princeton: Princeton University Press, 1996), 245–56.

7 For clear accounts of this point, see not only Mouffe, “Democracy, Power and ‘The Political’,” but also Chantal Mouffe, “Political Liberalism: Neutrality and the Political,” Ratio Juris 7.3 (Dec 1994): 314-24; Chantal Mouffe, “Deliberative Democracy or Agonistic Pluralism?” Social Research 66.3 (Fall 1999): 745-58; and “The Limits ofJohn Rawls’s Pluralism,” Theoria 118(Mar 2009): 1-14.

8 Thomas E. Hill, Jr., “The Stability Problem in Political Liberalism,” Pacific Philosophical Quarterly 75 (1994): 333-352; 345-6. Rawls’s own words,read in this light, confirm Hill’s reading. On the first point, Rawlssays that “[i]n one way, we view stability as a purely practical matter: if a conception fails to be stable, it is futile to try to realize it”; and, on the second, “[f]inding a stable conception is not simply a matter of avoiding futility,” since “what counts is the kind of stability, the nature of the forces that secure it” (PL, 142).

9 There would, of course, be a half-way house, of those who are motivable but not motivated. They would, perhaps, be other kinds of liberals—Millian utilitarians, for instance—whose vision of society is similar to that of the well-ordered Rawlsian state, but whose conception of justice is quite difficult from the fundamentally-Kantian political conception. I leave such parties aside, because thelogical problems this idea adumbrates would complicate my argument too horrifyingly—although I think that they would also strengthen it,because the logical problems arise for Rawls’s theory, and not for myargument.

10 Jon Mandle, “The Reasonable in Justice as Fairness,” Canadian Journal of Philosophy 29 (1999): 75-107; 76-7.

11 Ibid., 77.

12 Ibid, 93.

13 Ibid., 94.

14 Ibid., 96.

15 Ibid., 95. Mandle reiterates this conclusion at ibid., 102.

16 Note that making this response would require Rawls to withdraw the claim that all parties—actual and potential—possess reasonable dispositions.

17 Banu Kilan, “J. Rawls’s Idea of an ‘Overlapping Consensus’ and the Complexity of ‘Comprehensive Doctrines’,” Ethical Perspectives 16.1 (2009): 21-60; 50.