1. Introduction
When trying to negotiate how to live with others with whom we disagree, there are two helpful philosophical debates that could inform our response. First, in political philosophy, John Rawls places the challenge of reasonable disagreement between citizens as the central challenge for contemporary theorists. The framing question Rawls asks in Political Liberalism (PL) is: “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?” (Rawls Reference Rawls2005: 4). Second, in epistemology, theorists have investigated how someone should rationally respond when they become aware that an epistemic peer disagrees with them on a particular issue.
Often these two literatures have proceeded independently of each other. Indeed, Rawlsians have been critiqued for their failure to adequately engage the controversial epistemic commitments embedded in political liberalism (e.g., Raz Reference Raz1990, Wenar Reference Wenar1995, Estlund Reference Estlund, Norris and Elkins2012, and Enoch Reference Enoch, Sobel, Valentine and Wall2017). In response, some philosophers have eschewed Rawls’s reticence on epistemology and argued that political liberalism’s core commitment to public reason is (or should be) ultimately grounded in some core epistemic commitments (e.g., Gaus Reference Gaus1996, Leland and van Wietmarschen Reference Leland and van Wietmarschen2012, and Peter Reference Peter2013). In this article, I will focus on attempts to merge the epistemology of disagreement with political liberalism.
Despite the seemingly close connection between these two debates, I will argue that merging the epistemology of disagreement with Rawlsian political liberalism is less helpful than it appears. There are good reasons to refrain from using the epistemology of disagreement literature to develop the foundations of public reason. Instead, I will argue that Rawlsian epistemic abstinence about individuals’ beliefs is the more defensible stance.
To show where attempts to draw out the epistemology of Rawlsian disagreement go wrong, I will first (§2) pull apart three separate sets of questions that we face when investigating political disagreement, and show which questions are the starting points for epistemologists and political philosophers. I will then (§3) survey recent examples of scholars who merge the epistemology of disagreement with Rawlsian political liberalism. I raise two challenges to the ways scholars have merged the epistemology of disagreement with Rawls’s political liberalism. First, I argue that there are good reasons for Rawlsians to remain abstinent about reasonable citizens’ epistemic commitments (§4). Second, I suggest reasons for caution in applying lessons from the epistemology of disagreement to cases of political disagreement (§5), showing that there are key differences between how peer disagreement is conceived in each literature. I conclude (§6) by suggesting Rawlsian epistemic abstinence carries lessons beyond Rawls’s work when navigating political disagreements in contemporary life.
2. Three Types of Questions About Political Disagreement
I will focus on three sets of questions that arise when trying to responsibly navigate political disagreement: (i) the impact of disagreement on individuals’ own beliefs; (ii) the impact of disagreement on individuals’ actions; and (iii) the ways societies ought to manage disagreements within a political community.
Regarding (i): when encountering disagreement, an individual might ask how it should impact their own views. This question is at the heart of recent work in the epistemology of disagreement. To address it, one must first determine what makes disagreement epistemically significant. There might be instances in which individuals need to reconsider their own position (or their confidence in their position), and there might be instances in which individuals can justifiably hold fast to their view (see, e.g. Feldman & Warfield Reference Feldman and Warfield2010 and Christensen & Lackey Reference Christensen and Lackey2013). Matters become even more complicated in a political community. Should the number of people who hold a position in a disagreement have an impact on the justification of individuals’ beliefs (Lackey Reference Lackey, Christensen and Lackey2013)? In short, the first set of questions about political disagreements investigates whether, and if so how, individuals’ beliefs ought to change when they discover that other appropriately qualified people disagree.
The second (ii) set of questions related to political disagreement concern how individuals should treat those with whom they disagree. The impact of individuals’ beliefs on their actions is a central question that lies at the intersection of epistemology and political philosophy. People’s beliefs about which issues are politically relevant, which policies should be adopted, and which values should guide government action can impact their political actions, including their voting, political advocacy, and conversations with friends about politics. An individual, thus, must make a choice about how their beliefs should impact their political actions and their treatment of fellow citizens. Furthermore, one’s epistemic motivations for discovering the truth or acquiring knowledge may not align with the moral motivations of finding ways to cooperate with people who hold different views.
While these first two sets of questions ask what an individual should believe and how they should act, a third set of questions zooms out from the individual and instead seeks to manage disagreements within a political community. This approach, which is at the heart of debates about disagreement in political philosophy, starts from the collective (rather than individual) perspective. Political communities are typically made up of members who have deep and persistent disagreements among themselves about a wide variety of issues. Laws and policies apply to citizens at large, many of whom will disagree with at least some of the laws, policies, and policy-makers governing the political community. Political philosophers tackle the challenging question: how should a political community manage disagreement among its members? This question brings with it challenges for securing the legitimacy of laws and policies. The task of securing just, stable, and legitimate power in a society of citizens “who remain profoundly divided” by reasonable disagreements tied to their deeper moral, religious, and philosophical commitments is the central challenge that motivates Rawls’s turn to political liberalism (Rawls Reference Rawls2005: 4).
Recently, some philosophers have attempted to merge these three sets of questions as they arise in Rawlsian political liberalism. In the next section, I will investigate prominent examples of scholars who take this approach. In subsequent sections, I will return to the different categories of questions we face in the context of political disagreement to show why developing an epistemology of Rawlsian disagreement is less straightforward than these scholars suggest.
3. Developing the Epistemic Foundations of Political Liberalism
Some scholars have turned to the epistemology of disagreement to develop the epistemic foundations for political liberalism. Rawls argues that in the context of deep, persistent, reasonable disagreement, we ought to refrain from appealing to our controversial sectarian doctrinesFootnote 1 and instead justify political power using public reasons (Rawls Reference Rawls2005: xliv, 137, 216–220, 388–390). Citizens should bracket their beliefs about their sectarian commitments when engaged in political justification in recognition that they are too controversial to serve as a foundation for justifying coercive power. This is a demanding requirement that has been subject to extensive debate. In this section, I’ll investigate three prominent examples of how scholars have merged the epistemology of disagreement with Rawlsian political liberalism to answer the question: why should reasonable citizens bracket their sectarian beliefs when so much is at stake in political life? But first, we’ll look briefly at the way epistemologists have been tackling disagreement.
3.1. The Epistemology of Disagreement
The epistemology of disagreement literature is focused on how disagreement impacts the epistemic status of individuals’ beliefs. Of course, not every disagreement is an epistemic problem. Disagreement with someone who knows far less about a topic puts little epistemic pressure on an individual to adjust their beliefs. By contrast disagreements with experts or epistemic peers may challenge one’s views. Much of the epistemology literature focuses on disagreements between epistemic peers. Epistemic peers are roughly defined as individuals who are intellectual equals, have access to the same evidence, have roughly equal epistemic virtues, and there are no other defeaters or distorting factors impacting their assessment of the proposition in dispute (see, e.g., Christensen Reference Christensen2009: 756–757). The idealizations are meant to ensure epistemic symmetry between the parties to the disagreement with regard to the disputed proposition.Footnote 2
In the epistemology literature, an individual’s response to discovering that her epistemic peer disagrees fall into two broad camps (with many nuanced views straddling intermediary territory). These two broad camps differ on whether peer disagreement is epistemically significant. If I am justified in believing both proposition P and that my epistemic peer has a justified belief in not-P, I should either adjust my own confidence in P because the disagreement is epistemically significant (conciliationism) (see, e.g., Christensen Reference Christensen2007) or remain steadfast in my convictions because the disagreement is epistemically irrelevant (steadfastness) (see, e.g., Kelly 2005 and Wedgwood Reference Wedgwood2007). A wide range of stances span these two poles, varying on the extent to which one should conciliate,Footnote 3 what it means to be epistemic peers,Footnote 4 whether one or more epistemic stances could be rationally justified (Kopec and Titelbaum Reference Kopec and Titelbaum2016), whether disagreement offers additional (higher-order) evidence that one or both parties are wrong (see e.g., Feldman Reference Feldman2009, Kelly Reference Kelly, Feldman and Warfield2010, Kappel and Andersen Reference Kappel and Andersen2019), or whether disagreement gives one a potential defeater to one’s justification for their views (van Wietmarschen Reference van Wietmarschen2013 and Lackey 2019). But within these debates, the key question is determining whether and why disagreement with an epistemic peer is epistemically weighty enough to require an individual to change their credence in their views.Footnote 5
Let’s imagine John is justified in believing some proposition P and his epistemic peer, Susan, is justified in believing not-P. Once John and Susan discover their disagreement, both are confronted with the option to remain steadfast in their convictions or adjust their confidence in their beliefs in some way. Epistemologists often turn to relevantly different examples to motivate intuitions in the direction of each view (Simpson Reference Simpson2013). Conciliatory views suggest that until further evidence is revealed (independent of the disagreement itself) both John and Susan should give some weight to their peer’s position and lower their confidence in their initial beliefs. Classic examples used to motivate conciliatory views appeal to simple cases of assessing peerhood in the matter at hand. For example, calculating a tip at a restaurant (adapted from Christensen Reference Christensen2007). If John and Susan are both competent at basic math, neither is drunk, and they come to a different total, this disagreement undermines each party’s initial justification for their belief.Footnote 6 By contrast, steadfast views tend to appeal to more complex cases, like disagreements about moral, philosophical, and religious views. Instead of disagreeing about the bill, let’s imagine John and Susan disagree about the gendered division of labor. Steadfast views hold that peer disagreement need not change a person’s confidence in their justified beliefs (see, e.g., Kelly Reference Kelly, Feldman and Warfield2010, Reference Kelly, Christensen and Lackey2013; Goldman Reference Goldman, Feldman and Warfield2010; Sosa Reference Sosa, Haddock, Millar and Pritchard2010). John and Susan can each remain justified in retaining their initial credence levels despite acknowledging that they are both epistemic peers and disagree.
3.2. Merging the Epistemology of Peer Disagreement with Political Liberalism
Recently, political philosophers have drawn insights from the epistemology of disagreement and applied them to Rawlsian political liberalism. Yet, as we’ll see, there are different ways of drawing the literatures together. In this section, I’ll briefly survey three examples of scholars who merge the epistemology literature with Rawlsian political liberalism.
In Political Liberalism, Rawls seeks ways to rise above disagreements in political life and find fair terms of cooperation to govern a political society without requiring citizens to give up on or come to agreement about controversial sectarian commitments. Rawls argues that in the context of persistent reasonable disagreement (inevitable in any society protecting basic liberties), the justification of political power should not rest on controversial sectarian commitments. He argues that we should “recognize the practical impossibility” of agreeing on the deep truth of sectarian views in a political society that includes diverse moral, religious, and philosophical views (Rawls Reference Rawls2005: 63). In this context of deep disagreements, Rawls argues that legitimacy requires appealing to public reasons (rather than the full truth of one’s sectarian views) when justifying the coercive power of the government concerning “constitutional essentials and matters of basic justice” in the “public political forum” (Rawls Reference Rawls2005: 137, 442–443). Public reasons are shared by equal citizens insofar as they are derived from “a family of reasonable political conceptions of justice” (Rawls Reference Rawls2005: 442). Rawls explains, citizens have a moral duty to “explain to one another” on fundamental questions of justice and constitutional rights “how the principles and policies they advocate and vote for can be supported by the political values of public reason” (Rawls Reference Rawls2005: 217). He calls this duty to justify political power in terms of public reason, “the liberal principle of legitimacy” (Rawls Reference Rawls2005: 217).
However, a key challenge for political liberalism is defending the limitations on political justification imposed by public reason. To answer this challenge, several scholars have turned to the epistemology literature to explain why citizens ought to appeal to public rather than sectarian reasons in political justification.
R.J. Leland and Han van Wietmarschen (2012) have argued that political liberalism requires reasonable citizens to adopt a demanding form of intellectual modesty. They argue that citizens’ restraint in public reason would be strongest if each citizen believed that sectarian views were subject to disagreement at the highest levels of epistemic competence—that even the best reasoners disagree (Leland and van Wietmarschen Reference Leland and van Wietmarschen2012: 733). Without this intellectual modesty, citizens might mistakenly appeal to their sectarian commitments if they believe these commitments are uncontroversial among epistemically competent citizens. Thus, to show respect for one’s fellow citizens’ epistemic capacities and refrain from appealing to one’s sectarian doctrines in public reason, they argue, would require reasonable citizens to be intellectually modest.
Van Wietmarschen combines this view of reasonable citizens’ intellectual humility with a conciliatory view of peer disagreement, according to which peer disagreement undermines one’s justification because it shows that one’s belief is no longer well-grounded, even if each party has a body of evidence available to them supporting their stance (van Wietmarschen Reference van Wietmarschen2013). He presents a conditional argument showing that if a conciliatory view of peer disagreement is true, it poses a new skeptical problem for political liberals—the justificatory incoherence of reasonable citizens’ commitments (van Wietmarschen Reference van Wietmarschen2018: 487). He argues that peer disagreement either undermines citizens’ justification for their sectarian beliefs or it undermines their “belief that [their] sectarian beliefs are subject to peer disagreement” (van Wietmarschen Reference van Wietmarschen2018: 499). Either case demonstrates that a conciliatory stance on peer disagreement leads to skeptical problems for political liberalism.
If conciliatory views pose skeptical problems for political liberalism, perhaps steadfast positions will be more promising. Fabienne Peter develops an epistemic argument designed to explain why reasonable disagreement requires public justification (Peter Reference Peter2013). She defends an “Opacity View” of disagreement, drawing on the steadfast views of Ernest Sosa (Reference Sosa, Haddock, Millar and Pritchard2010) and Alvin Goldman (Reference Goldman, Feldman and Warfield2010) who seek to explain why it is epistemically reasonable for epistemic peers to disagree. Peter argues that reasonable disagreement is possible when “epistemic peers struggle to access and share evidence that is relevant for the justification of the beliefs they hold” (Peter Reference Peter2013: 611). People’s politically relevant beliefs can develop in ways that may not be fully transparent to them (Peter Reference Peter2013: 609, following Sosa Reference Sosa, Haddock, Millar and Pritchard2010). Thus, even good faith efforts at explaining one’s beliefs may fall short because people are often not able to access or share all their relevant reasons (Peter Reference Peter2013: 610–611). Furthermore, even if two people consider the same body of evidence, they may still be justified in holding mutually incompatible beliefs due to differences in their epistemic systems (Peter Reference Peter2013: 610, following Goldman Reference Goldman, Feldman and Warfield2010). Given these reasonable limitations in our ability to resolve our disagreements, “we have reason to be cautious about the epistemic authority that our beliefs about fundamental moral and/or religious truths give rise to” (Peter Reference Peter2013: 615). From this, Peter concludes that “the appropriate response to reasonable disagreement … is to accept pluralism” and embrace Rawls’s liberal principle of legitimacy (Peter Reference Peter2013: 618). Peter’s epistemic argument is designed to explain why legitimacy requires appealing to public reasons in political justification.
So far, we’ve seen two very different ways of drawing out the epistemic commitments embedded in public reason: appealing to intellectual modesty and appealing to limitations on accessing and sharing evidence. Not only does each rest on controversial epistemic commitments, they draw on incompatible theories from social epistemology (conciliatory and steadfast views of peer disagreement) to explain citizens’ commitment to public reason.
According to David Enoch, these incompatible and controversial epistemic stances are problematic for public reason theorists because political liberalism seeks to remain “independent of such ‘metaphysical’ theses” (Enoch Reference Enoch, Sobel, Valentine and Wall2017: 157). Enoch critiques public reason theorists for failing to develop or defend the many epistemological claims embedded in their theories (Enoch Reference Enoch, Sobel, Valentine and Wall2017). He draws out various possible epistemic commitments of public reason and argues they are highly controversial and largely indefensible (Enoch Reference Enoch, Sobel, Valentine and Wall2017: 132–133; see Tahzib Reference Tahzib2023 for a critique of Enoch’s view). The challenge, according to Enoch, is to determine what a reasonable person would need to believe to be able to continue to endorse their private sectarian commitments and see that these sectarian commitments cannot hold justificatory force for other reasonable citizens in public justification (Enoch Reference Enoch, Sobel, Valentine and Wall2017: 137). The core problem lies in the inconsistent epistemic positions an individual would need to endorse to embrace both their sectarian view and public reason. Either the epistemic commitments needed to embrace public reason would undermine any plausible way that citizens could continue to embrace their own comprehensive doctrine as true. Or, if reasonable citizens continue to endorse their own doctrine as true, this commitment would undermine their epistemic reasons for endorsing public reason. In short, Enoch argues, “it is very hard to make coherent sense of the epistemic-sounding commitments of public reason theorists” (Enoch Reference Enoch, Sobel, Valentine and Wall2017: 158).
I agree with Enoch that there need to be plausible, coherent ways to defend a reasonable person’s commitment to both their sectarian views and public reason. However, rather than doing more epistemology, I will argue that Rawls was right to remain silent about the epistemic commitments of individuals’ beliefs. Enoch’s tentative conclusion from his analysis of public reasons’ epistemic commitments is that public reason might be better off not relying on epistemology at all. However, stripping public reason of its epistemic-sounding commitments poses substantial challenges (Enoch Reference Enoch, Sobel, Valentine and Wall2017: 160). In what follows, I take the first step in this direction by defending a form of Rawlsian epistemic abstinence about reasonable citizens’ epistemic states.
4. A Case for Rawlsian Epistemic Abstinence
As we’ve seen, whether they are critiquing public reason or explaining the epistemic commitments that undergird a reasonable citizen’s commitment to public reason, the attempts discussed herein seek to apply the peer disagreement literature in epistemology to political liberalism. I will raise several challenges to this way of merging the epistemology of peer disagreement with Rawls’s Political Liberalism (2005). First, I will argue that there are good reasons for Rawls to remain abstinent about the epistemic commitments of reasonable citizens. To defend a renewed case for Rawlsian epistemic abstinence, I will briefly examine Rawls’s initial case for restricting appeals to the truth in political liberalism (§4.1). I will then show why a limited form of Rawlsian epistemic abstinence has an important role to play in political liberalism (§4.2) and argue that reasonable disagreement is best understood as disagreement between morally reasonable citizens (§4.3). Second (§5), I will suggest that merging the epistemology of peer disagreement with Rawls’s work is less straightforward than it appears due to key differences between the aims of each project and the idealizations used to specify peerhood in each case.
4.1. Rawls’s Case for Epistemic Abstinence
Before turning to my renewed defense of Rawlsian epistemic abstinence, let’s examine Rawls’s initial case for bracketing appeals to the truth and remaining silent about epistemic questions in political liberalism. When Rawls claims to limit “the truth” or make no use of “the truth,” this is generally shorthand for limiting appeals to the truth of sectarian doctrines in public justification. The shorthand commonly used in the literature on this front can lead to confusion because not all truths are excluded from political liberalism.Footnote 7
A key element of Rawls’s strategy in establishing fair terms of cooperation to govern a society characterized by reasonable disagreement is the limitation he places on appeals to the full truth of sectarian doctrines in political liberalism. Rawls explains, “while people can recognize everyone else’s comprehensive views as reasonable, they cannot recognize them all as true, and there is no shared public basis to distinguish true beliefs from the false” (Rawls Reference Rawls2005: 127). We should refrain from appealing to sectarian commitments in political justification because their truth is controversial among reasonable people. Instead, we should appeal to public reasons that are shareable by reasonable citizens. Citizens have a moral duty to justify political power in terms of public reasons because political power is “justifiable only when it is exercised in accordance with a constitution the essentials of which all citizens my reasonably be expected to endorse in the light of principles and ideals acceptable to them as reasonable and rational” (Rawls Reference Rawls2005: 217). Rawls explains, the concept of ‘reasonable’ should replace ‘truth’ as the “appropriate standard of correctness” for evaluating political conceptions of justice (Rawls Reference Rawls2005: 126).
However, in bracketing appeals to the full truth of sectarian doctrines for the purpose of political justification, Rawls does not deny that truth is relevant to issues of morality and politics (Rawls Reference Rawls2005: 150). Instead, limitations are set on appeals to the whole truth to ensure matters of constitutional essentials and basic justice are justified on the basis of publicly accessible reasons (Rawls Reference Rawls2005: 217). Foundational to Rawls’s theory is that political legitimacy is grounded in an ability to justify political principles in terms of reasons that “all might be reasonably expected to endorse” (Rawls Reference Rawls2005: 243). These limitations on appeals to the truth are designed to ensure that political power can be recognized as legitimate by people who disagree about the truth of foundational moral claims. In this way, Rawls argues that political liberalism does not “use (or deny) the concept of truth” rather, it “does without the concept of truth” (Rawls Reference Rawls2005: 94). The truth of comprehensive doctrines is simply not an appropriate basis for political justification in PL.
4.2. A Renewed Call for Epistemic Abstinence
Rawls’s epistemic abstinence holds that appeals to the whole truth of sectarian doctrines should be set aside to secure the legitimacy of coercive power in a society characterized by reasonable pluralism. This is not without its challenges.Footnote 8 Rawls acknowledges that the limitations placed on appealing to the full truth in public reason is “to many … a basic difficulty with the idea of public reason” that even “makes it seems paradoxical” (Rawls Reference Rawls2005: 216). Rawls defends the epistemic abstinence of public reason against its critics by appealing to moral reasons, explaining that limitations on the truth in public reason are required to support citizens’ basic rights and other central public goods in the context of reasonable pluralism (Rawls Reference Rawls2005: 219).
The renewed Rawlsian epistemic abstinence that I defend herein employs a similar strategy for a different locus of epistemic abstinence. Just as Rawls argues that limitations on appealing to the full truth of sectarian doctrines in political justification is required to support citizens’ basic rights in the context of reasonable pluralism, there is a similar moral argument available to motivate abstinence about reasonable citizens’ epistemic response to disagreement. Reasonable citizens may adopt differing epistemic stances regarding the epistemic significance of political disagreement.
Rawls defends his epistemic abstinence by drawing analogies with other contexts in which “we recognize a duty not to decide in view of the whole truth so as to honor a right or duty, or to advance an ideal good, or both” (Rawls Reference Rawls2005: 219). He reminds readers that there are “familiar cases where we grant that we should not appeal to the whole truth as we see it, even when it might be readily available” (Rawls Reference Rawls2005: 218). For example, to protect the rights of the accused to a fair trial, we limit the kinds of evidence that are permitted in criminal proceedings, excluding evidence from improper search and seizure, evidence gained by abusing defendants, and evidence gained by failing to inform defendants of their rights (Rawls Reference Rawls2005: 218).
There are two lessons to draw from Rawls’s argument. First, he argues that epistemic abstinence in public reason helps support citizens’ rights and secure public values (Rawls Reference Rawls2005: 253). Citizens should demonstrate respect for one another’s ability to reason responsibly and direct their lives in accordance with their own views. Rawls explains that citizens have a moral duty of civility, which asks citizens to explain their political advocacy in terms of public reasons, “listen to others,” and exhibit “fairmindedness in deciding when accommodations to their views should reasonably be made” (Rawls Reference Rawls2005: 217). While the willingness to listen to others and accommodate their views may be partially epistemic, it is motivated by a moral requirement of respect. He explains that reciprocity in public reason is important for securing the right kinds of political relations between citizens, ones of civic friendship, and shapes our institutions in ways that protect basic rights and liberties (Rawls Reference Rawls2005: xlix). Rawls also emphasizes the ways political and moral rights and values often place constraints on the types of reasoning and rules of evidence for a wide variety of exercises of our reason—this is not unique to the context of public reason (Rawls Reference Rawls2005: 220–221). Rawls argues that just as judges should not resolve difficult cases by appealing to their own political views, so too, citizens have a moral duty of civility to rely on public reasons when deciding on constitutional essentials and basic justice (PL: lii-liii). The duty of civility is a moral duty that overrides epistemic concerns.Footnote 9
The second lesson to highlight from Rawls’s defense is that the full justification for epistemic abstinence in public reason will come from within different reasonable comprehensive doctrines. Rawls does not give a full justification for epistemic abstinence because he expects that it may be justified differently by different reasonable comprehensive doctrines (Rawls Reference Rawls2005: 219). I extend this reasoning to show that different individuals may similarly be persuaded by different ways to justify their commitment to public reason. Some may be persuaded by moral reasons alone, others may turn to epistemic arguments, but there is no need for all citizens to commit to a single epistemic justification for their commitment to public reason alongside their sectarian views.Footnote 10
Rather than think that political liberalism hinges on the outcome of controversial epistemic stances on peer disagreement, political liberals should instead embrace a limited form of epistemic abstinence when it comes to evaluating the beliefs of individuals. Political liberal theory should not endorse a specific epistemic argument to explain reasonable citizens’ epistemic commitments. On this question, the theory should remain silent, leaving it up to particular individuals and comprehensive doctrines to fill in their preferred epistemic arguments, if any. There is no need to commit to any single controversial view about the implications of peer disagreement on a citizen’s own beliefs. A clear lesson to draw from a broad survey of this literature is that there are many different compelling explanations for what could motivate an epistemically responsible citizen to embrace public reason. Politically reasonable citizens can remain steadfast or conciliate—both are compatible with political liberalism because reasonable pluralism is not primarily about evaluating citizens’ private beliefs. It is about finding just and legitimate terms of cooperation to govern a diverse society. Rawls need not commit to just one epistemic response to disagreement just as he need not commit to one answer to how the central values of public reason are supported within each citizen’s comprehensive doctrine.
4.3. A Moral Basis for Reasonable Disagreement
Rawls appeals to the “burdens of judgment” to help explain what makes a disagreement reasonable (Rawls Reference Rawls2005: 55). Many argue that this embeds controversial epistemic commitments into political liberalism. In an early critique of PL, Leif Wenar argues that including the “burdens of judgment” as a qualification for reasonableness makes the theory too controversial to be a purely political conception of justice (Reference Wenar1995: 41–48). In particular, he argues that many religions will reject the burdens of judgment because they believe the truth of their doctrine is broadly accessible to those with “clear minds and open hearts” (Wenar Reference Wenar1995: 46) and will view the rise of pluralism as “’progress’ away from the truth” (Wenar Reference Wenar1995: 47–48). But does Rawls’s specification of reasonable pluralism and the burdens of judgment really require the significant epistemic commitments Wenar suggests are incompatible with many religious beliefs? Or epistemic commitments that are best explained by incorporating a controversial epistemological theory?
Far from requiring a specific epistemic theory to be accepted, I will argue that determining which disagreements qualify as reasonable disagreements for political purposes should be determined primarily on moral grounds, eschewing controversial epistemic commitments. I will show that Rawls’s own discussion of the burdens of judgment is at least consistent with remaining agnostic about which epistemic commitments a reasonable person must accept. I will also argue that Rawlsians should return to a limited form of epistemic abstinence about citizens’ private beliefs.
Rawls offers several explanations for why reasonable people may disagree, but remains agnostic in spelling out the detailed epistemic stance of reasonable individuals. Instead, he appeals to the “burdens of judgment” to offer a variety of conjectures about what could cause disagreement between people in a way that does not impugn the reasonableness of those who disagree (Rawls Reference Rawls2005: 55–56). Yet the reasonableness of the disagreement is defined in moral and not epistemic terms. The burdens of judgment are an account of “the sources, or causes, of disagreement between reasonable persons,” which Rawls defines as persons who possess the two moral powers “sufficient to be free and equal citizens” and “who have an enduring desire to honor fair terms of cooperation” (Rawls Reference Rawls2005: 55). These qualifications refer to Rawls’s moral qualification outlined in the “first basic aspect of reasonableness” (Rawls Reference Rawls2005: 48–54).
The burdens of judgment explain why citizens who already qualify as morally reasonable come to disagree. When defining the burdens of judgment, Rawls gives a variety of conjectures about how morally responsible people could come to different judgments in political life (Rawls Reference Rawls2005: 55–56). He explains, the burdens of judgment are not intended to be “a philosophical analysis of the conditions of knowledge,” rather they “simply list some of the circumstances that make political agreement … far more difficult” (Rawls Reference Rawls2005: 63). Rawls includes a range of familiar challenges that arise in making difficult political decisions when explaining the burdens of judgment. For example, the complexity of evidence, disagreements about which evidence is relevant, and disagreement about the weight of different normative and empirical considerations (Rawls Reference Rawls2005: 56–57). Rawls also emphasizes that “citizens’ total experiences are disparate enough for their judgments to diverge” on political questions because their life experiences shape their interpretations of moral and political concepts and the ways they assess and weigh available evidence and values (Rawls Reference Rawls2005: 56–57). The burdens of judgment are presented as an incomplete list of some of “the more obvious sources” of disagreement that can occur among morally reasonable people deliberating about politics (Rawls Reference Rawls2005: 56). Contra Wenar, I think this is best interpreted as an open list of suggestions that could explain why conscientious, morally responsible citizens may not always agree, rather than requiring reasonable citizens to endorse these as substantive epistemic commitments.
There are some epistemic elements in the way Rawls specifies disagreement between morally reasonable persons. However, these epistemic elements are simply capacities of morally reasonable people to engage in basic skills of reasoning, including their “powers of thought and judgement” and their ability to “draw inferences, weigh evidence, and balance competing considerations” (Rawls Reference Rawls2005: 55). These epistemic powers are not controversial epistemic stances (Rawls Reference Rawls2005: 62).
The burdens of judgment are meant to explain “the many hazards involved in the correct (and conscientious) exercise of our powers of reason and judgment in the ordinary course of political life” (Rawls Reference Rawls2005: 56). As Rawls emphasizes, in many areas of inquiry our correct and conscientious use of reasoning does not settle all disputes, this is not unique to moral and political inquiry (Rawls Reference Rawls2005: lii, 58). While each individual is free to judge others as mistaken or wrong, they qualify as reasonable insofar as they meet the “minimal conditions” of reasonableness (Rawls Reference Rawls2005 : 59–60).
What makes disagreements reasonable is that they occur between people who are sincerely attempting to engage in fair terms of cooperation that show recognition respect for their fellow citizens (Rawls Reference Rawls2005: 55–56, 58).Footnote 11 This is contrasted with unreasonable disagreements that arise from “prejudice and bias, self- and group interest, blindness and willfulness” (Rawls Reference Rawls2005: 58). It is also contrasted with disagreements that arise from people advocating for their narrow interests or making logical errors in their reasoning (Rawls Reference Rawls2005: 55). He argues that it “arouses mutual suspicion and hostility—to suppose that all our differences are rooted solely in ignorance and perversity, or else in the rivalries for power, status, or economic gain” (Rawls Reference Rawls2005: 58). As Rawls explains, pointing to self-interest, epistemic errors, or differences in intellectual abilities are reasons of the wrong kind when explaining why reasonable citizens disagree (Rawls Reference Rawls2005: 55).
I think there is strong evidence suggesting that Rawls remained agnostic about the epistemic commitments of reasonable citizens. But even if you reject this interpretation of Rawls—there are good reasons for political liberals to adopt a limited form of agnosticism about the epistemic reasons why people come to hold the variety of reasonable views they do.
5. Key Differences Between the Epistemology of Disagreement and Reasonable Disagreement
At this point, we have seen various attempts to apply the peer disagreement literature in epistemology to political liberalism. One way of understanding this project is to develop the epistemic commitments of reasonable citizens. I’ve suggested that Rawls has good reasons to remain epistemically abstinent about individuals’ beliefs, allowing different reasonable individuals to adopt different epistemic explanations when reconciling their sectarian beliefs and commitment to public reason. However, the success of applying the peer disagreement literature to political liberalism will hinge on how applicable peer disagreement is to the question of reasonable disagreement among citizens. In this section, I raise a further challenge in applying the epistemology of disagreement to political liberalism. These fields frame their questions at different levels of argument and use different idealizations in ways that challenge the applicability of epistemic peer disagreement to political liberalism (Edenberg Reference Edenberg, Hannon and de Ridder2021). In §5.1 I’ll focus on the different questions and levels of argument used in epistemology and political liberalism. In §5.2 I will highlight crucial differences between the epistemic symmetry of epistemic peers and the epistemic diversity of reasonable citizens.
5.1. Different Framing Questions and Idealizations
There are important differences between the ways epistemologists and political philosophers frame the problem of disagreement. While epistemic and moral disagreements are tightly connected in political disagreements, there are several relevantly different questions at stake that must be distinguished (see §2). There are epistemic questions about what individuals should believe, practical questions about how individuals should act, and political questions about how disagreements should be managed among many individuals jointly engaged in politics. An answer to one type of question may not readily transfer to another. Furthermore, each way of framing the problem of disagreement comes with its own set of assumptions and idealizations appropriate for the question at hand.
The epistemology literature concerns what an individual is justified in believing about a particular proposition. The core epistemic task is determining how to rationally resolve disagreement between two epistemic peers. What does responsible epistemic agency require: adjusting one’s confidence in one’s views, suspending judgment about the question at hand, or remaining steadfast in one’s own position? Could a given body of evidence justify one or more epistemic states? However, it is essential to highlight that these epistemic questions arise when faced with an epistemic peer who disagrees. It is not clear that we can transfer the implications of the epistemology of disagreement to political life unless we are justified in thinking reasonable citizens are our epistemic peers.Footnote 12
In contrast, Rawls frames disagreement as a permanent fact of society. The aim is not to get everyone to agree with one’s political or moral views. Nor is the aim for each individual to epistemically reconcile their beliefs in light of discovering disagreement.Footnote 13 Instead, Rawls asks what it takes for a political conception of justice to be legitimate for a diverse population that will inevitably disagree about moral, religious, and philosophical matters that usually ground the justification of principles of justice. We need to learn how to manage disagreements that will persist over time in order to protect freedom of thought and expression. In this context, determining ways to rationally resolve disagreement is not the proper aim. We should instead find ways to set aside areas of disagreement to make political progress. Furthermore, the idealization relevant to Rawls is not epistemic peerhood but rather civic peerhood, tied to a moral qualification of recognition respect for our fellow citizens (Edenberg Reference Edenberg, Hannon and de Ridder2021). Rawls sets aside each individual’s justification for their beliefs in order to find political solutions to manage persistent disagreement among morally reasonable citizens.
In short, unlike the epistemology of disagreement, political liberalism is not a view about what individuals should believe. It is a view that tackles how society as a whole should respond to disagreement in determining the justification of political power, and how individuals should treat one another given their status as free and equal citizens.
The philosophers surveyed above would likely respond by suggesting that their task is admittedly different from Rawls’s. They are attempting to bridge the epistemic and political questions to determine what responsible individuals would need to believe in order to embrace public reason (Enoch Reference Enoch, Sobel, Valentine and Wall2017: 135). This directs the focus away from the social response to disagreement and back to the justification of an individual’s own beliefs. In this sense, the philosophers might argue, the epistemology literature is directly relevant because each individual should seek rational justification for their beliefs once they acknowledge the ‘fact of reasonable pluralism.’
As I argued in §4, I think there are good reasons for political liberalism to remain epistemically abstinent about reasonable citizens’ beliefs. While there are a range of epistemic reasons individuals could embrace to reconcile their sectarian commitments with public reason, political liberalism can leave this open to each individual to decide for themselves. Rawls argues that it is essential to be able to maintain a distinction between one’s own justified beliefs, wherein appeals to the truth are perfectly appropriate, and the justifications that could be legitimately offered to others (Rawls Reference Rawls2005: 138). This is not done for epistemic reasons, but for moral reasons because he is focused on citizens’ political actions in their treatment of one another.
Rawls emphasizes that disagreement among morally reasonable citizens changes the way we should treat one another in political life. One way to treat those with whom we disagree as equals is to appeal to public reasons to justify political power. While justification is clearly an epistemic term, Rawls’s argument for this restriction is moral. Restricting political justifications to public reasons treats all citizens with respect and protects their freedom of conscience. Rawls argues that to insist on one’s own sectarian beliefs in political justification, without a public basis for establishing their truth, can be seen as “using upon [one’s fellow citizens] unreasonable force” and citizens are justified in opposing this out of “self-defense” (Rawls Reference Rawls2005: 247). Rawls’s focus is on the treatment of citizens in a political community rather than reconciling individuals’ beliefs.
The lessons from the epistemology of disagreement may be relevant in helping particular individuals reconcile their commitment to their own views alongside their commitment to public reason. But even this question is less straightforward than is sometimes assumed because one’s fellow citizens are often not one’s epistemic peers, nor is everyone apt to assume that disagreement persists at the highest levels of epistemic competence. I turn to these asymmetries between the political philosophy and epistemology of disagreement next.
5.2. Epistemic Symmetry versus Civic Peerhood
There is another crucial difference between the way peer disagreement is framed in epistemology and the way Rawls frames reasonable disagreement. For epistemologists, establishing the symmetry of epistemic qualifications, virtues, and access to evidence is crucial for determining the epistemic significance of a peer’s disagreement. However, understanding how a rational individual should respond to disagreement with their epistemic peer may have very little bearing on how they should respond to disagreement with someone who is not their epistemic peer. When evaluating which disagreements are politically significant, the epistemology of disagreement is less helpful than it initially appears because one’s fellow citizens are often not one’s epistemic peers (Edenberg Reference Edenberg, Hannon and de Ridder2021).
This comes back to a central difference between epistemologists and Rawlsians about which peers’ disagreements pose philosophically interesting challenges. While determining epistemic peerhood is relevant when specifying the disagreements that could impact the justification for individuals’ beliefs, societies need to manage disagreements between individuals of varying epistemic qualifications and capacities. Rawls’s approach to disagreement is designed to take up the broad range of reasonable disagreements in political life. For this purpose, specifying peerhood in terms of epistemic symmetry makes little sense. Given the epistemic diversity among citizens, a better approach is a limited form of Rawlsian agnosticism about individual beliefs. Instead of specifying civic peerhood through epistemic symmetry, Rawls’s idealization specifies a moral threshold qualification for determining which disagreements demand a response—reasonable disagreements.
In restricting the scope of the political liberal project to reasonable pluralism, Rawls recognizes that conscientious citizens who are reasoning responsibly can nevertheless come to different views about morality, religion, philosophy, and how to lead a good life. Securing a just and legitimate political order must be able to accommodate these persistent reasonable disagreements. Citizens’ treatment of each other should not hinge on whether someone is their epistemic peer. Political liberalism seeks a fair basis of cooperation among free and equal citizens in the context of reasonable pluralism. For this purpose, Rawls’s minimal moral qualification of recognition respect for people’s free and equal status is the crucial qualification because it meets the minimal standards appropriate for securing political justice (PL: 60–66). Morally reasonable citizens may not be epistemic peers, but judgments about epistemic peerhood should not generally play a role in political justification.
6. Lessons for Contemporary Political Disagreement
To close, I draw five lessons from these theoretical debates that are relevant for managing contemporary political disagreement. The Rawlsian epistemic abstinence I’ve defended herein will help us begin to build bridges across political divides on terms of respect for our fellow citizens.
First, it is worth keeping distinct the separate sets of questions that can arise when assessing political disagreement. The epistemic question about individuals’ justified beliefs is something best left to individuals. Citizens may inquire whether their beliefs are justified and when political disagreements should prompt them to reconsider their own views. For this, it may be relevant to specify the epistemic qualifications and virtues of the various parties to the disagreement. However, it oversteps the appropriate boundaries of political power for a political community to assess the epistemic status of individuals’ beliefs to determine which disagreements are politically significant. Likewise, individuals can assess how their political beliefs should inform their political actions, including their political advocacy, voting, and civil disobedience. In some cases, society may legitimately impose sanctions on individuals for actions that violate social norms of civility, toleration, and respect. But these sanctions are decisions made by a political community about how citizens should be treated and not about assessing their epistemic qualifications.
Second, if political liberals embed controversial epistemic positions into the qualifications for reasonableness, it would make political philosophy hang on the outcome of debates about epistemic peer disagreement, epistemic permissivism (see, e.g. Schoenfield Reference Schoenfield2014), and pragmatic encroachment (see, e.g. Brown Reference Brown2014). None of this should be good news for political liberals. Political liberals argue for a turn to public reason in order to avoid justifications for coercive power based on controversial sectarian reasons that cannot be shared by all reasonable citizens. However, in avoiding controversial metaphysical claims one may inadvertently step into controversial epistemic waters. Rawls seeks a public political conception of justice in order to find terms of cooperation that can rise above sectarian disputes. The purpose of the political conception of justice is to “avoid, so far as possible, disputed philosophical theses and to give an account of the burdens of reason that rests on plain facts open to all” (Rawls Reference Rawls2005: 57n.10). If this aspiration needs to wait until epistemology is settled, the history of philosophy suggests we shouldn’t hold our breath.
Third, when determining which citizens qualify as reasonable for the purpose of political debate, voting, and advocacy, there are good reasons to refrain from assessing individuals’ epistemic qualifications. Assessing the justification of individuals’ beliefs is the wrong starting point for assessing the public significance of political disagreement and risks a problematically narrow depiction of which citizens’ views should count within political debate. When disagreement is an epistemic problem, the first step in resolution involves assessing the relative epistemic capacities of our interlocutors and checking the verifiability of their evidence and the soundness of their reasoning. But note the consequences of using epistemic peerhood to determine which political disagreements are significant: if I determine you are not my epistemic peer in political life, your objections to a policy might not count. I can seek to impose my view despite your objections. I might also seek to change your mind by showing you the relevant evidence and teach you the error of your ways, but if you continue to disagree and I determine you are my epistemic inferior, I need not listen to you or be concerned about political legitimacy. This is exactly the stance Rawls cautions against in PL and deems to be patently unreasonable (Rawls Reference Rawls2005: 61). We do not get to override people’s judgments simply because we think we know better.
A moral threshold notion of recognition respect for our fellow citizens as reasonable is better suited to garner the kind of respect required for our fellow citizens to be considered our civic peers. In public life, we need to respect each person’s right to direct their lives as they see fit, provided they do not harm others. Part of this right to direct one’s lives in a democratic society includes the right to have a say in the government, laws, and policies through one’s political advocacy and voting (Neufeld Reference Neufeld2022). For this, defining the reasonableness of disagreement in terms of a minimal moral threshold (rather than epistemic peerhood) is significant.
Fourth, focusing on the epistemic divisions and qualifications of our fellow citizens is often counterproductive and leads to further divisions. Political beliefs often come in clusters of related ideas, bound together by key normative commitments. People are apt to judge those who share their political views as epistemically competent and those who disagree as getting important political issues wrong. Judgments of epistemic peerhood are not likely divorced from judgments of normative peerhood and shared partisan commitments (see, e.g. Elga Reference Elga2007, Kelly Reference Kelly, Feldman and Warfield2010, and Rini Reference Rini2017). We are more likely to trust those with whom we share important political and social affiliations. Viewing political disagreement in epistemic terms further entrenches this problem. Fair cooperation should not start from an assessment of the epistemic qualifications of our fellow citizens. This is the wrong kind of stance in political life and has the effect of further dividing our already fragmented political society. Epistemic assessments are likely to track the same kinds of sectarian divisions that characterize our political divisions.
Fifth, we have good reasons to remain agnostic about the epistemic capacities of our fellow citizens in political disagreements. In a toxic political environment in which partisans have a deep distrust of the epistemic credentials of their political opponents, a focus on the epistemic foundations of disagreements will not offer the olive branch needed to begin to rebuild the broken trust in our political community. In political life, respect for our fellow citizens is critical and this needs to include respect for them as both moral and epistemic agents. However, framing this epistemic respect in terms of the epistemology of disagreement begins from the wrong place.
To garner the appropriate respect, epistemic symmetry between parties to the disagreement is neither necessary nor always desirable. In fact, given the diversity of people’s lived experiences, we should both expect and embrace the epistemic asymmetry of knowledge relevant to political governance in our society (Anderson Reference Anderson2010: 98). A person’s experiences shape their knowledge of the policies relevant to governing a large and complex society. A person who has never been subject to police brutality or racial profiling knows far less about this subject than those who experience racial profiling daily and who have been taught from a very young age how to remain calm and respectful under extreme threats of violence by the police. Likewise, my students who have grown up with food insecurity and housing instability have wisdom to bring to policy conversations around social support systems that far exceeds the lucky members of society like myself (and most of our elected officials) who have never needed to rely on government support to fill our stomachs or put a roof over our heads. This knowledge is not symmetrically distributed, but is nonetheless essential for informing public policies about how best to secure individuals’ basic needs and rights in a complex and diverse society like ours.
Far more promising is an approach to bridging political divides through finding those points of commonalities across divisions. Despite the deep conflicts over politics, morality, and politicized facts about our world, I think there is some hope for building a shared basis of trust by drawing out the shared civic values of freedom, equality, and mutual respect for our fellow citizens. For this, adopting a limited Rawlsian epistemic abstinence about the epistemic qualifications of our civic peers can be a helpful starting point for bridging the deep divisions in our political society.
Acknowledgements
My thanks to the many people who have greatly improved this article through conversation and generously critiquing early drafts: Aaron Ancell, Brian Berkey, Paul Billingham, Helen Chang, David Enoch, Marilyn Friedman, Michael Hannon, Do Lee, Emily McGill, Blain Neufeld, Lina Newton, Rahul Pathak, Joan Robinson, Alison Suen, Bryan Weber, and Han van Wietmarschen. Thanks as well to audiences at the American Philosophical Association Pacific Division, Institute of Philosophy at the University of London, Political Philosophy Workshop at VU Amsterdam, and The University of Wisconsin-Milwaukee for helpful questions as I developed this project. Thanks also to the anonymous referees and Heather Battaly whose thoughtful comments greatly improved this article.
Open access
