Footnote 1Deliberative democracy requires some mechanism for shaping the contours of the deliberation. Some philosophers argue that reasons given in the course of democratic deliberation must draw on a public set of values (moral equality among all persons, for instance). As Seyla Benhabib describes this view, “to attain legitimacy, democratic institutions require the articulation of the bases of their actions and policies in discursive language that appeals to commonly shared and accepted public reasons.” Footnote 2 Because a democratic government wields coercive power, the least it can do is give a comprehensible reason to the coerced.

But which reasons are suitably “public?” In “Justice as Fairness: Political not Metaphysical, Footnote 3 ” Rawls suggests an approach to a public conception of justice that eschews any dependence on metaphysical conceptions of justice in favor of a political conception of justice. The upshot is that we don’t need to resolve intractable debates about meaning, purpose or value in order to have a functional deliberative democracy.

However, the fact that Rawls favors a political conception over a metaphysical conception means that if there is a metaphysical conception of justice that actually obtains in some philosophically meaningful way, then Rawls’ theory would not (and could not) Footnote 4 be sensitive to it. Rawls himself admitted in Political Liberalism that “the political conception does without the truth.Footnote 5” Similarly, in Law of Peoples, Rawls endorses a political conception of justice to govern the society of peoples that is not concerned with truth, but instead concerned with being sufficiently neutral so as to avoid conflict with any reasonable comprehensive doctrines.

In Joshua Cohen’s paper “Truth and Public Reason, Footnote 6 ” Cohen asserts that Rawls was mistaken in saying that his project “does without the truth.” Cohen is unconvinced by Rawls’ pronouncement that public reason can function without a conception of truth because, on Cohen’s view, a society that has no concept of truth is close to inconceivable. As he writes, there are many concepts without which political justification can remain perfectly comprehensible, like “personal autonomy, purity, courage, or honor.” Footnote 7

In contrast, the idea of locating a common ground of political reflection and argument that does without the concept of truth—like doing without the concept of an object, or cause, or thought, or reason, or inference or evidence—is hard to grasp. Truth is so closely connected with intuitive notions of thinking, asserting, believing, judging, and reasoning that it is difficult to understand what leaving it behind amounts to. Footnote 8

In this paper I will argue that public reason will not allow a conception of truth robust enough to provide a framework for the very functioning of public reason itself, nor does it allow for a conception of truth robust enough to allow constitutional liberal democratic peoples facing dilemmas that demand collective action to give reasons in favor of any particular response. I argue that the ongoing public health crisis presented by the COVID-19 pandemic makes this point especially salient, and I conclude that some metaphysical dimension is required to have a coherent and functional understanding of truth.

Public Reason

“One thing I failed to say in A Theory of Justice, or failed to stress sufficiently, is that justice as fairness is intended as a political conception of justice.” Footnote 9 Given the putative fact of reasonable pluralism, Rawls needs some process by which those in the public political forum can give content to debates concerning constitutional essentials and matters of basic justice without appealing to any particular comprehensive doctrine. Rawls continues, “One task of political philosophy in a democratic society is to focus on such questions and to examine whether some underlying basis of agreement can be uncovered and a mutually acceptable way of resolving these questions publicly established.”Footnote 10

Enter public reason. Public reason dictates that when one engages in official political discourse seeking to establish principles that govern the common, public domain, the reasons she gives ought to be recognizable to all members of the society. Public reasons must be justifiable in terms of commonly agreed upon principles that transcend any one comprehensive doctrine. This need for a conception of justice that can be widely shared and endorsed from within different reasonable views motivates his adoption of a political conception, rather than a metaphysical conception of justice. “The essential point is this,” Rawls says, “as a practical political matter no general moral conception can provide a publicly recognized basis for a conception of justice in a modern democratic state.” Footnote 11

Truth – Political Not Metaphysical

Rawls writes that “Only a political conception of justice that all citizens might be reasonably expected to endorse can serve as a basis of public reason and justification.” Footnote 12 This same rigorous criterion, the reasonable expectation that the conception of justice could be endorsed by all citizens, extends to all of the content of public reason, including truth. Furthermore, recall that Rawls gives us two conditions under which a person can be considered reasonable: “their willingness to propose and abide by fair terms of social cooperation among equals and their recognition of and willingness to accept the consequences of the burdens of judgment.”Footnote 13 Nothing in the forgoing definition of reasonable entails that a reasonable person will have such-and-such conception of truth. Public reason can only accept truth claims that do not conflict with any reasonable comprehensive doctrines. But the set of truth claims that lives up to that billing is very small, since Rawls’ definition of a reasonable person puts such minimal conditions on what views a “reasonable” person might hold about truth.

Cohen’s worry is that public reason, insofar as it has to avoid any appeal to metaphysics and any conflict with reasonable comprehensive doctrines, will not be able to utilize any conception of truth robust enough to allow public reason to operate. For Cohen, Rawls’ abandonment of truth “makes the idea of public reason hard to understand, because it proposes to leave the concept of truth behind while preserving notions of belief, assertion, judgment, reason, and objectivity, all of which are essential to an idea of public reason.” Footnote 14 That is, having a shared understanding of truth animates a certain set of cognitive concepts that are essential to the proper functioning of public reason. Cohen thinks that believing, asserting, reasoning and deducing all appeal to a conception of truth.Footnote 15 Beliefs, for example, aim at the truth, and without a conception of truth, we have no framework for evaluating our beliefs. If we had no conception of truth, we would have to do without believing, asserting, reasoning and deducing; but public reason cannot serve its purpose (namely, to provide a shared basis for political argument) without them. Therefore, we need a conception of truth available within public reason.

We might think that this worry can be dismissed by positing a political conception of truth. This is precisely what Cohen attempts. My view is that this project fails.

Cohen begins his analysis of a political conception with minimalism about truth, the view that truth has no nature and is not a substantial property; a truth apt claim is simply one which affirms the “truth schema.” Footnote 16 But Cohen admits he has already said too much. Minimalism about truth is an antimetaphysical conception, and to be compatible with public reason, we can only admit a nonmetaphysicial conception. Further, to the extent minimalism is presented as a rebuttal to correspondence or coherence or pragmatic theories, we have created unnecessary barriers to certain reasonable comprehensive doctrines.

Cohen hopes to rescue the political conception by positing a minimalism about minimalism. The political conception need only include four “commonplaces” about truth:

  1. 1.

    Truth is the norm governing beliefs.

  2. 2.

    True beliefs present things as they are.

  3. 3.

    Truth is distinct from warrant or justification.

  4. 4.

    Truth is important.

There are several reasons to be skeptical that Cohen’s minimalist version of minimalism about truth will be thick enough to provide the framework we need for deliberation in public reason. Firstly, a genuinely nonsectarian conception of truth would be so unlike any notions of truth that are presently in use that it seems inappropriate to consider it a conception of “truth” at all. A political conception of truth would need to avoid contradicting any part of someone’s reasonable comprehensive doctrine. For example, a political conception of truth would need to be silent as to whether in fact truth is achievable, whether in fact truth is valuable, Footnote 17 whether in fact truth is immutable, or whether in fact claims about value are truth apt. A conception of truth that is silent on all of these questions seems to me to be a conception of truth so minimal as to be incomprehensible, or at the very least not robust enough to be extended to any meaningful claims.Footnote 18 If both cognitivism and noncognitivism are reasonable, say, then a political conception of truth becomes useless if we hope to maintain our commitment to reasonable pluralism.Footnote 19

Furthermore, a political conception of truth needs to do the work that Cohen suggests public reason has forced us to abandon. Part of the original problem was that without a conception of truth, we can have no basis for believing, asserting, judging, etc. A political conception of truth needs to fix this problem, or we are back where we started. My view is that a political conception of truth cannot fix this problem. Here’s a reason to think this. Plenty of reasonable people think that empirical science tracks the truth. But many reasonable people (in the Rawlsian sense) also think that faith tracks the truth (in other words, that one’s believing something to be the case is sufficient warrant for that belief, or as one popular religion asserts, “faith is the substance of things hoped for, the evidence of things not seen.”)Footnote 20 These two views do not come from the same understanding of justification; they come from very different beliefs about what truth fundamentally is and how we might have access to it.Footnote 21 A political conception of truth seems unable to help us here, as it cannot contradict any  reasonable comprehensive doctrine. The political conception of truth cannot weigh in on whether science is a better guide to truth than faith. Consequently, public reason cannot officiate this dispute. A political conception of truth has no special resources to offer us a groundwork for believing, asserting, judging, and so forth, and for that reason it does not rescue public reason from the original problem.

Now, there is a distinction to be made between the political conception contradicting metaphysical conceptions and simply not containing them. An analogy to the American rules of evidence may be instructive. Courts apply strict rules to which types of evidence can be admitted at trial, and this conception of evidence is more restrictive that our ordinary, colloquial conception of evidence. For example, suppose a customer at a coffee shop is injured when they slip on an icy staircase and fall down because there was no railing. At trial, the customer wants to introduce evidence that the coffee shop has now added a railing, a warning sign and puts salt on the steps on icy days. Even though this evidence would tend to show that the coffee shop could have been doing quite a bit more to keep the stairway safe, this evidence will not be admitted as a matter of public policy. Footnote 22 After all, if the shop knew that taking remedial measures could come back to bite them at trial, they would not take precautions to secure the steps. How many unknowing victims might succumb to this icy obstacle? So here, the rules of evidence are not simply about filtering out data with low probative value, but even filtering out data that does have high probative value, but should nonetheless be inadmissible for reasons of public safety.

But does the legal conception of evidence contradict the whole truth about evidence? No. It simply says some things are inadmissible, which is not the same as saying they are unreliable. In fact, in the case we just examined, the evidence was reliable but nonetheless inadmissible. Likewise with public justification: the mere fact that something is inadmissible in public reason does not entail that it is also untrue.

Here’s the objection: some reasonable comprehensive doctrines might reject the empiricism that public reason seems to presuppose. Further, public reason relies on the claim that we can pursue political justification (deliberation, dialogue, reason-giving, etc.) in the absence of widespread agreement about what truth actually is. But that very claim (“we can pursue political justification in the absence of widespread agreement about what truth actually is”) is something we need to agree on, and take to be really true (in some robust, nonpolitical sense of truth). Therefore, public reason needs to rely ultimately on a nonpolitical truth about truth after all.

The problem goes deeper still. It would seem at first blush, since public reason needs to remain untainted by the influence of private religious considerations, Footnote 23 that liberal democratic peoples must reject (for purposes of political justification) suggestions that include prayer, prophecy, miracles, rituals or magic. But I suggest that the list of options that public reason will exclude is in fact much longer. For example, some reasonable comprehensive doctrines might reject appeals to induction or causation (Humeans, say). This means that scientific solutions to the foregoing problems might be viewed as hostile to some set of reasonable comprehensive doctrines, and hence inadmissible in public reason.Footnote 24

Pandemic Problem Solving

Consider an example occasioned by the pandemic: the closure of churches for public health reasons. Rawls extends the concept of public reason, at the very least, to constitutional essentials and matters of basic justice. Footnote 25 The power of the government to suspend religious services raises core constitutional questions, and would thus be in the purview of the public reason debate. Any data that officials could look to will be steeped in scientific assumptions, and any policy they pursued based on that data would inevitably make value judgments about the economic and moral worth of human lives. Plus, there is a vocal minority arguing all the while that whatever restrictions are implemented are in violation of citizens’ civil liberties. How would Rawls advise these city officials?

A Rawlsian may counter here that public reason need not be equipped to resolve any political issue; it need only set up the “basic structure”, which is Rawls’ jargon for the fundamental rights and institutions of society. But here, the question at issue is one of critical importance: religious freedom. A successful justificatory scheme does need to be able to guide us on such a core constitutional value. A conception of truth that is too narrow will preclude good deliberation. Rawls mentions that common sense and uncontroversial conclusions of science enjoy a privileged status in public reason. Footnote 26 But what is an uncontroversial conclusion of science? A reasonable person (in the Rawlsian sense) could maintain that science does not effectively track objective truth, supposing there even is such a thing as objective, mind-independent truth.Footnote 27 A reasonable person could also maintain that causation is not an actual relation that exists between entities in the world, but only an illusion.Footnote 28 It remains unclear how Rawls could respond to this problem.Footnote 29

Likewise, a reasonable person could claim that while nations have a duty to protect their citizens’ most basic rights and offer them the most basic liberties, there is no duty on the part of the state to protect its citizens from public health crises, nor is there a duty on the part of the state to work proactively to prevent future outbreaks. If public reason cannot admit reasons that conflict with reasonable comprehensive doctrines, then no serious debate about the causes, consequences or policy implications of a pandemic can take place in public reason. The same dilemma emerges for other national and international crises a people might face. Discourse on issues like how to combat terrorism, adapt to a changing climate or rekindle stagnated economies will be similarly stunted. Footnote 30

The Rawlsian can respond here in a variety of ways, but to my mind none solve the problem that Rawlsian neutrality about truth grinds the political procedure of public reason to a halt. The first objection is that when Rawls writes about neutrality about truth, he is referring to substantive truth, not a formal concept of truth. Rawls cares about the substance of the belief, not the architecture which gave rise to it. This is one of the fundamental premises of overlapping consensus. A cognitivist could believe in the value of free speech, or vaccine mandates, or voter identification requirements because it is entailed by her views on human rights. A noncognitivist could hold these same beliefs from the angle of their instrumental value to an organized society. So long as each view comes from a reasonable comprehensive doctrine, then Rawls is happy with this result. Therefore, it would seem, we can make claims about truth in public reason so long as they are simply formal, rather than substantive.

This distinction between substantive and formal truth does not avoid the problem because there is reasonable disagreement about the formal properties of truth. Should we replace the concept of truth itself with something else? Does belief aim at truth? Is truth an explanatory semantic notion or a normative one? Is the law of non-contradiction self-defeating? And so on.

Another objection is that we don’t need to resolve the biggest questions of epistemology or metaethics to answer first-order political questions like whether vaccinations should be required. Arguments in the public sphere will likely be between first-order moral views, and public reason can be robustly populated with reasonable first-order moral views. Naturally, arguments between first-order moral views often become appeals to second-order moral views (e.g. utilitarianism), and so long as these views are reasonable, this is permitted in public reason. But we rarely need to reach tertiary issues about the purpose of metaethical theories to resolve our practical political questions.

Indeed, we may not need to resolve philosophical debates between correspondence, minimalist and deflationary theories of truth before we can begin solving public health crises. But my concern is that reasonable people will have different background theories of truth, and this will spill over into their ideas about what counts as evidence and what counts as justification. Without a shared conception of truth we cannot have a shared conception of evidence, and certain subsets of reasonable people will not be able to have political dialogue at all because public reason cannot admit a concept of truth thick enough to answer any of the important questions required for the discussion.

Suppose you begin to debate the issue of vaccine mandates with a philosophical skeptic, who pushes on the very ideas of evidence, assertion, justification, belief, etc. First and second order moral views will not be enough to resolve the dispute; you will need to refer back to theories of truth. But if public reason does without the truth, then this dialogue would require us to move outside the bounds of public reason.

Another avenue of objection the Rawlsian will raise is that public reason doesn’t need to resolve every political issue (e.g. vaccine mandates, lockdowns, masks, etc.). The scope of public reason only extends to questions about the basic structure of society and fundamental issues of justice. But the central philosophical proposition in the founding documents of the United States is “we hold these truths to be self-evident,”. This is a substantive claim about truth, that human beings have innate liberties , the protection of which  justifies having a constitution constructed in such and such way. It almost seems that Rawls’ conception of public reason requires a conception of truth so thin that the philosophical claims of the Declaration itself would be inadmissible in public reason.

A final objection to my view is that the issue of eccentric epistemologies is resolved because Rawls expressly says that empirically based scientific knowledge and common sense are conditions of reasonableness. Footnote 31 But why should we take Rawls at his word on this? If the premise of public reason is to create a framework where no reasonable person is coerced by doctrines which are alien to them, then we cannot simply waive the problem away that some reasonable people do subscribe to eccentric epistemologies.

It is easy to say that a vaccine mandate is justifiable when objectors are saying that the vaccine contains government-created nanorobots which alter your DNA. But some vaccine objectors simply say that, on their view, the cost-benefit analysis of mandating a recently developed vaccine for young people, for example, who tend to be resilient against the virus, doesn’t pass muster. And even if they are wrong, it is hard to say such a view is so unreasonable that vaccine dissenters should simply be coerced by reasons they themselves cannot endorse. In short, even if we shoehorn our popular conception of science into public reason by fiat, it does not resolve as many practical issues as objectors might think.

Rawls thinks that “the great evils of human history” like famine, persecution, genocide and unjust war will disappear once a network of societies has adopted the Law of Peoples, but there is no reason to think that setbacks like global warming or terrorism won’t afflict the realistic Utopia. Footnote 32 These last two years of a surging  pandemic in the U.S. and across the world show that catastrophic damage can be caused by a virus. Containing an infectious disease requires a sophisticated and coordinated response. It is therefore deeply problematic that a society would be left with no just way for deliberating about such important issues. But since a political conception of truth is required by the principles of public reason, and since a political conception of truth cannot be sufficiently robust to deal with the aforementioned problems, the Rawlsian society is left completely unable to confront its most pressing challenges.

Public Reason in Crisis

The challenge for the public reason liberal is that there must be some overlapping consensus of comprehensive doctrines for deliberative democracy to proceed. But if public reason cannot admit any conception of truth, then deliberators will hardly be able to describe their problems, much less produce normative arguments toward solving them.

The tension between unified public concepts and reasonable pluralism is a recurring theme in the post-Rawls literature on public reason. For example, Elizabeth Brake Footnote 33 writes that a commitment to public reason entails the adoption of a minimal conception of marriage, a conception that does not privilege heterosexual, monogamous or even amorous relationships at the expense of others in the eyes of the state, as such practices might unduly disenfranchise a family of reasonable comprehensive doctrines. In the same way, Cohen himself argues for a similarly “thin” conception of human rights in the face of ethical pluralism and what he calls “global public reason” elsewhere in his work.Footnote 34 It may be that public reason, in principle, is not amendable to thick public concepts altogether. If this is so, the appeal of public reason seems to collapse. If a society is committed to the necessity of truth and the defense of human rights, but it can give no content to public conceptions of these ideas, then it will not succeed in either endeavor.

Toward a Solution

It may be that the problems I have described are caused by allowing too many doctrines to count as reasonable. The solution, then, may be to close the gates to a wider range of views. Some have argued for a wider concept of public reason, such that reasons need merely be intelligible or offered in good faith (that is, you can’t launder non-public arguments into the deliberative process through public arguments made insincerely). Others have argued for tighter constraints, for example, that political liberalism’s requirements of reciprocity and equal citizenship eliminate any conceptions of justice which do not yield substantive equality for all. Footnote 35 My own view is that public reason need not entertain reasons stemming from eccentric epistemic commitments, like commandments or promises purportedly made by supernatural beings beyond our plane of existence. Introducing such reasons, which are not comprehensible to others, disrespects and fundamentally misunderstands the deliberative process. Successfully demarcating the kind of reasons which should be admissible is far beyond the scope of this paper, but I point to this problem as being the upstream cause of our present question: can public reason contain any conception of truth?


Reconciling public reason with important public concepts, either by defending a compelling political conception (perhaps by renovating Rawls’ definition of a reasonable person to be less inclusive) or by finding a way to reintroduce the metaphysical, is perhaps the most important project for future research in public reason. My own view is that some metaphysical dimension is required to have a serviceable concept of truth. Of course, I also think that public reason requires us to jettison any metaphysical concept of truth. Therefore, it seems to me that public reason must be left empty of any concept of truth whatsoever.

I have argued (with Cohen) that approaching justice as political rather than metaphysical yields an unexpected project, that is, the project of designing a political conception of truth. I then argued (against Cohen) that this political conception of truth, which cannot conflict with any reasonable comprehensive doctrines, will need to be so minimal as to be useless. My worry is twofold. Firstly, a minimal conception of truth prevents us from using basic cognitive processes that are necessary to the proper functioning of public reason, and a political conception of truth cannot fix this problem. Secondly, public reason will not allow a conception of truth robust enough to properly equip societies for dealing with crises that demand a unified response.

I agree with Rawls that a political conception of justice is the most sensible way to organize a society with deep disagreements about value but a shared interest in stability and willingness to commit to reciprocity. But the political conception of truth is not similarly desirable, and my view is that public reason needs more than a minimalist political conception. I am aligned with Cohen when he writes, “Although I find the idea of public reason compelling, I disagree with Rawls’s claim that the concept of truth finds no place in it.” Footnote 36