OPR V.14 Part 2: Sincerity and Shared Reasons

I summarized Section 14 in a previous post; here I raise some critical points about the question of whether public justifiability should include a shared reasons requirement, and how this relates to sincerity in public deliberation.

Gaus rejects the requirement that deliberators deliberate in terms of shared reasons. To be a bona fide moral rule, a rule must be endorsed by each and every member of the appropriately (i.e. partly) idealized public, each based on the total set of reasons he or she accepts. “Mutual intelligibility” requires only that these personal evaluative standards pass some threshold of plausibility such that they can be generally recognized as genuine moral perspectives. But members of the public will still think that many of the reasons their fellows appeal to are bad reasons. They also think that this use of bad reasons for the assessment of moral rules is appropriate. In intellectual argument each will criticize the other for accepting bad reasons, and argue that others ought to change their views. When it comes to determining what count as valid moral rules, however, everyone accepts that each will assess proposed rules based on his or her own evaluative standards, and that rules will count as valid only if they meet with the unanimous approval from these diverse perspectives.

An alternate view would be that deliberators accept that they are to deliberate only on the basis of the reasons they share. What is wrong with the shared reasons view? Gaus’s answer in Section 14.4 (d) comes in the form of a response to Jon Quong’s argument for the shared reasons view. ((Jonathan Quong, Liberalism Without Perfection, OUP 2010, Chapter 9 “The Scope and Structure of Public Reason”)) Quong’s argument is based on the requirement that public reasoning be sincere. I will explain the dispute, then briefly argue that the underlying issue isn’t really about sincerity.

Insincere deliberation may be justified in some circumstances, but it is pro tanto morally bad because it makes public reasoning into a form of manipulation. If I argue that your beliefs commit you to supporting a particular proposal even when I don’t think they do, then I am not respecting your capacity for rational moral agency; I am treating you as a thing to be moved, not a person to be reasoned with. Conversely, sincerity in public reasoning expresses respect, helping to sustain civic friendship. Quong formulates the idea of sincerity in public justification in terms of three conditions, involving persons A(lf) and B(etty) and a proposal X.

  1. A reasonably believes he is justified in endorsing X,
  2. A reasonably believes that B is justified in endorsing X (…)
  3. A may only… offer arguments in favour of X to B that he reasonably believes B would be justified in accepting. ((Gaus cites the first two of these conditions on 288. All my quotes from Quong are from Chapter 9 of from Liberalism Without Perfection ))

Quong argues that Gaus’s convergence conception of public justification violates this “principle of justificatory sincerity” (PJS). The example he uses to illustrate this fact involves a public of only two persons A and B and two reasons Ra and Rb and one policy X. The beliefs as follows:

  • A believes Ra justifies X
  • B believes Rb justifies X
  • A does not believe Rb justifies X
  • B does not believe Ra justifies X

In this situation, there are no shared reasons for X, but each member of the public (A and B) believes X to be justified. The problem is that A and B cannot endorse X without violating PJS, because each person must believe that the other is not justified in endorsing X. “If either person were to vote for or otherwise support X, he would be supporting the imposition of a political decision on his fellow citizen that he does not believe can be justified to that citizen” (Quong).

In response, Gaus accepts that public justification presupposes a principle of sincerity, but distinguishes different possible specifications of this principle. A must believe B would be justified to support X, but under what conditions, he asks: (1) that she correctly completed all relevant reasoning; (2) that she deliberated at the level expected by members of the public; (3) that she deliberated at this level and arrived at a basis for X that is at least minimally plausible to Alf as a moral basis (even if incorrect in his view), or (4) that she deliberated at this level and arrived at a basis for X that she thinks sufficient, even if Alf does not think it intelligible as a moral basis? Gaus argues for C as opposed to D.

My first comment about this debate is that the issue is not best framed as a debate about sincerity. If I argue that correctly understood the Catholic doctrine of chastity supports same-sex marriage without acknowledging that I think this doctrine false, that can be insincere, if I am dissimulating my own fundamental disagreement with Catholicism. However, reasoning from conjecture need not be insincere; I can say “I don’t believe this doctrine, but I can see that it is a serious moral point of view, and I think it does imply that we ought to accept same-sex marriage.” If I truly believe that the doctrine supports SSM, I am not being insincere. Of course, if I think that opposite-sex-only marriage is a grave enough injustice, I may think it justified to engage in insincere deliberation. If at some point of injustice resistance is justified, at some lesser point of injustice insincere rhetoric is justified. ((On these points, see “The Sincerity of Public Reason” forthcoming in the Journal of Political Philosophy and “On Reasoning from Conjecture” forthcoming in the Journal of Moral Philosophy, both by Micah Schwartzman, who I trust will correct me if I’m misrepresenting his views)) But the mere fact that I am reasoning about what would be true if your fundamental beliefs were true does not make my reasoning insincere. Nor does the fact that I don’t think your fundamental beliefs justified mean that I can’t think the proposal in question is justified to you. The idea of justification to… is relativized to the variety of reasonable moral perspectives. This is true even when reasoning from shared reasons, as public reasons will themselves be based at some level on underlying religious and philosophical views that are not shared(overlapping consensus).

In my view, the main issue between the shared reasons or consensus interpretation of public justification and and the total reasons or convergence interpretation is whether we frame the demand for public justification as a constraint on rules directly, with a default of not having a rule, or as a constraint on the reasons we appeal to in deciding what kinds of rules to have, with a default of exclusion from consideration. But this is jumping ahead a bit; we need to wait for Section 16’s discussion of the default and the eligible set of proposals.

My second comment about this debate is that there seems to be simple answer to Quong’s A and B example, from Gaus’s perspective. If A thinks that B truly does not have reasons for X and B thinks that A truly does not have reasons for X then X does not satisfy BPPJ. The fact that there is a convergence of actual views is not relevant because, as Gaus makes clear on 225-6, it is up to each of us ultimately to determine whether we think others have sufficient reasons for X.

It is odd, however, to think that both A and B want to follow X and think X justified, but also think themselves not permitted to follow X. Quong says that if Alf supports X, in this instance, he would be supporting the imposition of a political decision on Betty that he does not think can be justified to her. Gaus actually agrees, if Alf thinks that Betty’s only reason Rb does not support X. But is there still imposition, in this case? Alf has qualms about “imposing” a rule on Betty that Alf thinks Betty ought not endorse, given her beliefs. But Betty points out that there is no imposition, because she disagrees, and thinks the rule justified given her beliefs. I am tempted to say that the demand for public justifiability simply doesn’t apply, in this situation. If there is no disagreement and no conflict, there is no imposition, and so no need for a principle to determine when we can justify imposition.

The remarks above defuse Quong’s worry that the convergence view licenses insincerity, I think, but they do not address Gaus’s objection that the consensus or shared reasons view involves disrespect for persons as moral agents. I want to disagree about this as well. Gaus claims that objection to the shared reasons view is that Alf should not have a veto over the reasons Betty employs in her own evaluations.

“Shared Standards implies a strong right of Member of the Public Alf to veto the use of Betty’s ?B in her own deliberations about whether x is justified to her. If Alf does not endorse ?B, then he has a right to block Betty from employing it in her own deliberations. This surely is objectionable, even if he cedes a similar veto to her” (286).

Gaus returns to this point on p.290, asserting that any principle allowing some to veto the reasonable considerations of others manifests disrespect for them as fellow moral agents. The thought here, I think, is that I as a free and equal individual ought to be sovereign over my own judgment. Yet the shared reasons view seems to give others authority over my own deliberative process. I don’t find this argument persuasive. Neither Alf nor Betty has any veto or even any say over what others will think justified according to ?A or ?B. Each has the right to make up her own mind. The veto is only over inclusion in the set of grounds that we acknowledge as a valid basis for our common decisions about moral rules. Any view involving a unanimity requirement gives a veto of some kind to someone or indeed anyone, potentially; the only issue is what form this veto takes. According to the shared reasons view, any reasonable rejection of a reason disqualifies it as a basis for collective decision-making about social rules; according to Gaus’s preferred alternative, any reasonable rejection of a rule disqualifies it (though we have yet to learn exactly how the default or baseline is set). Which is more disrespectful of moral agents as free and equal; to give individual members of the public a veto over proposed rules, or to give them a veto over proposed reasons for decisions about rules? Gaus’s brief argument here against the shared reasons view has not resolved the issue.

Be Sociable, Share!
This entry was posted in Books, Posts, Reading Group and tagged , , , , , . Bookmark the permalink.

7 Responses to OPR V.14 Part 2: Sincerity and Shared Reasons

  1. Hi Andrew,

    Thanks for this thoughtful set of comments on my sincerity-based critique of the convergence model. Let me explain why I’m not persuaded by your critique of my view.

    First, you say that you don’t think the dispute actually has much to do with sincerity. You point out that there’s nothing necessarily insincere about Alf telling Betty ‘I don’t adhere to your doctrine myself, but I believe that your doctrine clearly supports a commitment to rule X’. Of course this is true, but I don’t think it touches the point I am making. My claim is that, under many forms of convergent justification, Alf cannot sincerely say the following to Betty ‘I believe your doctrine clearly supports a commitment to rule X, and I believe you are justified in believing the relevant part of your doctrine’. If Alf cannot sincerely make this latter claim (and if there are no other grounds that Alf believes justifies X for Betty) then Alf cannot endorse X as a rule for Betty without falling foul of PJS. In brief, a principle of justificatory sincerity (under the conditions I specified) requires not merely that Alf sincerely believe Betty’s doctrine somehow supports X, he must also sincerely believe that Betty is justified in believing the relevant parts of her doctrine that provide this support. Without some version of this stronger requirement, Alf would be able to be “sincerely” engaged in convergent forms of justification with people who hold obviously silly doctrines (e.g. astrology) by saying ‘I don’t believe you are justified in accepting astrology, but you do accept it, so let me explain why rule X can be imposed on you in light of your beliefs’.

    Second, you say that there’s an easy way for Jerry to deal with my example, which is to simply point out that X fails the Basic Principle of Public Justification (BPPJ). If Alf does not think Betty has sufficient reasons to endorse X, that’s the end of the story as far as BPPJ is concerned, and so the whole dispute about sincerity is superfluous. In a sense, this is the point I’m making with my sincerity argument: if Alf doesn’t believe Betty has sufficient reasons, then he cannot sincerely support the imposition of X on Betty. But there are reasons why the argument is better framed in terms of the sincerity requirement rather than by directly appealing to the BPPJ. By focusing on sincerity, I wanted to emphasize that in conditions where people disagree about what reasons Betty has, it’s not acceptable for Alf to simply defer to what Betty herself believes—Alf must ask whether he can sincerely justify this rule to Betty regardless of what Betty may herself believe. In this sense I think Jerry and I are not in any disagreement. But this is why sincerity is a good way to capture what’s troubling about the convergence form of justification. We cannot suppose some Archimedean point from which we “know” whether Betty has sufficient reasons to adopt X; all we have are real people’s conflicting opinions about whether Betty has sufficient reasons. And under these real conditions of disagreement, I think a good way to explain what’s wrong with the convergence view is to say that Alf cannot sincerely believe Betty is justified in endorsing X, and so Alf cannot support X consistent with the PJS. It doesn’t matter whether Betty is or is not justified in endorsing X from some objective view—all that matters is whether the real people who must interact with Betty can sincerely believe she is justified. If they cannot sincerely believe X is justifiable to Betty, then they cannot support it, regardless of whether X “truly” passes the BPPJ.

    Third, you suggest that the example I use is odd since it’s not really clear how the rule is imposed on Betty if she believes that the rule is justified. If both people want the rule, you ‘are tempted to say that the demand for public justifiability simply doesn’t apply, in this situation. If there is no disagreement and no conflict, there is no imposition, and so no need for a principle to determine when we can justify imposition.’ I don’t agree. There are plenty of cases where a person might believe that rule X is justifiable, but nevertheless find the existence of a rule to be an imposition when the time comes to follow it (I think plenty of people feel this way about the collection of taxes, for example). Just because we both believe a rule is justified doesn’t mean the rule need not be imposed at some later date, and so the question of public justification does arise I think (I also think it’s false to suppose the question of public justification only applies in contexts where a rule is imposed, but that’s another story).

    Finally, I’ll just briefly say that I am not persuaded by Jerry’s response to my position because I think Jerry’s new version of the sincerity requirement is too weak. Jerry’s new version of the sincerity requirement only demands that Alf see Betty’s deliberation as intelligible, but intelligibility is much weaker than justification. If it’s OK to impose rules on people provided one believes that those on whom the rule will be imposed have reasons that are intelligible and support the rule, then this no longer seems to be an account of public justification, but an account of public intelligibility (The Order of Public Intelligibility doesn’t have quite the same ring to it…). Many rules or laws might be intelligible without being justifiable. The reasoning of people who believe in astrology, for example, is intelligible to me, but I don’t think that fact means I can appeal to the ‘laws of astrology’ when justifying the imposition of laws on people who accept astrology.

  2. Thanks for your response, Jon. You make a good point about the level of idealization, which is that “intelligibility” seems to be much weaker than “reasonableness.” I downplayed this difference by emphasizing that Jerry is talking about intelligibility as a moral doctrine, assuming that classifying something as a genuine moral doctrine implies that it passes some minimal standard of reasonableness. However, some views are intelligible as moral doctrines but don’t accept the basic moral ideas that a Rawlsian would take take as criteria of full reasonableness: free and equal moral persons, burdens of judgment, reasonable pluralism, society as a fair scheme of cooperation. We can use Estlund’s generic term “qualified acceptability” to refer to different possible standards or levels of idealized acceptability. You’re suggesting that Jerry has substantially weakened his standard of qualification. Are we sure about this?

    Looking back over his 2007 “New Wine” paper, 2010’s “Classical Tilt,” and the forthcoming “Religious Belief” paper, I see Gaus specifying the degree of diversity in evaluative standards of members of the idealized public with the phrase “reasonable pluralism.” In the “classical tilt” paper he uses the phrase “intelligible and reasonable:”

    “I cannot think that your deliberation based on standard ΣX provides you… with a reason to endorse a law as binding unless in my view ΣX is an intelligible and reasonable basis for deliberation. That your unreasonable standard lead you to endorse L cannot lead me to think you have a reason to endorse L: garbage in, garbage out. Plausible justificatory liberalisms, then, must at least accept what we might call “mutually intelligible evaluative pluralism” at the level of members of P [i.e. the public, in Jerry’s technical sense]. Members of P will see themselves as deeply disagreeing about the basis for a law’s acceptance, but will acknowledge that the bases of others’ reasoning is intelligible and is appropriate to the justificatory problem.”

    I’m not sure how far Jerry now wants to weaken this standard. It seems clear based on his comments on reflexivity on p.228 that views need not accept the principle of public justifiability in order to count as qualified. ((I think in some earlier works he did assume that qualified points of view accept the principle of public justification in some form, but I can’t find the relevant quote at the moment, so maybe my memory is wrong)) That’s fine; it’s not crazy to care about acceptability to Joseph Raz even if Raz doesn’t care about reasonable acceptability. But what about acceptability to people who don’t share more basic notions that underpin the idea of public justification? Am I really supposed to care about the acceptability of moral rules to people whose doctrines are intelligible as moral doctrines but don’t recognize burdens of judgment, or reasonable pluralism, for example?

    More on sincerity and shared reasons later, but I thought I would just underline this question of interpretation so that we can get clear on what exactly Jerry now thinks about the appropriate level of idealization for delimiting members of the “public.” ((To avoid any confusion for those who may be following the discussion but not reading the book, “the public” is not the body of citizens; Gaus is not suggesting that we banish or otherwise deny the rights of citizenship to people simply because they are unreasonable. “The public” is the set of persons or moral perspectives that we recognize as sufficiently plausible as moral perspectives that we care about whether they can accept a given moral rule. If it still seems obviously wrong to exclude people from the public as unqualified even in this technical sense, then consider that people who reject the principle of public justification exclude everyone but themselves as unqualified. Correctness justification (to use Steven Wall’s term) is just public justification with a public of one – me or us, the person / people with the correct view.))

  3. About the two-person example; if we were only ever going to be two people, then any mutually acceptable imposition would be self-imposition. However, in any real situation, we are considering rules or laws for many people and future as well as existing people, and someone is bound to disagree with any given rule, so in any real situation, the principle of public justification will apply. It probably would have been better if I hadn’t raised this issue of whether application of the principle is triggered by disagreement, because it distracts attention from the main issue, which is best examined in the simplest possible i.e. two-person setting. I want to defend Jerry’s view, for the sake of argument, at least to make sure I’ve got it right.

    Jon, you say that in order to support a rule X, Alf must believe that Betty is justified in believing the parts of her doctrine that support X. Without some version of this requirement, the convergence standard allegedly allows us to engage in justifying rules on the basis of obviously silly doctrines such as astrology. I’m not sure why such examples can’t be taken care of by our reasonableness or qualification standard, whatever that is. Jerry’s standard may be too weak, so let’s accept your standard of reasonableness in doctrines, and suppose that our two doctrines pass this standard. Suppose that each doctrine consists of only two reasons: one shared reason Rp (for public) and one non-shared reason Ra or Rb (for Alf’s and Betty’s respective non-public reasons). The convergence standard is that the two total balances of reasons must be positive ( Rp + Ra, and Rp + Rb). The consensus standard is that the balance of public reasons must be positive (which in this case is just Rp, since there is only one public reason).

    Complications arise because people can disagree about whether these standards are met, but if there is a problem with convergence justification, it should appear in the core case in which Alf and Betty agree about what meets the standard. The problem should be evident when the two standards diverge, as for example when the convergence standard is met but the consensus standard is not. This happens when both total balances are positive but the public balance is negative. So we’re to imagine that Rp is negative but outweighed by the positive Ra (for those who believe Ra), or by the positive Rb (for those who believe Rb). If we decide based on the consensus version of public justifiability, we opt for not-X, on the grounds that Rp opposes X; if we decide based on the convergence version of public justifiability, we opt for X, on the grounds that Ra and Rb each separately outweigh Rp. Alf thinks Rb false, but not so clearly false as to be unreasonable. Alf also believes that if Rb were true, Rb would outweigh Rp. The same is true for Betty with respect to Ra. What’s wrong with convergence, in this case?

  4. Hi Andrew,

    Thanks for the response. It might be helpful if I just briefly restate my objection to Jerry’s position here, as well as how I understand Jerry’s response, and this might help clarify a couple of things in light of your further comments.

    I don’t deny that convergence justification can work and be consistent with the PJS in isolated cases. But in order for it to work and be consistent with the PJS in any given case, I argue that each person involved in the convergence must believe that all the other people involved are justified in endorsing the rule (X) at issue. I’m assuming a pure version of the convergence view where there are no shared reasons: each person believes X is justified purely for reasons grounded in their own comprehensive doctrine. So we’re imagining a case where Albert thinks Betty is justified in believing her different doctrine (even though he doesn’t adhere to this doctrine himself), and this is why Albert can, consistent with the PJS, support X even though he and Betty do not share any reasons which support X (and vice versa for Betty).

    The problem is that in order for Albert to believe that Betty can be justified in accepting a doctrine that is different from his own (and vice versa for Betty), he must hold a theory of epistemology or axiology that allows for this kind of justificatory diversity. He must agree with the claims Jerry makes (or that other philosophers make) that rational people can be justified in holding very different comprehensive doctrines. Obviously some people believe this is possible, but lots of reasonable people don’t think this kind of justificatory diversity is possible (e.g. those philosophers who adhere to some kind of externalist theory about what reasons a person can have). Just to clarify: thinking that someone else is justified in holding the doctrine that she does is not the same thing as thinking that someone else is reasonable or intelligible in holding the doctrine that she does. I think, for example, that religions based on literal readings of the Bible can be both reasonable and intelligible, but I also think such views are unjustifiable for almost any competent adult in the modern world to accept (I elaborate on this distinction on pp. 270-71 of my book). And so when we’re considering liberal societies generally, we cannot expect convergence justification amongst sizeable populations to be consistent with the PJS since expecting this would require everybody involved holding a particular philosophical position about justificatory pluralism (and also believing that all the other people involved in justificatory process have doctrines which happen to be justifiable), and this is too demanding an expectation in light of the fact of reasonable pluralism. I summarize this argument as follows (p. 272 of my book):

    1. Convergent justifications amongst people adhering to different comprehensive doctrines can only be made consistent with PJS provided each person involved sincerely believes that the other people involved are justified in adhering to their different comprehensive doctrines.
    2. The belief required in (1) is generally not possible unless citizens accept certain epistemological or axiological doctrines (e.g. Gaus’s).
    3. The fact of reasonable pluralism means we cannot and should not expect citizens in a liberal society to adhere to any particular epistemological or axiological theory.
    4. Therefore, as a general rule, we cannot expect convergent forms of justification to be consistent with PJS in a liberal society.

    I understand Jerry’s response to my argument as follows. He rejects my argument by denying the first premise. He denies the first premise because he now insists that convergence justification does not require that each person sincerely believe that the others are justified in adhering to their doctrines (and hence justified in endorsing X as their doctrines apparently require): he only requires that each person see the others’ doctrines as intelligible, and hence offering intelligible reasons in support of the X. Intelligibility, I take it, is a weaker standard, and so Jerry can then go on to say that it’s not too much to expect all the members of a liberal pluralistic society to see most other members as having intelligible doctrines. Albert does not have to hold any specific epistemological or axiological theories to see Betty’s doctrine as intelligible, which means the second and third premises of my argument are rendered irrelevant, and so naturally the conclusion no longer follows.

    My response is to say that intelligibility, when construed in this weaker way, seems too weak. By weakening the standard of sincerity to intelligibility, Jerry can defuse the argument above, but only at the price of no longer achieving rules that are justified to each person. If Albert thinks Betty’s comprehensive views are intelligible but unjustifiable, and if there are no shared reasons in favour of X, then Albert cannot believe that X is a justified rule for Betty, and it would be wrong of him to support it. If, on the other hand, intelligible means roughly the same thing as justifiable, then my argument above still works.

    Regarding your second comment. In my book I consider a pure version of the convergence view where there are no shared reasons in favour of the rule: each person has only their own ‘private’ or comprehensive reasons. I think this is the best way of evaluating the plausibility of the convergence approach—the convergence view (as I understand it) is not committed to the idea that the balance of shared and non-shared reasons for each person must be positive since there can be cases without any shared reasons where each person believes there are sufficient non-shared reasons to endorse the rule. So I don’t think we need to worry about shared reasons in order to assess the convergence view. Setting that point aside, hopefully what I’ve said above clarifies what I think is wrong with the convergence view. I don’t deny convergence models of public justification are possible and consistent with the PJS, but I do deny this is something we can expect to be the case in liberal societies characterized by reasonable disagreement.

  5. After reading the back-and-forth above, I’m wondering whether PJS conflicts with Rawls’s idea of reasoning from conjecture. Suppose Al believes that a nonpublic reason, NRA (no pun intended), supports policy p, and that Al believes there is no sufficient public reason for p. In Rawlsian terms, Al has no pro tanto justification for p. Suppose further that Al decides that NRA is strong enough to outweigh his commitment to public reasoning. On the basis of NRA, he is willing to impose p, even though he knows there is no public reason to support it.

    Now suppose Betty thinks that Al has misunderstood his own comprehensive view. She thinks that, rightly understood, Al’s view requires the rejection of NRA. But Betty also thinks Al’s comprehensive view isn’t justified. Al’s view is a form of theism, and Betty is an atheist. She thinks all forms of theism are unjustifiable. But she also believes that certain conclusions follow from Al’s theism, one of which is the rejection of NRA. Can she say to Al, “I think your comprehensive view is unjustifiable. It’s not just that I disagree with your view. I think it’s silly. But since you persist in maintaining that view, I think it commits you to rejecting NRA.”

    There isn’t anything obviously insincere about this claim. But it does seem to violate PJS. Does that mean that we shouldn’t try to convince those with (what we believe are) silly views that they can still be reasonable (in the Rawlsian sense), i.e., that they can accept the burdens of judgments, the ideas of freedom and equality, the value of reciprocity, and the commitment to public reason that follows from these ideas? Provided we disclose that we think others’ views are unjustifiable, and provided we are clear about our intentions in arguing for a commitment to public justification, is there something wrong with engaging others on their own terms, even if we think those terms are, all-things-considered, unjustifiable? Another way of putting the point might be to ask whether we should care how citizens reach the conclusion that they ought to be reasonable. Provided that they are reasonable, should we care about their reasons for affirming reasonableness? (There might be good answers to this question; e.g., we might want to know whether others’ disposition toward reasonableness will be stable over time, and so we might want to inquire into the sources of that disposition.)

    There is a deeper concern here. Political liberalism is based on the idea (or the hope) that reasonable comprehensive views can converge on a set of moral ideas that form the basis of a commitment to public reason. But that convergence might be based, in part, on doctrines that some citizens believe are unjustified. If that case, when we offer others public reasons as pro tanto justifications, we have to consider that they may achieve full justifications on the basis of what we think are unjustifiable comprehensive views. Is that something that political liberals ought to worry about it, or is it a virtue of the theory that citizens need not be concerned about the deeper (nonpublic) justifications others offer for their commitment to public reasoning? If the latter, how does PJS fit into this picture?

  6. Hi Micah,

    I think this is a really interesting issue. Here’s my off-the-cuff thoughts about what you say.

    I think Rawlsian reasoning from conjecture can be consistent with the PJS under at least two conditions. First, if Betty believes there are sufficient shared reasons for rule X, but she also believes Alf’s unjustifiable comprehensive doctrine supports X, then I think she can engage in sincere reasoning from conjecture with Alf where she tries to show him his own doctrine commits him to X (but she probably ought to publicly articulate the shared reasons too if we believe in a strong publicity condition). Second, in the case you describe, Betty is only trying to dissuade Alf from supporting the imposition of a rule by appealing to Alf’s comprehensive doctrine—she’s not advocating in favour of any particular position, and so this might also seem consistent with the PJS in the sense that she is not supporting any rule, and so it might seem the PJS doesn’t apply—provided that Betty does not try to persuade Alf to support rules for what she believes are reasons that Alf is not justified in accepting, and provided she does not herself believe that the rule in question (NRA) has a sufficient shared justification, then there’s no conflict. I think the first case is more important than the second since it’s much more likely to arise.

    Regarding whether the PJS causes problems for political liberalism (PL) when we consider the broader issue of the overlapping consensus amongst reasonable people. My views on this are complicated, but in brief, I think an essential part of PL’s strategy of epistemic abstinence involves taking it as given that all reasonable persons are justified in endorsing the fundamental political values, and accepting the burdens of judgement. PL does not directly enquire into the question of whether individuals are justified in accepting those values in order to abstain from controversial epistemological theories over which we assume reasonable people disagree. We just take the fundamental political values as given, and then can appeal to those shared values in any instance of public justification and stipulate that in doing so, we can assume our starting premises are justified to reasonable persons. Making good on that assumption is something political liberalism as a theory remains silent about: we leave that up to individual citizens as part of the background culture or comprehensive philosophy.

  7. I think the distinction in Andrew’s original post is important: between
    giving individual members of the public a veto over proposed rules, or
    giving them a veto over proposed reasons for decisions about rules.

    I think it is important because it implies that certain theoretical and
    practical problems surface at different “locations” in the different
    theories. Gaus thinks that the first kind of veto “surely is objectionable”
    (286); and I think other authors, e.g. Christiano, have made the same kind
    of point against a duty of civility or a principle of reasonableness. Yet I
    think similar reasonable objections can be made against a veto over proposed
    rules, or – with respect to the real world – supermajoritarian legislative

    I deal with these issues in more detail in a paper that focuses on Gaus’
    Justificatory Liberalism but now also discusses some aspects of OPR and the
    relation between the two books. We will get to some of these issues. If
    anyone is interested, the paper is available here (comments and criticism
    would of course be welcome):


Leave a Reply