Sincerity, Individuation and Classical Liberal Public Reason: Comments on Chapter 4

Let me begin by thanking Blain for an excellent recap of Chapter 4 of the book, which is arguably the centerpiece chapter. He raises five important concerns, but I’m going to set two aside. First, Blain raises the question of my Rawls exegesis. I suspect that is something better dealt with in a journal format or conference proceeding. It is interesting and important, but my main arguments do not depend on it. I will say, briefly, that yes, the process of generating convergence justifications can encourage revision of pieces of certain comprehensive doctrines. The second issue I set aside concerns my indirect model of public justification and the idea that restraint (of a certain sort) applies to legislators but not to citizens. That is one of the two main questions at issue in Chapter 6, so I’d like to push discussion to that post. But briefly, a lot of the case for the indirect model is based on the fact that citizens complying with restraint is neither necessary nor sufficient to promote publicly justified outcomes given all the other stages between a popular vote and the passage of legislation. 

So let’s focus on sincerity, individuation and the question of how classically liberal public reason is. 

I. Sincerity 

As Blain notes, in Chapter 4 I argue that sincerity requirements cannot be used to refute convergence views, as Jon Quong maintains, because sincerity requirements’ normative force is entirely derivative from a general commitment to offering one’s reasons for supporting a law and an account of what those reasons are (convergence, consensus or whatever). Blain is worried that sincerity still has some unique normative remainder that is not entirely derivative from other principles. 

Blain asks us to imagine a particular case of what Rawls called reasoning by conjecture, where A tries to convince B to support a law by appealing to B’s comprehensive reasons. However, in Blain’s case, the B-reason A appeals to best supports L2, a law that A rejects or finds sub-optimal, though the B-reason nonetheless (though to a lesser extent) supports L1, the law that A proposes. Now suppose that A appeals to the B-reason as a reason to support L1 but doesn’t raise the fact (of which A is well-aware) that the B-reason better supports L2.

To put this more concretely, imagine that secular A is a fan of single-payer healthcare and wishes to convince Catholic B to support single-payer. A appeals to B’s Catholic reasons to believe that God has a special preference for the poor and that poor will do better under single-payer than our present system in the US. However, let’s suppose secular A knows that a two-tiered system would actually work better for the poor on the whole, as single-payer often prohibits private insurance. Secular A supports single-payer because it is more egalitarian than a two-tiered system, but A knows that Catholic B doesn’t share that egalitarian concern. So A tells B, “Well, Jesus said that what you do for the least of these, you do for Him, right? And single-payer helps the least of these, right?” But A then omits a more moderate possibility, that of a two-tiered system. Is there insincerity here? 

I think that there is a clear potential for insincerity in this case, but that the fact of insincerity depends on A’s assessment of the, for lack of a better phrase, “justificatory spread” between a two-tiered system and single-payer. If Catholic commitments genuinely support single-payer, but less well than a two-tiered system, then I don’t think A is insincere in offering an argument for her position. But if A thinks that Catholic commitments support single-payer far less well than a two-tiered system, then I can see the insincerity. In politics, I don’t think public discussion requires total disclosure of one’s views in political discussion. 

But suppose there is an insincerity, would that be a problem for my line on sincerity? I don’t think so. There are lots of ways to be insincere, dishonest or deceptive in politics that are forbidden by ordinary moral rules. My claim is merely that the restraint that is typically said to fall out of a commitment to public justification cannot be justified by appealing to a sincerity requirement derived from a commitment to public justification. 

II. Rough-grained Individuation 

One of the most important issues in figuring out the implications of a commitment to public justification is determining which unit of coercion must be justified. Is it constitutions? Parts of constitutions? Laws? Bits of laws? The question Andrew Lister originally raised, and Jerry Gaus and Jon Quong have further analyzed, is how to individuate coercion, and whether to do so finely or coarsely.

I am on the fine-grained end of the spectrum: public justification is aimed at the justification of coercive laws, and the justification of all other coercion is to be understood in terms of the justification of laws. For instance, I will argue in a forthcoming paper and in what I hope will be my next book that we should choose constitutional rules in public reason based on the relative frequency that they output publicly justified law and prohibit or reform publicly unjustified law in comparison with other constitutional rules. So we don’t publicly justify constitutional rules directly, but in terms of their effects on laws (with an important exception we can discuss later).

What Blain proposes is a very ‘rough-grained’ individuation of coercive measures, such as only publicly justifying a society’s constitutional essentials and basic structure as an overall system. This is Rawls’s view. But recall that Rawls allowed that you might be able to apply the restrictions of public justification to smaller bits, like laws. He just didn’t require public justification at that level once the system was in place.

There are two ways to try and show Rawls is wrong: one path charted by Quong, the other by Gaus. I prefer the Gausian line (surprise), though I think Quong is right as well. The basic Gausian argument in OPR is that public justification always must bear clearly in mind the members of the eligible set of potentially justified proposals. That is, when we propose to publicly justify constitution C, what are the alternatives? Does a model of public justification always vindicate only one option? Well, Rawls certainly denied this. Justice permits either property-owning democracy or liberal socialism. We could institutionalize either one. Another thing we must bear in mind is that when we insist on public justification we are in effect asking whether some proposed arrangement is better than no arrangement at all. That is, all publicly justified laws are thought to be better than nothing. But this suggests something unsettling: if publicly justified arrangements are those that are better than no arrangement, then we may end up with a really big eligible set of constitutional orders whose members include lots of really illiberal regimes. For instance, surely various forms of autocracy, if properly limited, are better than no constitutional regime, and this would be so from all reasonable points of view, so why can’t we publicly justify autocracy? In sum, the worry with going coarse-grained is that you can publicly justify really illiberal stuff because illiberal stuff is better than a whole lot of nothing.

One could respond that we’ve already publicly justified principles that serve to rule out such regimes, but the question of “ruling out” is just the one we’re asking. Suppose we have the true principles of justice on the table and autocracies mildly realize the principles, whereas anarchy doesn’t realize them at all. Then what would render autocracy illegitimate? We do have sufficient reason to endorse it, after all, since it is better at realizing justice than no regime.

To give sense to our intuition that we can insist that the eligible set of justified regimes be more strictly limited, I think we have to tighten up our individuation of coercion. The reason autocracy can’t be publicly justified in general is that it generates lots of bad laws and policies that can’t be justified to free and equal people. It would be better to not have autocracy because it would be better not to have the laws distinctively passed by autocracies than to have them. In short, finer individuation can help us capture our sense that it isn’t enough to justify a regime to show that it is better than anarchy.

III. Classical Liberal Public Reason

Gaus has argued that public reason has a classical liberal “tilt” in that publicly justified orders will tend to have limited government not just in protecting individual civil rights but in protecting relatively extensive private property rights in capital. In a classical liberal regime, states can restrict property rights more than in libertarian regimes, in large part to provide public goods and establish limited social insurance.

Egalitarian liberals don’t like classical liberalism, but they think classical liberal regimes are better than nothing. However, the reverse does not hold. Many classical liberals would rather try a much more limited legal regime than go for egalitarianism. If so, then egalitarian liberalism falls out of the eligible set, leaving classical liberalism by itself. Blain, however, rejects this asymmetry. Either classical liberals come to believe that egalitarian liberalism is better than some forms of anarchy, or the egalitarian liberals come to believe that anarchy is better than classical liberalism. Whatever direction we go, “the cases are symmetrical.”

Now, for all I say in the book, Blain could be right. I’m trying to show that merely adopting convergence does not entail classical liberalism. So let me be clear: merely adopting convergence without other modifications to Rawlsian public reason does not entail classical liberalism. It will entail a more limited government than on the consensus view, in many circumstances, anyway, but not a classical liberal state.

However, I think there are other modifications to public reason that begin to push public reason in a classical liberal direction. In particular, combining convergence with a fine-grained individuation of laws will take you a ways towards classical liberalism, along with the conception of coercion we settle on (or at least the core cases of coercion that we all agree are coercive). A lot will also depend on the justification of basic rights (what in my next book I plan to call “primary rights” to resemble Rawls’s notion of “primary goods”), and about how extensive rights to property can be justified at this first stage.

In sum, I thank Blain for the really helpful summary and comments. Lots of good issues here and lots of important questions that need to start getting addressed in the literature. Public reasoners, let’s get writing!

This entry was posted in Reading Group and tagged , . Bookmark the permalink.

9 Responses to Sincerity, Individuation and Classical Liberal Public Reason: Comments on Chapter 4

  1. Thanks very much to Kevin for his interesting replies to my comments! I have some questions about them, though, and so would like to push him on them here. (I’ll do so in three different posts, one devoted to each of Kevin’s replies.)

    I. Sincerity

    It seems that Kevin concedes that in cases like the proposed health care law scenario that he helpfully outlines (at least in the version of that scenario in which there is a significant “justificatory spread” between the two rival proposals in question), he “can see the insincerity.” Such insincerity, though, is not a problem for his position, he avers, as “there are lots of ways to be insincere, dishonest or deceptive in politics that are forbidden by ordinary moral rules.”

    I found this reply somewhat surprising, given the central role that a concern with personal integrity plays in Kevin’s defense of the convergence account. The version of ‘integrity’ that Kevin employs refers to the ability of citizens to act upon their overall considered convictions (values, life-projects, beliefs, and so forth), and thus not be forced to restrain (or ‘divide’) themselves in their political lives. But presumably it is a common feature of many people’s moral codes that they conduct themselves with honesty and sincerity in their relations with others. Consequently, to countenance insincerity and dishonesty in politics when justifying laws would seem to threaten (many) citizens’ integrity.

    Also, it strikes me that it remains an advantage of the shared reasons view that it does not encounter this problem. If citizens A and B share public reasons C, they both can appeal simply to C in justifying whichever policy they favour in a straightforward, sincere manner (of course, they may disagree over whether law X or Y is most justified by C, but that is not a problem for the shared reasons view). So the shared reasons account avoids this problem of integrity altogether. Ironically, the shared reasons account consequently may cohere better with (many) persons’ moral views than the convergence account that permits insincerity and dishonesty in political deliberation.

    (I assume that Kevin’s view is that even if I am right about this feature of the shared account of public reasons, the other features of the shared account and convergence account ultimately weigh in favour of the convergence account.)

    • Blain, I agree with you in criticizing Kevin’s claims that “most sincerity requirements decompose into a reasons requirement” and that our approach to sincerity will “fall out of” our views on justificatory reasons and restraint (124). But I’m also sceptical of the view that we can justify restraint by appealing to sincerity. The example you give clearly involves sincerity. But it also seems possible to imagine sincere deliberation in a convergence framework. It just involves frank, full-disclosure argument from conjecture, along the lines of “I don’t believe X, but you do, right? And it seems to me that X supports policy Y…” There need not be anything manipulative about that kind of debate, so long as I really do think that X would, if it were true, support Y (more than whatever the alternate policy is).

      Does the restraint requirement rule out insincerity? That’s what you seem to be suggesting in saying that the shared reasons approach avoids the problem. If we both recognize the need to appeal to shared reasons, then we have no reason to make arguments from conjecture, and hence no insincere arguments from conjecture. There are probably other kinds of insincerity, however. Even if I recognize a restraint requirement (i.e. decide based on public reasons), I might argue strategically, making claims that I think false or illogical but which I calculate may persuade others to support the policy that I think publicly justifiable. If the stakes are high enough, that might even be the right thing to do, despite the insincerity involved. For one thing, I could make empirical claims I take to be false, to get you to support the policy I think publicly justified, when you think a different policy is publicly justified.

    • The way I understand integrity considerations is that liberals shouldn’t impose restraint because it violates integrity, but the theory is not in the business of requiring people to act with integrity, which is what it seems to me you’re suggesting in your criticism. Public reason doesn’t require that people be sincere in their political deliberations, in my view, and insincerity might lead them to undermine their own integrity, but I can’t see a case for restraint at work here.

      Also, I think some insincerity has to be permitted in politics just to accommodate strategic considerations, say when forming a political alliance with another group. You don’t have to disclose all your motivations, at least.

  2. II. Holism versus Independency in Justification

    While I found Kevin’s comments here interesting, I did not see how they addressed my concern with coherence. Perhaps I’m simply missing something, but let me try to press my concern again.

    In On The Order of Public Reason, Gaus notes a criterion of “justificatory dependency” for individuating issues. According to this criterion (very roughly), issues A and B are independent only if no citizens (‘Members of the Public’) have rankings in which proposals concerning A depend upon the laws and policies concerning B. As Andrew Lister notes (correctly, in my view) in his review of OPR, “This criterion introduces a much stronger degree of holism into Gaus’s model of public justification than was initially apparent.” Indeed, if (some) citizens regard all fundamental political issues — all constitutional essentials and matters of basic justice — as interdependent, such that they cannot rank any proposal concerning one essential or matter without taking into account the other essentials and matters, then Gaus’s own criterion of justificatory dependency means that we should employ Rawls’s unit of justification!

    Moreover, one reason why Gaus favours justifying individual rules (which Kevin seems to appeal to in his reply to me) is that doing so is not as cognitively demanding as ‘holistic’ justifications of entire ‘systems’ (OPR, pp. 272-273), such as Rawlsian ‘basic structures.’ But given the criterion of justificatory dependency, if many citizens’ views of particular laws and policies depend upon what other laws and policies exist, then I’m not sure the fine-grained approach to justification favoured by Gaus and Kevin really is less cognitively demanding.

    Consider the following citizens:

    Amy: supports L1 only if L2, L3, and L4 exist.
    Bernie: supports L1 only if L4, L5, L6, L7, and L8 exist.
    Cathy: supports L1 only if L2, L5, L9, L10, and L11 exist.
    Denis: supports L1 only if L5, L7, L12, L13, and L14 exist.
    Ellen: supports L1 in isolation.

    Now in this situation, given the criterion of justificatory dependency, sorting out whether L1 is actually justified looks very cognitively demanding! It might actually be easier for citizens to evaluate and rank Package 1, Package 2, and Package 3 (with each package including a set of internally consistent and coherent laws, say, basic structures organized in accordance with different reasonable conceptions of justice).

    Also, I don’t see how the example of the autocracy that Kevin brings up actually helps the fine-grained view. If we are concerned only with evaluating individual laws, then the fine-grained liberal may end up supporting a lot of what the autocracy does! Say the autocracy passes and enforces laws that respect religious liberty, allow for private property, permit some measure of freedom of expression (just not expression that threatens the authority of the regime), and so forth. If we look at the laws one-by-one, the fine-grained liberal might find herself endorsing a lot of the laws passed and imposed by this autocratic regime. In contrast, a liberal ‘holist’ could maintain that since all of these laws were passed by an autocratic regime, the entire system is rotten and illegitimate.

    • Some good thoughts here, but I suppose a lot of your first concern will depend on how interconnected most folks’ views really are in the context of imposing, repealing or reforming a law. My impression, and I thought this was commonsense, is that we’re capable of evaluating lots of issues independently of others. My views on the first amendment don’t have any strong dependence on my views on healthcare economics.

      As for the autocracy stuff, it is true that the law-by-law liberal could support some of the same *laws* as the autocrat, they won’t end up being forced to *support autocracy* rather than anarchy. So it seems to me that the commitments of the law-by-law liberal are much more secure than the commitments of the holistic liberal.

      • I suspect that a significant reason why it appears that most citizens evaluate proposed laws ‘in isolation’ is that they simply do not make explicit the ways in which their evaluation of those proposed laws presupposes the system of already existing laws. If (many of) those existing laws also were made open for revision, then I think that considerable interdependency would be revealed (at least for more fundamental laws). Indeed, I think that we see something like this when societies design new constitutions, or consider significant amendments to existing constitutions. In such cases, there often is careful attention to how the various features fit together into a unified, coherent system.

        Moreover, interdependency can be quite complex. To use your example, while my commitment to freedom of the press and my commitment to (some form of) publicly provided healthcare might not seem to be directly interdependent, they may turn out to be indirectly interdependent. My commitment to freedom of the press as a basic liberty (to be considered a ‘constitutional essential’ within my society) requires that it be rendered compatible with the other basic liberties, so that all of the basic liberties can be exercised adequately by all citizens. And I also think that access to basic healthcare is necessary for the adequate exercise of those basic liberties by citizens over the course of their lives.

  3. III. Coercion, Fine-Grained Justification, and Justificatory Liberalism’s Classical Tilt

    I have two questions based upon this comment from Kevin: “combining convergence with a fine-grained individuation of laws will take you a ways towards classical liberalism, along with the conception of coercion we settle on.”

    In a previous discussion, Kevin helpfully pointed out to me that Gaus’s argument for justificatory liberalism’s ‘classical tilt’ does not depend upon any shared conception of coercion. Thus even if egalitarian liberals find a classical liberal regime to be highly coercive — given the widespread coercive enforcement of the unequally distributed private property that such a regime would entail, and the absence of public spaces that it would contain (imagine how much more coercion homeless people would face in a classical liberal regime than they do even now) — they nonetheless would concede that such a regime would be preferable to anarchy. So Gaus’s argument does not require any shared conception of coercion amongst classical and egalitarian liberals. Yet Kevin’s reply here suggests that there is (at least potentially) a shared conception of coercion. So I find myself wondering, again, whether Gaus’s argument for a ‘classical tilt’ is supposed to work without a shared conception of coercion.

    Second, Kevin’s point about the “fine-grained individuation of laws” moving justificatory liberalism towards classical liberalism brought to mind the following question: do justificatory liberals (like Kevin and Gaus) support the fine-grained individuation of laws because — at least in part — they think that it makes it easier to support classical liberal political arrangements? I realize that there are other considerations that they adduce support of the fine-grained approach (such as the ‘cognitive burdens’ claim). But I wonder if this is not one reason. (Are there any egalitarian or even non-classical liberals who support the fine-grained individuation of laws approach to public justification?) If this is the case, then perhaps it would be better simply to justify classical liberalism directly, rather than employing this more indirect strategy to do so?

    Conversely, does the ‘holistic’ (or ‘rough-grained’) approach to justification have any ‘egalitarian’ tilt? I don’t see how it does, but perhaps I’m missing something. (When I was an ardent classical liberal, back in the days of yore, I was happy to try to justify classical liberalism on ‘holistic’ grounds. But perhaps my own apostasy indicates that the holistic approach indeed ‘tilts’ against classical liberalism!)

  4. About this:

    “Gaus’s argument for justificatory liberalism’s ‘classical tilt’ does not depend upon any shared conception of coercion. Thus even if egalitarian liberals find a classical liberal regime to be highly coercive… they nonetheless would concede that such a regime would be preferable to anarchy. “

    That’s a situation in which the two parties agree on the zero-public -coercion point, which is functioning as the default. It’s true that they don’t have to agree how regimes rank beyond that point, in terms of coerciveness. That’s because we only have to publicly justify some coercion as opposed to none, not more as opposed to less. All coercion requires justification, of course, and more coercion requires more powerful reasons in its favour. But the idealized unanimity requirement doesn’t have the form “more coercive policies must be unanimously acceptable as improvements over less coercive policies, otherwise we should default to less coercive policies.” That would be an incremental version of the principle. Disaggregation into finer-grained policy choices is closely related to incrementalism.

    I think you’re right that a more holistic, joined-up approach does not have an egalitarian tilt. If we are reasoning at the level of whole regimes, both egalitarian and aristocratic regimes might pass the better-than-nothing test.

    Gaus’s test has two levels, however, if I remember correctly. For example, at 436 he distinguishes between a system of property being a desirable social convention compared to the alternative of a Hobbesian state of nature, and a system of property having moral authority. To have moral authority, a system of property has to pass a higher threshold than simply being better than a war of all against all; it has to be better than the no-rules scenario that would obtain among members of the modestly-idealized Public, I think, though I can’t find the page reference at the moment. The idea is to imagine that a set of somewhat idealized but still very diverse people who in the absence of a common rule would each act according to their own conscientious convictions. If a givne rule is an improvement on that, it is authoritative, if it is in fact the rule we have settled on.

  5. Thanks for the questions, Blain. I could have been clearer when I suggested that “we” might settle on a conception of coercion. I was imagining that public reason theorist setting out a conception of coercion that is sufficiently loose to count as eligible for all members of the public, a conception that will include lots of core cases but leave out some controversial stuff. Someone might argue that we should rule out work at low wages as coercive (on the grounds that there’s no such thing as a coercive offer), whereas another theorist might deny this. But they’d both agree that theft is coercive. Depending on how that process works out in the end, we could end up with more or less classical liberalism.

    As for the second concern, I sure *hope* that I don’t support fine-grained individuation because it leans towards conclusions I prefer vis-a-vis other public reason liberals, but of course it is hard to introspect your ultimate motivations on the subject. It is certainly not one of my *explicit* reasons. I like to think many of the cited reasons are sufficient to establish more fine-grained individuation.

    And while I don’t think coarse-grained coercion has an egalitarian tilt (I’m not sure *what* tilt it has, if any, partly for reasons Andrew cites), it seems to be to the left of the fine-grained view.

Leave a Reply