(Philosophy of the Social Sciences 31 (1) March 2001)

Terrence Kelly

California State University, Hayward

Department of Philosophy

25800 Carlos Bee Blvd.

Hayward, CA 94542

(510) 885-3566

tkelly@csuhayward.edu

Like many critics of Rawls, Habermas doubts that the Original Position (OP) can do the work expected from it without implicitly utilizing substantive philosophical assumptions. A good example of this is the priority of the first principle of justice. To Habermas, it seems unlikely that the parties of the OP could give priority to rights and liberties over and against other primary goods, and by extension give priority to the first principle of justice. Such a judgement seems unattainable given the basic equipment that Rawls provides for the parties of the OP. Armed only with instrumental rationality and the primary social goods, it is hard to imagine how the parties of this hypothetical decision-making process would come to favor some goods, such as liberty, over others, for instance wealth. Would not such a judgement require precisely the kind of substantive evaluative criteria that has been denied to the parties of the OP?

Rawls defends the priority of the first principle by arguing that liberty is "among the social conditions necessary for the development and exercise" of the individual’s ability to fashion a conception of the good life itself (Rawls 1996: 313). Thus rights and liberties will be given special weight in the deliberations of the OP because they represent the structural underpinnings of the pursuit of the good life itself. As the parties of the OP recognize the unique role that rights play in the utilization of the other primary social goods, they will come to accord priority to the first principle.

However, such a judgement is only possible if the parties of OP already assume that they possess a robust practical rationality that is capable of developing a conception of the good life—one that orients the individual’s enjoyment of the primary social goods. Furthermore, the parties of the OP would have to be able to recognize the unique structure of rights. They would have to recognize that rights are not actually goods at all, but are rather special guarantees or, to use Ronald Dworkin’s term, "trumps" that protect the individual as a free and equal member of society (Dworkin 1977). This in turn requires that the parties of the OP recognize themselves as free and equal members of a well ordered society, because only under such conditions would rights play the critical role that Rawls assigns to them.

The problem with these suppositions is that the parties of the OP possess only an egoistic rationality that calculates efficient outcomes, but does not make substantive evaluations. Furthermore, it would seem that the parties of the OP would need to know much more about their situation (i.e. the condition of free and equal citizenship) than the veil of ignorance allows. If Rawls is going to build substantive assumptions into the description of the Original Position itself (i.e. citizens as free and equal, the forms of practical rationality), then he must provide a defense for such assumptions. However, given Rawls’ desire for a "freestanding" theory that avoids philosophical controversies, such a defense is not forthcoming. This leads Habermas to conclude that "Rawls introduces normative contents into the very procedure of justification." (Habermas 1995: 119) If true, this charge of circularity would certainly undermine the freestanding nature of the principles of justice.

Rawls can muster a robust defense against the charge of circularity by empirically grounding his account of the Original Position through the interpretive social sciences. While substantive in nature, the normative contents of the OP can be empirically identified and reconstructed as shared features of democratic life. Such an approach, I argue, can blunt the charge circularity. However, because Rawls adopts an impoverished form of the interpretive social sciences, his account ultimately falls prey to contextual relativism.

In his response to Habermas, Rawls admits that the Original Position is built around basic concepts that appear to be substantive: citizens as free and equal in a well ordered society, the principles of practical rationality, and the public role of a political conception of justice. However, Rawls rejects the contention that these are substantive concepts—at least in the sense that they depend on any particular worldview, metaphysical, or philosophical doctrine. While these concepts may not be entirely formal or purely procedural, they are nonetheless "political" concepts. As political, these concepts are drawn from the "common fund" (Rawls 1995: 135) of democratic political culture that is "familiar and intelligible to the educated common sense of citizens." (Rawls 1996: 14) For the purposes of moral theory (as opposed to moral or political philosophy), the justification of these concepts comes only from the fact that they are based on "presently accepted general beliefs and forms of reasoning found in common sense…" (Rawls 1996: 224f).

What is interesting about Rawls’ defense is that it rests on a straightforwardly empirical claim— the concepts behind the OP are, in fact, already accepted features of the common sense possessed by democratic citizens. What is even more interesting, however, is the methodology used by Rawls in identifying the shared elements of democratic common sense and converting them into the ideal concepts that make up the OP. It is, I argue, a methodology that bears striking similarity to Habermas’ conception of "rational reconstruction." If successful, Rawls can appeal to this methodology to defend the OP from the charge of circularity.

Rational reconstruction is a joint endeavor between the social sciences and philosophy that identifies various intuitions that are embedded in particular practices and thematizes them into basic conceptions that make explicit the implicit formal aspects of the practice. It then idealizes those concepts into general rules or norms of social interaction. This makes rational reconstruction both descriptive and prescriptive. It descriptively explicates the implicit intuitions or pre-theoretical know-how agents utilize in their practices by thematizing intuitions into formal concepts or rules. The formal concepts and "exact arguments" that replace implicit "common sense" (Habermas 1979: 13) help agents recognize their own practices in a more articulate way. From this explicit and thematized description of embedded intuitions, rational reconstruction also constructs idealized norms that guide practices. The normative force of these idealized norms comes, in part, from the fact that practitioners already, at least implicitly, accept them. But more than that, the idealized norms of rational reconstruction represent the "formal" features of the practice-- they constitute nothing less then the inescapable presuppositions of a practice

In utilizing this methodology, Habermas focuses on those intuitions that are embedded in the practice of communication. Through social scientific research, for instance Harold Garfinkel’s studies on everyday communication (Garfinkel 1967), Habermas (Habermas 1984) identifies the operative intuitions of successful communicative acts and thematizes them into a number of idealized concepts (i.e. the validity claims of truth, rightness and sincerity). These concepts are then part of a second idealization that constructs a legitimate decision-making procedure— justificatory discourse that pragmatically concretizes the norms of an "ideal speech situation."

When Rawls thematizes his "political" concepts and their idealization into the description of the Original Position he uses a similar methodology. In Political Liberalism, Rawls rejects the suggestion that the OP is the product of construction (as the two principles are) as well as Ronald Dworkin’s argument that the OP is the product of reflective equilibrium. Instead, the OP is, for Rawls, simply "laid out" as the best description of the formal conditions of practical rationality given our moral sensibilities.

This laying out, I suggest, takes the form of a rational reconstruction of the intuitions embedded in the structure of our moral sensibilities as deployed in democratic practices. These intuitions are thematized through a series of idealized concepts that serve as the corner stone of an idealized decision-making procedure—the original position. This reconstruction can be broken down into three stages: identification, idealization and proceduralization.

The first stage of this reconstruction of democratic sensibility begins by appealing to the "common fund of political culture" (Rawls 1995: 135) in an attempt to identify those intuitions that are fundamental to the practice of democracy. Rawls argues that one can here rely on an analysis of the "leading representative writers," (Rawls 1999: 288) "historical documents" (Rawls 1995: 135) and fundamental institutions of the constitutional democratic tradition. By analysis of the central liberal democratic thinkers (i.e. Hobbes, Locke, Kant), the American Constitution, The Federalist Papers, U.S. common law, etc., one can distill those intuitions that are most central to democratic practices. The identified intuitions then serve as the raw material for the second, and most critical, stage to Rawls’ reconstruction.

Once the implicit intuitions of democratic culture have been identified, they are idealized or thematized into a number of basic concepts. In one of his more extended comments on his methodology, Rawls notes in "Justice as Fairness" that:

I think of justice as fairness as working up into

system of cooperation over time from one generation

As in any good reconstruction, these idealized or thematized conceptions (i.e. persons as free and equal) are both descriptive and prescriptive. They are descriptive in that they clarify and formalize the often unreflective or pre-theoretical know-how that constitutes democratic common sense. But more than that, Rawls argues that the idealized concepts reconstructed here must represent the best overarching or systematic representation of the myriad intuitions that make up democratic common sense.

If the reconstruction has properly identified the intuitions of democratic common sense and if they have been adequately represented in explicit idealized concepts, then those concepts can help democratic citizens recognize their own practices in a more articulate way. Furthermore, because these concepts have gained the implicit acceptance of competent democratic citizens, they can be used normatively in providing ideals that are based on the guiding assumption of social contract theory—legitimacy is produced through the agreement of citizens.

Rawls proposes to use these concepts in the third stage of his reconstruction—proceduralization. Once agreed upon concepts have been worked up, the theorist can utilize them in a demand for a basic consistency between actual democratic practices and the self understanding of democratic common sense. Rawls does this by using the fundamental concepts of democratic common sense as the building blocks for an ideal normative decision-making procedure. In a move strikingly similar to Habermas’ reconstruction of the "ideal speech situation," Rawls takes the idealized concepts from stage two and through a second idealization lays out a procedure—the OP-- by which principles of justice can be legitimated. This decision-making process serves as a "representational device" which "makes vivid" (Rawls 1971: 18) the idealized concepts of free and equal persons, a well ordered society and the public role of a political conception of justice. Those concepts in turn systematically represent the fundamental intuitions embedded in democratic common sense.

Thus, like the idealized concepts themselves, the OP is both descriptive and prescriptive. Rawls notes in Political Liberalism, "As a device of representation the OP serves as a means of public reflection and self clarification. It helps us work out what we now think, once we are able to take a clear and uncluttered view of what justice requires when society is conceived as a scheme of cooperation between free and equal citizens from one generation to the next." (Rawls 1996: 26) If the rational reconstruction has been successful, the OP can be said to be "political" because it come from the shared common sense of democratic citizens.

By construing his project in reconstructive terms, we can fill out a Rawlsian defense against the charge that the OP is constructed around unacknowledged substantive assumptions. On the one hand, the fundamental concepts of citizens as free and equal, a well ordered society and the public role of a political conception of justice are "substantive" in the sense that they have a specific concrete content. For instance, Rawls cannot help but rely on a specific account of freedom (i.e. freedom as liberty) and moral reasoning (i.e. two distinct moral powers) when assigning priority to the first principle. In this sense Rawls is not a pure proceduralist. However, in a different sense—and this is the sense Rawls uses the term-- the idealized concepts of the OP are not substantive because they are drawn from the common fund of democratic culture. By appealing to rational reconstruction Rawls can shift the justificatory burden of his theory. "Political" concepts do not require a metaphysical defense, only a social scientific one—namely the kind of reconstructive argument that I have tried to outline above. As reconstructed, Rawls’ idealized concepts may have concrete content, but it is a content that is accepted in the implicit common sense of all competent citizens of democratic societies—thus making them, at least in the context of Western Democracies, "free standing." Thus, contra Habermas, these concepts are not in need of a prior justification as they already enjoy the (at least implicit) agreement of citizens.

However, shifting the burden of justification and relieving the burden of justification are not the same thing, though Rawls sometimes proceeds as if they were. With this reconstructive defense in hand, Rawls now has the burden of producing an empirical justification for how, in stage one, he identifies the embedded intuitions of democratic common sense. He must also muster a methodological defense of the particular way he takes those embedded intuitions in stage two and idealizes them into the thematized concepts.

So far, I emphasized the similarities between Rawls and Habermas’ use of rational reconstruction to highlight the fact that Justice as Fairness, like Discourse Ethics, is grounded in the reconstructive social sciences. However, there are important differences in the ways Habermas and Rawls deploy rational reconstruction and these differences are the source of new problems for Rawls.

An initial concern is Rawls’ failure to live up to his own methodology. While I have outlined the methodology that Rawls says he relies on in developing the OP, he rarely, and with little rigor, actually does it. His work, for instance, contains little social scientific research that actually seeks to identify the embedded intuitions of democratic practice. While Rawls examines some of the leading representative writers of the liberal democratic tradition and some leading institutions of constitutional democracy, most notably the United States Supreme Court, he fails to adopt a rigorous social scientific study of democratic culture. For instance, a systematic analysis of significant texts in democratic cultures could be useful in culling out common themes and ideas that are basic to democratic common sense. This could help support a claim that particular intuitions are fundamental aspects of democratic common sense. Perhaps to avoid social scientific controversies, Rawls instead opts for a kind of "armchair" history of the themes that are, by his own lights, at the core of democratic common sense. But such a weakened form of the interpretive social sciences simply will not do once Rawls is committed to the empirical claim that the Original Position best represents the intuitions that democratic citizens actually have. Without more research, Rawls’ central empirical claim remains a thesis to be proven.

But this is not necessarily a fatal problem. Rawls would certainly be willing to bet that on further research the leading texts of liberal democracy would reveal a number of common intuitions that are captured in his reconstructed idealizations. After all the intuitions he is appealing to are very common, and on first blush many leading writings of the liberal democratic tradition certainly do seem to reflect them.

This, however, highlights the problem of relying on leading historical writings as an evidentiary basis for identifying those intuitions that are shared by democratic citizens. While historical writings may serve as helpful exemplars, they cannot replace actual research on the intuitions that citizens actually hold. While Madison and Hamilton may exhibit the intuitions that Rawls is reconstructing, it is surely a leap to say that this is evidence that all democratic citizens share these intuitions. Certainly feminists and critical race theorists would rightly point out the "leading representative writers" of the liberal democratic tradition tend to reflect certain class, gender and ethnic backgrounds. For instance, Charles Mills’ work highlights the ways in which the fundamental ideas of the liberal tradition (i.e. social contract theory) have typically been used (and are still used) in the name of domination and privilege (Mills 1997). When properly understood in the context of historical analysis, the intuitions of privileged members of society may turn out to be less egalitarian and more problematic than realized on first blush.

Furthermore, when one moves beyond Rawls’ favored writings, there appear to be many forms of cultural expression which indicate that at least some citizens of democratic societies do not share the liberal intuitions captured in the OP. For instance, some forms of "Gansta Rap" represent such a deep cynicism about "justice" that Rawlsian intuitions about fairness have been replaced by the intuitions of a modern day Thrasymachus. N.W.A.’s "Fuck tha Police," for example, expresses a "might makes right" idea of justice. In taking on this theory of justice, N.W.A. argues that if the white supremacist state is to be overthrown, appeals to fairness must give way to appeals to force, strength, and violence.

On the other side of the spectrum (in more ways than one), Roland Barthes’ argues that the popularity of professional wrestling can be attributed to a deep rejection of procedural justice (Barthes 1972). In professional wrestling, the "good guys" and "bad guys" are clearly marked with salient symbols drawn from the lifeworld. For the spectators, the match becomes a drama between good and evil in which justice is served only if the "good guy" is victorious. Anything less constitutes tragedy and injustice. Of course, as is well known, precisely how the good guy comes to defeat his opponent is unimportant. The rules only serve as a pretense for the match and can be quickly discarded if that is what is necessary for good to triumph. If, for instance, the good guy requires the assistance of a metal folding chair in his match, then that is exactly what justice requires. For the followers of professional wrestling, justice is fundamentally substantive and not procedural. It is not about impartiality, or equality. It is about good triumphing over evil.

I offer these examples simply to raise some questions regarding Rawls approach. A great deal more research would have to be done to show that N.W.A. truly represents a rejection of Rawlsian intuitions, or if anyone, particularly oppressed groups, really shares N.W.A.’s vision of justice. The same is true for wrestling. Would fans of wrestling also apply their substantivism to politics as such?

The point is that answering such questions requires a commitment to broader social scientific research on democratic culture. Rawls himself only provides prima facie evidence that his reconstruction has identified those intuitions that are implicitly accepted by citizens of democratic societies. Gangsta rap and wrestling offer some interesting prima facie evidence to think that this might not be true. In other words, it is by no means obvious that the citizens generally in democratic societies share those intuitions that Rawls identifies and idealizes in his reconstruction of democratic common sense. Indeed, what is particularly distressing about the N.W.A. example is that if it is the case that a significant number of marginalized citizens no longer accept liberal intuitions about justice (as certainly many feminists have argued), then Rawls’ reconstruction may well suffer from gender, race and/or class biases. This should come as no surprise. A great deal of the motivation behind rigorous social scientific method is precisely the avoidance of ethnocentric and biased cultural analysis. Because Rawls eschews a rigorous methodology, his reconstruction seems particular prone to precisely those dangers.

This problem has not gone unnoticed. In fact, the criticism that I offer here is something of a social scientific version of R.M. Hare’s complaint that "if (as will certainly be the case) [Rawls] finds a large number of readers who can share with him a cozy unanimity in their considered judgements, he and they will think that they adequately represent "people generally" and congratulate themselves on having attained the truth" (Hare 1989). While the later Rawls makes it clear that justice as fairness is not a realist theory of justice (and is thus not about finding the true nature of justice), the problem of representing people generally remains.

However, I believe that for all their seriousness, these objections are also not entirely damning. While Rawls appeals to the leading historical writings of the democratic liberal tradition, there is no reason why he could not supplant such an analysis with more straightforwardly sociological studies of the embedded intuition of agents "on the ground." Of course, if such studies show that citizens, or even particular groups of citizens, do not share Rawls’ intuitions about justice and democracy, then the description of the OP has serious difficulties. But again, Rawls would surely be willing to bet that further social scientific research would cash out the basic concepts of democratic common sense he has identified.

For instance, there is a rich body of literature on procedural justice that indicates that many people do, in fact, have procedural intuitions regarding conflict resolutions—particularly when such conflicts occur in the political/legal sphere. Tom Tyler’s work suggests that people do tend to believe that individuals should be treated as free and equal group members (whether citizens, institutional or community members) and that goods should be distributed equitably. Furthermore, Tyler argues that there is good research that indicates that people generally believe that fair procedures should be used in making equitable decisions, and that impartiality is a key component of any fair procedure. There is even research that indicates that the veil of ignorance is a fairly intuitive condition of fair procedure. As I will argue, this work can prove to be invaluable for Rawls in his effort to establish an empirical basis for the OP. The social scientific literature on procedural justice can go a long way in establishing the generality of the intuitions that support the OP.

Even with more research in hand, Rawls’ approach to rational reconstruction still faces the serious problem of contextualism. Again a comparison with Habermas is instructive.

For Habermas, the outcomes of rational reconstruction are normative because they capture practical intuitions that are shared by competent agents. However, they are also normative because they thematize the formal conditions of a given practice. This means, in Habermas’ Neo-Kantianism, that one identifies the inescapable universal presuppositions of a particular practice.

This Neo-Kantianism motivates a research program that is universal in scope and exhaustive in depth. In terms of scope Habermas argues that rational reconstruction should "relate to pre-theoretical knowledge of a general sort, to universal capacities, and not to the competencies of a particular group (e.g. the ability to utter sentences in the Low German dialect.)" (Habermas 1979: 15). Of course, one cannot rule out the possibility that the reconstruction "is valid only for specific, anthropologically deep-seated behavioral systems…" (Habermas 1979: 15). Thus the regulative ideal of the ongoing empirical research that supports rational reconstruction is universality. Any particular study of the competency of practitioners should be able to be expanded and continually confirmed in ever widening research contexts. Only in this way can the theorist increase the confidence in the universality of the proposed reconstruction.

In terms of depth, Habermas argues that rational reconstruction must capture all of the relevant intuitions utilized by competent agents in the practice being studied. This translates into a subtle methodological decision on Habermas’ part. He argues:

A proposal for reconstruction…can represent

The shared intuitions of competent agents cannot be criticized with the reconstructive process because they are the raw material, as it were, by which one grounds the normative basis of criticism itself. To criticize intuitions here would be to smuggle evaluative standards into the process of rational reconstruction itself. And these evaluative standards could not, without some serious circularity, be said to be the product of rational reconstruction.

More importantly, Habermas requires such a stringent standard if rational reconstruction is to capture the inescapable presuppositions of the practical know-how that agents must master to successfully engage in a practice. This means that a good rational reconstruction can identify the relevant intuitions of competent practitioners and reconstruct thematized and idealized concepts that capture all of them.

For instance, formal depth grammar rules should be interchangeable with any of the intuitions utilized by competent speakers in particular language acts. Inadequate reconstructions that cannot account for all of the intuitions of competent practitioners cannot be said to be transcendental because they cannot be applied to all instances of the successful practice. On this score Habermas approvingly sites Carnap’s requirement of explications that "the explicans should be like the explicandum, that is, from now on the explicans should be able to be used in place of the explicandum in all [emphasis mine] relevant cases" (Habermas 1979: 11).

For Rawls, the methodology of reconstruction is quite different both in scope and depth. While the OP is supposed to represent the "formal features" of the moral powers of citizens, Rawls does not view his theory of justice as capturing the inescapable presuppositions of democratic practices. This is so because Rawls’ use of rational reconstruction is narrow in scope. In practice, his reconstruction focuses on the intuitions embedded in the common sense of democracy as practiced in the United States.

Rawls is also not committed to the exhaustiveness that Habermas demands for rational reconstruction. For instance, it is very clear that while the basic concepts that make up the Original Position capture our intuitions in a "systematic" or "overarching" way, they by no means account for all of our relevant intuitions about justice. This point has been well made by critics and even supporters of Rawls (Dworkin 1989: 24). For instance, neither the basic concepts of the Original Position, nor the Original Position itself, capture the commonly held intuition that fair decisions are fully informed decisions. Thus, while the veil of ignorance captures our intuitions about the role of a political conception of justice and moral powers related to it, it does so precisely at the expense of other widely held intuitions. In the face of such conflicting intuitions, how can rational reconstruction proceed?

Rawls has had little to say about this issue. However, Ronald Dworkin has taken up this problem at some length. Dworkin argues that the Original Position is based on a coherency model of our democratic intuitions. While moral theory begins with the intuitions embedded in democratic common sense, it cannot absorb them all. In pluralistic societies there are simply too many varied intuitions to reconstruct a procedure for choosing principles of justice that could account for them all. Instead, Dworkin argues that Rawls begins with common sense intuitions about justice and organizes them into a coherent or "consistent plan of action" (Dworkin 1989: 28). While Habermas also does this, what makes Rawls work different is that some intuitions will be rejected within the process of reconstruction in the name of consistency. In other words, as the theorist organizes intuitions about justice into series of coherent and consistent concepts, some intuitions will have to be rejected because they are simply at odds with the bulk of our other intuitions about justice (Dworkin 1989: 38). In other words, Dworkin would call the rational reconstruction of the OP a unique kind of reflective equilibrium.

Rawls has been ambivalent about this reading of the OP. While he rejects, he also has been clearly influenced by Dworkin in this matter. Rawls himself notes in PL that the OP "serves as a mediating idea" which enables us "to establish greater coherence among our judgements" (Rawls 1996: 26).

If Dworkin is correct then a fuller account of Rawls’ use of rational reconstruction emerges. For Rawls rational reconstruction begins with the embedded intuitions of democratic common sense and attempts to arrange them into a coherent and consistent order through a number of idealized concepts. These concepts will not account for all of the intuitions embedded in democratic common sense, simply the bulk of them in a coherent fashion. These concepts are descriptive in that they describe some of the basic features of democratic common sense, but are also normative because agents already, at least implicitly, agree with them, and because these particular concepts represent the most coherent way of organizing the bulk of our varied intuitions.

Again the merit of this approach is that it can provide for a fair procedure in the face of conflicting intuitions about justice. Unfortunately, this approach also raises the possibility of contextualism. On this reading, the OP might only represent the formal conditions of one particular way of organizing our intuitions. While Rawls argues that there is no better way then the OP to elaborate the basic concepts of democratic common sense, there is no reason why Rawls’ basic concepts are the only ones that could be thematized from democratic common sense. Once the theorist is allowed to discard some intuitions in the name of coherence and consistency, multiple accounts of justice are certainly possible—and each one could claim to be a consistent and coherent rendering of our intuitions.

For example, I noted that Barthes’ analysis of professional wrestling points to a deep intuitive basis for substantive models of justice. Obviously, this is one of the intuitions that Rawls must discard in the name of coherency and consistency. But one could certainly redeem these substantive intuitions and build a coherent theory of justice by rejecting Rawls’ favored intuitions. Indeed, communitarian theories of justice often claim to their merit a certain authenticity to our moral sensibilities.

For these reasons, coherency is too weak a standard to show the superiority of Rawls’ reconstruction of democratic common sense. Instead, he must claim against his competitors (i.e. communitarians) that (a) his theory presents a coherent and consistent account of our embedded democratic intuitions and (b) that in doing so it captures a larger share of our intuitions than other theories. In other words, Rawls must argue that his reconstruction of democratic common sense is the most consistent and coherent arrangement because it represents a greater bulk of our intuitions than its competitors.

Even if this claim could be cashed out, it is all the more sensitive to the specific culture one is studying. Now, even the differences between different democratic cultures are relevant to Rawls’ reconstruction. What might be the most coherent account of one democratic culture might not be in another. Because Rawls’ account of the OP has clearly been built on a survey of American democratic culture, there is no reason to expect it to be the optimum account for other democratic cultures as well. For instance, Habermas argues that on a contextualist reading "[Rawls] would have to admit that the two principles of justice do not claim to be valid for, say, Germans, because formative equivalents to the American constitutional tradition cannot be found in German culture and history" (Habermas 1996: 62).

This problem is illustrative of the interrelationship between philosophy and the social sciences in rational reconstruction. Given his strategy of avoidance, Rawls seeks to avoid controversies in both philosophy and the social sciences. He therefore avoids the neo-Kantian ambition of reconstructing practical rationality as such. This more cautious philosophical stance motivates Rawls’ modest social research agenda. In practice, Rawls’ use of the interpretive social sciences extends little further than an articulation of a coherent account of democratic practice within American culture. The problem is that in such a weakened form, the interpretive social sciences are too impoverished to help Rawls maintain his uneasy position between Neo-Kantianism and contextualist relativism. The narrow empirical basis of the OP provides little confidence that the validity of the OP extends beyond American culture. Indeed research on procedural justice indicates broad acceptance of basic norms of procedural fairness, the emphasis given to these norms often varies from culture to culture. For instance, Tyler point out that studies comparing German and American procedural intuitions found that Germans placed much more importance in the norm of correctability than did Americans (Lind and Tyler 1988). These kinds of differences give good reason to doubt that the OP can be unproblematically seen as the most coherent arrangement of democratic intuitions for all democratic cultures. Because democratic common sense is not homogeneous from culture to culture, there is no reason to think that there is a single most coherent account of all of the them.

Of course, this might not be a bad thing. Richard Rorty would certainly count it to Rawls’ credit that justice as fairness works pragmatically in creating a "socio-historical description of the way we live now" (Rorty 1991: 185). It is precisely in avoiding ponderous and unattainable transcendental goals that Rawls emerges as a superior alternative to Habermas in post-metaphysical philosophy.

Despite its attractiveness to Rorty, Habermas is at least correct in noting that contextualism is deeply problematic for Rawls. Rawls himself hardly admits to such contextualism in his work and has rejected Rorty’s interpretation of justice as fairness. Furthermore, works such as The Law of Peoples (Rawls 1999c) seem inextricably problematic if, for instance, the Original Position cannot be extended at least to other democratic societies (let alone to non-democratic well ordered societies).

Expanding the validity of the OP requires nothing less than a stronger philosophical commitment to the interpretive social sciences. Only through a true reconstruction of democratic common sense as such, Rawls can avoid the problem of contextualism. Philosophically, a reconstruction of democratic common sense could make use of formal pragmatic arguments which, going beyond mere coherence, attempt to establish the inescapable preconditions of democratic practice as such. Such arguments provide a better basis for a non-culturally specific decision making procedure in Rawls’ theory of justice. And while more demanding, such arguments are better suited for establishing the superiority of Rawls’ theory of justice—even in the face of coherent and consistent alternatives. The OP could then be viewed as more than a coherent arrangement of common intuitions, but could a representation of the inescapable preconditions of pluralistic democratic practice. The intuitions not captured by the OP, Rawls could argue, are simply not essential to the practice of democracy.

Formal pragmatic claims also motivate, as in Habermas, a more aggressive social scientific research agenda. A formal pragmatics of democratic practice requires a systematic study of democratic culture that expands beyond the confines of one’s immediate context. As I have noted earlier, there is already interesting empirical research that indicates that the procedural intuitions behind the OP are much stronger and widespread than Rawls acknowledges. Research on procedural justice such as Tyler’s could help establish a strong empirical basis for a philosophically richer reconstruction of democratic common sense (and perhaps to an even broader moral sensibility). Indeed, Tyler suggests procedural intuitions are not culturally specific, but may extend broadly into well-ordered societies generally.

Armed with a more robust social scientific and philosophical program, Rawls would at least posses a methodology that could cash out his claim that the OP represents the formal conditions of democratic common sense itself. Such an approach could provide him with an empirical basis for the description of the OP while developing a general theory of democratic practice in pluralistic conditions. This would give the OP a quasi-transcendental status that could avoid both circularity and contextualism.

REFERENCES

Barthes, Roland. 1983. Mythologies. London: J. Cape.

Dworkin, Ronald. 1989. The original position. In Reading Rawls. Ed. Norman Daniels.

Palo Alto: Stanford University Press. 16-52.

Garfinkel, Harold. 1967. Studies in ethnomethodology. Cambridge: Polity Press.

Habermas, Jürgen. 1979. What is universal pragmatics?. In Communication and the

evolution of society. Tr. Thomas McCarthy. Boston: Beacon Press.

Habermas, Jürgen. 1984. The theory of communicative action, volume I. Tr. Thomas

McCarthy. Boston: Beacon Press.

Habermas, Jürgen. 1995. Reconciliation through the public use of reason. The journal of

philosophy, XCII (3). 109-31.

Habermas, Jürgen. 1996. Between facts and norms. Tr. William Regh. Cambridge: MIT

Press.

Hare, R.M. 1989. Rawls’ theory of justice. In Reading Rawls. Ed. Norman Daniels. Palo

Alto: Stanford University Press.

Lind, E. Allan and Tom Tyler. 1988. The social psychology of procedural justice. New

York: Plenum Press.

Mills, Charles. 1997. The racial contract. Ithica: Cornell University Press.

Rawls, John. 1971. A theory of justice. Cambridge: Harvard University Press.

Rawls, John. 1995. Reply to Habermas. The journal of philosophy, XCII (3). 132-80.

Rawls, John. 1996. Political liberalism. New York: Columbia University Press.

Rawls, John. 1999a. The independence of moral theory. In John Rawls: collected papers.

Ed. Samuel Freeman. Cambridge, Mass.: Harvard University Press. 286-302.

Rawls, John. 1999b. Justice as fairness: political not metaphysical. In John Rawls:

Rawls, John. 1999c. The law of peoples. Cambridge: Harvard University Press.

Rorty, Richard. 1991. The priority of democracy to philosophy. In Objectivity,

Relativism and Truth. Cambridge: Cambridge University Press. 175-97.