Anti-Constructivist Constructivism

I was reading Hayek's essay “The Errors of Constructivism” today (in New Studies), and I was struck by, well, by just how great Hayek is, and how his thinking is both well and ill suited for integration with contractarian normative theorizing, which is my favorite kind of normative theorizing. Please allow me to waive any implied guarantees of cogency, and to indulge in some musings about Hayekian anti-constructivism . . .
Hayek is clear, and I'm sure right, on the fact that the principles or rules of behavior that account for regularities in individual conduct and macro-level social order are to some degree inscrutable. We are guided by “knowledge how” and not “knowledge that”

What I want to show is that men are in their conduct never guided exclusively by their understanding of the causal connections between particular known means and certain desired ends, but are also by rules of conduct of which they are rarely aware, which they have not consciously invented, and that to discern the function and significance of this is a difficult and only partially achieved task of scientific effort.

Now, Hayek is talking here about the theorist, too. The contractarian theorist, for instance, can't just kick back in her chair and think about what rules her imaginary rational agents would choose to be governed by, for we, the theorists, are not especially aware of the principles of our own conduct, and can only partially understand how those principle contribute to social order, if they do, and so surely we cannot say what our fictional fully rational proxies would endorse.
So Hayek leaves us properly skeptical of the possibility of churning out principles of just conduct from a constructive procedure, i.e., an elaborate Rawlsian thought experiment.
Nevertheless, Hayek is incredibly strong on what I take to be a primary element in distinctively contractarian reasoning, probably exemplified best in Hobbes and Hume (and in Buchanan, Rawls, and Gauthier in the last century)–the idea that individual actions can combine according to a particular logic (the logic of interpendent expectation and choice–game theory) to create normatively interesting macro-level properties, and that we should evaluate common rules of conduct in terms of their role in supporting or undermining these properties.
Hayek:

The rules we are discussing are those that are not so much useful to the individuals who observe them, as those that (if they are generally observed) make all the members of the group more effective because they give them opportunities to act within a social order.

And later:

It was the great achievement of economic theory that, 200 years before cybernetics, it recognized the nature of self-regulatiing systems in which certain regularities (or, perhaps better, 'restraints') of conduct of the elements led to constant adaptation of the comprehensive order to particular facts, affecting in the first instance only the individual elements.

So Hayek endorses the contractarian way of thinking about the relation of rules to order, micro to macro. Rules are justified in terms of their contribution to order. And our reasons to adhere to these rules is not grounded in their immediate benefit to us, but in the benefit we derive from participation in the order they establish. He is, however, skeptical about our ability to know what that relation is in particular cases. And so he counsels that we not mess with them–much. Hayek's conservativism flows from his ideas about social evolution, which I take to be the weakest part of his social philosophy.
However, Hayek doesn't think he really helps the conservative too much in the end:

I must at once warn you, however, that the conservatives among you, who up to this point may be rejoicing, will no probably be disappointed. The proper conclusion from the considerations I have advanced is by no means that we may confidently accept all the old and traditional values. Nor even that there are any values or moral principles, which science may not occasionally question. The social scientist who endeavours to understand how society functions, and to discover what may be improved, must claim the right critically
to examine, and even to judge, every single value of our society. The consequence of what I have said is merely that we can never at one and the same time question all all its values

The last is a Neurathian/Quinean “can't get off the raft” point that I rather like. Hayek stresses it again later:

The only standard by which we can judge particular values of our society is the entire body of values of that same society. More precisely, the factually existing,imperfect order of actions produced by obedience to these values provides the touchstone for evaluation.

And then, immediately afterward, one can catch a whiff of Rawlsian reflective equilibrium (this was a lecture from 1970, by the way):

Because prevailing systems of morals or values do not always give unamibiguous answers to the questions which arise, but often prove to be internally contradictory, we are forced to develop and refine such moral systems continuously. We shall sometimes be constrained to sacrifice some moral value, but always only to other moral values we regard as superior. We cannot escape this choice, because it is part of an indispensible process.

However, given Hayek's skepticism about our ability to kick back and weigh and work through the relevant trade-offs in our rather limited heads, we need to substitute some external social process that can test and adjudicate conflicting moral principles. (Even Rawls, it is very much worth pointing out, says that reflective equilibrium is only achieved at the ideal limit. I wonder why he doesn't take this point seriously, as Hayek did.)
So here is a Hayekian argument for political federalism (little experiments in governance) and rights of exit, along with a well-functioning system of evolving common law. These are mechanisms that we might think have a tendency to produce principles in a kind of (not reflective, in the head, but social, enacted) equilibrium with one another.
I take some of the deliberative democrats to have been similarly looking for a procedure, a mechanism, for arriving at principles of just conduct–for a way to get the the method of reflective equilibrium out of the head and into the world. But democratic choice procedures are singularly well-suited for delivering bad answers. (Do deliberative democrats, Kerry voters all, I'm sure, really think the problem is just that there's not enough deliberation?) The Hayekian post-Rawlsian looks instead to federalism, exit, and the common law. There may not be many of us, but there should be.