Let me begin by saying that, according to any sensible system of classification, John and I share the same political philosophy, and that I consider Free Market Fairness (FMF) the best explication and defense of that philosophy to date. Our agreement is vast, our disagreement small. John is very much a man after my own heart, and his fusion of Rawls and Hayek in FMF warms this Rawlsekian’s cockles. Rawls and Hayek truly are two great tastes that taste great together. However, John’s particular blend doesn’t go down quite as smoothly as I’d hoped. I’d like to try to say why. Let me beg forgiveness in advance for going on so long. I couldn’t quite find the time to write something shorter.
John repeatedly expresses his desire to defend FMF on specifically moral grounds. That is to say, John wants FMF to win hearts and minds not on the grounds that it works better than the alternatives, but that it specifies a structure of institutions that would bring out the best us in and realize our highest moral ideals. John distinguishes “three levels of political argumentation.” They are: political philosophy, political theory, and public policy, in descending order of abstraction. By political philosophy, John has in mind a “purely moral discourse about political questions” which “focuses narrowly on the task of identifying the morally appropriate standard of evaluation of a society’s basic institutions.” Later, John argues, following Rawls, that the task of identification requires that we invoke “a series of modeling assumptions” which “require that we abstract away from many of the most stubborn facts of political life.” This includes, among other things, assuming that “the principles of justice, whatever they might turn out to be, will win the full motivational compliance of all citizens.” That is to say, “we run the evaluation by considering what social world those principles would tend to produce if everyone was willing to follow the rules correctly.” It follows that when doing political philosophy, as John conceives of it, questions of feasibility fall entirely by the wayside. The question is how people ought to act, not how we think they will act. John follows David Estlund in seeking a path between “complacent realism” and “moral utopianism.” According to John, “the project of identifying normative political principles is the project of determining how things should turn out.” Fixating on how things actually will turn out — we’re looking at you public-choice economists — misses the point. It is to suffer “utopaphobia.”
Okay! Well, then. I guess we’d better start with an account of how people ideally ought to act. Where do we get one of those?
As Deidre McCloskey is right to point out, John doesn’t help us much here. He doesn’t first lay out detailed, substantive theory of the good and the right. Rather, he cobbles together a rather austere but nevertheless attractive moral ideal, “responsible self-authorship,” out of materials found in Rawls and other high liberals. But mostly in Rawls. Indeed, John takes on board a good deal of Rawlsian moral psychology. There’s the “two moral powers,” the “higher-order interest” in the development of moral personality, the need for the various “primary goods” among which the “social bases of self-respect” looms especially large. And so on. FMF is very Rawlsian. But also, I’m afraid, not Rawlsian enough.
As it happens, John doesn’t dwell overmuch on the really deep structure of Rawls’ theory of justice. It’s hard to blame him. It’s tedious, there’s a confusion of versions, and it’s not clear any of them are worth the trouble. Still, John’s relative neglect of the structure of Rawls’ theory leads him to take a bit of a Chinese menu approach to Rawlsian concepts. Thus, we’re left largely in the dark about the source of our normative ideals of right behavior, and the method of their justification because John omits Rawls’ moral anthropology, moral epistemology, the theoretical grounds for the various modeling assumptions, and the part of Rawlsian ideal theory that is concerned with feasibility.
I hate to do this, but I’m going to talk about Rawls in a simplified way perhaps insulting to the assembled panel of august scholars, but I swear it leads somewhere.
I find it useful to imagine the Rawls’ Original Position thought experiment as a sort of computer simulation with Rawls in the position of the programmer. The programmer’s first job is to whip up a group of simplified model people inside the computer. Rawls calls this the “model conception of the person” but I like to call the model person a “sim,” like the virtual folks in Sim City. So Rawls programs the sims to look like us in some ways – they have plans and project, hopes and dreams — and rather unlike us in others – the are purely instrumentally rational and have no sense of justice. Rawls then designs a virtual situation–the Original Position–with certain unusual, special features, such as the “veil of ignorance”– a device meant to model the alleged impartiality of our alleged flesh-and-blood moral sensibility working at its best. Next, the programmer throw the sims into the Original Position and offers them a choice between competing conception of justice. The task of the sims is to choose, “once and for all,” the principle of justice they prefer to govern the “basic structure” of the society they will live in once they emerge, unveiled, from the Original Position.
Now, as you might suspect, what comes out of the simulation is a function of what goes in. What’s the point of this elaborate exercise. Why should we trust its results?
So here’s how the we’re to validate the principles chosen by the sims. We (in the role of the programmer) check it against our real-person “considered moral judgments.” If the principles from the simulation clash with our actual moral judgments, that’s a sign we’re on the wrong track. But we can’t know in advance if the problem lies in our moral judgments or in the way we’ve set up the simulation. So we probe our moral judgments and revise the ones that seem, on reflection, more questionable than the principles generated by the simulation. And we go back into the program and tweak the parameters of the simulation to better reflect those judgments that seem to us most reasonable and secure upon consideration. We keep at this process of mutual adjustment until the simulation finally produces principles of justice that are in “reflective equilibrium” with our considered moral judgment.
According to Rawls, flesh and blood humans (but not the sims in the Original Position) have a “sense of justice” or “moral capacity,” which is the fundamental to the structure of his theory and his method of justification. (There’s some empirical evidence for a moral sense not that different from the one Rawls describes.) The sense of justice is supposed to have two functions. First, it produces moral judgments, such as “It is wrong to steal” or “It is obligatory to help this drowning child.” Second, the sense of justice motivates us to set aside self-interest, act according to our moral judgments, and to comply with fair terms of social cooperation. For now, don’t worry about moral motivation. Focus on the sense of justice as the source of moral judgments.
Rawls’ larger argument hinges on the idea that our sense of justice is sensitive to the institutional conditions under which we, and our moral capacities, have developed. We’re assuming from the start that our society is not already ideally just, otherwise we wouldn’t need a theory of justice. But the non-ideality of our current condition implies that our actually-existing sense of justice has not developed under ideal conditions, and is thus somewhat less than the best that it could be. This is one reason why it’s not good enough simply to sit around noodling on our moral intuitions until we arrive at principles of justice that just feel right. For one thing, different people are going to come up with all sorts of different answers, but one of our goals is to come up with a principled way of concluding some important disagreements about justice. Furthermore, our untutored, purely intuitive theorizing will tend to reflect assumptions, norms, and conventions that we have absorbed in the daily business of life in a less than ideally just culture. If our principles simply mirror existing injustice, they’ll be useless. We need to find a way to stand far enough outside our conditioned moral sensibility in order to see clearly how the status quo falls short of justice. Sounding familiar, yes?
So the Original Position thought experiment and the process of reflective equilibrium is one ingenious way to try to thread the needle between complacent conservatism and utopian irrelevance. We try to model ourselves as we might be if we were a little more rational, as if we had grown up under conditions a little more favorable to the development of our moral sensibility. There is, of course, no way to avoid reliance on our moral intuitions, as tainted as they may be by the non-ideality of the status quo. But we can try minimize our dependence on them, and add rigor and structure to our reasoning, by building a device of the imagination that idealizes our rational and moral capacities in ways that seem reasonable, and then asks what kind of society such beings aspire to live in.
But even all that jazz that doesn’t get us all the way to the sort of “realistic utopia” Rawls, or John, is looking for. To fully justify our normative principles we’ve got to test them for a sort of feasibility: stability. For Rawls, this involves showing, at least, perfect compliance with the demands of justice do not stymie the individual’s ability to achieve her personal aims. Principles of justice that systematically interfered with our projects and plans would not win the affirmation and willing compliance needed for a stable, “well-ordered society.” Rawls’ arguments for stability depends crucially on the idea, again, that the institutions under which we are raised affect the development and calibration our moral sensibility and the content of our conceptions of the good. “[E]veryone recognizes,” Rawls contends, “that the institutional form of society affects its members and determines in large part the kind of persons they want to be as well as the kind of person they are.”
That is to say, we are to a large extent creatures of our institutional environment. If we have come up within and operate daily in a partially unjust set of institutions–and we most certainly have–then chances are that who we are to some degree reflects the injustice of our social milieu. Rawls goes on to observe that “an economic regime, say, is not only an institutional scheme for satisfying desires and aspirations but a way of fashioning desires and aspirations in the future. More generally, the basic structure shapes the way the social system produces and reproduces over time a certain form of culture shared by persons with certain conceptions of their good.”
Great, but whoa! The upshot of this sort of thing would seem to be that our de facto moral judgments, even if hammered into reflective equilibrium with the help of clever thought experiments and modeling assumptions, are also going to be, to a very large extent, creatures of our environment. But what if we think our socio-political environment is, from a moral point of view, very bad indeed? Rawls’ impressive machinery is going to produce plausibly authoritative normative principles only if (a) we think the gap between our debased, non-ideal moral judgments in reflective equilibrium and the lofty moral ideals worth baking into our normative political principles is not too big , or (b) we have very modest expectations about identifying true or justified-once-and-for-all moral principles, and we’re happy to settle for marginal improvements toward the frontier of our meager moral imagination.
I’m okay with (a) because I’m happy with (b). But I think most Rawls fans aren’t happy with (b). It smacks of complacent realism. I suspect G.A. Cohen’s complaints to the effect that Rawlsian ideal theory isn’t all that ideal reflects something like the rejection of both (a) and (b). Now, I have no idea why Cohen thinks he can trust his gut, or his mind’s penetrating eye, or whatever it is he’s got. I haven’t got it. (I suspect it’s confidence.) It seems to me that if you take Rawls’ point about the endogeneity of moral personality to institutional structure really seriously, which I do, we’d conclude that identifying normative principles very far afield from the ones we already recognize–in our fallen, non-ideal condition–is beyond our powers.
Here’s where Hayek comes in with a vengeance. And here’s where I’d like John to take Hayek more seriously. John loves spontaneous orders, as one should. Economies are spontaneous-ish orders. Natural languages are spontaneous orders. And so are the bodies of moral convention that provide the raw material for the heady normative content of high-altitude ideal theory. Spontaneous orders are so unpredictable because there are so many of them and they’re so often mutually determining. Economies are influenced by language and morality. Morality is influenced by language and economic structure. Language is shaped by economy and morality. And so forth. Open up a spontaneous order and find another spontaneous orders spontaneously ordering it, spontaneously.
One reason Hayek and Rawls can blend is that both reject rational choice or homo economicus assumptions about human action. Rawls and Hayek both endorse the idea that rational action in the broadest sense is not wholly instrumental, but rather requires the heeding of rules, norms, and conventions that constrain the pursuit of our interests. Both suspect that the general practice of heeding these sorts of constraints turn out to advances our interests on the whole. Rawls and Hayek even agree that following rules is not rational because it pays, but that following rules is, to some extent, what rationality is.
However, Hayek is much more radical than Rawls on this score, which speaks to important differences in their fundamental philosophical inclinations—differences which are reflected in their views of moral psychology and social order. When it comes to the nature of practical rationality Rawls is very much a Kantian, and accordingly displays typically Kantian instincts about the universality of reason and moral personality as well as typically Kantian skepticism about the relevance of empirical theories of human nature and sociality.
In rather dramatic contrast, and in a testament to his good judgment, Hayek is an ardent Humean, and was given to repeating Hume’s distinctly un-Kantian dictum that “the rules of morality are not the rules of our reason.” As opposed to the sometimes primly rationalist Rawls, Hayek positively reveled in the messiness and indeterminacy of empirical theories of human nature and sociality. But he brought to the subject his economist’s rigor and anticipated a great deal of the best contemporary thinking about the natural basis of morality. Moral rules, Hayek argues, emerge from a process of cultural evolution.
In a nutshell, the Hayekian story is that rules that helped human groups meet the unique challenges of their particular physical and social environment allowed those communities to grow in number, widening the population guided by those rules. Conversely, groups acting according to less successful rules tended to dwindle relative population and become assimilated into more fecund and flourishing bands with more successfully adaptive rule. So their rules, their moralities, died out. We are cultural creatures, Hayek observes, biologically adapted to absorb culture through imitation. We follow moral rules basically for the same reason men in Texas talk about football: it’s just the done thing. But that doesn’t mean we’re naturally adept at identifying or explicitly articulating all the rules that guide us minute to minute. Of course there are explicit rules. And we know violations when we see them. But often we cannot say exactly what the relevant rules are. Even when we can, we generally don’t grasp how common adherence to a rule generates creates patterns of social activity that interact in complicated ways with other patterns created by common adherence to other rules. And if the moral rules work to coordinate behavior in a way that successfully serves the needs of our community, we can’t necessarily say how or why they were successful.
Hayek believes that the rule-bound texture of social life is much subtler than we are inclined to imagine, and that the basis of our social order is to some extent inscrutable. If he’s right, this doesn’t bode well for the prospects for moral of political theorizing that attempts to transcend the facts and constraints of the contingent current dispensation. That’s why Hayek tends to flip his lid when puffed-up radicals come around with big ideas for rapid advances in social justice. As imperfect as our society may be, we’ve got a pretty good thing going, but mostly we don’t actually understand what’s keeps it going, or what makes its going so good. So we need to be careful not to screw it up. When we think about making a difference, we need to think small, or else anticipate a high probability of failure.
But Hayek takes it even further. He argues that we won’t necessarily find reason in the wild – at least not in the grand capital ‘R’ sense of the Enlightenment philosophers. It is neither universal nor inbuilt. It has a history. Reason is a set of largely inarticulable culturally-selected rules and constraints that guide the operation of thought, much as morality guides action. This provides a truly striking gloss on Hume’s dictum. The rules of morality cannot be the rules of reason because both sets of rules emerged from a similar process of cultural selection, and they have developed side by side, not one after the other. If morality were a spinoff of reason, reason would have had to come first. But Hayek says reason is a relative newcomer, while adherence to some system of moral rules is ancient and primordial. Thus it is that Hayek says: “It is probably no more justified to claim that thinking man created his culture than that culture created his reason.”
How’s that for endogeneity? And where does that leave our attempts to rationally reconstruct our considered moral judgments into the regulative principles of realistically-utopian ideal regime types? Not so far from where we started.
In light of all this, I hope my misgivings about John’s sorta-Rawlsian, not-sufficiently-Hayekian method makes some kind of sense. I think I have a fairly solid feel, based in my no doubt idiosyncratic exegesis, of how Rawls’ argument hangs together, more or less. But I don’t yet have a fully clear sense of how John’s argument hangs together, because I’m not entirely clear about which Rawlsian notions he’s taking, leaving, and why. Why some but not all of Rawls moral psychology? Why responsible self-authorship but not, say, GLORY?
I completely understand why a useful political philosophy needs to aim for a sweet spot between complacent realism and pointless utopianism. Yet Hayek (and a certain strand of Rawls) makes it hard for me to see the point of bracketing most well-known regularities of social, economic, and political life when the normative content of our ideal political principles is inevitably drawn from regularities of conventional moral judgment produced by the very same non-ideal cultural and institutional milieu. Why think that under conditions of full information, perfect compliance, common knowledge of preference profiles, zero transactions costs, or whatever you like, any of us would still care about, say, “responsible self-authorship” in the way some of us do now? Why try to envision the socio-political realization of our homely ideals under fanciful conditions? If our institutional arrangements were more than marginally different from the status quo, chances are some of our favorite moral principles would be more than marginally different, too. Do we want to achieve our some of moral aspirations or to use them as symbolic motifs in the design of highly innovative institutional schemes that would, if realized, probably make them obsolete? I doubt we can even comprehend what it would mean if we all were to behave a lot better than we do. A little better, we can imagine. Isn’t that ideal enough to concoct a useful, non-complacent picture of moral aspiration?
That said, I certainly admire the rhetorical beauty of John’s dialectical jiu-jitsu. By exempting social-democratic high liberals from classical liberal feasibility objections, he takes away social-democratic feasibility objections to FMF. So what if actual capitalists poison the air and inflict constant indignities upon humiliated workers? Who cares? Ideal theory for all! My favorite thing about this move is that it may force some high-liberal ideal theorists to consider whether this sort of political philosophizing doesn’t have the virtue of theft over honest labor.
Puzzles of moral justification aside, if we think it would be simply terrific were folks to behave more cooperatively, more generously, more respectfully, more autonomously, and to enjoy a greater sense of personal dignity and responsible self-authorship (because that’s what we learned at the dinner table, or in the third grade, or at Brown), well then that’s great. So what do we do about it? Try this. Look to see under what conditions, as a matter of empirical fact, things tend to turn out the way we like, and then promote those conditions–at least when that would actually help. I suspect Rawlsekian market democracy would hold up just fine in those terms. Maybe the dreamy high liberals wouldn’t be so impressed, but that scene’s busted anyway.