Gaus' "The Order Of Public Reason"
Let me start by saying that The Order Of Public Reason, or something quite like it, is the book I would have hoped to write had I decided to go into philosophy. I don't know that it's going to rise to the level of A Theory of Justice (only time will tell there), but it really is an amazing work which I hope will become part of the canon. Gaus' ability to synthesize a huge number of topics into a single, coherent system really cannot be understated.
Given the effusive praise above it should be obvious that I'm generally in agreement with what Gaus' has to say. The main difference, I think, is that I'm a lot less optimistic about the outcome than he is when all is said and done. So, rather than try to summarize and comment on the entirety of the book I'm instead going to focus on those areas where we diverge or where I have other salient commentary.
I'd like to start with Gaus' discussion of instrumental morality. This section, by itself, makes the entire book worthwhile just by virtue of its (seemingly) definitive treatment of the subject. People who are more well-versed than I in these things may find reason to object, but it looks to me like Gaus makes a solid case that instrumental rationality does not lead to anything resembling a comprehensive morality.
The basic argument is that people have reasons to follow the rules when they can be reasonably assured that they will be detected and punished. Repeated rulebreakers will suffer future harm as well because their reputation will cause people to be less likely to interact with them. However, as societies become more diffuse these reputational effects diminish to the point where they're effectively non-existent. For example, he talks about the relatively tight knit group of (100k or so) Chaldeans in Detroit and how they find it difficult to track individual reputations directly, relying instead on proxies like family name (pp. 92-93). The United States as a whole has long since surpassed the point where such individual attribution is possible, which blows a big whole in instrumentalist theories.
As a tangential note I have to say that I ultimately find discussions of instrumentalism to be a little unsatisfying. It would be great if some variant thereof worked, since it would address all sorts of problems regarding secular morality, but at the same time it almost seems like its missing the point. The success of the various flavors of instrumentalism, to the extent that they work at all, is based on a happenstance alignment of our pre-existing intuitions regarding morality with the utilitarian calculation of "what's in it for me?". For example, Daniel Fincke's answer to the question of why murder is wrong is that it constrains the potential excellence of the murderer. This raises two issues:
- "Ought"-less morality: Instrumentalism's strength is that it doesn't rely on "oughts"/"ought nots", only an individual's rational calculation of eir own welfare. Is a system without any "oughts" "moral" in the typical sense of the word?
- Why is it important for murder to be "bad"? Instrumentalists go out of their way to show that various and sundry anti-social behaviors (murder, rape, theft, etc.) are off-limits. The fact that they care at all indicates that they acknowledge a pre-existing list of "bads" which must be addressed. Rather than trying to show that an instrumentalism theory covers all these bases wouldn't it be more efficient to cut to the chase and figure out where that list comes from?
Gaus addresses the latter observation in his discussion of the "moral emotions". For Gaus, the whole purpose of social morality is to allow individuals to co-exist as a society, which goes a very long way towards explaining why murder is bad: Societies with rampant murder very quickly cease to function. However, there's a really big disconnect between this observation and the system of public morality which he develops over the course of the book.
Gaus' linchpin observation is that individuals internalize social norms, leading to the moral emotions of guilt (experience by rule breakers) and indignation (experience by people who detect the rule breaking). This system, having evolved over a long time, is very good at (and perhaps even indispensable for) ensuring social stability:
Our current moral practice is made intelligible and sensible once we understand that human society depends on a social morality based on Rule-following Punishers. For such a system to have arisen, we must care about the moral actions of others, care about making demands on them, and hold that we have standing to make these demands. (p. 193)
Which is absolutely true. However, the evolved morality of Rule-following Punishers is also demonstrably irrational (in the sense of being logically incoherent) and regularly beats down people who deviate from the status quo. What if humanity has managed to evolve itself into a stable, but ultimately unjust, morality? To his credit Gaus recognizes this conundrum:
Still, the worry gnaws: have we simply landed in a confused practice that we cannot reason ourselves out of? That may be a recipe for despair rather than contented resignation, much less justification. We not only wish to know where we have landed but to have some reassurance that it is a destination worth arriving at, and not the result of being marooned with no hope of rescue. (p. 193)
Unfortunately, Gaus doesn't ever confront that question head-on. Rather, his answer to the problem seems to be that we merely need to internalize rules based on the principle that all persons are free and equal and then the moral emotions will do the rest:
Once a society of free and equal persons has coordinated on specific moral rules and their interpretation, the point of invoking moral authority is to police this equilibrium selection against "trembling hands" - individuals who make mistakes about what rule is in equilibrium - and those who otherwise fail to act on their best reasons. In these cases the overwhelming social opinion concurs in criticizing deviant behavior. An individual who violates the social equilibrium will not simply be able to check demands on her, for she will meet the same demand form almost all others. In Mill's terms, the deviant will not simply confront the opinion of other individuals but the judgment of "society." This, I shall argue, renders decentralized authority effective in inducing compliance with social morality. (pp. 47 - 48)
Color me skeptical. "Don't steal" is a very concrete prohibition; it's easy for the moral emotions to come into in such situations, for people to feel indignant or guilty as appropriate. "All persons are free and equal", by contrast, is highly abstract, which makes its interpretation and enforcement an intellectual exercise. It might be the case that the "specific moral rules" on which we have coordinated are sufficiently concise and clear-cut as to accurately trigger the moral emotions, but I'll need a lot more convincing in that regard.
This is the first place where I think the edifice that he has so painstakingly constructed starts to fall apart. Because he places an emphasis on the practical, lubricating aspects of social morality, it is necessary for him to endorse a system of moral reasoning which is readily accessible to the bulk of the populace:
As Baier says, social morality is for everybody. Such weak publicity is absolutely essential for the rules of social morality ot perform their job of coordinating a mutually beneficial social existence. Rules that are too complex to be reliably taught to children could not be followed by the population in general. A social morality composed of abstract principles that required sophisticated philosophical reasoning to be applied simply cannot form the basis of a shared public morality; neither could a social morality that required sophisticated case-by-case moral judgment. Perhaps such moral theories apply to some aspects of moral discourse, but insofar as the cognitive demands far outstrip the abilities of normal moral agents, they cannot provide the framework of our shared social existence. (p. 296)
Then he turns around and acknowledeges that there are "better" moral reasoners:
But better moral reasoners may be able to extend reasoning in ways that others cannot. For example, it is widely agreed today that a person with a normal concern for her own welfare and values could not be expected to care for a rule that, say, enslaved her, and so such a rule could never generate a justified internal ought for her: it could not be a binding moral rule... A better moral reasoner, though, may point out that the same considerations show that the slave cannot be bound to show that a similarly reasonable person cannot be bound by an oppressive property regime. Such people may serve us all as moral critics and reformers; they do not disqualify the rest of us from being competent moral agents. (p. 217)
There is an apparent tension between Gaus' desire to democratize moral reasoning and his recognition that there are better ways to reason, so it's worth a detour to consider what precisely Gaus means by "better". He endorses, with caveats, Kohlberg's six-stage model (pp. 214 - 217), and talks approvingly of "post-conventional" reasoners at several points. However, a criticism that can be levied against Kohlberg, Gaus, and moral philosophers in general (which I do not necessarily think is valid) is that post-conventional reasoning is the secular equivalent of asking how many angels can dance on the head of a pin, a fancy language game that has nothing to do with right or wrong. Why, then, should we equate "more sophisticated" with "better"? Gaus never addresses this objection directly, perhaps coming closest in the following passage:
Kohlberg hypothesized that these stages form a universally invariant sequence of increasing cognitive sophistication; higher stages were more adequate than lower stages, providing solutions to issues left unresolved by lower stages. (p. 216)
Whether he fully explains himself or not is beside the point; my main purpose for this digression is to establish that he's using the term "better" in the way that it is typically understood. Furthermore, this "betterness" is (mostly) objective rather than subjective, being characterized by a fuller, more competent assessment of the pertinent facts and their implications in any given situation. So my question to Gaus is, if post-conventional reasoning is objectively better than that displayed by the typical individual, shouldn't we grant post-conventional reasoning/reasoners more weight in those situations where they have the superior argument?
I honestly don't know what to make of this, because it seems to be a glaring and obvious contradiction. To recognize a superior mode of reasoning, but then to deny it any formal weight, seems manifestly irrational. Ultimately this may be a symptom of Gaus' aversion to charges of moral authoritarianism (pp. xv - xvi); to give formal weight to post-conventional reasoning would allow some people to dictate the terms of social morality to others. But if this weight derives from better application of the rules of logic, rules which Gaus is quite happy to invoke elsewhere, then it is hard to see how this complain can stand.
More likely, I think, is that Gaus has a vision of what he hopes a just world would look like, and is glossing over those bits that make his vision less plausible. He would like to see a "Great Society" (p. 268) of free-and-equal individuals who converge on principles of social morality which all can endorse. However, there are at least two other major conflicts, apart from the problem I just described above, inherent in Gaus' system:
- "All persons are free and equal" is not a widely-accepted principle.
- Obtaining universal asset to any rule at the level of the Great Society seems impossible.
With respect to the first point: A significant fraction of the American public, perhaps a plurality or even a majority, have not and never will internalize the concept that all persons are free and equal in the manner which Gaus requires. I'm not just talking about supremacists of various stripes, but rather the ordinary citizens who collectively form the foundation of society. They may pay the it lip service, but deep down in their hearts they still regard their own judgments as superior to that of others. These individuals are not "committed to the moral enterprise and treating others as free and equal" (p. 282) and thus "our moral relations with such people [are] transformed into the relations that obtain between us and those who are not capable of moral autonomy" (p. 283). Consistently applying the rules that Gaus has laid out would result in the widespread exclusion of individuals from moral society; this does not appear to be his intent, but it seems to be an unavoidable result.
As for the second point: I think that his fundamental approach to the issue is correct, I'm just less sanguine than he is about the result. Gaus requires universal assent to any moral rule:
The socially eligible set, then, consists in all those proposals that are unanimously ranked by all Members of the Public as strictly preferred to blameless liberty - this is, rules that all have reasons to endorse as authoritative. We thus must eliminate as a possible moral rule among "citizens of the realm of ends" any proposed moral rule that is in the ineligible set of any Member of the Public. (p. 322)
This, by itself, would not be a problem if there were any sort of restrictions on the constitution of the ineligible set. However, consider the example of a "strictly unacceptable" proposal which Gaus provides on p. 311:
Given the plural basis of the parties' deliberation, we cannot preclude that Alf, some other Member of the Public, will advance proposals that Betty finds objectionable in this way. To be sure, given that all are reasoning on the basis of evaluative standards that all accept as relevant to moral deliberation, and further that all Members of the Public hold that their proposals conform to the formal constraints on moral rules, there will not be out-and-out dictatorial, selfish, or manifestly oppressive proposals. Yet their differences in evaluative standards can still lead some to propose moral rules that others find strictly unacceptable. For example, when deliberating about rules concerning freedom of religion, Alf may sincerely propose (w) a rule of freedom of thought and exercise, though with an obligation to support (at least financially) a single national Church, on the grounds that this will promote community spirit and dampen sectarian conflicts. As an adherent of a particular faith, Betty may insist that such a rule is strictly unacceptable.
If proposals may be offered and rejected on religious (i.e. fundamentally untestable) grounds then there seem to be essentially no constraints on the contents of the ineligible set. Thus, as more Members of the Public are introduced it becomes more and more likely that any given proposal put forward will be found in some Member's ineligible set and thus must be removed from the socially eligible set. It seems a virtual certainty to me that the socially empty set will become empty well before we reach a Public representative of the population of the United States. Gaus recognizes this as a possibility, but doesn't have much in the way of a rejoinder:
We cannot exclude the possibility that an interpretation may be strictly unacceptable to some Members of the Public (Betty may certainly have sever doubts about Charlie's Freudian proposal). A proposal might, in the opinion of a Member of the Public, be such a bad interpretation of the abstract right of freedom of speech that it does hardly anything to secure the core interests and concerns of agency that are being interpreted, or else it severely and unnecessarily attacks her own evaluative standards. This certainly should not be a commonplace, since all are devoted to these core interests and all agree that it is of the first importance that the basic claims of agency be secured. To end up with a null set of eligible interpretations would be a moral disaster for the agency of all. And we must remember that the constraints on proposals still apply, including reversibility and the weak common good requirements (§15.2). Members of the Public are thus confronted by a set of socially optimal eligible interpretations of an abstract justified right, just as they were confronted by a socially optimal eligible set of possible rules, x, y, z. We have compelling reasons to think that the set would be neither null nor a singleton.
I call bullshit; Gaus has given nothing in the way of "compelling reasons" to think that, given a large enough group, the socially eligible set will be anything other than null.
This isn't a fatal blow to his theory as a whole, but it really does crimp his aspirations to a social morality which operates on a grand scale. Taking the US as an example, I expect that the application of his method would lead to rapid fragmentation into largely-autonomous sub-communities bound together at the "top" by little more than realpolitik. Consider the following, fundamentally-irreconcilable divisions:
- Archists vs. minarchists vs. anarchists
- Naturalists/materialists/atheists vs. theists
- People committed to the proposition that "All persons are free and equal" vs. people who think that's liberal bilge.
And so on. Which goes back to Nozick's observation that, given reasonable diversity of opinion, the will not be a single Utopia but rather utopias. The search for one social morality to rule them all may be fundamentally misguided.
Now, since this may be the last blog post I write for awhile (or, perhaps, ever), I want to end on a personal note. Gaus has neatly encapsulated a lot of the ideas that I've had that have caused me to take up the mantle of libertarianism. The fundamental intuition is that we cannot prove the truthfulness of moral rules and, thus, cannot foist them on other people without their consent. This standard, combined with the reality of evaluative diversity, means that there will not be one monolithic, "just" society. Rather, societies will form spontaneously when enough individuals with enough overlapping concerns get together. The boundaries of theses societies, and the sub-societies (and sub-sub-societies and so on) of which they are comprised, will be defined by this sphere of common concern and its associated commitment to a shared set of norms.
What I believe this means in practice is that any nation state which makes a pretense of being just (provided that such a state is even possible) will necessarily have a very lightweight set of rules that cover "the basics" (don't murder, don't steal, etc.) and not much else. Further rules are certainly possible, but the scope of their application must be narrowed in order to obtain the requisite level of acquiescence. Ultimately:
Do not do unto others what you would not have them do to you.