This post provides a summary of Section V.15 and then raises some preliminary questions.
In the last section (V.14), Gaus advanced the Basic Principle of Public Justification and developed a Deliberative Model for determining whether social morality satisfies that principle. Andrew Lister summarized the parameters of the model here. Most of section V.15 is devoted to specifying the model further by constraining the set of rules that may qualify as proposals within it.
Before discussing particular contraints, however, Gaus describes the Deliberative Model as reflecting the Kantian idea of “legislation in the realm of ends” (292). Every person represented in the model is both subject to the law and the legislator of it. But, as Gaus notes, this is a fairly sophisticated idea. What if normal moral agents (NMAs) can’t understand or appreciate it? We can’t expect most people to grasp the complex philosophical arguments offered for reasoning from within the Deliberative Model. As Gaus says, not everyone is a stage 6 Kohlbergian reasoner. The question, then, is how to develop the Deliberative Model so that justifications for proposed rules of social morality (or “proposals”) are accessible to all NMAs. The answer is that the model itself need not be accessible to all NMAs; only the reasons offered for proposals must be accessible (and acceptable) to them. You don’t have to be able to read (or understand) the Order of Public Reason to have sufficient reasons for endorsing rules of social morality, but you do have to be able to understand the reasons that justify those rules.
Having addressed this concern about the accessibility of the Deliberative Model, Gaus next specifies and defends a number of formal constraints that qualify proposals as moral rules, rather than rules of some other kind. These constraints are as follows:
1. Generality: Many philosophers (including Rousseau, Rawls, and Scanlon) have argued on grounds of fairness and nondiscrimination that proposals should be general in the sense of not including classifications based on proper names. But understood in this way, generality is a weak constraint, since laws that do not classify by proper names may treat other classifications (e.g., those derived from social roles) unfairly. Still, following Lon Fuller, Gaus says that a modest restriction on proposals is that they must regulate by using general descriptions.
2. Weak Publicity: Proposals are subject to a weak publicity condition, which requires that everyone in society can be taught them, including children. Following Baier, Gaus calls this a “teachability requirement” (296). Only rules that satisfy this requirement can serve the function of coordinating social behavior. Any proposals that are more complex, or that require case-by-case judgments, are therefore excluded from the Deliberative Model.
Gaus distinguishes weak publicity from two other publicity conditions. Whereas weak publicity requires only that proposed rules be accessible to all NMAs, strong publicity requires that the reasons for those rules be accessible to all. As we have seen, Gaus believes that all NMAs must have sufficient reason to endorse proposals. This explains why strong publicity is central to a publicly justifiable morality (296). But on Gaus’s view, it does not follow that strong publicity is a formal constraint on proposals within the model (297). The purpose of formal constraints is to distinguish moral rules from non-moral ones. Since a rule can be moral (or at least a candidate for morality) even if its reasons are not accessible, strong publicity is not included in the model as a formal requirement. Yet, as Gaus explains, a “great advantage” of the Deliberative Model is that it yields proposals that satisfy strong publicity. This is a result of the model, rather than a condition built into it.
In contrast with weak and strong publicity, complete publicity requires that rules and their reasons — the reasons that justify them — must be common knowledge, such that everyone knows that everyone else knows that they know, and so on. It is obvious that this state of affairs cannot be guaranteed by the Deliberative Model, and so complete publicity is not a formal constraint on proposals (though it would certainly be desirable for social morality to achieve it in the real world).
3. Conflict Resolution and Claim Validation: For Rawls, moral persons are viewed as “self-authenticating sources of valid claims,” and the point of a decision procedure for selecting moral principles is to resolve conflicts between those valid claims. Gaus rejects this view for failing to respect others’ moral autonomy. Moral claims are only valid when others have sufficient reason to accept them. The Deliberative Model is a device for determining when others have sufficient reason, or, in other words, when we have a valid claim against others. The proposals considered in the Deliberative Model may contain priority rules for resolving conflicts, but these are not conflicts between valid claims. They are conflicts between claims simpliciter; the rules selected in the Deliberative Model tell us which of them is valid.
4. Rules as Requirements: Proposed rules cannot be merely guidelines; they must be based on reasons weighty enough that they usually override other goals and values (298). In Razian terms, we might say that this constraint does not require that rules amount to non-defeasible exclusionary reasons, only that they normally obligate compliance.
5. Universalizability as Reversability: Gaus rejects the idea that moral rules must be universally applicable in the sense of holding for all moral persons. Since the Deliberative Model represents Members of the Public (MoPs), who are members of a particular society (sec. V.14.3), this condition is too strong. Instead, Gaus argues that a rule is moral only if it is universal in the sense that a person’s endorsement of it does not depend on knowledge that he or she will occupy a specific role or position (e.g., race, sex, class, etc.) subject to the rule (300). Furthermore, this conception of reversibility does not require bracketing one’s evaluative standards, or asking whether one would endorse a rule if one had another’s evaluative standards. In this regard, Gaus’s conception of universalizability is different than Rawls’s. Gaus’s deliberators are not under a veil of ignorance with respect to their own evaluative standards. They can, and indeed must, appeal to their evaluative standards to determine whether they can accept a rule from the perspective of the various roles and positions governed by the rule.
6. Modest Common Good Requirement: A rule does not count as moral if it would undermine the good of other Members of the Public (301). More specifically, although moral rules may require “episodic sacrifices” of others’ good, such rules may not pose threats to their “core interests,” including protection of bodily integrity, basic liberties to make important life decisions, control over basic resources, etc. (302-03). If Members of the Public believe that a proposal would undermine fundamental aspects of a person’s good, they must conclude that at least some others would not have sufficient reason to endorse it.
Thus, to count as a rule of social morality, a rule must have (1) generality, (2) weak publicity, (3) the aim to resolve or validate claims, (4) weighty requirements, (5) reversibility, and (6) conformity with the basic good of all Members of the Public. Only a rule that satisfies these constraints can qualify as a proposal in the Deliberative Model.
A general question to ask about this section is whether Gaus has correctly identified the constraints on what counts as a moral rule, as opposed to a publicly justifiable moral rule. Are his constraints too strong or too weak, or perhaps both? Just to get things started, here are some preliminary questions along those lines:
- Is weak publicity too strong a constraint in specifying which rules count as moral? Why couldn’t a stage 6 Kohlbergian reasoner propose a rule that isn’t accessible to everyone but that only applies to stage 6 Kohlbergian reasoners? Suppose the rule places no demands on anyone who hasn’t reached that level of sophistication, and, further, that it satisfies all the other constraints. The only defect in this rule is that some normal moral agents cannot be expected to follow it (by “ought implies can”). Gaus says that “[r]ules that are too complex to be reliably taught to children could not be followed by the population in general.” But this is probably false. Think of rules involving sexual relationships. Many of these rules can’t be taught to children (or should not be taught to children), and yet they qualify as moral rules. Are there other examples that do not involve children? It might be possible to identify role-based obligations that are not accessible to those unfamiliar with the relevant roles. In this case, the stage 6 MoP might say, “You might not understand this rule, but it does not require anything from you, though it does impose requirements on others who can understand it.” This might sound patronizing, but then we are dealing with different levels of moral ability. If everyone capable of reasoning in the relevant way has sufficient reason to comply with a rule, why should the failure of those who are incapable of grasping the rule make it such that the rule is not a moral rule? True, it isn’t a moral rule for them, but why not for the others? More generally, we might wonder whether a society could have a “tiered” morality, such that all MoPs must conform to all rules accessible to them. MoPs capable of higher levels of reasoning would be subject to all the rules accessible to those less capable, but they would also be subject to additional rules accessible only to those with higher capacities. This tiered model would still provide a baseline of morality to coordinate “mutually benefit social existence” (296), while allowing for “improvements” from the baseline through the additional coordination of capable MoPs. Of course there might be feasibility problems in identifying the relevant capable MoPs, but would such problems disqualify a rule from consideration at the outset? (To be sure, I’m not endorsing the idea of a tiered morality — far from it; I’m only asking about how Gaus might address this possibility.)
- Gaus says that Rawls’s idea of persons as “self-authenticating sources of valid claims” is puzzling because it “implies that others are bound by moral claims that they cannot authenticate through their own reason” (298). But I wonder if Gaus is reading Rawls correctly here. In Political Liberalism, Rawls uses the idea of “self-authenticating sources of valid claims” to describe a person’s freedom. People are free when they can make claims on others based on their own conception of the good. They do not need to rely on the good of others, or on their particular social role in society, to assert their claims. Rawls gives the example of slaves whose rights against maltreatment do not originate with them, but with the interests of slaveholders or with society in general (PL, p. 33). Unlike slaves, MoPs are self-authenticating sources of valid claims because they do not need to rely on others to advance their claims. Indeed, Gaus seems to recognize something like this point in discussing his common good requirement. MoPs must reject any rule if they have an undefeated reason for believing that the rule threatens the basic good of any MoP (OPR, p. 303). The idea that each MoP has certain basic interests that serve as a check on others’ claims seems “self-authenticating” in exactly the sense that Rawls describes.
- Gaus’s common good requirement precludes the selection of social rules that threaten basic aspects of the good. This requirement seems to build into the concept of a moral rule – or the concept of right, in Rawls’s terms – consistency with a thin theory of the good. No rule that violates the thin theory can be acceptable in the Deliberative Model. But is this plausible? Some MoPs may have evaluative standards that are demanding in the sense of requiring support for certain social or economic conditions, which might be necessary for the continued existence of their evaluative standards over time. In a diverse society, it may not be possible to secure agreement on such support. These MoPs might then complain that the failure to select certain rules threatens to undermine their good. Rawls recognizes that some conceptions of the good – including valuable conceptions — may not endure in a well ordered society. Although it may be regrettable, there is no “social world without loss” (PL, p. 197 n.32). It seems implausible that Gaus would want to deny this, but, despite its purported modesty, does the common good requirement prevent him from doing so?