Two Kinds of Apriori

In this essay (also here) philosopher Barry Smith writes about two kinds of apriori – impositionist and reflectionist.

That was new to me. I knew only one kind – impositionist or Kantian. Indeed, my search for the term reflectionist in the online Stanford Encyclopedia of Philosophy returned nothing. Anyway, Smith attributes the reflectionist apriori view to Aristotle and Carl Menger. He attributes it less so to Ludwig von Mises, who explicitly endorsed Kant’s apriorism.

“On the one hand are what we might call impositionist views, which hold that a priori knowledge is possible as a result of the fact that the content of such knowledge reflects merely certain forms or structures that have been imposed or inscribed upon the world by the knowing subject. Knowledge, on such views, is never directly of reality itself; rather, it reflects the `logical structures of the mind', and penetrates to reality only as formed, shaped or modelled by a mind or theory.

“On the other hand are reflectionist views, which hold that we can have a priori knowledge of what exists, independently of all impositions or inscriptions of the mind, as a result of the fact that certain structures in the world enjoy some degree of intelligibility in their own right. The knowing subject and the objects of knowledge are for the reflectionist in some sense and to some degree pre-tuned to each other. Direct a priori knowledge of reality itself is therefore possible, at least at some level of generality knowledge of the sort that is involved for example when we recognize the validity of a proof in logic or geometry (where it is difficult to defend the view that the character of validity would be somehow imposed upon the objects in question by the epistemic subject).”

The Kantian influence on Mises qua methodologist is very clear. On the other hand, Smith says: “When once we examine Mises' practice, however, then a quite different picture emerges, and we discover that Mises, too, was not at his best in his methodological self-interpretations. For we are forced to recognize that there is a veritable plenitude of non-logical primitive concepts at the root of praxeology.”

“Consider, however, the concepts causation, relative satisfactoriness, reason, uneasiness, valuation, anticipation, means, ends, utilization, time, scarcity, opportunity, choice, uncertainty, expectation, etc., etc. The idea that one could simultaneously and without circularity reduce every one of the concepts in this family to the single concept of action, that they could all be defined by purely logical means in terms of this one single concept, is decisively to be rejected.

How much better would it be to accept that we are dealing here with a family of a priori categories and categorical structures which would be, in the jargon, not analytic but synthetic.”