Symposium Introduction

Preliminary Note: As many of our readers may be aware, and as I am deeply saddened to report, Gerald (“Jerry”) Gaus (1952–2020) passed away last fall, as this symposium was still in development. I encourage our readers to read the touching tribute that Jerry’s many former students and mentees, including Ryan Muldoon, coauthored for the PPE Society in August 2020. We are fortunate that Jerry so generously agreed, as he always did whenever he was asked, to share his erudition, passion, and remarkably kind spirit with us in these pages.

Notwithstanding the marked difference in their answers, Ryan Muldoon wrestles with a similar question to the one that gripped John Rawls from Political Liberalism until his death: “How is it possible for there to exist over time a just and stable society of free and equal citizens, who remain profoundly divided by reasonable religious, philosophical, and moral doctrines?”1 In the several years since Muldoon published his book and this symposium has come to fruition, the domestic and global divisions that have troubled us all have only worsened, and categorically new ones have appeared on the scene. Anti-immigrant movements in the West have proliferated. Widespread protests and counterprotests have gripped American cities in the wake of the tragic killings of African American citizens, such as George Floyd and Breonna Taylor, at the hands of police officers. These demonstrations have further elucidated the depth of systemic racial injustice in the United States, but have also been the site of mutual enmity and violence among ideologically divided citizens. Amid global suffering throughout the COVID-19 pandemic have been bitter conflicts over various governments’ efforts at managing the virus’s spread, including locking down businesses, mandating masks, closing schools, and the manufacture and equitable distribution of vaccines. And finally, in the first week of a new year, violence erupted in the United States capital, as far-right conspiracy theorists and insurrectionists invaded the halls of Congress, fueled by claims of corruption and illegitimacy at the heart of their system of government.

It’s difficult to imagine a year that better vindicates the occasional doubt we see in Rawls’s work that such a stable and just society is possible at all, doubts with which Muldoon evidently sympathizes. At the heart of Muldoon’s book, however, is the claim that Rawlsian political liberalism provides an ineffective solution to social instability. Much of social contract theory from Locke through Rawls aims for institutions and dialogical practices that better harness, manage, or tolerate diversity. On the Rawlsian framework, we are encouraged to engage in political debate by reference to political values that we share (i.e., public reasons), rather than those that are unique to our separate religious and moral doctrines. Certain of our laws and distributional schemes must therefore be publicly justifiable. By justifying our political structure through public reasons only, we demonstrate to our differently minded neighbor that we: (a) respect her as a political equal, and (b) are committed to the socially cooperative arrangement that we share with her. Laudable as this might sound, this is a doomed attempt at achieving stability in an increasingly diverse climate, as something like toleration is (unfortunately) a paltry incentive for most citizens to welcome diversity and stably continue our joint democratic enterprise.

Two of the chief problems with the Rawlsian contract, Muldoon argues, are its substantial reliance on our (i.e., flesh-and-blood citizens’) sameness and our epistemic access to a complete (and final) theory of justice. These assumptions are not only unrealistic, but counterproductive. As other critics of the Rawlsian approach have argued, the extent of our shared political values might be too thin to supply innovative solutions to our deepest sources of conflict. Worse still, diverse citizens don’t even conceptualize the conflict itself in the same manner: they categorize the world differently through their myriad perspectives. Citizens occupying different perspectives have markedly different ontologies, procedures for evaluating situations of social import, and adeptness at solving different kinds of problems effectively. As two further and particularly important consequences, those who occupy different perspectives come with very different preference-orderings and, quite often, different skillsets. While this might make the problem of stability initially seem more intractable than the Rawlsian formulation, Muldoon contends that such radical differences provide the key to its solution.

A social contract model that fails to engage the whole of citizens’ diverse perspectives exacerbates several social harms and misses several social opportunities. Unless we better incentivize citizens to engage with and learn about one another’s deepest interests, there is limited opportunity to develop empathy, bargain with one another, or capitalize on our diverse, specialized skillsets. Under the current approach, we tend to isolate ourselves among like-minded citizens, and encounter only a superficial or caricatured version of those who are different.2 We are more likely to see members of other groups as threats in a zero-sum economy and less likely to realize the mutually advantageous social surplus that diverse and specialized societies produce. If instead we come to see our lives as enriched by others’ perspectives—which requires our encountering and even subsidizing them—we can encourage various communities to engage in small-scale social innovations akin to J. S. Mill’s notion of “experiments in living.” Those innovations or unique ways of living that bear fruit will gain traction outside of the communities that tried them, as Muldoon is optimistic that other perspectives will see widespread advantages that certain of these innovations bring. Muldoon likens this, on a smaller scale, to the comparative advantage that we clearly enjoy in developing complementary professional skillsets in those around us.

To begin to realize such changes among the citizenry, Muldoon argues that we must first eschew the “impartial” epistemological standpoint he calls the View from Nowhere, which he associates with the likes of Rawls and Thomas Nagel, and the static political solutions that it produces. In its place, he encourages our adoption of his View from Everywhere, which (in the mold of Thomas Hobbes and David Gauthier) makes political decisions based on an aggregation of citizens’ various perspectives and permits dynamic solutions suitable to changing circumstances. While the panelists’ contributions and Muldoon’s replies will highlight some principles of justice and institutional changes that emerge out of the View from Everywhere, I will briefly elaborate on the bargaining process that stands at the center of it.

Bracketing perspectival differences, as one would do in the View from Nowhere, leads to a uniform scheme of rights, liberties, and public goods distributions. As much as this sounds precisely like something we want in the abstract, such a scheme is frequently unresponsive to citizens’ varied interests, as those occupying many perspectives are eager to concede certain of their rights or public goods in one area in order to enjoy a more extensive right or public good distribution in another. On the current approach, especially in the case of minority perspectives, the dominant vocabulary is sufficiently restrictive that certain citizens are unable to adequately articulate their dissatisfaction in public fora. For example, an adherent to traditional Lakota beliefs will place significantly greater weight on temporary, exclusive access to sacred land, such as a ritual space atop of Devil’s Tower in Wyoming, than the right to evangelize in public places. She is also willing to make significant concessions regarding some other public good so that recreational climbing groups will halt their climbing activities near Devil’s Tower during important rituals. But since current constitutional jurisprudence cannot cognize such a bargaining process, nor incentivize the climbers to engage in one, the Lakota are stuck with a Free Exercise jurisprudence that generally favors the right to doxastic expression (e.g., proselytizing) over the right to performative rituals on sacred land. If indeed such scenarios are sufficiently common, we can recognize why Muldoon sees a perspective-blind model as one that inevitably ends in disappointment and enmity.

Muldoon’s bargaining model, by contrast, would allow for such perspective-specific innovations to our rights, liberties, and public good distributions. While this iteration might not work on a global scale, he cites the Belgian model—wherein rights for various cultural groups are managed locally to tailor them to local interests—as a version of this that has enjoyed some success. Muldoon is optimistic about bargaining because it does not require its participants to speak one another’s language, yet it incentivizes us to understand what the other wants. Since a proper bargain is more effective in optimizing each party’s interests than an a priori distribution scheme is, a bargaining model also provides a more reliably stable social glue than alternative approaches. At an institutional level, then, Muldoon imagines institutionalized mechanisms whereby various groups can: (1) State their interests and willingness to pay for the opportunity to produce a unique good or other social innovation; (2) Enjoy small-scale state funding for that new activity (a kind of experiment in living) where an adequate coalition supports it; and (3) Prove (or fail to prove) the merit of their ideas (or activities) such that they enjoy wider public support. Importantly, the public need not see themselves as endorsing each such experiment, as its sanctioning is not fixed, but they will certainly end up recognizing new solutions, interests, and self-conceptions that they otherwise wouldn’t have.

Despite this optimism about this bargaining-and-experimentation process, Muldoon recognizes the need for two important constraints on such a bargaining-and-experimentation process. First, each party must be made better off by participating in the social bargain—where “better off” is cognizable from each perspective’s point of view—than they would be on an alternate model. Second, distributions should be proportionate to a group’s social contribution, such that each group benefits from the fruits of their efforts. With these caveats in mind, these conceptual and institutional changes will demonstrate the positive-sum game that more diverse societies provide and encourage citizens to subsidize social experimentation among their compatriots. In the end, as Muldoon succinctly puts it at the end of his book, what is offered is a model of the social contract centered around experimentation, discovery, and dynamism. While it is possible that some societies can’t be made to see the positive-sum game of greater diversity, and will need to separate, Muldoon demonstrates why he sees promise in his bargaining model for keeping us together.

This symposium is fortunate to have a panel with varying points of agreement with Muldoon’s project and critiques of it from a variety of philosophical traditions. True to the message of his project, the reader will notice that Muldoon’s replies to each panelist evidence his belief in the great enrichment that each of their perspectives provides. For my part, I would like to express my deep gratitude to Muldoon and each of the panelists for their deeply reflective contributions and for their extraordinary patience during the lengthy period that this symposium, for various reasons, hit roadblocks, delays, and hiatuses of all sorts. The fortunate byproduct of this lengthy development, however, is living proof that both Muldoon’s and each contributor’s concerns were remarkably prescient: both need for such a reimagining of the social contract, and the rationale behind the contributors’ varied criticisms, can be located in one or more of the recent conflicts mentioned at the beginning of this introduction.

Arnold Farr applauds Muldoon’s project as an attempt to meet the diverse world as we find it, but worries that Muldoon encourages a limited, whitewashed form of diversity. Farr compellingly argues that Muldoon fails to address the circumstances that lead to the formation of various perspectives, and their interests, in the first place. First, his bargaining mechanism does not distinguish between those demands that are simply in pursuit of that perspective’s interests, and those that are needed as a corrective to past oppression. Second, some groups are simply the product of past oppression (and perhaps those who are most marginalized can claim no group at all). Demands in each category ought not be treated the same, as correcting past injustice is a necessary precondition to there being an equal bargaining table. He leaves room, of course, for the possibility that such distinctions could be built into the contours of the kind of process that Muldoon intends. Moreover, Farr worries that the capitalist mechanisms at the heart of Muldoon’s project will fail to produce the sort of empathy that he suspects it will, as there is much evidence from the history of capital-driven behaviors to suggest the contrary.

Fred D’Agostino shares Farr’s skepticism regarding the rosy consequences that Muldoon predicts for his bargaining process, in large part because of the antipathy various would-be bargainers feel, from within their perspectives, toward one another. The great challenge of Muldoon’s bargaining model is to create conditions that could foster attitudes among the myriad perspectives that are sufficiently conciliatory to want to bargain. Though D’Agostino agrees with Muldoon that everyone would be better off were they to embrace diversity, perhaps including the bargaining and experimentation mechanisms Muldoon proposes, the notion of “better off” that is at work throughout the book is perspective-dependent, not perspectiveless nor universal. Thus, what would be needed for certain of Muldoon’s central mechanisms to demonstrate the positive-sum game of diverse economy—and D’Agostino directs the reader to p. 91 especially for the book’s methodological turn—is for citizens to first be willing to set aside those elements of their perspectives that give rise to such antipathies. Despite his skepticism about the ultimate success of Muldoon’s central argument, however, D’Agostino is emphatic that Muldoon’s project is in several senses on the right track.

Cynthia Stark provides a compelling articulation of the dilemma all social contract theorists face. As one option, they may constrain the inquiry to what principles citizens would hypothetically accept if only they met this or that moral qualification, as Rawls (and more recently, Jonathan Quong) by only including “reasonable” and “rational” persons in what Marilyn Friedman has aptly dubbed the “legitimation pool.”3 This buys moral normativity for the resulting principles at the cost of its excluding the interests of numerous flesh-and-blood citizens. Or alternatively, they may base the contract on the actual agreement of real and diverse persons, which buys actual acceptance at the cost of moral justification. Since Muldoon opts for the latter, Stark contends, the bargainers in his model have prudential reasons (i.e., realizing their maximal advantage) to agree to the terms, but not moral justification.

Gerald Gaus describes himself as a “fellow-traveler” with Muldoon because, among other reasons, his later work similarly aims at a “new diversity theory” and he concurs with Muldoon that we find agreeable terms of social cooperation through an evolutionary, bottom-up process (which Gaus develops most comprehensively in his Order of Public Reason), rather than through ideal theory. Nonetheless, Gaus offers several criticisms that he takes to be “intramural.” First, Gaus worries that one of the claims to which Muldoon is committed—which he deems the Master Epistemic Principle—is a controversial principle that is not widely shared and likely favorable to the intelligentsia (which is very much the opposite of what Muldoon intends). Second, Gaus questions the bright line that Muldoon draws between discovery and justification-based social contract models. Certain of the metaphors for discovery that Muldoon employs depend on the explorers agreeing (to some extent) on the landscape to be explored, and what counts as a discovery, which is precisely what is lacking in society comprised of radically diverse perspectives. The necessity of such a common landscape, the precise terms of which were to be settled by “you and me” trying to locate our shared political values, is precisely what Rawls was after. Thus, Gaus contends that Muldoon somewhat misconstrues Rawls on this point and, for this reason, misses an important sense in which justification must be primary. Ultimately, however, Gaus believes that Muldoon’s core point is the right one: the task of political philosophy is not for a moral philosopher to lay out the right blueprint for just terms of social cooperation, but perhaps to conceive a process through which we might discover those terms we can all live with.

  1. John Rawls, Political Liberalism, expanded ed. (New York: Columbia University Press, 2005), 4.

  2. For recent discussions of such phenomena, see Robert Talisse, Overdoing Democracy (Oxford University Press, 2019) and C. Thi Nguyen, “Echo Chambers and Epistemic Bubbles,” Episteme 17.2 (2020) 141–60.

  3. See Marilyn Friedman, “John Rawls and the Political Coercion of Unreasonable People,” in The Idea of a Political Liberalism, ed. Victoria Davion and Clark Wolf (Lanham, MA: Rowman & Littlefield, 2000), 16–33, at 16.

Arnold Farr


A New Social Contract Theory with the Same Old Problems for a New World


I. Diversity as a Challenge to Social Contract Theory

I must say that I read Professor Ryan Muldoon’s book Social Contract Theory for a Diverse World: Beyond Tolerance with mixed thoughts. First, I greatly appreciate Professor Muldoon’s attempt to construct a version of social contract theory that attempts to meet the challenge of a diverse world. Second, I find myself disappointed that Professor Muldoon was not able to escape certain social and theoretical trappings that caused problems for classical social theory and the Rawlsian version as well. In this section I will focus on the need for a work like Muldoon’s. In the following sections I will explain my reservations.

Muldoon correctly observes that economic and political globalization have changed the composition of various societies. He writes:

Economic and political globalization has encouraged waves of immigration that have changed the social composition of countries, leaving once homogenous cultures to deal with their newfound diversity. Unfortunately, Western countries have not handled this ongoing transition with much success. As Western Europe has seen immigration rise, it has seen support for liberal policies decline, and civil unrest decrease. (1)

He goes on to say of the United States: “Even with the United States’ longer history of ethnic diversity, immigration and affirmative action polarize political discourse” (1). Although I find this last statement problematic my critique of it will have to wait until later. My point here is to just recognize the importance of Muldoon’s attempt to rethink social contract theory in light of the rapid restructuring of Western societies due to globalization and increased immigration.

The changing composition of a society does bring with it certain difficulties that demand a rethinking of the assumptions on which that society is based. These difficulties can lead to forms of social unrest that destabilize that society. Muldoon’s new version of social contract is an attempt to help us avoid such destabilization by offering us resources with which manage these difficulties. Muldoon says of his new model:

In particular, it establishes three key ideas that differ from traditional social contract theory. First is the idea of perspective in political reasoning. Rather than assume that all citizens view moral and political problems the same way, it articulates a way to productively aggregate disparate views to find potential areas of agreement. The second key idea is a bargaining model that requires virtually no agreement amongst participants. And finally it presents a model of how agents can be motivated to seek out more diverse societies, even in the face of distributive disputes. (3)

Since we’ve all read the book I will not go into any detailed analysis of these three key ideas. In my critique I will challenge the last two. However, that occurs within a critique of the entire project. At the end of the day I am not convinced by Professor Muldoon’s new model. His new model might provide help later down the road after we address some more fundamental issues. In other words, Professor Muldoon’s new model is premature.

II. More Ideal Theory

I claim that Professor Muldoon’s model of social contract theory is premature because the model is constructed as if the people involved are on a level playing field. Muldoon is aware of social and economic inequalities and the possibility of inequality increasing. He even cites Charles Mills and Carol Pateman in a few places. This suggests that he is quite aware of the relationship between social contract theory and domination. This is what is surprising. Muldoon goes on to construct what Charles Mills calls ideal theory without adequately addressing the real causes of inequality and social domination. I suspect that Muldoon believes that he has not fallen into the trap of constructing ideal theory due to his attention to the real problems of globalization and immigration. Nevertheless, the theory remains ideal.

In their book Contract & Domination, Carol Pateman and Charles Mills argue that the history of race and gender subordination requires that we rethink the way we do political theory (79). Mills argues that most political philosophy is ideal theory, meaning that the philosophical quest for principles of justice on which a well-ordered and fair society is to be founded begins from ground zero. The political philosopher (social contract theory included) ignores the real injustice and systems of domination already present in society. Mills argues that a political philosophy that is merely ideal theory is not helpful in addressing real concrete social and political problems. Mills points out that while Rawls concerns himself with the basic structure of society, he constructs his theory of justice by ignoring the basic structure of society. That is, Rawls seeks to discover principles of justice while paying no attention to real structures of injustice.

While Muldoon does address the very real problem of increasing diversity, like Rawls, he fails to address structures of injustice. One example is when Muldoon writes, “Even with the United States’ longer history of ethnic diversity, immigration and affirmative action polarize political discourse” (9, the racial contract). There are two things that I want to challenge here. First, Muldoon is guilty as are so many others of disconnecting the concept of diversity from real social practices of injustice and exclusion. Secondly, Muldoon’s observation that in the United States immigration and affirmative action polarize political discourse ignores the fact that the polarization in the United States is prior to political discourse. This initial polarization has its roots in the real contract of exclusion and domination in the United States. With regards to the first problem, my worry is that the concept of diversity is often presented in a very weak and whitewashed form. It simply signifies the importance of accepting those who are different from those persons who are considered to be in the mainstream in a given society. The emphasis is on celebrating difference. However, the concept of diversity much like that of tolerance arose out of a struggle of oppressed people to be accepted and treated equally in the society that has oppressed them. Therefore, it is not simply a matter of difference. We must talk about socially produced difference that is the direct result of oppression and domination. We must talk about real injustice and its rectification, we have to talk about how group identity gets constituted within the context of oppression and exclusion, we have to have a discussion about the ways in which group identity and difference are a response to forms of systematic dehumanization, etc. So, when we talk about diversity in the United States, is it a form of natural diversity that is created by people coming together from different parts of the world, or is it a form of diversity that is created by one group responding to domination by another? If we fail to adequately address issues like these, are we a place where Muldoon’s version of social contract can help us or even have any meaning?

The nature of ideal political theory is to construct a form of social contract theory that ignores the fact that the society that is being addressed has its origins in a contract of domination. The contract of domination has long-term consequences that shape the development of our institutions, values, identities, and our social and personal narratives. The United States is a prime example of a society that has its origin in a contract of domination. The founding fathers sought to protect the desires and advancement of white male property owners. The soil of American society is saturated with the blood of Native Americans and Africans. There is a huge body of literature on how European immigrants such as Irish and Jewish people adopted a white identity so that they could reap the benefits of whiteness in a racist society. One of the problems with recent diversity initiatives is that they fail to recognize the way in which some of our differences were forged under the contract of domination. For example: much of black culture in the United States is a response to US racism. Further, racism still impacts black life in the United States. Therefore, any social contract theory must begin by dealing with the long-term effects of the contract of domination.

III. The Contract of Domination and Perspective Formation

In chapter 3, Muldoon’s distinction between the view from nowhere, the view from somewhere, and the view from everywhere has the potential to be helpful. However, at the end of the day the avoidance of non-ideal theory (which actually considers the way in which our society has actually been structured by a contract of domination) derails Muldoon’s project. Muldoon’s intuition with regards to the false objectivity of the view from nowhere, as well as the narrow focus on the perspective of the individual in the view from somewhere, is correct. Both of these views are problematic for a number of reasons. Muldoon attempts to save the good intent of both views with his concept of the view from everywhere. In his own words:

The view from everywhere is an attempt to reconcile the positive features of both the view from nowhere and the view from somewhere. In particular, it aims to achieve the neutrality of the view from everywhere while maintaining the epistemic feasibility of the view from somewhere. To do this requires a significant departure from both views: most notably in that the moral-political stance is not something to be taken by a lone individual. Instead, it is fundamentally an aggregative social viewpoint. In addition, the view aims to be a project of discovery rather than justification. Rather than build in and justify substantive moral claims, the view is designed to allow these claims to arise from collective deliberation. (45)

What I appreciate here is the attempt to move beyond the moral-political stance of the lone individual toward something more social or communal. This is a kind of democratic coming together. Muldoon goes on to explain that the view from everywhere has three basic elements. First there is the collection of everyone’s beliefs and desires. Second, there is the collection of everyone’s perspective. Third, there is a method for taking these beliefs, desires, and perspectives as input, and output impartial statements about the evidentiary status of particular moral perspectives (46).

While I appreciate Muldoon’s efforts here, the missing piece is the non-ideal social context and the social process by which perspectives and beliefs are formed. The sharing of beliefs and perspectives must include a critique of the process wherein beliefs and perspectives are formed. That is, if our society has already been shaped by a contract of domination are there not social mechanisms in place that support such a contract and also shape beliefs and perspectives so that they conform to or reproduce the contract of domination? To be fair to Muldoon, maybe he anticipates the emergence of critiques of the contract of domination during the sharing of perspectives.

IV. Beyond Tolerance within the Framework of Capitalism?

My final concern regarding Professor Muldoon’s book is his attempt to replace the deliberative approach to the social contract with the bargaining approach. His concerns about the deliberative are legitimate within the framework of social contract theory in its ideal form. His worry is similar to that of feminists and philosophers of race. That is, the centrality of public reason in Rawls’s version or the General Will in Rousseau’s version seems to demand a likeness among us that puts difference under erasure. As we enter public space for deliberation our private interests must be put out of play. Muldoon is right to rescue private interests here and to see them as important elements in the process of bargaining. He also believes that bargaining encourages empathy. However, my worry here is that he presupposes a form of rationality that is merely ideal. At the end of the day, Muldoon’s theory is still based on the problematic belief in atomistic individualism. Individuals (even as members of social groups) come together with their interests and perspectives intact and bargain. The problem is that many of these individuals come together as victims of a prior contract, the contract of domination.

For example: Muldoon supports his bargaining approach by appealing to an economic model of bargaining. He even uses unions as examples of successful bargaining within capitalism. First, in a capitalist system, unions may bargain for various improvements in the workplace but they have not been able to successfully challenge day-to-day exploitation. Further, unions are constantly under attack in the United States and are quickly fading. Secondly, economic groups are not natural groups but are socially produced groups. Therefore, we must ask, “What form of social contract produced these social groups?” Given the tendency of capitalism to produce wealth and power for an elite and very small segment of our society I hardly think that it provides us with the tools for constructing a social contract that is fair and just. Further, there is nothing in Muldoon’s theory to convince me that the capitalist will somehow consider the interests of others.

Finally, we must come back to the issue of diversity. Muldoon seems to be under the impression that the capitalist is more inclined to champion diversity when he sees that it can benefit business. The variety of perspectives leads to more creativity. This is the conclusion of ideal theory. In the real world where we actually live we see capitalists effectively using diversity to maximize their wealth by creating conflict between various social groups. This is not accidental. Groups are put against each other then manipulated. I can give examples of this but I’ve gone on too long. I look forward to discussing these matters with Professor Muldoon and others.


  • Ryan Muldoon

    Ryan Muldoon


    Response to Arnold Farr

    I am delighted that Farr has brought up concerns around inequality, domination, and challenges around what we might call transitional justice. Pateman and Mills are important touchstones for me in how I conceive of social contract theory, and so I am glad to see their arguments brought out in service of a critique of my approach.

    There are a few different points of contention that I’d like to try and address here. The first is perhaps the crux of the matter: whether there is a role for ideal theorizing when we live in a world marked by massive injustices, and as a related matter, how we should understand what ideal theory is. Second, I’d like to consider Farr’s concerns around “natural” compared to “constructed” groups and perspectives. Lastly, I’d like to think a bit about bargaining, and some of the assumptions that might come with it that Farr is worried about.

    Let’s start by considering ideal theory, and the role of political philosophy. Ideal theory as I understand it is a system in which people are motivated by, and indeed act in compliance with, the dictates of justice. Ideal theories are also typically idealized in the sense of imagining rational actors, reducing sources of disagreement, and assuming some kind of equilibrium state. Philosophers often defend ideal theory in its fullest form by saying that we should identify what kind of society we want, or what kind of society would embody justice, before we deal with the world as it is. G. A. Cohen is a good example of this, where he valued developing a conception of justice even if it were something that might not be realizable. Rawls, as I understand him, wanted to make sure that his theory could be for people like us, so he was worried about things like whether a Well-Ordered Society was stable. Like Gaus, I have concerns that we are even able to identify what an ideal society would be like, and if we were to identify a purported ideal arrangement, we very well may be quite wrong about what society would be like if we tried to establish those ideals. Hence my theory’s focus on discovery and change, rather than identifying some ideal end state. While I don’t think we can identify what the best society would look like (if there is such a thing), I do think we can make strides on thinking about a more just way of determining what an improvement would be.

    A focus for me in developing this account was to turn down the dial on normative idealizing as much as is possible while still articulating a normative theory that offers novel procedures. So there is some compliance required, there is an assumption that if you agree to be in a political community with others you need to do so on equal terms (though I provide a way to split the community apart), and there are some rationality assumptions, but those only go so far as to say that you’ll not choose to make yourself worse off by your own lights. I don’t imagine ideal agents in a hypothetical contract, I don’t imagine a stable population isolated from other political communities, and I don’t assume that the starting conditions are just. I take those as they are, and ask where we can go from here.

    Another project—one that can be and has been fruitfully taken up with others, would focus on an analysis of existing injustices and the structural features and individual actions that brought them about. That’s an important task, but one that I think is separable from this one. That is for two reasons. First, much of the design of the contract theory I offer is meant to facilitate getting us out of unjust states. This is why I spend time articulating how this approach can help engender fellow-feeling and a sense of equal worth of persons rather than just assume it upfront, and why I argue for an account of equality that doesn’t depend on some account of “sameness.” The theory is meant to offer procedures for building up the preconditions for fairer social arrangements. I start from a position where I assume people have a variety of prejudices, and try and show why we can still build something better, and give those people a reason to come along. The second reason is that on my account, existing structures might be the things that we know the most about, but they do not have special priority for staying fixed, unless a broad spectrum of perspectives agrees that they should. In a bargain over rights, everything except the things we have consensus over are up for grabs. So this approach allows for radical change from the status quo. All this is not to say that the examination of particular institutions and their relation to some account of justice isn’t valuable—of course it is—but this is just a different task than what I am up to here. I’m interested in the question of how we can try and make continual improvements, where everyone has an independent reason to endorse them.

    All that said, I think there is a great deal of room for imagining a synthesis here. Using the more abstract procedures discussed in the book, I think it would be a wonderful project to try and adapt those into concrete policy tools for particular problems. For instance, what would a revised approach to (local) housing policy look like if we had a bargain amongst stakeholders on equal terms? If we were procedurally bound to do far more than we do to distribute benefits and burdens more fairly, there are real advances that could be made such that everyone would benefit, while reducing inequality. Raj Chetty’s empirical work on economic mobility points in this direction—economic and racial segregation harms the wealthy and whites as well as the poor and minorities, even if the burdens are substantially larger on the poor and minorities. As I discuss in chapter 5, though, people are attuned to far more than just material considerations, and so there are deep challenges to getting racists on board with policies that support integration.1 So I think there could be a great deal of productive work done trying to work out a concrete instantiation of the iterative procedure I lay out in various policy domains, working to be sensitive to what got us to the status quo.

    I would love to hear more about Farr’s concerns that my account of diversity is insufficient. It is worth, then, saying a bit more about how I conceive of diversity. I focus my attention on perspectival diversity rather than a particular social group affiliation. I do this for a few reasons. First and foremost, as Farr rightly notes, one is hard-pressed to find “natural” social groupings. Most, if not all, social groups and social identities are socially created, and these categories change through time and place. It’s not useful to pick some subset of these categories or identities and reify them. So, I conceive of these (for my purposes) as one way in which a perspective might be formed. Other mechanisms for perspective generation are present as well—one’s education, life experiences, ideological tilt, and other factors can likewise encourage someone to start categorizing the world in particular ways and not others. These perspectives impose an ontology on the world, and it is useful to note that there is no “neutral” or “natural” ontology, our descriptions of the world are going to be incomplete, and potentially in conflict with one another. I’m more interested in the fact that these perspectives exist and inform our understanding of the world, such that they enable us to easily see some things and hinder us from seeing some other things than having a comprehensive account of how particular perspectives came into being. This is just because there is so much exciting work to do in trying to understand political conflict with just this much added into the picture, beyond different preference profiles or resources. I think this helps unlock how we can understand a more diverse community as being a benefit to all, and not merely a problem to overcome. The perspectives framework demonstrates why there can’t be a privileged position from which to reason. It helps show how there can be productive gains to be had from political conflict, and indeed I think it helps elucidate how “tolerance arose out of a struggle of oppressed people to be accepted and treated equally in the society that has oppressed them,” as Farr correctly notes. I think Farr and I are in greater agreement than he realizes. The point of remaking social contract theory around an iterated procedure of discovery instead of some particular end goal is precisely because I think we should understand liberalism as the result of a process of struggle and conflict amongst competing interests and values. Those values remain imperfectly realized at best. I think political philosophy would benefit from putting more of that process of struggle into the philosophical accounts themselves. Liberalism wasn’t developed in a calm cool hour of reflection, it was born of a series of conflicts and compromises, and there should be clearer accounting of that in our philosophical engagement with these ideas. Even if we suppose we can perfect some of those ideas using our reflection, doing so builds in the supposition that we are “done.” These struggles are ongoing, and may uncover other important normative ideals yet to even be partially realized.

    Lastly, Farr is discomfited by the role of bargaining in my approach, and some of the individualistic and market-based ideas that might come along with that. To speak to that, let me say a bit about why I favor a bargaining approach to describe a process of coming to an agreement rather than a public reason sort of approach. There are three main reasons. First, insofar as public reasoners are meant to share public reasons and shun private ones, it means that it is very hard for minorities of various kinds to participate on equal terms. Their reasons won’t count as public. So they will suffer compounded disadvantages. Not only will fewer people be interested in what they want, they will have a harder time coming up with a means of expressing their interests on terms that others will even admit into discussion. Second, public reason typically requires that we not only come to some consensus about what we are going to do, but we also must agree on the reasons that we have chosen it. This is an unbelievably high bar, and incredibly unlikely in a diverse community. It is far better, in my view, to allow people to agree on what to do, but disagree about why they should do it. This allows more paths to success, and takes more views seriously. Finally, public reason, or just public deliberation, imagines that we are mostly won over by argument. I think this is just false. If you look at, for example, the slow and then very rapid rate of improvement in public acceptance of the LGBTQ community in the United States, it’s not due to a new, more powerful argument emerging. It’s just because more people found out that they knew a gay person, and realized that they were just normal people who wanted to live their lives. Changes in attitudes around, for instance, police violence against black people happened less because of new arguments, and more because of widely available video evidence. Deliberation makes a difference at some margin, but I think a better way of representing what we’re doing is that we’re trying to work out what rights we all want and what burdens those impose on others. Bargains are a very good model of just that.

    Farr points out that in bargains in which participants have quite unequal bargaining power, that imbalance is represented in the outcomes. True. But that does not mean all bargains need to be structured in this way, and the bargaining model I presented supposes that everyone has equal veto power in the bargain, which would drive the participants toward a fairer deal. With an appropriate set of rules, which I believe I have offered, selfish jerks still can be party to a fair deal. Likewise, I don’t think that bargaining models are unique in having situations that generate unequal outcomes. More charismatic (or even just louder and more obnoxious) people can shift real deliberations more than introverts. There are whole hosts of biases that cause some to favor or disfavor particular speakers because of their social identity rather than the content of their thought. Bargains at least have the virtue of allowing us to focus in on different private interests and how to balance those in such a way that everyone takes themselves to be made better off by agreeing.

    Another concern of Farr’s here is that this supposes an atomistic individualism. I suppose here that people are bargaining, seeing the world through particular perspectives, and trying to maximize their utility as they understand it. So yes, I treat individuals individually, but they can share common interests, they can share (or not) a particular perspective, and they can care about other people: their interests don’t have to be selfish. This is an area of some disagreement across social sciences. I am not in a position to resolve these methodological disputes in a few sentences, but I can say that I chose what I take to be the most flexible modeling option. Either you can try and capture community-level interests via shared perspectives and shared interests across distinct individuals (which would be my preferred approach), or you can imagine agents here as stand-ins for different non-overlapping communities, in which case groups are now the unit of analysis.

    While my theory doesn’t start with a careful accounting of the various sorts of injustices in the past and in the present that have shaped our current society, it does offer an account of how to go from an unjust state to one that we believe will be more just. It is a theory of social contract revision, and is meant to operate indefinitely, so as we learn more about how we ought to live with each other, we can try it out and see how well we do.

    1. I discuss the challenge of segregation and political polarization in much more depth in “Diversity isn’t what divides us. Division is what divides us.”

Fred D'Agnostino


What Counts as a Gain to Trade?

It is forthright and courageous, intellectually, that Muldoon’s book begins with a frank acknowledgement that the current situation is very difficult one for the kind of political theorizing that he wants to engage in. He wants to take people as he finds them, but what he finds are, alas, attitudes, widespread and entrenched, that create difficulties for liberal-democratic aspirations and institutions. As he puts it, “Around the world, right-wing, anti-immigration, anti-minority parties and political candidates have seen robust political support” (1). (Note that this was written before the Brexit referendum and before the election of Mr. Donald J. Trump as president of the United States.) As Muldoon acknowledges, “minority groups are often depicted as subhuman, demonic, craven, or somehow alien” (90). Indeed they are, and not just so-called minority groups, but many of those, of all sorts and conditions, who might be on opposite sides of highly polarized issues—e.g., (cartoon) red state and blue state voters in the United States.

Certainly, it is no longer merely that the members of such cohorts have different ways of looking at the social world, different values, different principles. It is, rather, that at least some members of each of these two cohorts think of many members of the other cohort that they are not on an equal moral footing, are not worthy of moral regard and, indeed, are not the sorts of beings that you can or should bargain with about fundamental rights or the distribution of the costs and benefits of social cooperation. (This point could also be developed, in the currency of preferences, in terms of the idea, mentioned but not exploited by Muldoon [cf. 44 and 59n7], of tuistic, or other-regarding preferences: blue staters [strongly] prefer that (various of) the preferences of red staters not be satisfied, and vice versa.)

Here the point is not just difference, but, rather, difference of a kind that divides, that excludes, that precludes engagement for the purposes of bargaining toward a system of social cooperation for mutual benefit. People so divided lack cogent reasons to engage in various forms of social cooperation that might deliver mutual benefit because even were each to benefit in some way from cooperation, the very fact that their antagonists would also benefit would be repugnant enough to undermine the welfare calculus that exhibits the so-called net benefit, that dictates cooperation on the basis of that benefit, and that therefore justifies the bargain. The whole arrangement unravels itself in the face of this kind of, increasingly common, intransigent Other-disdaining perspective. As Muldoon puts it:

In this imagery [of Ronald Reagan’s “Cadillac-driving welfare queens”], the Other has a series of negative qualities for which they are morally blameworthy. This can not only serve to remove our moral obligation to ensure that they have adequate resources to conduct a life that they find valuable, but it can serve to instigate a desire to view resource allocation as a zero-sum game in which whatever minorities receive is loss to the rest of us. We then find ourselves with the view that we must protect ourselves from this parasitic group, and work to deny them any resources that we might otherwise have been able to allocate to ourselves. (89–90, FD’s emphasis)

All this, it seems to me, is: (a) empirically plausible as a description of much of the current political situation in many first-world democracies; (b) relevant to Muldoon’s proposed approach to political theorizing; and, unfortunately, (c) a severe challenge to Muldoon’s ambition to identify a plausible method, as he puts it, “to lead people there from here” (95).

Notice, furthermore, Muldoon has denied himself an easy “get out,” namely, that this is no longer a philosophical, but, rather a therapeutic problem. For he has set himself the task of delivering a method that “allows for political agreement amongst individuals who hold any arbitrary set of perspectives” (65, FD’s emphasis), including, presumably, perspectives that are utterly antipathetic to one another in the sense illustrated. He therefore cannot address, let along solve, the problem posed by this sort of diversity simply, as many theorists would do, by normalizing it away, at the philosopher’s desktop, on the grounds of its wrong-headedness or moral bankruptcy.

This makes Muldoon’s task a heroic one. That said, Muldoon, in effect, starts over when he reaches the point I’ve just described, seeking to identify a mechanism that would, not on the desktop but in “the real world,” effect a partial normalization of citizens so that, despite their (perhaps continuing) substantive diversity of perspectives (or preferences) they could come to see each other as worthy bargaining partners. Such a mechanism would, it’s posited, facilitate the gradual rehabilitation of relations between antipathetic groups, leading, in due course, to growing tolerance, empathy, and mutuality.

We begin to encounter this part of the larger argument on p. 91, the most important page in the book in my view, where Muldoon takes a sharp methodological turn and increasingly relies on an argumentative manoeuvre of whose bona fides I am nevertheless not entirely convinced.

First, the methodological turn. Muldoon has acknowledged, in effect, that there is no rational basis for successful bargaining between individuals who view each other as anathema and that there is no way available to him of blocking the inclusion, in a bargaining scenario, of perspectives that deliver those anathematizing judgments. As a consequence, the theorist’s job is to identify a social mechanism (as I’ll call it) that can create the conditions in which bargaining can occur . . . create those conditions not on the desktop by imagining away antipathy, but, rather, by inducing, in the society, a social process in which people’s attitudes are altered at least to the extent necessary for bargaining to seem sensible to them.

As Muldoon puts it, “It becomes necessary to design the basic structure of society in a way that aims to foster the attitude that all should be treated as moral equals, rather than design institutions that already assume that such an attitude exists” (91, FD’s emphasis). A “basic structure” designed to this specification is, then, the mechanism of partial normalization that I mentioned. The normalization is partial, of course, because a lot of diversity in first-order values or preferences will survive the moderation of antipathy (or of strong negative tuistic preferences). And the normalization is not accomplished, on this model, by the stroke of the theorist’s pen (all the virtues of theft over honest toil!), but, rather, through concrete social processes. To adapt a trope familiar in Brisbane, Australia, where I write this, it is “normalization for the real world.”

In particular, we aim on this account to show Ego and Alter that, despite their mutual antipathy, there is something at stake for both of them in some form of mutual regard. Specifically, and as Muldoon puts it, “what we need . . . is a demonstration that diversity promotes an individual’s own ability to conduct her life as she wants” (91) from which we might, in turn, show how initially antipathetic individuals “rationally learn to become more accepting of others” (95). This manoeuvre, if successful, would transform the zero-sum game of mutual antipathy into a positive-sum game where there was potential net benefit to both Alter and Ego and thus something to be pursued via bargaining, notwithstanding a historical antipathy that is, via this iterative process, gradually left behind.

Indeed, most of chapter 5 of Muldoon’s lovely book is devoted, precisely, to sketching such a demonstration, based on the proposition, as he puts it, that “a more diverse population is better at creating wealth than a homogeneous one” (91)or that “diversity leads to greater economic productivity [because] the economy benefits from the division of labor, and a diverse society offers more opportunities for further division” (98). It is, in short, the diversity that produces the surplus and the surplus that provides the incentive to bargain notwithstanding initial antipathy that is gradually eroded by the increasingly positive “gains to trade” that acceptance of diversity yields on this account.

Notice, of course, that this mechanism, as Muldoon himself fully understands, cannot be guaranteed to succeed. As he puts it, “individuals may well remain stubbornly against diversity” (101). He continues, “They could simply have deep-seated fears or prejudices against other kinds of people.” He concludes, opening up another line of argument, but one, in my view, of the same character and hence likely to peter out in the same way: “Though I cannot provide an argument that each of these people will come to decide that they are wrong and begin to see the intrinsic value of human differences, I can argue that the incentives are aligned against their views.”

We can, I think, see how this is supposed to work. What we don’t yet have, however, is a model of attitudinal change that enables us to feel confident that such a mechanism, if implemented, would work. What is it, in particular, about the repeated demonstration to Ego and Alter that their cooperation would produce divisible net benefits that enable them to “get over” their antipathy, on largely non-material grounds, to one another? And how confident are we that this mechanism, whatever it turns out in detail to be, can in fact be deployed in current circumstances?

Certainly, I understand what Muldoon means when he says, of the Antipathetes, that “the incentives are aligned against their views,” but this assertion is only true if, by “incentives,” we mean purely and narrowly economic incentives. For, if we mean by “incentives” all the considerations that move a person to action or attitude, then, surely, the economistic “gains to trade” associated with a yet-to-be-achieved acceptance of diversity may, as an empirical matter and no matter how many iterations of the demonstration, continue to be outweighed by the “costs” to the participants of recognizing the moral equality of those individuals who are “Othered” in their perspectives. As my title suggests, it matters, for the functioning of Muldoon’s mechanism of partial normalization, what counts as a “gain to trade” and that is something that itself is perspective dependent. From the perspective of a “hater,” material gains to trade simply do not “trump” (haha) the existential losses associated with recognition of “the Other.”

There is another matter, surfaced in the 2016 US presidential election, which cuts across the first point from another angle. One thing we learned from the election (and other events) is that there is, among wide sections of the populace, increasing scepticism about even the narrowly economistic calculus of “gains to trade” that is supposed to show the advantage of a diverse society from the point of view of material well-being (and thus, gradually, to undermine the antipathy between opposed cohorts of citizens). A promissory note was issued some time ago that precisely such gains might “trickle down” from the so-called liberalisation of economic regulatory arrangements and now, increasingly, precisely those groups that are antipathetic to diversity are also increasingly sceptical about delivery against that note, and, accordingly, are no longer in the mood to be moved by gains-to-trade arguments of the kind Muldoon deploys. (Indeed, their scepticism about gains-to-trade arguments may well be a factor in the liberation or development of the very antipathetic attitudes toward Others that Muldoon, and all of us, are struggling to theorize or deal with.)

Muldoon, it would seem, has arrived where we can, a priori, expect someone to arrive who has sought to accommodate all the empirically available perspectives that guide human thinking and action and who nevertheless seeks to identify a system of coordination for mutual benefit. He has arrived at the point, as he forthrightly puts it, where “though there are many incentives for promoting diversity in society, there are also limits on how far these incentives can take us” (109) . . . perhaps more stringent limits than Muldoon himself believes.

To repeat: Muldoon is right that people who reject diversity would be “better off” if they accepted it. The problem, however, is that the notion of “better off” that sits at the center of this argument is one that is itself perspective-dependent, rather than perspective-neutral or perspective-universal. For those perspectives which build in, as side-constraints on other forms of calculation, some sort of antipathy to certain forms of human difference (or that are built around strong, negative tuistic preferences), the “whole equation” simply doesn’t add up. Sure, I might be better off in some thin sense of material comfort, though even that is now doubtful after repeated failures of benefits to “trickle down.” But the promise of such benefit depends, precisely, on my abandoning some of my existential commitments—namely, to refuse recognition as moral equals of those who are, in my perspective, moral pariahs.

Every approach in political philosophy has its limits. To say that Muldoon’s approach doesn’t fully succeed is therefore not a differentiating claim, let alone a dismissive one. We are in Muldoon’s debt, even if his main argument doesn’t (quite) succeed, not least because of his engagement with and development of the idea of perspectives, because of his introduction of the gains-to-trade argument that I have been busy criticizing, and, in my view, most importantly, because of his refusal to do on the desktop by theoretical fiat what has to be done in “the real world” if we are to salvage liberal-democratic orders.

Though this wasn’t, I think, his intention, which is present and future oriented, I think that, whatever its merits as a solution to contemporary problems, Muldoon’s fundamental mechanism can be understood as a rational reconstruction of the recent (say mid-twentieth-century) historical experience of individuals in the great manufacturing and trading societies, and hence can help us understand something about the potential for social change. In those societies, large segments of whole populations did indeed experience improvements in material well-being that were and were widely seen to result from a division of labor that was underpinned by a diversity of individual and cultural types. It does seem once to have worked that initially diverse groups of people came to see one another as partners in the production of net social benefit and hence as legitimate participants in bargaining over the distribution of that benefit. So real-world mechanisms of partial normalization haven’t always been impotent, and Muldoons’s analysis helps us understand their efficacy.

Whether there is a second act, however, is, in the face of our current political realities, a vital question. What, if anything, is it, empirically, that distinguishes the current situation from superficially similar situations in the past? What are the prospects that we can, again, find a way through from polarization to partnership? What more than merely material benefits has to be at stake to bring obdurately antipathetic parties to the bargaining table? I welcome further discussion of these matters.

  • Ryan Muldoon

    Ryan Muldoon


    Response to Fred D’Agostino

    The challenge of responding to D’Agostino is that I more or less agree with him. He very accurately characterizes the problem that I aim to address—deep divisions in diverse polities—and charitably reconstructs the approach that I take to ameliorate the problem. Upon doing so, he accurately points out that a perspectives-based account is going to have to account for the fact that gains to trade are themselves perspective-dependent, and there are going to be limits to what my approach could achieve in highly polarized societies where various groups are dug in against each other. Since I don’t normalize these problems away in my account, I do have to try and deal with them. As he suggests, I think the best I can do is find different spots to chip away at the problem, and hope that these efforts combine together such that the basin of attraction of iterated improvements encouraging more people to work with others on cooperative terms expands outward to cover more cases. As I will discuss a bit, much of my work since the book has been an effort to think through these problems more carefully.

    The problem, as both D’Agostino and I understand it, is that there has been a sharp rise in polarization in the liberal West, or at least it has become a far more visible driver of national politics. This polarization has tracked pretty well with increases in immigration flows, especially from countries that are perceived as more culturally distant. In the United States at least, even though racial attitudes are better than what they were a few decades ago, we are paying a very high price for racialized policies like those that encouraged or even mandated housing and schooling segregation, which has led to any number of other problems, most prominent of which at the moment are policing and hugely disparate health outcomes. As I argued in “Diversity Does Not Divide Us. Division Divides Us,” segregation and polarization work hand in hand to undermine democratic discourse, and perhaps most insidiously, make segregation appear to be a solution to a problem, rather than a very serious problem itself. It is one thing to show diversity is beneficial in a well-functioning market-liberal society with a steady flow of immigrants. It is quite another to show it is beneficial when people view their political opposition as the enemy. As D’Agostino points out, plenty of people would deny themselves a benefit if it means denying something to their opponent. This kind of scorched earth approach to politics pushes us away from reconciliation.

    So, what can be done about this problem? In my book, I offer a couple different avenues for reconciliation. The initial approach is to just better understand the nature of disagreements within a perspectives framework. When people’s preferences clash, there is a limited set of options available. When people disagree and it’s also the case that they understand the world differently than each other, there is at least the possibility that there is room for learning from one another, or finding an option where both parties get what they want. This is not a guarantee of course, but the perspectives framework gives us more degrees of freedom to find a resolution to conflict, and reveals the possibility of everyone walking away content. Of course, perspectives can also reveal the depth of our conflicts. Issues like abortion help reveal that we disagree not only what we should do, but what the world is like, and who counts as a person. The perspectives framework helps to clarify why conflicts might persist.

    Despite different perspectives, we may well share areas of common belief, even if we hold those for disparate reasons. It is helpful to notice where we might agree, and use that to constrain our bargains a bit. This can allow us to more easily find at least some minimal set of rules that can allow us to cooperate, or at least avoid conflict. The bargaining approach I have outlined doesn’t specify how big the bargain has to be, so a possible outcome could be a fairly limited set of social rules, and perhaps an emphasis on being left alone. In “Exploring Tradeoffs in Accommodating Moral Diversity,” I explore this idea a bit, and show that once a lot of perspectives are at the table, constructing these more minimal social rules is messier than people might expect, but it is possible.

    As D’Agostino notes, once we have a minimal set of social rules arrived at via a bargain, I turn my attention to a basic structure that fosters a recognition of others as equal citizens. I don’t quite see this as a break with what came before it, in the way that D’Agostino does, but rather the last step in a three-step sequence. This look at the basic structure comes at a stage when we have already established some areas of moral agreement, and we have completed a bargain over the rules, and now we are going to try and find ways of living with each other on those terms. This last stage is meant to be about constrained experimentation, so that we can learn about how this arrangement works, and can take what we learn back into the first step of the social contract process. So, this stage is where the real work happens. It’s one thing to come to an agreement, it’s quite another to live with it, and learn about the inevitable unexpected consequences. This is why I think I can’t just assume people will be on their best behavior and act from a sense of justice. They are trying to live their lives in a manner that comports to their sense of a good life, while constrained by the social rules. So, rather than assuming success, I try and think about what would make success more likely.

    On this set of assumptions—we are at a stage where the bargain has been made, the rules are roughly set, and we have to actually start cooperating—we really do need to take people as they are. As I argue in the book, I am unconvinced that we can rely on abstract arguments about justice to convince people to treat others well. And so we need to think about how to set up institutions such that one does better by their own lights when they treat others fairly and with respect. My goal—or perhaps hope—is to demonstrate that repeated interactions in nonpolitical contexts, such as market exchange, will help to show that while diversity can generate conflicts about how to divide the pie, it undeniably makes the pie bigger. Rather than needing to make explicit arguments, the idea is that people will come to notice that more diverse places do better on the dimensions that they care about. There will be greater cultural production, more innovation of various kinds, more wealth, and more ideas. This gives a lot of people more reason to favor diverse environments. For those that are unconvinced, it’s more likely that their children will be more thoroughly exposed to these ideas and take them on as their own, not unlike patterns of assimilation with immigrants and their children.

    D’Agostino is right to point out that this is not going to compel everyone, and likely not some of their children either. There is an empirical question of numbers here, but clearly the number is well above zero. I try and wrestle with this idea in “The Paradox of Diversity.” A diverse society is better off in part because of conflicts that force people out of their comfort zones, treat ideas with more skepticism, and try harder to outdo others. That’s all well and good for people whose dispositions favor this kind of dynamism, but for people with conservative dispositions, or those who value some kind of cultural purity, this is going to be a deeply unsettling way to live. They can try and live within the public rules but otherwise withdraw to like-minded communities, but as I argue in “Local Diversity and Polycentric Democracy,” this is not easy to do, as it is hard to avoid serious spillover effects with other communities.

    Within the confines of the project that I’ve undertaken, I don’t think there is an easy way to get out of these problems. I am convinced that there is no silver bullet. Instead, the best sort of reply is going to be multiple, overlapping attempts at demonstrating the value of diverse others, on people’s own terms. This is the slow boring of hard boards, as Weber would say. If people are unconvinced by greater productivity and a finer division of labor, perhaps they like more interesting food. If they don’t like different food, maybe they like variety in music and culture. If they don’t want that, then maybe they value the opportunity to try and convert people to their views. If they don’t want that, then perhaps they like the ability for their community to hold a purer vision of a particular kind of life thanks to the ease with which nonbelievers can leave to pursue a different conception of the good. Perhaps the presence of such a variety of different ways of life increases the (opportunity) costs on people who would try and cheat the rules of the particular community, making it safer for rule followers. None of these alone solves the problem, but in combination (and perhaps along with other arguments that I haven’t yet considered), they work to narrow down the scope of people who are simply unmoved.

    As D’Agostino has written about with great insight across numerous works, social systems are complex adaptive systems. The question is less about whether we convince everyone of a particular set of rules at a point in time, and more about how wide the basin of attraction is for a notion of dynamic stability for one’s social system. It is for this reason that I’ve tried to fixate on adjusting incentives to favor participation on fair terms in a diverse society. These can pull people along even if they are not someone’s top priority. But D’Agostino is right to note that incentives themselves are perspective-dependent, and so there is an extra kind of dimensionality to the political challenge, and one that is going to be dependent on the particular set of perspectives active in the population in question. Which is why I believe only a piecemeal approach is available as a response.

    D’Agostino makes the interesting observation that my approach may be useful as a framework for understanding the expansion of rights and pollical equality that was realized over the course of the twentieth century. I had not really considered that when writing, but I like the idea, and I think he might be right that using this lens might help us get a better sense of what helped expand the basin of attraction for greater appreciation of diverse others. I do think that by making the process of change part of the account, we can gain a better appreciation for how liberal values were shaped in the way that they were, and expanded to cover more people. It also highlights that this process is incomplete, or rather, there isn’t a notion of “complete” that is relevant. It’s not something that will end. We will always find new dimensions along which we can realize our common interests, and new dimensions along which we can see each other as importantly different.

    Perhaps unsurprisingly, I am more optimistic than some that we will find a way out of the polarized mess that we find ourselves in, in part because I think that many of the benefits of diversity come out of the tensions that it can generate. But very clearly the institutions that we have that are supposed to help manage our conflicts have either failed or were not meant to handle the situation that we find ourselves in. So it remains a question how we can make adjustments to these institutions to start expanding the basin of attraction for iterated improvements encouraging more people to work with others on equal terms. But it is a question that can be fruitfully engaged with within the framework I present.

    • Fred D'Agnostino

      Fred D'Agnostino


      Reply to Muldoon

      Yesterday, in thinking about what Professor Farr had said about Ryan Muldoon’s bargaining approach, I tried to conceptualise a social process that might lead to the amelioration of structural injustice and do so in a way that applied Muldoon’s basic ideas. Here’s how it went. (This is my take, not necessarily Muldoon’s or Farr’s.)

      We can imagine a series of bargains. At each stage in the series, there is on offer a gross increase in the constituents of well-being that will arise from cooperation that exploits diversity. Those who have historically been dominated will accept a bargain in which their gains from cooperation reach a certain threshold and so too will others, their counterparties. Provided that this series is long enough extended, the position of those historically dominated can be in due course significantly improved. Is this a classic “gradualist” story about the pursuit of social justice? Maybe, but that’s a side issue. What I was concerned about yesterday reflects what I’ve said in my own comments about Muldoon’s basic argument … namely, that there might well come a point in this imagined series (perhaps even very early in the series) when the counterparties (i.e. those not historically dominated) will not accept the bargain because, even though it delivers additional value by their lights, it also delivers additional value to those whom they despise. They would, at this point, “cut off their noses to spite their faces”; they would rather forego gains to their own well-being than see those who have historically been dominated realise such gains as well.

      Now this worry is a real worry. We see the evidences of this every time we open a news article. But what I didn’t see clearly yesterday was that there was another mechanism also at work (and probably at work in the production of spiteful attitudes). It has to do, not with spite, but, rather, with distribution. Indeed, this is a case of my being trapped by a model. (I’m sure Muldoon isn’t trapped, but I certainly was.)

      Notice how I said above that each bargain in the imagined series was one where the parties share in the gains delivered by inclusivity. But what if one of the parties weren’t going to share in these gains? Then, with or without “spite”, they would walk away from the table … they would not willingly accept a bargain in which the counterparty would be better off but they wouldn’t. And that is what a lot of people think has happened over the past few decades. Sure, there have been gains in well-being arising from more inclusive attitudes in respect of diversity. As Muldoon puts it in his Reply to my Response (FD’s emphasis), “diverse places do better”. But while previously excluded parties may have benefitted from those gains, and while structurally well situated entrepreneurs of diversity have massively benefitted from those gains by appropriating the surplus value (if I may), a host of “little guys and gals” have not and may indeed have lost relative to others and perhaps even absolutely, so that it does not quite follow, I submit, that such a process of social change “gives a lot of people more reason to favor diverse environments”, and not, as I might have thought and as Muldoon says, because the unrewarded counterparties have “conservative dispositions” or “value some kind of cultural purity”, but, rather, simply because they’re being asked to contribute to social change from which they do not benefit even in narrowly materialistic terms. Notice that, on this account, while there is indeed cultural polarization, that is an epiphenomenon on top of economic exploitation. As a lifelong socialist I could have been expected to see this more clearly and earlier than I have.

      The problem for this thought experiment about gradualism in the pursuit of enhanced social justice then becomes one, not of imagining or engineering away spiteful attitudes, but, rather, of thinking about how to ensure that all relevant parties are going to be beneficiaries … that we’re not expecting parties to continue to be part of the bargain even though they do not themselves actually get anything out of it. It’s not enough that the gains from inclusion are large enough so that all could be compensated (Kaldor-Hicks efficiency); it is, rather, that all the relevant potential parties have to be compensated … and that, arguably, hasn’t happened. And its not happening is, arguably, a significant source of the spiteful attitudes that I, trapped by the model, was so focused on.

      I see that Muldoon is to participate in a PPE Working Group Event on Joshua Priess’s book Just Work for All, and note this from the “blurb” about the book (FD’s emphasis):

      A “renewed political and policy commitment to Just Work … is essential to combat the negative moral externalities of an economy where the fruits of growth are increasingly claimed by a relatively small portion of the population: slower growth, rising inequality, declining absolute mobility, dying communities, the erosion of social solidarity, lack of faith in political leaders and institutions, ethnic and nationalist backlash, the rise of authoritarian politics and parties, and the rapid rise in what economists Angus Deaton and Anne Case call deaths of despair.”

      On this account, the pathway to social justice via more inclusive engagement with diversity is crucially dependent on finding institutional means for ensuring that all the relevant parties to the social bargain are indeed actual and not merely potential beneficiaries (by their own lights). It is, of course, an empirical question whether, if this had more completely been the case over the past few decades, the “spiteful” attitudes that I initially focused on would have been less in evidence, but it is, of course, a prerequisite to bargaining per se that there be gains for all parties. Again, I’m sure Muldoon understands all this (and much better than I), but it was something that I missed and may be something worth reflecting on further.

      I note, by way of conclusion, that I believe and said in my Response that the “gradualist” story quite likely characterised mid-20th century social progress in the North Atlantic democracies. And if it did, then, I suggest, this was because it did to a greater extent than has recently been the case deliver gains from inclusion of diversity widely enough to sustain its own progressive unfolding, something that has not happened over the past thirty to forty years. How was this engineered? How, if not engineered, did it happen? What changes to institutional arrangements might be feasible in our current circumstances that could raise the probability of this happening again? I look forward to lerning more about these crucial matters. In the meantime, thanks to Ryan for his work and to the curators of this symposium (and the other participants in it).

    • Ryan Muldoon

      Ryan Muldoon


      Reply to Fred D’Agnostino

      Fred raises a very interesting question, building off of Farr, of what we can say about distributional questions. That is, how do we ensure that all parties to a bargain have a reason to see themselves as better off? And why would they support such a bargain if they don’t?

      I have a boring answer to this, and then I’ll gesture at the more interesting answer.  The boring answer is that, at both the bargaining stage and the “experiments in distributive justice” stage, I’ve put in procedural elements that encourage, if not outright require, widely spread gains.  In the bargaining stage, all parties can reject a bargain where there are substantial gains and they’re not getting any.  Second, in the experiments stage, while I do not specify an account of distributive justice, I suggest that any account of distributive justice has some boundary conditions, ensuring that there’s no identifiable party who fails to benefit from the cooperative enterprise.

      This second condition on distributive plans is modeled after a formula used in New Deal spending to ensure that gains were widely shared.  There is a nice discussion of this in Peyton Young’s excellent book, Equity, which is where I drew my inspiration for this criterion.

      So that is the boring answer: I’ve tried to introduce procedural elements that ensure everyone will expect to do better than before, everyone is better off in this system of cooperation than some narrower system of cooperation (that is, no one is being exploited by some other subset of the parties that would be alleviated by no longer cooperating with them), and while I allow for a great deal of experimentation over possible systems of distributive justice, this experimentation is constrained to those for which there are (in expectation, at least) gains for everyone.

      This obviously does not assure that everyone reaches equal material holdings, but formally the procedures I outline will favor distributions that help expand social surplus and those that split that social surplus more equally. This just stems from the mathematics of Nash bargaining and its variants, along with the constraints on a distributive scheme.

      So what’s a more interesting answer?

      I think a more interesting response requires a lot more elements.

      First, I think it’s really important to separate absolute vs relative gains.  If we start from an unjust position, which Farr is right to say that we are, we absolutely cannot rely on a mechanism that assures no relative losses.  That would mean that we would be forced to preserve unjust inequalities and existing systems of hierarchy that are manifestly unacceptable to an inclusive society.  There is no way around this.  A goal of preserving relative positions forces us into a zero-sum competition.

      In my work I try and focus on absolute gains and losses.  What I present in my book is something that assures absolute gains for everyone, but I think it’s reasonable to suppose that what may be promised in a theory wouldn’t be assured in an imperfect institutional instantiation, and sometimes gains in expectation don’t actually materialize.  There, what we would want to know is whether the system can get itself out of a bad outcome.  This is a big goal of my approach. I make no promises that any particular bargain will work out how people wish, but what this approach does allow for is course correction, and a great deal of social learning.

      But, to get closer to the ground on some of these considerations, we might think that while it’s bad for a theorist to ensure no relative losses, it’s totally psychologically normal for citizens to be very sensitive to relative losses.  I think it’s an interesting normative challenge to think this through in a way that respects everyone as an equal citizen.  Especially if we start out in an unjust distribution of a host of material, social and political goods, justice demands relative losses, but it doesn’t mean that people aren’t experiencing real losses – they just never held a real entitlement.  But their sense of loss can shape their political choices, even if they think that increases in civic equality are good.  The answer to this has to be somewhere between “you didn’t lose anything because you never should have had it” and “you shouldn’t have to lose anything” or even the more stringent “there should be some kind of compensation for this loss.”  This has to be dealt with alongside the people who may now be getting relative gains, and have quite legitimate reasons to demand compensation for previous harms.  I don’t pretend to have a good answer worked out here.

      The other thing that I think is worth noting here is that I think it’s unwise to treat the mid 20th Century as a useful baseline for successful inclusive growth.  It was (at least hopefully) a massive anomaly. Much of the productive capacity of the major world powers had been bombed out, and the wealthy world was engaged in a form of catch-up growth, while relatively high on patriotic fervor.  Not only that, but relatively quick wage and wealth gains also cover over a massive range of debts that were “off book” so to speak.  In the US, the Post-War period saw massive expansion in racial residential segregation as federal policy, widespread lead poisoning, a federal highway policy that juiced growth and gutted cities and cut through minority communities that lacked adequate political voice, and a whole array of environmental harms, leading to huge increases in carbon pollution, spoiled waterways, and air pollution that’s directly implicated in the huge rise in heart and lung disease that’s a main source of “deaths of despair.”  The decline in growth and collapse in solidarity roughly track our political attempts to respond to those problems. At the same time, the post-70’s environment, and particularly the last 20-30 years, has seen a number of substantial (though incomplete) gains in civic equality, institutional access, and political voice.  There are also nontrivial economic gains, even for those at median or below.

      My point in raising these issues is not so much to say that the 1950’s were bad, or that the present circumstances are good, but rather that insofar as we’ve seen progress, it’s been in fits and starts, and across many different dimensions.  I don’t think we should see this as a steady collapse in political success, as much as a widening scope of democratic engagement, a globalizing economy, and enormous technological changes that have led to a much more complex social-political order.  Even just considering our information environment now compared to then should give us reason to pause on whether we can say whether spiteful attitudes have gone up or down, or changed in their distribution.

      What I do think we can say is that these important issues should give us reason to think that we need to find more room for local and regional experimentation in how we might better institutionalize generalized gains.  A challenge is that, as Fred has rightly emphasized, these gains are perspective-dependent and stem from many different dimensions, and that absolute gains may be paired with relative losses.  But I don’t think this is an insurmountable problem – it just means we need to have the space to explore our options and try things out.

Cynthia Stark


Diversity and Justification

The central problem addressed by Ryan Muldoon’s Social Contract Theory for a Diverse World is one that surely needs solving: how do we arrive at principles of justice for the basic structure of society that are suitable for individuals with radically different perspectives? What principles for distributing political rights might stably govern a scheme of social cooperation composed of e.g., Catholics, libertarians, progressives, white nationalists, descendants of slaves, disabled people, feminists, queer people, “incels,” unskilled workers and trust-funders? Muldoon’s solution is a variant of social contract theory. Morally binding principles of justice, Muldoon claims, are the outcome of an actual agreement among diverse individuals who are assumed to have roughly equal bargaining power and who choose on the basis of personal preference. The outcome is a set of mutually advantageous, and hence stable, principles. These are likely to change over time, as circumstances evolve, and they would likely be subject to two constraints. The first is that the principles would pick out a distribution located on the Pareto frontier. The second is that gains from cooperation would be distributed in proportion to contribution where one’s contribution is the difference in production one makes by participating in the productive process.

As a “morally unconstrained,” actual consent theory, Muldoon’s view can be contrasted with another dominant strain of social contract theory: “morally constrained,” hypothetical consent theory.1 In that tradition, the parties to the contract and their circumstances are morally idealized. For instance, on Rawls’s theory,2 to which Muldoon is providing an alternative, the parties are characterized as ignorant of their capabilities, social position, values and preferences. They are, on that ground and to that extent, impartial. Moreover, they are described as regarding one another as moral equals. Their consent is capable of producing morally justified principles, on Rawls’s view, only under these moralized conditions. Furthermore, the principles are, for Rawls, the object of hypothetical agreement: they are the principles that such morally idealized agents would consent to. By contrast, Muldoon’s bargainers are simply regular people with their perspectives, and the principles of justice are, for him, the object of an actual agreement among such people.

Morally constrained hypothetical consent views are not without their problems: the moral force of the principles they produce seems to depend not upon the fact that they would be agreed to but rather on the moral ideals informing the agreement. Hence the contract seems superfluous. Moreover, it is not clear why the fact that individuals would agree to a principle requires them to abide by it—it seems that we are required to do something only if we have actually agreed to do it.

Muldoon’s account is free from these problems. However, this strength, I contend, comes at a cost: the sacrifice of moral normativity. Below I will explain why the two features of Muldoon’s view that enable it to accommodate diversity—actual consent and purely descriptive circumstances of agreement—keep his principles from having content that is morally normative. My argument reveals a rather serious dilemma for social contract theories: they cannot both be inclusive and yield morally justified principles.

Consider, first, the issue of actual agreement. Suppose you and I agree that every Monday I will bring you a case of mineral water. I am now morally bound to bring you the water and I arguably do something morally wrong, ceteris paribus, if I fail to turn up with the water, for I have violated our agreement. But the fact that we agreed that I will bring you water every week does not justify the policy of my bringing you water—it does not make my bringing you water a morally good thing to do. In other words, my consent does not give me a moral reason to bring you water, it gives me a moral reason to keep my promise, which happens to be to bring you water. The wrongness of my failing to bring you water consists in my violating an agreement, not in my depriving you of water. Indeed, we might have agreed that I will bring you a stolen bicycle every week. I may be thereby bound to do so, but the rule, as it were, of supplying you with stolen bikes is not justified by our agreeing to it.3 The general problem is this: people might actually agree to almost any principle but this does not justify the content of the principle: if everyone agrees to a principle that subordinates Christians, it seems implausible to conclude that principle is thereby morally justified.

Muldoon’s account, to be sure, precludes agreement on such things as suppressing a certain religious group: individuals are equal in bargaining power, on his account, and some are Christians. Those individuals would never agree to such a principle, as it would be to their disadvantage. Principles produced by actual agreement among people with a variety of outlooks, who have equal bargaining power and who desire rules that are to their advantage will likely preclude mistreatment. So, the sorts of social phenomena generally viewed as unjust—housing discrimination, forced labor, voting inequality—will not likely emerge from the social contract that Muldoon envisions. Nevertheless, this is a contingent feature of his view: these practices are not ruled out in principle.

Compare bargaining theories that deploy hypothetical consent.4 These accounts are capable of giving people, no matter what their perspective, reasons to do the acts prescribed by the rules that are the object of hypothetical agreement. Suppose it is the case that ideally rational individuals with equal bargaining power who are seeking to maximize their utility would agree to a principle of religious freedom. This fact gives actual agents, who are not ideally rational, a reason to adopt and abide by a principle of religious freedom, for surely they ought to do what it is rational for them to do. Nevertheless, it does not obligate them to abide by that principle, as they have not, in fact, agreed to it. Notice that the “ought” in this case is prudential: if the aim is to maximize one’s utility and following a principle of religious freedom is a means to fulfilling that aim, then, as a matter of prudence, one ought to follow that principle. Furthermore, one has a motive to do so, if we assume that actual people are inclined to maximize their utility. So, even if Muldoon, to solve the problem of justification, were to rely on the idea of a hypothetical bargain, his theory would still not produce principles of justice, but rather principles of prudence with which people, no matter what their perspective, are motivated but not obligated to comply.

The problem of the lack of moral normativity is solved by morally constrained hypothetical consent theories. This type of theory can produce principles that are morally justified—principles the content of which actual people have a moral reason to follow. (However, this type of theory, like hypothetical bargaining theories, cannot explain why people are obligated to obey such principles.) Suppose that the parties to a hypothetical contract are impartial with respect to people’s ideas of the good life and regard one another as moral equals. Suppose, further, that the following counterfactual claim is true: such individuals would agree to a principle of religious freedom. The reason they might do so is that those individuals accept all (morally permissible) ways of life and they believe that people are equally entitled to live according their own values. It follows that actual people have a moral reason to obey a rule requiring religious freedom: the reason is that the rule was adopted in conditions that were fair. This constraint makes it the case that people ought morally to follow the rule. To be sure, people may lack a motive to obey this rule, or may indeed have a motive to not obey it, especially if it is not in their interest to do so. They have, nonetheless, a moral reason to comply.

We can see here a deep dilemma for social contract theorists: in order to provide morally justified principles—that is to say, genuine principles of justice—they must resort to morally constrained hypothetical consent. But this approach cannot accommodate the diversity in perspectives that serves as Muldoon’s starting point. Those who believe the state should promote a certain way of life or who endorse a caste system will not have a reason to follow principles constrained by the ideals of impartiality and the equal worth of persons. It seems one must choose, then, between a justificatory structure that infuses principles with moral normativity or one that is maximally inclusive. The very features of Muldoon’s view that make it inclusive—actual consent and morally unconstrained circumstances of agreement—prevent it from generating principles with morally justified content, despite the fact that it generates principles that people may be morally obligated and also motivated to abide by.

  1. Christopher Morris, “Justice, Reasons and Moral Standing,” in Rational Commitment and Social Justice: Essays for Gregory Kavka, ed. Jules L. Coleman and Christopher W. Morris (Cambridge: Cambridge University Press, 1998), 189.

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

  3. It is not obvious that when people consent to the terms of a social contract they are making a promise in the standard sense. This is because the device of consent, in social contract theory, is designed to determine what is just. People typically make promises, however, in the context of already existing moral principles. This allows to arise the question of whether a promise to do something immoral is a genuine promise. See, e.g., Seana Valentine Shriffin, “Immoral, Conflicting, and Redundant Promises,” in Reasons and Recognition: Essays on the Philosophy of T.M. Scanlon, ed. R. Jay Wallace et al. (Oxford: Oxford University Press, 2011), 160.

  4. A theory of this type is offered by David Gauthier in Morals by Agreement (Oxford: Clarendon, 1986).

  • Ryan Muldoon

    Ryan Muldoon


    Response to Cynthia Stark

    Stark frames her insightful commentary around a dilemma that she thinks all social contract theories face: either be inclusive such that they can accommodate diversity, or have morally justified principles. The moves that diversity-minded contract theorists such as myself make to ensure that everyone’s interests are taken into account preclude the moves that allow for moral content to be present in the contract. I think Stark is correct in demonstrating that there is a tension between infusing moral content into the core principles of justice and having a social contract that accommodates all (or at least most) comers.

    Where Stark and I disagree is whether this represents an actual problem, or insofar as it is a problem, whether it is at all unique to social contract views rather than just a core problem of political philosophy as a distinct normative enterprise from moral philosophy. Social contract views just help to lay plain this core problem, because they are organized around the idea that there are legitimate conflicts of interest and values in societies that need to be overcome in order to be able to effectively cooperate with each other, and have institutions and roles that can act with legitimate authority. Principles of justice derived by social contract theories are meant to do the grounding work for the authority that those institutions can wield, and the conditions under which the authority is legitimate. Social contract theories, as Stark deftly illustrated, differ in their setups, which can cause them to vary along whether people take themselves to be bound by the principles of justice that the theory produces. As Stark argues, there are theories such as mine that rely on actual agreement, theories that rely on hypothetical agreements from prudential reasoners, and theories that rely on morally constrained hypothetical agreements. Actual agreement theories, and some thinly idealized hypothetical consent theories, can accommodate all kinds of people and views, but Stark is right to point out, other than the normativity generated by the fact of agreement, there isn’t much moral content. Morally constrained hypothetical agreements have obvious moral content, of course, but the cost of that is a reduction in the scope of who feels bound by the agreement. After all, morality is contested ground, and people differ both on what they take to constitute a good life and what they take rightness to consist in. The structure of social contract theory makes this plain, but this is obviously a problem for any liberal political theory. The more a theorist infuses moral content into the mode of justification, the more the theorist narrows the theory: either in terms of what the political theory can do, or who feels morally bound by it.

    Political philosophers have approached resolving this tension in a couple different ways. One is just ignoring the disagreement, and proceeding with a thick moral conception. The extreme version of this is simply “determining” the true moral theory, and then scaling it up to just constitute justice. To me this is obviously unworkable. People who are convinced by some other moral theory will think they are being unfairly coerced into following a set of moral views that they reject. There will be no sense of morally binding them to act on the dictates of justice, because they reject the moral foundations. The reasons that could be deployed to argue for the correctness of the political authority being wielded would not be widely shared, and so people would resist the state coercion that would go along with it.

    This is unworkable because it fails to recognize the distinction between the task of ethics and the task of political philosophy. While ethics can focus on determining the right way to live, political philosophy needs to deal with how people who, along with other reasons for conflict, can peaceably cooperate with each other on fair terms. Rawls recognized the importance of separating moral theory from political theory, which is why he emphasized the idea that the two principles of justice were “political, not metaphysical”—they were free-floating conceptions that were not attached to any particular moral theory. He attempted to ground them in an overlapping consensus of liberal moral theories. And this is another approach that liberal theorists can take—find the substantive core of our moral agreements, despite our many moral disagreements. This is clearly superior to simply picking a privileged moral theory and scaling it up, but as I argue in my book it needs to presuppose that there is, in fact, robust moral agreement, not merely about what the right thing to do is, but what reasons we have for acting and how we understand the extensions of the concepts that we purportedly agree on. For instance, it is very hard to find someone who disagrees with the claim that murder is wrong, but it is very easy to find enormous disagreements about which cases this claim covers, because we understand “murder” differently than each other. Some people think that embryos can be murdered, others think not. Some people think that animals can be murdered, others think not. Some people think this makes war unjust, others think not. And so on. Our areas of agreement might be quite robust, or they might be rather thin and patchy. As I discuss at length in my book, it’s not merely that we can disagree about what actions or outcomes we value or abhor, it’s that we disagree on how to characterize those things in the first place. We can see the world quite differently from each other, and this makes moral agreement much more complex.

    On my approach, agents try and find out if they have any common ground, and if they do, whether that common ground is due to just sharing a perspective, or that common ground is supported across many perspectives. These areas of agreement can help ground any bargains that happen later, as they provide a shared set of moral constraints if they are present. If there isn’t this kind of moral agreement, then a political theory shouldn’t be able to help itself to it. The political problem—how can we get people whose interests and values conflict to peaceably cooperate with each other—is present no matter what. So a solution to the political problem can’t be contingent on a solution to the problem of moral agreement. This is what the procedure in the “View from Everywhere” is about. Any bargaining that happens occurs after we’ve established the possible grounds of preexisting normative agreements. In an easy case, there is a deep well of moral agreement to draw from to help normatively constrain the bargain. In a harder case, there’s very little, if any, normative agreement, but we still need to live together. It is true that my theory treats this kind of moral agreement as contingent, but that’s because I think it really is contingent. Different populations of people will have different areas of agreement. So we proceed with what we have, and that means bargaining on equal footing.

    Stark is right to point out that the bargains will almost certainly block oppression or manifest unfairness or things that we would normally consider unjust, just because everyone has the ability to reject things that don’t advance their interests. Stark takes this to be cold comfort because these things are not explicitly ruled out. But I see this as a feature rather than a bug. Bargaining on equal footing is a great way to embody the idea of political equality. It is a very real political constraint on the process, relative to what might otherwise just be bargaining with threat advantage. I proceed from the idea that if people want to live in a system of cooperative relations with others, everyone should be willing to do so on equal terms. If they feel that this fails to advance their interests, they are free to break off and form their own, smaller system of social cooperation amongst political equals, but in doing so they will lose out on the benefits of wider cooperation. This basic idea—that we have to reason as equals about what the terms of our cooperation should look like—helps to capture liberal processes rather than just liberal outcomes. It also helps us see liberalism and the rights it is associated with not as things that were understood as constraints from the get-go, but rather solutions to conflicts. On my account, today’s uneasy compromise could be tomorrow’s agreed upon moral constraint. Or, perhaps we find that the compromise had unintended adverse consequences, and so we try again. Liberalism did not emerge fully formed, but instead slowly built up a set of rights that solved problems for real people, and slowly and imperfectly expanded its domain to cover more people.

    Political philosophers start from the fruits of this process of liberalism—namely, the set of rights we now recognize as core liberal rights, and the associated values that make liberalism so vibrant. While I obviously value these liberal outcomes, I am much more interested in the processes by which these things can emerge, especially if we can reflect on how they could emerge in a more just way. The history of the emergence of liberal ideas is a history of incomplete access to liberal protections, the social discovery of new political needs, and efforts to balance between competing interests. I worry that if we treat some particular moment in time, suitably idealized, as the appropriate set of side-constraints that liberalism ought to impose from the get-go, we will miss out on the huge value of the process of social discovery that is always underway. The process of rights expansion, and the social discovery of better ways to live as equals, is something that won’t end anytime soon, nor should it. A theory focused on this idea of discovery opens new avenues in political philosophy that I think have been closed off by traditional forms of social contract theory in particular.

    It’s true that a theory like mine does not offer a robust moral foundation for political authority, but that is because I think, at least insofar as we have a generally liberal disposition about treating diverse people as political equals, we simply can’t assume that there is one. That is not to say that we don’t have a variety of normative appeals to make here. Agents aren’t merely making random agreements, they are making agreements with others that include any available basis for moral agreement, along with a process that ensures everyone has equal ability to pursue their own interests (whether those are material, moral, cultural, or something else) as constrained by everyone else pursuing theirs. Agreements are on equal footing, and are made by parties who want to find some way to peaceably cooperate with each other. This gives us prudential reasons to support the agreement, politically normative reasons to support the agreement, and moral reasons to support the agreement. There are prudential reasons, just because agents will see agreements as in their personal interest (otherwise they would have rejected them). The agreement is the best that someone can get, given the options everyone can come up with, and given that everyone else is trying to get more of what they value as well. It’s the best compromise that we can find. There are political reasons, as this gives grounding for us to live together peaceably. Everyone has reason, on their own terms, to endorse the bargain, even if they disagree about what the reasons to support the bargain are. The bargain secures everyone as a cooperator and helps to manage conflicts such that we can work past them. And there are moral reasons, as you gave your word in a procedurally fair situation where your interests were taken into account on par with everyone else’s, subject to any moral constraints that could be agreed on in advance. You made a promise to others in good faith, and that should be honored.

    I think Stark is right to say that a more diverse political community is one in which there will be a sharp tension between a rich moral foundation to a political arrangement and a political arrangement that is appropriately inclusive. As I’ve tried to argue, I think this is not so much a problem that social contract theory faces uniquely, but something that social contract theories more broadly help make apparent, and mine in particular aims to address more head-on. The more that political philosophy helps itself to thick moral concepts, the less grip it will have in a political community marked by robust disagreements. The answer is then obvious: political philosophy has the task of finding ways of dealing with disagreements, and not imagining them away. If that means leaning on other sources of normativity to help ground our political commitments, then that is what political philosophy ought to do.

    • Cynthia Stark

      Cynthia Stark


      Bargaining on a Footing of Equality?

      I greatly appreciate Muldoon’s careful response to my comments on his work. If anyone could convince me of the merits of bargaining contractarianism, it would be Muldoon, but, alas, I am not convinced. Muldoon’s notion of a perspective explains this irreconcilability: in the end, he and I have different understandings of the extension of the concept “liberal justice.” However, I do have a worry about some of his remarks that is not attributable to the distance between our perspectives; it is a worry that accepts much of his approach. He says, “Bargaining on equal footing is a great way to embody the idea of political equality. It is a very real political constraint on the process, relative to what might otherwise just be bargaining with threat advantage.”

      I am concerned about this distinction between bargaining on an equal footing and bargaining with threat advantage. I believe this binary relies on an implausible “leverage-free perspectives” assumption: Muldoon states that differences in how we “value goods and services, or even currency itself, does not affect how we account for relative bargaining power (92).” For Muldoon, differences in bargaining power are external to one’s perspective. They reside in one’s social location—say as a member of the capitalist class—and Muldoon argues that that sort of bargaining advantage is short-lived. In the long-run he argues, individuals’ positional threat advantages tend to evaporate. So, we can assume that in bargaining for the division of rights and goods, the parties to contract are on a footing of equality with one another in the sense that they are equal in bargaining power.

      Now consider this criticism of Braithwaite, offered by Rawls. He states:

      On the analysis…it turns out that the fair division of playing time between Matthew and Luke depends on their preferences and these are connected to the instruments they wish to play. Since Matthew has a threat advantage over Luke, arising from the fact that Matthew, the trumpeter, prefers both of them playing at once to neither of them playing, whereas Luke, the pianist prefers silence to cacophony, Matthew is allotted twenty-six evenings of play to Luke’s seventeen. If the situation were reversed, the threat advantage would be with Luke…. But we have only to suppose that Matthew is a jazz enthusiast who plays the drums and Luke a violinist who play sonatas, in which case it will be fair on this analysis for Matthew to play whenever and as often as he likes, assuming as it is plausible to assume that he does not care whether Luke plays or not. Clearly something has gone wrong. (A Theory of Justice, 134, ftnt 10.)

      This example seems to challenge Muldoon’s leverage-free perspectives assumption. As Rawls sees it, Matthew has a threat advantage simply in virtue of his preferences—the instrument he enjoys playing and the context in which he likes to play it. His perspective gives him a threat advantage over Luke, who in the worst case, rarely if ever has an opportunity to play his violin without a drum “accompaniment” which he would prefer to avoid. His bargaining power is not due to something external to his perspective, for instance, to his having power over Luke due to being Luke’s employer or landlord. Thus, for Rawls, theories of bargaining are always theories of bargaining from threat advantage—there is no bargaining on a footing of equality so long as people’s preferences differ. For this reason, Rawls goes on to say in the passage quoted above, “We cannot take various contingencies as known and individual preferences as given and expect to elucidate the concept of justice (or fairness) by theories of bargaining.”

      I am not sure if the dispute between Muldoon and Rawls is empirical or a matter of how to define a threat advantage (or something else). Nevertheless, if Rawls is right, then the parties to Muldoon’s contract, with their differing perspectives, bargain on an equal footing only in the sense that they have an equal chance to participate in the bargain and not in the stronger sense that they are equal in bargaining power once they participate. So, the nature of their political equality is even more minimal than Muldoon claims.

    • Ryan Muldoon

      Ryan Muldoon


      Reply to Cindy Stark

      I am very happy that Stark has brought up this concern about the fairness of bargaining approaches.  She suggests that the structure of one’s preferences can induce unfair advantages in bargaining.  If we imagine, as Rawls does in response to Braithwaite, that two musicians are trying to fairly divide time spent playing, their preferences about playing can shape their threat advantage.  In Rawls’ example, if Matthew is simply indifferent to other playing, whereas Luke needs silence from others to play, Matthew can simply ignore Luke and play as he wishes.  Luke is out of luck.

      From this, Stark, draws the conclusion that bargains are always bargains with threat advantage, such that a bargaining model simply can’t serve to embody a way of conceiving of citizens on equal terms.  I think this is mistaken, and I think it’s usefully drawn out by considering whether this musician example is a good model of what would be going on in a bargain about allocating rights and obligations.

      The first thing to note is that the “solution” Rawls suggests is that Matthew could simply ignore Luke.  This is obviously not what a Nash solution would recommend, but the idea is that a rational agent could just not care about the interests of other agents in the pursuit of their own interests.  But if we take this to be a model of determining the rules that we are bound to – the sets of rights and obligations that we take ourselves committed to – the reply to Braithwaite here is just to assume that no agreement is made, because there is nothing that Matthew needs from Luke.  Matthew can try to be Robinson Crusoe, but even Robinson Crusoe found himself in such a scenario because of a shipwreck, not a considered choice.  Insofar as Matthew would choose a strategy of simply barreling over the interests of others instead of reaching an agreement, Luke would be unbound by any rules of behavior toward Matthew, because he didn’t agree to such an outcome.  One may think that Matthew wouldn’t have his drums for very long.

      The more interesting response, however, is that I think these limited thought experiments obscure as much as they reveal if we want to understand how a bargaining-oriented view would work.  First, there are lots of people, not just 2, who are trying to work out rules of living together.  Matthew may be able to run roughshod over Luke, but perhaps not over Luke, Mark and John.  Second, and far more importantly, our bargaining would be, as I discuss in the book, very high-dimensional.  There are simply a lot of different kinds of things we do together.  Let’s say that Matthew values cacophonous playing extremely highly. As per the setup, this would be nontrivially burdensome on Luke.  In the book I discuss the “price” of a right (understood as a set of permissions and restrictions in a set of circumstances) as the externalities it imposes on others – their loss of being able to do something in some set of circumstances that they find valuable.  Matthew may want to, in this domain, impose costs on Luke (or others).  For him to get a stable right to do so, he’s got to find a way of compensating them for their perceived losses.  Luke may accept his loss of playing sonatas in some contexts if he can gain some right (or collection of them) that is at least as valuable by his own lights.  This may or may not place an imposition on Matthew that Matthew notices – he may hold perspectives that make it hard for him to “see” what Luke wants. But for Matthew to be able to impose costs on others, this has to be compensated for.  If not, or if Matthew insists on burdening others without accepting burdens himself, others can simply cut Matthew out of the agreement.  If Matthew can’t accept the tradeoffs that facilitate commodious living in a system of social cooperation, he loses its benefits.

      I think a crucial insight of the contractarian tradition that eschews a thick normative starting point is that we depend on others to achieve our own goals, whatever they might be.  It is in our rational self-interest to cooperate with them on terms that they find acceptable.  We may be able to press an advantage in some particular domain, but we still have many other dimensions of interaction and cooperation that we rely on.  The costs we impose in one area can easily come back to us in another where we have a less favorable position.

      On the social contract framework I’ve developed, it is possible that we arrive at the bargaining stage with fairly robust ethical agreements that serve to constrain our bargain.  In those cases, we will have more robust guarantees about the nature of the bargain.  But in those situations where we lack that normative consensus, simply imagining that it’s present would not serve to be convincing to those who disagree. What I suggest, though, is that all is not lost: the very nature of our sociality, and the overwhelming benefits of social cooperation give us powerful tools to keep each other in line. The complexity of social life, and the richness of the rules that we select for ourselves, gives us ample opportunities to find tradeoffs and compromises that ensure that everyone has equal capacity to better realize their conception of the good.  Of course, as with any agreement model, we may make good faith errors in what would best advance our conceptions of the good.  Or we may find that the world has changed enough that the rules that used to work for us no longer do.  This is why, on my approach, we can learn from those experiences and try again.

Gerald Gaus


Diversity, Discovery, and Justification

Ryan Muldoon’s Social Contract Theory for a Diverse World is strikingly original, bold and important.1 It advances an array of new ideas and concepts: moral perspectives, objectivity as the “view from everywhere,” and a radically new view of bargaining between different perspectives. And at the heart of the entire work is the clarion call for moral and political philosophers to take our deep diversity seriously. I count myself as a fellow-traveler, seeking, like Muldoon, to develop a “New Diversity Theory,” which appreciates not only the depth of, but the opportunities presented by, the diversity of our contemporary world.2 My disagreements with Muldoon are thus intramural ones, about the best way to articulate a large body of shared commitments and concerns.

1. The Priority of Discovery?

An important difference between our two approaches is that, while Muldoon’s point of departure is an expansion of John Stuart Mill’s justificatory framework (chap. 2), mine is an expansion of Rawls’s public reason liberalism. As Muldoon sees it, “Mill’s approach to experiments in living offer us an account of social discovery. On this kind of account, justification becomes subsumed to iterated discovery, which includes a permanent competition of perspectives” (30). “This,” he goes on to say, “is a more thoroughly empiricist (and evolutionary) model of political justification—rather than pointing to a regulative ideal and comparing ourselves against that a priori standard, we try competing approaches out, and see what works in our circumstances” (30, emphasis added). Muldoon is crystal clear that discovery and experimentation have “primacy” for political justification in a changing world (35). Thus the fundamental problem of “Rawlsian public reason”: it is “ultimately about justification, not about discovery” (29).

The language of “experiments” and “discovery” suggests a scientific analogy, and indeed Muldoon explicitly draws on current accounts of the division of cognitive labor in scientific inquiry (30). Now, as I understand them, models of the division of scientific labor typically assume diverse teams exploring different parts of an agreed-upon “scientific landscape.”

Scientists are imagined to be “hill-climbers” on an unknown “landscape.” The landscape itself is interpreted as a topic of scientific inquiry. The X and Y dimensions represent potential research approaches. The Z dimension represents the epistemic significance of any findings to be had given the research approach indicated by the (X, Y) position. . . . At the beginning of inquiry, scientists have no knowledge of the landscape—that is, they do not know anything about the comparative significance of any research approaches. They discover this only by traversing the landscape.3

These models assume that when one competent scientist reports the Z value of a specific coordinate (Xi, Yj), others generally concur. They are exploring essentially the same landscape in different ways. If the diversity of perspectives leads them to different search strategies on the same landscape, their diversity supports an efficient cognitive division of labor. However, if their perspectival diversity leads them to explore different landscapes—such that when Alf is at (Xi, Yj) he observes a value of 100 on the Z dimension while when Betty is there she reports a Z value of 0—their searches will not be of much value to each other.

Scientists sharing roughly the same paradigm are exploring roughly the same epistemic landscapes: they share the similar problems, standards, and categories such that when one discovers a solution to problem P others will generally agree that it is indeed a solution to P. The Hong-Page theorem to which Muldoon refers (52), has a similar feature: a diverse group of agents who have different ways of looking at a common problem, and who agree on the value of any given solution, will, under rather demanding conditions, necessarily find the best solution. Again, they share an agreement on the value of any point on the landscape. Given this, my worry should come as no surprise: in our deeply diverse societies, it is seldom if ever the case that we all share the same “epistemic-value” landscape about politics and justice. As I argued elsewhere, what one person considers an ideal point that perfectly solves the problem of justice (perhaps market socialism), another might see as a manifestation of grave injustice.4 Indeed, points that Alf sees as discoveries (say, a society of universal love without any self-interest) may strike Betty as simply impossible, so not scored at all. To be sure, in a diverse society some groups will share sufficiently enough “epistemic-value” landscapes with some others so that what one reports as a discovery will be taken up by like-minded others. I have called these “communities of moral inquiry”; these are critical features of free and diverse societies. Certainly some experiments of some quite different others really do constitute discoveries for me. However, only if social diversity is severely restricted will an entire society constitute a single community of moral inquiry, such that one group’s experiment will constitute a public discovery, which can be claimed to be publicly justified, showing “what works for [all of] us.” Contrast this to science, where the experiments of one team often do constitute common, public, findings about what works.

2. Objectivity

There is an obvious way to avoid this problem: hold that, after all, we really are searching the same landscape, even when we don’t know it. For any (X, Y) coordinate, it can be claimed, there is an objective Z score—what we might call the “true Z score.” In this vein Hélène Landemore and Scott Page appeal to the idea of an “oracle” who can announce the true values of points on the landscape.5 This brings political philosophy back towards the scientific model, for we are again searching a common landscape and seeking to discover the true value of various points, which we then can share with others. This is indeed in the spirit of John Stuart Mill’s understanding of experiments in living and the discoveries they yield. Mill was a perfectionist, and believed that individuality, intellectual development and fellow-feeling (such as national feeling) were features of a developed human being.6 The experiments of intelligent and mature humans would help us all discover the sorts of lives well-suited to the perfection of our nature, and we would ultimately converge on these. Thus, for example, Mill was convinced that experiments with different modes of industrial organization would lead intelligent workers to abandon the wage employment of capitalist firms in favor of new experiments in worker-owned and managed cooperative enterprises; in the end only the least intelligent and least energetic workers would remain as wage laborers.7

Muldoon’s complex analysis of objectivity as “The View from Everywhere” (chapter 3 of Social Contract Theory for a Diverse World) is thus critical in understanding his analysis of moral and political discovery and, by extension, his understanding of public justification. I cannot hope to do full justice to the complexities of this rich chapter in a couple of pages, but as I read it, eight claims are critical.

  1. Like Mill, the aim is to provide a test that gives evidence about what is (objectively) valuable, good or right (47);
  2. This test is to filter out, or guard against, moral relativism (55);
  3. We have no direct way to test for objective correctness of moral principles (or value) (47);
  4. (a) Beliefs, attitudes and interests are “correlated” with moral principles (47),

(b) Beliefs, attitudes and interests are “evidence” for our moral theories (47);

  1. (a) “Perspectives . . . help make sense of our moral beliefs and interests” (52);

(b) Perspectives can be understood as “lines of support for beliefs” (54);

  1. So by 4b, beliefs and interests support (are evidence for) moral principles, and by 5b perspectives are lines of support for beliefs.
  2. The Master Epistemic Principle: The more perspectives (lines of support for moral beliefs that, in turn, support/are evidence for) moral principle P, the more confident we should be that P is (objectively) correct (re: 1 and 2).
  3. “The goal [of the preferred aggregation procedure for determining objective value/moral correctness] is to determine the set of beliefs that have the most independent lines of argumentation supporting them [as per 7, the most perspectives], not those that are most widely held” (55).

Now my concern is whether an account of discovery based on correct or objective Z values along the lines of 1–8 can ground a plausible conception of what is publicly justified in a deeply diverse society.8 I have trouble seeing how: 1–8 express precisely the type of controversial metaethical view that Rawls (I think powerfully) argues that we must avoid in public justification.9 As far as I can tell, moral relativists are simply excluded from the justificatory public by 2. We cannot expect moral relativists to endorse principles which are justified via a discovery test intended to exclude moral relativism. More generally, the Master Epistemic Principle10 is itself highly controversial. While Muldoon stresses the number of independent lines of reasoning that lead to moral beliefs, a Christian might appeal to John 14:6: “I am the way, the truth, and the life: no man cometh unto the Father, but by me,” which rather suggests that one and only one route is worth paying attention to. There may be many paths to damnation, but only one to salvation.

I don’t think it is only relativists and Christians who might draw back from the Master Epistemic Principle—I worry that it is based on a controversial theory of reasoning and is unwittingly biased in favor of the intelligentsia. The crux of the principle is that the more different lines of reasoning we can find for a conclusion, the more confident we should be in it. The supposition is that the justification starts with beliefs which then, by inference, give us conclusions; the more inferential lines we can find, the more confident we should be in the conclusion. But a good deal of evidence indicates that reason often goes the other way around: we start out with intuitive conclusions for which we find reasons. We know that people are very adept at coming up with many lines of reasoning supporting erroneous intuitions.11 The more people reason on their own, the more lines of reasoning they find for their prior beliefs.12 The intelligentsia of a society are its professional reasoners, so in any society we should expect that there will be the most independent lines of reasoning for whatever their moral intuitions are. After all, their job in the cognitive division of labor is the production of increasingly refined and differentiated lines of reasoning for their intuitions. Think, for example, of all the different lines of reasoning supporting egalitarian principles of distributive justice in political philosophy. That professional philosophers start off with leftish moral intuitions almost guarantees the proliferation of leftist perspectives in our society. Whereas the populace is more content to take over a modest number of existing perspectives (and so each will be widely subscribed to), the business of the intelligentsia (a requirement for tenure?) is to arrive at new, often idiosyncratic, ones, albeit usually supporting the old intuitive positions. Unless we can admit into public justification the assumption that the intelligentsia is also more likely to be morally aware—not, I reckon, an admissible claim—the Master Epistemic Principle, and so claim 8, strike me as too controversial as basis for justification.13

3. The Priority of Justification

I thus find it hard to see how discovery and experimentation can have “primacy” for political justification—if that means justification to all the members of the public—in a deeply diverse world. A reasonable section of the public (e.g., the “moral relativists”) do not believe that there is a common moral landscape to be discovered, and even those who do accept some notion of the objectively correct “Z scores” deeply disagree about the method for uncovering them. These types of worries lead the Rawlsian to insist on the priority of justification over discovery of the moral landscape in thinking about political principles and institutions in a diverse society. Because we so deeply disagree about what is objectively correct (or best) and how to uncover it, deeply diverse societies cannot be organized on the basis of a competition to discover it. That, as I have been saying, is a matter for different moral communities to approach in their own ways. This, though, does not mean the basic moral and political constitution of a diverse society cannot be justified to all: drawing on their diverse perspectives, we can ask whether all members of the public have reason to endorse the basic structure of our social relations. That, I think, is why most readers of Muldoon’s book will focus on chapter 4, “Justice Without Agreement.” The innovative proposal of distributing rights through bargains between diverse groups, who reason from different perspectives, brings justification back to center stage.14 They do not agree what moral objectivity is to be discovered, yet seek common grounds for living together.

Muldoon, however, offers an alternative interpretation of the Rawlsian project. Rawls too, he suggests, adopts a view of correct moral reasoning and objectivity; and rather than Muldoon’s empirically informed “view from everywhere,” Rawls seeks moral discovery via a “view from nowhere.” “Rawls . . . opts for a procedure that . . . aims to strip away those features of ourselves that might bias us in our process of moral reasoning. The procedure Rawls has in mind is deliberation in the Original Position, behind the ‘veil of ignorance’” (39–40). As I see it—and I think Rawls is clear on this point—the aim of the original position is not for us to evaluate our society from a perfectly impartial and detached “nowhere,” but to develop a shared perspective of democratic citizens. It is not constructed from nowhere, looking down at our social world, but by “you and me” as democratic citizens, trying to identify a perspective (or, rather, what we might call a “partial perspective”) that we all share.15 As such, it must filter out any information that would allow a person to draw on parts of her perspective not shared by others. “The difficulty is this: we must find some point of view, removed from and not distorted by the particular features and circumstances of the all-encompassing background framework, from which a fair agreement between persons regarded as free and equal can be reached.”16 Indeed for Rawls, finding a suitable method to narrow our disagreements so that we concur in our judgments “normally suffices for objectivity.”17 In this way the original position as a device of justification establishes the basis of objectivity via the priority of justification—not the other way around.

4. The Discovery of Justification

Because Muldoon understands the original position as a view from nowhere, he depicts the principles that parties arrive at in the original position as “a priori” (30). As I have been arguing, I cannot see how Rawls’s principle of justice are knowable before experience; Rawls would claim that they are based on the shared experiences and political values of citizens in a democratic society. However, while overstated, Muldoon’s core point is sound: the point of view from the shared perspective of the original position is overly abstract and informationally impoverished. This, I think, is for two reasons. First, if the perspective is genuinely to be shared among all good-willed and competent citizens of a democracy, it must be abstract indeed, and so it is difficult to see how it can go beyond abstract principles or platitudes for social living. These are important, but at best they only identify the very broad contours of the terms of our moral relations. Rawls is only able to generate more substantive results from the original position because he implicitly assumes a fairly egalitarian perspective, and so in fact excludes a good portion of the democratic public.18 Second and relatedly, the theory is based on Rawls’s understanding of what is shared. The results in the inevitable biases of a single philosopher seeking to articulate a moral blueprint for the construction of social and political order that all members of an extensive, deeply diverse, society can endorse. Is someone with the limited life experiences and knowledge of any single philosopher really competent to devise such a plan for a deeply diverse free and open society?

The great merit of Muldoon’s pathbreaking book lies, I think, in his reflections about how we might discover, not the optimal, best, most adaptive, correct, or true, but the terms of association that we can all live with. Although discovery in a Millian sense is not prior to public justification, it is true that we must engage in an ongoing process of discovery to see what can be justified to all members of the public. Philosophic constructions are not up to the task. Hayek showed us that markets are ways to discover information, but he also insisted on the fallacy of thinking that the aim was to reveal the socially most valuable system of ends (or the system of ends with the most perspectives supporting it).19 What markets reveal to each individual is how to effectively secure her ends in a world in which others are trying to secure theirs. Or, to put it in terms of justification, each is searching for terms of engagement (“bargains”) that others find acceptable. This is not a society-wide competition of perspectives (30), but a search by perspectives for ways to reconcile their diverse ideas of an acceptable framework for social living.20 And here I enthusiastically concur with the message of Muldoon’s wonderful chapter on “Justice Without Agreement”: finding these terms is a bottom-up social process, not the discovery of any moral philosopher.

  1. Ryan Muldoon, Social Contract Theory for a Diverse World (New York: Routledge, 2016). All parenthetical page references in the text refer to this work.

  2. See my essay, “The Complexity of a Diverse Moral Order,” Georgetown Journal of Law and Public Policy, forthcoming.

  3. Ryan Muldoon, “Diversity and the Division of Cognitive Labor,” Philosophy Compass 8 (2013) 117–25, at 120; emphasis added.

  4. See my Tyranny of the Ideal: Justice in a Diverse Society (Princeton: Princeton University Press, 2016), chap. 3.

  5. Hélène Landemore and Scott E. Page, “Deliberation and Disagreement: Problem Solving, Prediction, and Positive Dissensus,” Politics, Philosophy & Economics 14 (2015) 229–54.

  6. I argued this a (distressingly) long time ago in The Modern Liberal Theory of Man (New York: St. Martins, 1983).

  7. John Stuart Mill, The Principles of Political Economy with Some of Their Applications to Social Philosophy, in The Collected Works of John Stuart Mill, ed. J. M. Robson, vols. 2 and 3 (Toronto: University of Toronto Press, 1965), bk 4, chap. 8.

  8. Of course putting the issue in terms of “Z values” is not necessary.

  9. See John Rawls, “The Independence of Moral Theory,” in John Rawls: Collected Papers, ed. Samuel Freeman (Cambridge: Harvard University Press, 1999), 286–302.

  10. This is my label, not Muldoon’s.

  11. For example, in the Wason selection task, where subjects are seeking to test the truth material conditionals, people who arrive at the wrong answer are excellent as formulating a bevy of reasons for their erroneous choices. See, e.g., Hugo Mercier and Dan Sperber, The Enigma of Reason (Cambridge: Harvard University Press, 2017), 213.

  12. Mercier and Sperber, Enigma of Reason, 247. For an argumentative theory of reasoning such as Mercier and Sperber’s, it is the confrontation of conflicting reasons in argumentation that helps us sort out the good from the bad, not the production of lines of reasoning supporting one’s intuitions. I certainly do not wish to suggest that Muldoon ignores that in deliberation there is an “exchange of perspectives that might help to settle” moral disputes (56). My limited concern is with what he calls “modified version of the Condorcet Jury Theorem” (57), as articulated by claims 7 and 8, and their role is establishing the priority of discovery.

  13. Sam Spade, though, might accept 7: “All those [reasons are] on one side. Maybe some of them are unimportant. I won’t argue about that. But look at the number of them. Now on the other side, we’ve got what? All we’ve got is the fact that maybe you love me and maybe I love you.” Dashiell Hammett, The Maltese Falcon, in The Novels of Dashiell Hammett (New York: Knopf, 1965), 438.

  14. Bargaining about rights allocation is the second stage of Muldoon’s “three-stage process” (62). My concern in this essay is the first. The second stage, I think, must do all the work in securing public justification.

  15. John Rawls, Political Liberalism, expanded ed. (New York: Columbia University Press, 2005), 28. On partial perspectives, see The Tyranny of the Ideal, 105ff.

  16. Rawls, Political Liberalism, 23.

  17. Rawls, Political Liberalism, 120.

  18. See Tyranny of the Ideal, 150–54.

  19. F. A. Hayek, “Competition as a Discovery Procedure,” in The Market and Other Orders, ed. Bruce Caldwell (Chicago: University of Chicago Press, 2014), 303–13, at 307–8.

  20. As I argue in “Self-Organizing Moral Systems: Beyond Social Contract Theory,” Politics, Philosophy and Economics 17 (May 2018) 119–47. Muldoon’s appeal to the Ricardian model of trade (33–35) is spot-on here, for just as individuals come to appreciate the great gains from trade, so too do we come to appreciate how justified terms of engagement secure the shared normative goods of mutually recognized moral claims and accountability.

  • Ryan Muldoon

    Ryan Muldoon


    Response to Gerald Gaus

    I am deeply fortunate to have Gaus as a fellow traveler. The incipient “New Diversity Theory” is incredibly exciting to me, and I have gained immensely from trying to consider his and other fellow travelers’ critiques and suggestions. Whenever he suggests a criticism of my approach, it is definitely reason to pause and see where I’ve gone wrong, either in thinking about or expressing my approach.

    As Gaus points out, while we share several goals, we go about trying to achieve them in different ways. I am most interested in the idea of discovery, and my book is in many ways an effort to see how far one can push that concept to do useful work in political philosophy. Gaus has a refined take on public reason liberalism. My interest in discovery, and concern about public reason, comes from two concerns: first, I think that public reason is a poor way to capture anything about political change or political progress, and second, I think it is very hard to avoid a fair amount of normalization in a public reason approach. The first concern is driven by something like a pessimistic meta-induction across past theories of the social contract: each presents itself as a complete theory of justice, but each is heavily shaped by trying to solve some particular problem of a time and place. So the available reasons and the salient problems are indexed to some particular set of circumstances. Of course there is nothing wrong with that—my work is clearly responding to problems of the day as well—but where I think I’m doing something different, and hopefully better, is that I recognize that this is what I’m doing and take steps to pull that process of reconsideration and revision in light of new information and new situations into the process of justification itself. This is meant to be much less ambitious in one sense—I am giving up on the claim that the theorist can articulate an ideal endpoint or equilibrium—but perhaps more ambitious in another, in that the goal of this procedure of revision itself can remain fixed across new challenges as they arise. Public reason, I think, is a good model of the ideal of what can be justified in some particular set of circumstances, but this forces all reconsideration out of the model, and into a new one, which is more or less silent on what’s come before. This strikes me as leaving a lot of the important stuff out.

    My second concern is that public reason demands normalization. Gaus’s own model of public reason is admirably more constrained in how much it normalizes, but public reason demands a shared perspective. The shared perspective is difficult to attain, and pretending that we can, or that we have, attained it will leave people out, and the resultant method of justification is not going to have any grips on them. More generally, the more normalization one has, the less we’re looking at diversity, especially perspectival diversity. This means that the idealizing choices made by the philosopher become much more important, as they may be inadvertently narrowing the scope of their theory. A shared perspective certainly simplifies things, but as Bentham says, hunger is not bread.

    So, let’s turn to Gaus’s very thoughtful concerns about the discovery approach. Gaus suggests that I am making a kind of scientific analogy, which is quite right—my work on search problems in science, and the role of the division of cognitive labor in scientific communities has informed my work here. Gaus worries that the landscape-search kind of approach requires the sort of normalization that I was just criticizing public reason models for. That is, when a scientist makes a discovery, there’s broad agreement as to the importance of those findings. Diversity helps in coming up with new search strategies, but if it permits multiple overlapping landscapes, then there’s no shared world to share information about. This is a great thing to push on. Conveniently, I think there is more room for different interpretations of the Z dimension than Gaus appreciates.

    As Gaus points out, in my work on epistemic landscapes, there is an (X,Y) position on a landscape that indicates a particular research approach, and the Z dimension indicates the significance of any findings using that particular research approach. Agents traverse the landscape to discover more about this Z dimension, which they do not know in advance. Gaus worries if Alf and Betty report different Z values for the same point, then they won’t be able to learn from each other—they will just treat the other’s discoveries as noise. This is right. However, that is only if we normalize too much. The problem is if Alf and Betty report competing values for epistemic significance. But if epistemic significance is really a kind of summary statistic, a weighted average of some set of more basic elements, then it’s not a problem at all. Alf might just care about things that have immediate industrial application, Betty might care about testing the Standard Model in physics, Carol might care about application to other fields of science, and so on. Z isn’t a single number, but a tuple, and each agent can collapse that tuple to a single value given what they take to be important or valuable. So long as the reporting on Z includes information on the decomposed tuple, each will benefit from each other’s work, even if they have different reasons to care about it.

    This model can readily extend to social and political life. Our experiments in living can inform people with quite different values or perspectives precisely because these experiments can be reinterpreted and evaluated on different terms. This happens all of the time in cultural production and consumption—ideas and concepts flow quite freely across people with very different interests and ways of looking at the world. Their new uses may look very different from their original intent, but so be it. More structurally, we can also learn a great deal about, say, stable social forms or successful mechanisms for behavior change across very different areas of life. Many of the formal tools and models that Gaus and I regularly use in our work are useful exactly because they help theorists see common patterns in otherwise very different kinds of social situations. One of the valuable features of models is that there is frequently a many-to-many relationship between models and the world. Different models impose different structure, and we learn about different aspects of the same thing, and find new similarity relations across various contexts. What actually matters, as I see it, is not that everyone shares the same exact landscape, or is sensitive to the same set of items, but that there is the expectation that information will flow through the social system, such that there can still be social learning. I think we have lots of reasons to think that it can, even with a great deal of disagreement about what matters. This does give us some sense of what the hard boundaries are going to look like—there has to be enough shared perspectives that the bits and pieces can make their way around. If there are no overlaps of these decomposed features across Gaus-style communities of moral inquiry, and no reinterpretation, then Gaus is right that this would create serious limitations for what could be justified to the political community at large.

    It’s for this reason that I think we can back away from the need for an “oracle” kind of model that Landemore and Page appeal to. We don’t need an oracular central planner, we need a more Hayekian look at the flow and use of information. A key idea here is that we don’t need common knowledge or common methods of evaluation, but instead there needs to be sufficient overlaps of various kinds that we can glean the information we need from people whose efforts might be quite different from our own.

    Gaus continues his sharp analysis of my approach to objectivity (or perhaps more mildly, intersubjective agreement) for the purposes of public justification by offering an 8-step reconstruction of my “View from Everywhere” approach. I think what he’s laid out is quite right, with the exception of 2, which I think is my fault. He notes that on p. 55, I say, “Robustness against alternatives is the sole check the view has against moral relativism.” I was too quick here in the book, as my intention was to express the idea that robustness against alternatives helps prevent the resulting political outcome from being considered relativist, not that it excludes relativist positions from the public. What I was concerned with was that since the theory doesn’t specify a particular end state, and what trajectory any given society takes with this theory would depend on the details of the perspectival mix in the population, a ready critique was that this was just a theory that would just be a random walk. To put it in evolutionary terms, it would be mostly drift rather than selection. So, what I (clearly poorly) meant to express was that there were guardrails for the process itself to be responsive to evidence, rather than the idea that the process would or should fail to account for the evidence evinced by relativists.

    All that said, I think Gaus’s next concern, that the “Master Epistemic Principle” is going to be less compelling to people who see evidence of others disagreeing with them as nothing more than evidence that others are wrong, is just to a large extent a bullet I have to bite. But I don’t think that, at least compared to rival liberal accounts, the bullet is all that big. If we take seriously the Christian example that Gaus introduces, it would seem that, for the holder of such a perspective, the only fully justified outcome would be some kind of integralist theocracy, which would be unacceptable to everyone else. If the Christian, on the other hand, is looking to find rules that we can all live by, Christians and non-Christians alike, where the start of that process is trying to find reliable touchstones that are supported by many others in the political community, then the approach I lay out is quite reasonable. Here is where the softer idea of intersubjective agreement, as opposed to objectivity, might help. The goal in the view of everywhere is to find stable points of agreement prior to bargaining. As the procedure iterates and we come to gain new social evidence, we might find that those points of agreement change. We don’t need to respond to the Christian that we are seeking to identify the truth about morality, we can simply say that we are looking for stable rules that can be justified across perspectives. It is true that I think it far more likely that cross-perspectival agreement is much likelier to indicate a firmer foundation for moral beliefs, but that doesn’t have to be part of the justificatory story.

    I had never even considered Gaus’s worry that this system will favor the intelligentsia, at least a subset of whose business is to come up with new perspectives. Perhaps, but that will likewise be weighed against religious sects that deviate in interpretation, cultural and social movements that develop new modes of social understanding, and so on. I think Gaus’s worry is perhaps strengthened by suggesting that rather than specifically favoring the intelligentsia, it likely disfavors conservatives. Broadly speaking, they disfavor new perspectives and want to rely on well-established ones. My account gives them full ability to participate, of course, but the theory I develop is one that prioritizes dynamism and diversity, and so this is going to be less appealing to a conservative. They still will participate in a bargain in which they are equals, and so can ensure that they take themselves to be better off, but as Gaus argues, they are likely to face a large number of novel perspectives. I’ve argued in “The Paradox of Diversity” that we need to devote more serious attention to making sure that conservatives see the benefits of a more diverse society, or have a way to locally preserve a bit more of the stability that they prefer.

    Gaus is right to say that I push my critique of Rawls a bit too far when I call his approach a priori. But, as Gaus points out, this is perhaps an overstatement of my core worry that Rawls requires too much normalization for a shared citizen perspective, and that a lot of the action is left out of the model—namely, all of the experience and learning that led us to favor some reasons over others. It’s too easy for the biases or preferences of the theorist to just get baked in under the guise of a shared view of idealized citizens. By pulling discovery into the justificatory framework, we can avoid this problem, and better find out the terms on which we can peaceably live together as equals.

Verified by ExactMetrics