In "Degrees of Explanation", Hayek defends a system of analysis that approaches knowledge in a deductive manner. His proposed methodology is unique in that it is not Misesian, though it bares some similarities. Core knowledge need not be "apriori" in the sense that Mises employs the term. As Kant claimed in his Critique of Pure Reason, apriori claims must be established in empirical observation. Since they are core claims within an argument or, more generally, within a system, we must be especially confident about these claims.
Mises begins his analysis with a single fact. Humans act. Of course he does more than this. In contextualizing his analysis, Mises argues that,
Complex phenomena in the production of which various causal chains are interlaced cannot test and theory. Such phenomena, on the contrary, become intelligible only through an interpretation in terms of theories previously developed from other sources. In the case of natural phenomena the interpretation of an event must not be at variance with the theories satisfactorily verified by experiments. (31)This is true for the study of society as well.
Praxeology is a theoretical and systemic, not a historical, science. Its scope is human action as such, irrespective of all environmental, accidental, and individual circumstances of the concrete acts. Its cognition is purely formal and general without reference to the material content and the particular features of the actual case. It aims at knowledge valid for all instances in which the conditions exactly correspond to those implied in its assumptions and inferences. (32)So far, we see nothing unreasonable. Research in one field should be consistent with, or at least take account of research in other fields. Mises then goes on to make more extreme statements.
Its statements and propositions are not derived from experience. They are, like those of logical and mathematics, a priori. They are not subject to verification or falsification on the grounds of experience and facts. They are both logically and temporally antecedent to any comprehension of historical facts. (32)A consistent reading of Mises suggests that these arguments are necessary as an analytical starting point. The trouble is that he does not frame them as such. In attempting to defend the scientific nature of social analysis, which includes economic analysis, Mises makes an argument that is stronger than the facts themselves merit. The facts that we employ are, as far as agents with finite knowledge can know, more true or less true. The facts support analysis of social order must be well-supported by prior research. The task of a good theorist is to identify those facts that are implied by the facts that one takes as given. This should lead to new areas of research that can either support or invalidate the extrapolation. This isn't what Mises is saying, though I wish it was.
Not too far later (39), Mises makes the claim that is perhaps the most problematic for his treatise:
Action and reason are congeneric and homogeneous; they may even be called two different aspects of the same thing. That reason has the power to make clear through pure ratiocination the essential features of action is a consequence of the fact that action is an offshoot of reason. The theorems attained by correct praxeological reasoning are not only perfectly certain and incontestable, like the correct mathematical theorems. They refer, moreover, with the full rigidity of their apodictic certainty and incontestability to the reality of action [emphasis mine] as it appears in life and history. Praxeology conveys exact and precise knowledge of real things.Two problems arise here.
1) Mises does not explicitly direct interpretation away from social atomism. Perhaps all action is the result of reason, but we must ask, "Whose reason?" Must action and reason be coterminous? Can the reason of one man or many lead to action in someone not directly connected to them? What is the range of action, local or non-local, that results from reason exercised at some time t. The reason of humankind echoes throughout society in its institutions. These are established through a combination of reason and blind groping, through the process of trial and error. Without the use of reason, an agent might take action in the same way that wolves learn the social norms of a wolf pack. Perhaps this fits Mises's definition of reason as applied with a means/ends framework, but I doubt that Mises would apply it in this manner as he argues that "animals are unconditionally driven by the impulse to rpeserve their own lives and by the impulse of proliferation (19)." The relationship between reason and action is more complex than Mises admits here; I suspect that this is the result of Mises's ideological emphasis and the collectives zeitgeist that was in the air of academic conversation at the time.
2) Mises claims that "the theorems attained by correct praxeological reasoning" are "perfectly certain and incontestible." Mises is playing a logical trick here. His statement is true to the extent that conclusion arise from "correct praxological reasoning". But in what situation can I confirm that I have engaged in "correct" analysis? Every theory must always have a disclaimer attached to it that says, "we consider fact A implied by our theory true as long as facts B, C, D, etc... hold as true." We don't live in a world with perfect knowledge, but if we did, we would be apodictically certain. Just as this statement about apodictic certainty is a tautology - useful as it may be - so is Mises claim, but he does not explicitly identify it as a tautology. And his defense of economics that follows is a defense contingent on the truth of the facts upon which praxeology relies. A more honest approach admits that we cannot know that the claims of economists are true with apodictic certainty, even if they apply the system that Mises uses. The best that one can say is that he or she employs a system that makes full use of available, time-tested knowledge (even this claim may be too much due to the asymmetries of interpretation).
In "Degrees of Explanation", Hayek engages a softer claim.
It is, no doubt, desirable that in working out such deductive systems the conclusions should be tested against the facts at every step. We can never exclude the possibility that even the best accredit law may cease to hold under conditions for which it has not yet been tested. But while this possibility always exists, its likelihood in the case of well-confirmed hypothesis is so small that we often disregard it in practice. The conclusions which we can draw from a combination of well-established hypotheses will therefore by valuable though we may not be in a position to test them.Later in the same paper, he claims,
We shall here have to proceed in our deductions, not from the hypothetical or unknown to the known and observable, but - as used to be thought to be the normal procedure - from the familiar to the unknown.There is no need to claim that we have apodictic certainty of any kind. Our analysis is a certain as the facts that represent the core claims that support the analysis. In this sense, theory represents facts and arrangements of facts that we are most certain about, assuming that deductions were correctly carried out.
In this spirit, Hayek writes in "Economics and Knowledge",
My criticism of the recent tendencies to make economic theory more and more formal is not that they have gone too far but that they have not yet been carried far enough to complete the isolation of this branch of logic and to restore to its rightful place the investigation of causal processes [emphasis mine] , using formal economic theory as a tool in the same way as mathematics.Hayek points the way toward forming a more perfect praxeological system. If we are to take Popper seriously, we must accept that we will never have such a perfect system. Theory represents knowledge in most general and most perfect form that we can expect. It is from this light that Daniel Klein's revised iteration of Peter Boettke's claim that "The best reading of Mises is a Hayekian one and the best reading of Hayek is a Misesian one" holds (2012, 30):
I might concur with Peter Boettke that the most charitable reading of Mises is a Hayekian one. The most charitable reading of Hayek, however, is not a Misesian one (32).
