In Leviathan, Thomas Hobbes wrote that “the doctrine of right and wrong is perpetually disputed, both by the pen and the sword: whereas the doctrine of lines and figures is not so” (Reference Hobbes and Waller1904: 67). According to Hobbes, factual truths – such as the postulates of geometry – rarely provoke conflict because they are generally unrelated to human “ambition, profit, or lust,” the usual sources of disputes. Only if a “man’s right of dominion” became somehow dependent on a factual claim – for example, that “the three angles of a triangle should be equal to two angles of a square” (Reference Hobbes and Waller1904: 67) – Hobbes acknowledges that his previous indifference would shift to concern. To the extent that such a claim became disputed among individuals, resolving it would again require both the pen and eventually the sword.
Like Hobbes, most political theorists have often assumed that the only relevant controversies are moral ones. This assumption is especially prevalent among public reason liberals, who have sought – since their Hobbesian origins – “to identify an agreed-upon public judgment or public reason that allows us to overcome the disunity and conflict that would characterize a condition in which each followed her own private judgment or reasoning about morality and justice” (Gaus Reference Gaus and Wall2015: 112). Since these conflicts primarily arose from religious disagreement, the foundational question for public reason liberalism is “whether a society faced with intractable struggles and irreconcilable conflicts of absolute depth can share a common social and political existence on terms that are acceptable to all” (Gaus Reference Gaus and Wall2015: 113). Such intractable struggles and irreconcilable conflicts have rarely been thought to involve factual beliefs, as if factual disagreements lacked the depth that would justify their privatization, isolation, bracketing, or any other mechanism employed by public reason liberals to establish political legitimacy.
In recent years, however, a growing body of literature within liberal political theory has questioned the status of science in contemporary liberal democracies (Anderson Reference Anderson2011; Badano & Bonotti Reference Badano and Bonotti2020; Bellolio Reference Bellolio2018; Fowler Reference Fowler2019; Jønch-Clausen & Kappel Reference Jønch-Clausen and Kappel2016; Kappel Reference Kappel2021; McKinnon Reference McKinnon2012; Pamuk Reference Pamuk2021; Reid Reference Reid2019; Torcello Reference Torcello2011; Ward & Creel Reference Ward and Creel2024). Whether spurred by the creationist comeback in the form of Intelligent Design theory, the new generation of anti-vaxxers galvanized by the COVID-19 pandemic, or persistent skepticism about climate science, these authors have debated whether mainstream scientific knowledge merits any special place in the structure of public justification. But there is a prior question as to whether science works as public reason in the Rawlsian sense or whether any other epistemology serves us to justify claims of an empirical nature in political debates. That question asks whether liberal political theorists should be concerned about factual justification at all.
This paper addresses a question that, with few exceptions (Kappel Reference Kappel2017; Kogelmann & Stich Reference Kogelmann, Stich, Sobel, Vallentyne and Wall2021), has received insufficient attention in normative political theory – especially within public reason liberalism, the branch most directly concerned with justifying norms whose empirical premises are increasingly contested in an era of posttruth, alternative facts, and I-pistemologies (van Zoonen Reference Van Zoonen2012). If liberals remain committed to ensuring the legitimacy of coercive norms in pluralistic societies, and yet the relevant disagreements increasingly occur at the factual rather than at the normative level, they should also address these factual disagreements explicitly. Thus, the central question posed by this paper is whether public reason liberals have a duty to publicly justify the factual claims underpinning coercive norms, in the same manner as they feel compelled to justify their normative content. In other words, it asks whether they have what I will call a Liberal Duty of Factual Justification (LDFJ) – regardless of how precisely they would fulfill this duty, whether by appealing to mainstream science or any other epistemic framework.
The paper proceeds as follows. Sections I and II examine two main reasons why factual disagreements differ from moral disagreements, and thus why liberal duties of public justification might not extend to factual claims: the Not-Too-Important Objection and the Easy-to-Solve Objection, respectively. Sections III through V then propose three grounds for establishing a potential LDFJ: the Inseparability Argument, the Equivalence of Public Reasons Argument, and finally the Argument from Epistemic Pluralism. Section III through VI concludes by asserting the pertinence of recognizing an LDFJ.
1. The not-too-important objection
The first reason why factual disagreements have been downplayed by liberal political theorists refers to their relative irrelevance compared to the significance of moral disagreements in our lives. Returning to Hobbes, little is at stake in the “doctrine of lines and figures,” as the latter does not inflame the passions of “ambition, profit, or lust.” Going back to Gaus, the kind of “intractable struggles and irreconcilable conflicts of absolute depth” that motivated the emergence of public reason liberalism are moral through and through. Hence, it seems there is no need to employ sophisticated justificatory devices in conflicts that do not have the potential to destroy our peaceful coexistence or threaten our lifestyle. This is what I call the Not-Too-Important Objection: factual disagreements do not arouse the kind of passions that call for an LDFJ.
For instance, Harry Brighouse (Reference Brighouse1998) has argued that the rules of public justification and the liberal commitment to neutrality prohibit appeals to controversial moral claims but not to controversial empirical premises. Only the former, the argument goes, are intimately connected with our conceptions of the good. According to Brighouse, we do not find the same intimacy in the case of empirical and factual beliefs. Indeed, “while significant revision of our moral beliefs supports revision of our identities, this is not true of our empirical beliefs” (Brighouse Reference Brighouse1998: 738). In a similar vein, Andrew Lister claims that empirical controversies, such as “the dynamics of supply and demand … [do] not create the moral problem to which public reason is meant to be the solution” (Reference Lister2007: 13). Economic theories, or scientific theories for that matter, do not raise the “moral concern about legislation based on a particular conception of salvation” (Lister Reference Lister2007: 13), such as those that characterize comprehensive religious doctrines. In Lister’s view, the justificatory constraints pledged by public reason liberals only apply in matters that “speak in a fairly direct way to a basic spiritual or philosophical question having to do with fundamental values” (Reference Lister2007: 13), akin to Brighouse’s notion of “intimacy.” Thus, the LDFJ that holds in moral controversies is relaxed when it comes to factual controversies. It seems a prima facie sound reason to explain why most liberal political theorists have been hitherto unconcerned by fact-dependent policy disagreements, to wit, the assumption that factual beliefs are not constitutive of a person’s conception of the good or are rather irrelevant to their identity.
The implications of the Not-Too-Important Objection are at sight: to the extent that factual controversies do not raise the kind of moral concerns that motivate public reason liberals to resort to the rules of public justification, then fundamental laws or policies which turn out to be grounded over empirically disputed premises – or overtly crazy, for that matter – are entirely permissible. In other words, public officials and government agents can freely invoke the facts they see fit to activate or deactivate the coercive muscle of the state. From Donald Trump to Jair Bolsonaro and Javier Milei, they can all declare that they are not taking any political action against climate change as they are positive that it is a Chinese hoax, a Marxist conspiracy, or a socialist lie – as all have indeed said, respectively. Insofar as the rules of liberal justification constrain the range of permissible arguments in the discussion of relevant political issues, the absence of an LDFJ implies that there is no restriction regarding the empirical arguments that can be offered in public deliberation. Everything goes. If the Not-Too-Important Objection holds, public reason liberals have no other remedy than to put up with this – perhaps bitter – conclusion, and then redirect all their efforts to the normative side of the debate.
2. The easy-to-solve objection
Another possible explanation for why public reason liberals have not paid attention to factual justification is that empirical disagreements are easier to resolve than moral ones. Given the human tendency to favor one’s own cause – as Locke early recognized – moral conflicts are exceedingly difficult to settle, prompting public reason liberals to invent a variety of impartial procedures, umpiring devices, and public standards of adjudication. From the outset, John Rawls (Reference Rawls1951) was seeking a “decision procedure for ethics,” not a procedure to resolve disagreements over factual reality. Indeed, he appeared to assume that science could provide the necessary factual premises. In A Theory of Justice, parties in the original position under the veil of ignorance retain general knowledge of established scientific facts (Rawls Reference Rawls1999). In Political Liberalism, public officials may appeal to the methods and conclusions of science as legitimate public reasons (Rawls Reference Rawls2005). Facts, it seems, are simply facts, and thus questions about justice or legitimacy do not arise. This aligns with Hannah Arendt’s (Reference Arendt2006) distinction between facts and politics: facts are despotic, coercive, and domineering, while politics is the realm in which things can always be otherwise. Therefore, facts appear given, whereas political agreements – particularly moral ones – must sometimes be painfully negotiated. This reasoning is what I call the Easy-To-Solve Objection: since factual disagreements are easier to settle than moral ones, we do not have a LDFJ.
In many formulations, the liberal ideal of public justification has been linked to impartiality or even neutrality. Impartial or neutral justifications are expected to be acceptable to all, at least insofar as they are not intentionally devised to favor one party and harm another (Larmore Reference Larmore1987). If factual claims are not subject to the same rules of public justification as moral controversies, then we should not be neutral between established scientific facts and seemingly absurd claims or conspiracy theories. As Alan Montefiore exemplifies, “no geographer or astronomer would be considered as biased or narrow-minded simply because he refused to spend time on careful exposition and consideration of the views of the Flat Earth Society” (Reference Montefiore1975: 18-19). Similarly, Brian Barry observes that “the notion of neutrality has been invoked by the partisans of so-called ‘creationist science’ in the United States to argue that the Darwinian theory of evolution should be taught as a mere hypothesis and forced to share equal time with the theory that the species was created in a succession by God,” which, in his view, is equivalent to demanding that the National Science Foundation’s budget for astronomy “should be split in two, with half going to astrology” (Reference Barry1995: 161). To put it differently, liberal justificatory constraints would not apply to controversies where one party appeals to established facts and the other holds patently wrong or mistaken views. The crucial assumption here is that one claim is evidently correct, whereas the other can be effortlessly recognized as false. Given the cultural authority of the scientific enterprise in contemporary societies, the correct claim is determined by the relevant community of experts. In other words, the Easy-To-Solve Objection states that factual controversies can be forthrightly adjudicated by the respective epistemic authority, while ethical matters lack a comparable authority – at least in secular pluralistic societies. As Federica Liveriero sums it up,
“When dealing with the interpretation of factual beliefs and in many circumstances of our daily life it is very often possible to find an agreement about who are the epistemic authorities on determinate matters (e.g., a doctor for establishing how to treat a patient; an engineer for building a non-collapsing bridge; a lawyer in a courtroom). On the contrary, when dealing with disagreements about ethics, politics, religion and so on, it is extremely difficult to solve our disagreement by referring to the opinion of an expert” (Reference Liveriero2015: 523).
To be sure, this argument does not imply that epistemic authority straightforwardly translates into political authority, typically described as technocracy, also characterized as a “linear model” of government in which unmediated expert advice dictates public policy (Pielke Jr. Reference Pielke2007). The Easy-To-Solve Objection represents a much modest claim: factual disputes possess a universally validated mechanism for adjudication, while moral disputes do not. Thus, the challenge of liberal justification emerges only in the latter case.
In this sense, the Easy-To-Solve Objection implies that we make special efforts to justify value-dependent policies to our fellow citizens because disagreements in these domains are reasonable and even expected, as described by Rawls’ “burdens of judgment” (Reference Rawls2005). In contrast, what Kappel terms “fact-dependent policy disagreements” (Reference Kappel2017) are viewed differently: one side is getting its facts right, while the other is epistemically unreasonable. According to an influential interpretation of Rawlsian public reason liberalism, reasonableness involves the willingness to accept overwhelming evidence and revise beliefs in light of new information (Freeman Reference Freeman2007: 350). Hence, skepticism, rejectionism, or outright denialism in areas supported by robust scientific consensus – such as Darwinian evolution, the climate crisis, or vaccine safety – is unreasonable in the relevant Rawlsian sense. In this vein, Catriona McKinnon argues that denying either (1) the global average temperature warming observed over the twentieth and twenty-first centuries or (2) its anthropogenic causes falls “outside the limits of reasonable disagreement in climate change science” (Reference McKinnon2012: 25-26). Similarly, Torcello asserts that while experts might sometimes err, “it is never reasonable to disregard expert opinion in favor of non-expert” (Reference Torcello2011: 206).Footnote 1 From this, Nebojsa Zelic concludes that factual controversies among the lay public are not the kind of reasonable disagreements that necessitate public justification, but instead constitute “mere disagreements” (Reference Zelic2018: 104). In sum, the Easy-To-Solve Objection states that public reason liberals have no duty to justify the empirical premises of laws and policies (thus, a LDFJ) to citizens who unreasonably reject the facts established by the relevant epistemic authority.
In the remainder of the paper, I survey three arguments. The Inseparability Argument and the Equivalence of Public Reasons Argument can be read as an answer to the Not-Too-Important Objection. The Argument from Epistemic Pluralism mostly replies to the Easy-To-Solve Objection. Let us take these in turn.
3. The inseparability argument
The Inseparability Argument states that whatever reasons we offer to our fellow citizens to justify coercive laws and policies, these reasons always mix empirical and normative factors. Put differently, it might be just impossible to detach factual from value claims and epistemological from moral claims. This seems to work both ways: some of our moral views are rooted in certain factual beliefs, and some of our factual claims derive from our comprehensive doctrines.
The former case is familiar. In the long-standing tradition of theological realism, concrete events that took place in the real world are foundational to decisive religious beliefs of metaphysical or moral character (Bellolio Reference Bellolio2025; Polkinghorne Reference Polkinghorne1998). For Christians, the resurrection is a case in point. As St. Paul wrote to the Greeks, “if Christ be not raised, then your faith is vain” (1 Corinthians 15:17). Even the authors who have lambasted the New Atheists for treating religion as a set of factual propositions instead of about meaning and community, have conceded that this is not a poetical expression but a very precise historical claim which, if proven false, puts Christianity in trouble (Eagleton Reference Eagleton2009: 116; Gray Reference Gray2018: 15). In Bruce Reichenbach’s account, the “phenomenal dimension of the incarnation” speaks about a human Jesus “who was born in Roman-occupied Palestine, had the occupation of rabbi or teacher, had followers whom he taught, and died for blasphemy at the hands of the Romans at the behest of Jewish religious leaders” (2010: 1045).
These are typically factual claims that merge into the kind of religious narratives that Rawls paradigmatically identified as comprehensive doctrines. Even if religion is above all a way of life, “a way of life presupposes beliefs about the nature of reality and cannot be sustained if those beliefs are no longer credible” (Barbour Reference Barbour2000: 37). These beliefs are so entangled that any effort to identify common ground from which to derive legitimate laws and policies ends up screening metaphysical positions, religious ideas, philosophical principles, and factual beliefs all together. Thus, the latter cannot be ignored, especially if they are foundational with respect to any of the former. Against the Not-Too-Important Objection, we reply that some factual claims are part and parcel of a person’s conception of the good. As Adrian Bardon puts it, “because we are not dispassionate about our ideological commitments, it is exceedingly difficult (indeed, almost unheard of) to be entirely dispassionate in the way we account for them using facts and evidence” (Reference Bardon2019: viii). Therefore, fact-dependent policy disagreements, as Kappel (Reference Kappel2017) names them, must be dealt with using the same justificatory tools that public reason liberalism has at its disposal. The LDFJ stands on the grounds of the Inseparability of comprehensive doctrines.
The mirror perspective is considering that our factual beliefs derive from our worldviews. Instead of assessing the nonmoral data on its own evidential merits, individuals are inclined to accept the information that fits with their worldview while rejecting those factual claims that collide or strain their preexisting beliefs. This is the phenomenon of “cultural cognition” widely discussed in recent years (Kahan & Braman Reference Kahan and Braman2006; Kahan, Jenkins-Smith & Braman Reference Kahan, Jenkins-Smith and Braman2011). Translated into the Rawlsian parlance, this means that people’s views about empirical matters are embedded in their comprehensive doctrines. Many public controversies regarding well-established scientific facts – in areas such as climate change, biological evolution, vaccination, nuclear power, genetically modified food, or water fluoridation, to name a few – have been socio-psychologically explained through the lens of cultural cognition. Research shows that laypeople do not disagree over the empirical premises of laws and public policies because they lack information, fact-checking or are otherwise ignorant – much to the contrary (Hamilton 2011; Kahan et al. Reference Kahan, Peters, Wittlin, Slovic, Ouellette, Braman and Mandel2012; Levy 2019) – but because they force factual views to fit with their “mental models” (Sterman Reference Sterman2011), “social meanings” (Kahan & Braman Reference Kahan and Braman2006), or “value similarities” (Cologna & Siegrist 2020). Therein, the Inseparability Argument: facts cannot be detached from our overall evaluative standpoints.
Recently, building on Gaus’ insight that “some of our deepest and most intractable disputes are not about values or principles of justice, but about the world to which these principles apply” (2016: 162), several authors have developed a political theory of perspectival diversity (Muldoon Reference Muldoon2016). This approach begins from the premise that although individuals confront the same external world, they interpret it through distinct conceptual frameworks, leading them to classify or categorize phenomena differently. The abortion debate illustrates this point. Both pro-life and pro-choice advocates may agree on fundamental moral commitments – such as the sanctity of life or the value of autonomy – but they diverge over “how to categorize objects in the world” (Chung & Kogelmann Reference Chung and Kogelmann2020: 843). The former holds that the fetus is a person and a bearer of rights; the latter sees it as a cluster of cells. They disagree not on whether personhood matters, but on what counts as a person. A similar dynamic plays out in debates over Mill’s Harm Principle. Disagreement rarely turns on whether harm is bad, but rather on what qualifies as harm. Consider the case of cannabis: advocates typically concede that marijuana use may affect the user’s health but argue that it should not be criminalized because it does not harm others. Yet a parent whose teenager smokes heavily and disrupts the family’s shared expectations might perceive that harm differently. In such cases, values and factual judgments are not separate layers but fused within a single perspective.
Finally, an influential strand in cognitive psychology emphasizes that human epistemic capacities did not evolve primarily to track truth, but to facilitate social interaction – especially by avoiding cognitive dissonance and asserting group membership (Klintman Reference Klintman2019; Mercier & Sperber Reference Mercier and Sperber2017; Sloman & Fernbach Reference Sloman and Fernbach2017). We are driven more by the need to belong than by the need to get the facts right. In this light, factual and moral beliefs function inseparably as coordinates of identity. Consider climate change: although a scientific matter, it has become so politicized that it now occupies a central role in the “culture wars” defining political identity in many societies. In the United States, for example, one’s stance on the climate crisis is no longer just an epistemic position – it aligns with broader ideological packages that include views on abortion, gun rights, and immigration. As with these other issues, climate change has become a litmus test – a “marker of group allegiance” (Levy 2019: 322), a “measure of social identity” (Fiorino Reference Fiorino2022: 807) – within an increasingly sorted partisan landscape (Dunlap et al. Reference Dunlap, McCright and Yarosh2016). Similar dynamics were visible during the COVID-19 pandemic, where, as Reid notes, “it is difficult to disentangle the purely factual epistemic aspects of existing within such social networks from the other valuable aspects of interpersonal relationships” (Reference Reid2024: 17). Contra Brighouse, empirical beliefs do not float free from identity: they operate as signs of affiliation and belonging. The Not-Too-Important Objection, which seeks to downplay the role of factual disagreement, therefore collapses.
4. Equivalence of public reasons argument
While the Inseparability Argument packs facts and values, the Equivalence of Public Reasons Argument disaggregates the content of political decisions into factual and moral components. Recall that public reason liberals believe that to be legitimate, coercive policies should be justified to all (reasonable) citizens, so public officials must offer reasons that are public and do not solely rest on a particular understanding of the good life. But for every relevant law and policy, more than one public reason is available. The public character of a reason does not directly define the content of the norm but operates as a condition of admissibility into the political debate. The latter will be limited to the reasons that passed the filter. A law or policy is legitimate to the extent that its justification comes from one or more of the reasons found in the pool of admissible reasons – what Kevin Vallier has called the “justificatory pool” (Reference Vallier2011: 372). The Equivalence of Public Reasons Argument states that this pool is comprised of both empirical and normative reasons, of both factual premises and moral evaluations. Once within the pool, these public reasons are weighed against each other, and a decision is finally taken. Depending on the issue, some decisions will take special account of empirical considerations, while in other cases, all action is rooted in the moral or normative dimension of the debate. The bottom line of the Equivalence of Public Reasons Argument is that, insofar as factual claims enter the justificatory pool in the same way that moral reasons do, the former must be justified in the same way as the latter.
Arguably, few relevant political decisions can be made by appealing to just empirical premises. But few decisions can be taken without any factual foundations whatsoever. When these are relevant, it can happen that we are stuck in an empirical disagreement. Kappel (Reference Kappel2017) invites us to imagine a situation in which two persons sharply disagree about a policy, and the only reason why they are opposed is because they diverge about a nonnormative factual proposition. Their values do not differ in a way that influences their ranking of policy options. If both were to agree on this factual proposition, they would also agree about what policy to adopt. Their disagreement is located at the factual level but has crucial implications for political decision-making: the content of the rule depends on its resolution. In line with Kappel’s point, the Equivalence of Public Reasons Argument states that the liberal principle of legitimacy should be upheld in these fact-dependent policy disagreements, too. This often understated feature of public reason liberalism has been recently emphasized by Kogelmann and Stich (Reference Kogelmann, Stich, Sobel, Vallentyne and Wall2021) in their critique of the administrative state. Since administrative agencies routinely make policy decisions that involve “scientific and social scientific considerations” – and since these decisions can have even more direct and coercive effects on citizens’ lives than legislation or constitutional essentials – such considerations, they argue, “must be such that all citizens accept or endorse them” (Kogelmann and Stich Reference Kogelmann, Stich, Sobel, Vallentyne and Wall2021: 162). In other words, insofar as the policies that public reason seeks to justify rely on both normative and empirical (or “positive,” as they term them) reasons, there is no principled basis for applying public reason’s justificatory standards to the former but not the latter.
Government responses worldwide to the latest COVID-19 pandemic provide fresh examples. In this case, the factual premises were mostly offered by the scientific community: this is how the virus operates, this is how we can keep it at bay. By themselves, these empirical inputs are idle. To activate the coercive muscle of the state, they need to be joined by moral assessments, such as weighing the value that we assign to the life of a ninety-year-old person compared to the value of an entire generation of children losing a year of schooling. The interplay between both factual and normative considerations motivated a wide range of interventions such as lockdowns, mask mandates, school closures, and general restrictions on freedom of assembly and movement. Some have argued that most governments acted without “strong epistemic grounds,” at least for justifying the sweeping scale of pandemic-related restrictions (Winsberg et al. 2020: 216), implicitly reinforcing the Equivalence of Public Reason Argument – that empirical claims, no less than normative ones, require justification acceptable to all.
In turn, political leaders often deflected responsibility by claiming they were simply following the science. Yet as Winsberg et al. (2020: 230) point out, scientists – by the nature of their disciplinary focus – tended to prioritize the harms caused by the virus over those resulting from economic disruption or reductions in political freedom. But the problem here, if any, is that political authorities implicitly assumed a uniform set of values across the population. They effectively chose to go full Kantian rather than Utilitarian, treating the preservation of life as overriding, without democratic deliberation on that prioritization.Footnote 2 This episode vividly illustrates how political decisions are composed of both normative and empirical elements. Returning to Kappel’s example, if political representatives already agree that saving lives is paramount, but disagree about the relevant nonmoral facts, then the disagreement is fact-dependent and policy-relevant. In such cases, public reason cannot ignore empirical claims; rather, it must address them through legitimate, public-facing justification – thus triggering an LDFJ.
The popular historian Yuval Noah Harari (Reference Harari2016) offers another example in this sense. Harari argues that the “practical guidance” often issued by various religions conflates factual statements with ethical judgments. Without factual claims, no practical guidance is possible. To prove the point from a historical perspective, Harari reminds us of the so-called Donation of Constantine. According to the tale, the Roman emperor Constantine signed an official decree granting the current and future Popes perpetual control over the Western Roman Empire. For a thousand years, the Catholic Church invoked this document to ground its right to political rule in Europe and beyond. The practical guideline that all Europeans ought to obey the Pope – was the result of conflating an ethical judgment – people ought to respect ancient decrees – with a factual statement – on 315 BC, Constantine issued a decree granting the Popes dominion over Europe. While science or any other empirical method has little to say about whether people ought to respect ancient decrees, it has something to say about the veracity of factual statements. Indeed, researchers discovered that the Donation of Constantine was forged in the papal court sometime in the 8th century. Consequently, the practical guidance collapsed. Bringing the case to our terms, both the factual statement and the ethical judgment share the “justificatory pool” from which the practical guideline emerges. In this strict sense, both have equivalent value as public reasons, as the argument contends.
5. The argument from epistemic pluralism
The Inseparability Argument and the Equivalence of Public Reason Argument jointly respond to the Not-Too-Important Objection. On the one hand, factual beliefs are often inextricably bound up with moral commitments, making it impossible to isolate them in deliberation. On the other hand, even when empirical and normative elements can be analytically distinguished, both contribute to the content of coercive political decisions and therefore must be held to the same standards of public justification. What remains is to address the Easy-To-Solve Objection, which claims that, unlike value-based disagreements, empirical disputes are straightforward to resolve because citizens ostensibly recognize and defer to the same epistemic authorities.
But increasingly, this no longer seems to be the case. In recent years, trust in traditional epistemic authorities – especially in the pronouncements of mainstream science – appears to be weakening in the US and elsewhere (Hamilton & Safford Reference Hamilton and Safford2021; Kennedy et al. Reference Kennedy, Tyson and Funk2022; Nichols Reference Nichols2017). Even before the COVID-19 pandemic, some citizens felt compelled to mobilize in explicit defense of science, as if its epistemic authority could no longer be taken for granted (Krause et al. Reference Krause, Brossard, Scheufele, Xenos and Franke2019). Some scholars now argue that the long-standing “science-society contract” is under strain (Zapp Reference Zapp2022). While several studies indicate that public attitudes toward science are increasingly polarized along political lines (Rekker Reference Rekker2021), others point to the rise of populist epistemologies that vilify experts and elevate folk ways of knowing (Bellolio Reference Bellolio2024). A “fundamental epistemological cleavage between lay and expert forms of knowledge” appears to be widening in contemporary liberal democracies (Staerklé et al. Reference Staerklé, Cavallaro, Cortijos-Bernabeu and Bonny2022: 916). The emergence of social media and its “echo chambers” has only deepened this rift (Del Vicario et al. Reference Del Vicario, Bessi, Zollo, Petroni, Scala, Caldarelli and Quattrociocchi2016), and some pessimistically predict that “denial of scientific information will worsen through increased partisanship in traditional and digital media” (De Cruz Reference De Cruz, Baghramian and Martini2022: 450). While some authors interpret this trend – and its populist roots – as a welcome assertion of “counterknowledge” aimed at redistributing epistemic power rather than simply rejecting expertise (Meyer Reference Meyer2024), the unavoidable implication is that the once-incontestable authority of science is no longer secure. As a result, the Easy-To-Solve Objection loses much of its practical force.
Beyond this contingent erosion of trust, a more principled argument can be made against the Easy-To-Solve Objection. This is the Argument from Epistemic Pluralism. Recall that the liberal duty of justification arises in response to the fact of pluralism in contemporary societies. Yet pluralism is not only moral but also epistemic: citizens differ not just in their conceptions of the good, but in the epistemic outlooks through which they interpret the world. As Robert Talisse observes, “many popular comprehensive doctrines prize loyalty, humility, obedience, and even forms of ignorance over those cognitive virtues associated with truth-seeking and error-avoidance” (Reference Talisse2008: 114). If public reason is to take such diversity seriously while still justifying coercive policies, it must supply factual premises that are acceptable across a wide epistemic range. Deference to expert or scientific consensus, then, cannot be assumed or treated as unproblematic – it must itself be subjected to justificatory scrutiny.
It might be counterargued that public reason liberalism is not concerned with unrestrained pluralism, but only with reasonable pluralism (Rawls Reference Rawls2005: 36). While the notion of reasonableness is itself contested, most liberals treat it as a fundamentally political criterion. At its core, a reasonable citizen is one who views society as a fair scheme of cooperation among free and equal persons. However, as Jønch-Clausen and Kappel note, “epistemic qualities play a very little part in the make-up of the reasonable person” (2016: 129). In other words, political reasonableness imposes minimal epistemic demands. Any more inflationary account risks exceeding the bounds of public reason liberalism – at least as Rawls conceives it. In this spirit, Tim Fowler argues that many citizens who endorse Intelligent Design over Darwinian evolution “are reasonable in the relevant sense” (2019: 36), and even “individuals who believe in a flat earth [can be] law abiding and politically engaged citizens” (Reference Fowler2010: 372). The same could apply to those who accept pseudoscientific claims such as homeopathy, astrology, crystal healing, detox diets, psychic powers, or quantum mysticism: while their empirical beliefs may be highly implausible, they may still qualify as reasonable in the strictly political sense – sufficient to live as free and equal members of a liberal democracy. Indeed, in their recent and extensive work on how Rawlsian liberalism might respond to the rise of illiberal views – such as those promoted by radical right-wing parties worldwide – Gabriele Badano and Alasia Nuti (2024) continue to treat reasonableness primarily as a political criterion. There is little to no mention of substantive or robust epistemic requirements, such as adherence to the methods or conclusions of established science. Instead, individuals who endorse “doctrines built around the idea that racial minorities are biologically inferior” are deemed unreasonable not because of the epistemic implausibility of those beliefs, but because they hold that such minorities “should not share equally in the benefits of social cooperation” (Badano & Nuti Reference Badano and Nuti2024: 56).
Critics might respond that Rawlsian reasonableness is not purely political, but also tied to the acceptance of the burdens of judgment – the idea that reasonable people recognize disagreement as “the inevitable outcome of free human reason” (Rawls Reference Rawls2005: 37), rather than as a sign of ignorance, misinformation, or bad faith. Disagreement, and particularly factual disagreement, arises from multiple sources: the complexity of the evidence, the difficulty of assessing it, and the influence of one’s background experiences and comprehensive worldview (Rawls Reference Rawls2005: 56-57). Reasonable citizens, then, acknowledge that even sincere and competent individuals may reach different conclusions – not because they are epistemically deficient, but because of the inherent limitations – and possibilities – of human reasoning in pluralistic societies.
However, the objection continues, tolerance for factual disagreement has its limits. When individuals deny or reject well-established scientific facts, they are no longer considered reasonable – not because they weigh evidence differently, but because they are simply mistaken (Long Reference Long2011: 505). Climate change is often cited as a paradigmatic example, as “the weight of scientific evidence makes… a complete rejection of the reality of climate change, or an unwillingness to alter a view to account for such evidence, clearly mistaken and hence unreasonable” (Long Reference Long2011: 508–9). As shown in section II, this argument is endorsed by McKinnon and Torcello. If climate science deniers – so often labeledFootnote 3 – persist in their views despite overwhelming empirical data, they are assumed to be acting in bad faith or driven by ideological commitments. Since such disagreement cannot be explained by the burdens of judgment, it no longer qualifies as reasonable. On this view, public reason liberals are under no obligation to justify the empirical premises of coercive policies to those who reject them.
However, many individuals who reject mainstream climate science participate in these debates in good faith, putting forward sincerely held beliefs. Recent research suggests that conspiracists are not only overconfident but also “massively overestimated how much others agree with them,” which “indicates that conspiracists are genuinely unaware that their beliefs are on the fringe” (Pennycook et al. Reference Pennycook, Binnendyk and Rand2025). Moreover, many citizens who reject the scientific consensus on a specific issue are not rejecting science wholesale; rather, they believe that science is on their side. From modern-day creationists to anti-vaxxers and climate skeptics, such actors engage in factual disputes with what they regard as evidence-based arguments (Fischer Reference Fischer2019; Nagel Reference Nagel2008; Phillip-Muller et al. Reference Philipp-Muller, Lee and Petty2022). This phenomenon is increasingly interpreted through what some call the epistemological turn in populist studies (Nawrocki 2024). According to Lütjen, those who resist mainstream science often embrace a “narrative of individual self-empowerment,” typically expressed through the motto do your own research (2022: 77). This ethos is visible in the widespread use of the “red pill” metaphor in online spaces, where epistemic dissent is cast as a civic virtue. These communities portray themselves as “independent minds that need no elite guidance to find their way through the jungle of the 24/7 news cycle” (Lütjen 2022: 83). Their defiance of conventional wisdom is not merely contrarian – it is framed as a method of uncovering how things really are. While these beliefs may be deeply misguided – as they often are – it is not obvious why we should withhold public justification from those who hold them, particularly when the laws and policies in question will coerce them as well. If public reason is meant to justify coercion to all reasonable citizens, and if these individuals are not acting in bad faith but are epistemically sincere (even if mistaken), then they may still be owed a justification of the factual premises that underpin public policy.
As for ideological commitments – the other commonly cited reason for deeming science deniers unreasonable – it is difficult to see how these fall outside Rawls’s own account of the burdens of judgment. Rawls explicitly identifies as sources of reasonable disagreement both the formative role of personal experience and the fact that individuals reason from within distinct comprehensive doctrines, which shape their interpretation of both factual and normative claims. In this sense, Rawls anticipates insights later developed by cultural cognition theory. Consider again the case of climate change, where consistent research shows that “attitudes toward climate science are marinated in people’s worldviews” (Hornsey & Fielding Reference Hornsey and Fielding2020: 10). The distinction that Samuel Freeman attributes to Rawls – between, on the one hand, factual beliefs that are revisable in light of new evidence and, on the other, comprehensive doctrines “which do not admit of change, in spite of changed conditions and evidence that contravenes their major doctrines” (Freeman Reference Freeman2007: 350) – is more porous than it may first appear. Factual beliefs, when embedded in a person’s worldview, are rarely revised, and mistakes are not easily admitted, precisely because doing so would incur high ideological or identity-related costs. For many climate change deniers, the price of accepting the scientific consensus is not just intellectual – it compromises their social standing (Brown Reference Brown, Crow and Boykoff2014; Klein Reference Klein2014). A similar dynamic plays out in the evolution-versus-creationism debate within the “curriculum wars.” As Fowler argues, “belief in ID is a reasonable conclusion given the commitment to an activist God” (2019: 37), which is itself part of a broader – and reasonable – comprehensive theistic doctrine. Intelligent Design may be deeply mistaken as a scientific theory, but it does not appear epistemically unreasonable for committed theists to maintain it. If this is right, then public reason liberals cannot simply dismiss mistaken factual views as ipso facto unreasonable. To the extent that these factual disagreements “result from the burdens of judgment… [they] are thus reasonable disagreements” (Kogelmann & Stich Reference Kogelmann, Stich, Sobel, Vallentyne and Wall2021: 167-8). Hence, they activate the LDFJ.
Rather than treating those who selectively reject established scientific facts as epistemically unreasonable, we might instead view them as epistemically unlucky. As Levy explains, “conservatives who reject the expert view on global warming or on evolution do not reason any worse than liberals who accept it” (2019: 315). Indeed, “those who reject the consensus most strongly may be on average better informed and more capable reasoners than those who accept the consensus” (Levy 2019: 315). The crucial difference lies not in reasoning ability, but in patterns of epistemic deference. Both sides defer to those “they rightly take to be more knowledgeable than, as well as benevolent toward, them” (Levy 2019: 322). Yet because issues like climate change and evolution have become deeply politicized, this otherwise reasonable disposition to defer is mediated by group identity. As a result, liberals tend to trust mainstream scientific experts, while conservatives defer to contrarian voices. In such cases, Levy concludes, liberals are merely “epistemically luckier” (2019: 322), in that they happen to defer to those who are – arguably – correct. But this asymmetry is contingent, not principled. Patterns of epistemic trust may shift over time, and on other contentious topics – such as genetically modified organisms, nuclear power, or gender-transition therapies – liberals themselves may find themselves epistemically unlucky. Then, those who reject the scientific consensus in particular domains are not necessarily unreasonable – just unfortunate in where they place their trust. In that case, it seems unjustified to exclude them from the scope of the LDFJ. If they are to be coerced by public policy, they remain entitled to factual justifications they could reasonably be expected to accept.
To be sure, the Argument from Epistemic Pluralism that grounds the LDFJ does not imply that public laws or policies must accommodate fringe factual views – such as the belief that climate change is a hoax, that biology textbooks should include creationism, or that vaccines cause autism. Rather, it holds that individuals who espouse such views are, like everyone else, entitled to a justification. So long as these factual disagreements are deemed reasonable enough – that is, traceable to the burdens of judgment – the demands of public reason continue to apply. As Kogelmann and Stich argue, just as “reasonable disagreement over normative considerations forces a retreat to shared ground,” so too must “reasonable disagreement over scientific questions” (2021: 170). What such epistemic shared ground should consist of is beyond the scope of this paper. As noted at the outset, there is a flourishing literature on whether the methods and conclusions of mainstream science qualify as public reasons – perhaps, even paradigmatic ones – and whether this justificatory threshold is satisfied by expert consensus within the scientific community or instead requires acceptance by the broader public.
6. Conclusion
This paper explores whether public reason liberals have an obligation to justify the factual claims that underpin coercive norms and policies. Traditionally, liberal theorists have focused on justifying moral principles, assuming that empirical facts either (1) are not deeply relevant to people’s lives in the way that moral beliefs are, or (2) can be easily resolved through expert consensus. However, growing public disputes over scientific facts and recent insights from cognitive psychology challenge these assumptions.
What I have termed the Not-Too-Important Objection can be answered by combining the Inseparability Argument and the Equivalence of Public Reasons Argument. The former holds that factual and moral claims are deeply intertwined: many empirical beliefs are embedded within broader worldviews that shape people’s identities. The latter maintains that, insofar as factual claims play a decisive role in shaping coercive policies, they must be subjected to the same justificatory standards as moral claims. In turn, I have addressed the Easy-to-Solve Objection with the Argument from Epistemic Pluralism, which posits that liberal societies are not only morally but also epistemically diverse, with citizens holding divergent views about what counts as reliable knowledge. To the counterargument that fringe factual views – such as scientific denialism or conspiratorial thinking – should not be considered reasonable (and thus do not activate a LDFJ), I have replied that most public reason liberals treat reasonableness primarily as a political criterion. And even if we extend that requirement to include epistemic elements, many mistaken factual views can still be explained by Rawls’s burdens of judgment. The proponents of such views often hold them sincerely, and defend them in ways shaped by their life experiences and comprehensive doctrines. Rather than labeling them epistemically unreasonable, I have argued that they are better understood as epistemically unlucky – but still entitled to public justification.
In sum, I have defended a LDFJ, that is, the idea that public reason liberals must publicly justify the empirical premises of coercive policies, just as they justify moral principles to people from diverse avenues of life. This duty arises because factual disagreements in democratic societies are often as deep as moral ones; ignoring them risks undermining public legitimacy. Given the rise of misinformation, populism, and skepticism toward scientific expertise, liberal theorists cannot assume that factual claims will be universally accepted or that deference to experts will suffice. Instead, they must engage in the public justification of facts to ensure that policies are not just normatively sound but also epistemically legitimate. This conclusion challenges a long-standing oversight in public reason liberalism, suggesting that a theory of legitimacy must account for both moral and empirical justification. If public reason liberalism is to remain viable in an age of contested facts, it must take epistemic pluralism seriously and develop strategies for engaging with factual disagreements in a way that respects democratic legitimacy.
To be sure, the existence of this LDFJ is analytically different from how we fulfill it. Here, I have made the case for the former. The latter is part of an ongoing discussion that interrogates whether scientific claims, which are certified by the relevant expert community, are suitable to serve as public justifiers. But that is another – and subsequent – conversation.
Funding
This research was partially funded by Chile’s National Association for Research and Development (ANID), Project #1240042 on “The Climate Challenge to Liberal Democracy.”
Acknowledgement
An early draft of this paper was presented in the British and Ireland Association for Political Thought (BIAPT) Annual Conference, held at the University of Oxford in January 2020, just weeks before the pandemic. The key speaker was Gerald Gaus, who attended the panel and provided me with his insightful comments. It was his last in-person conference, as he passed away later that year. This paper is dedicated to his memory.