Steven Horwitz has an essay up at Cato protesting that, contrary to common perception, "modern" Austrian economics is actually very empirical. Much of his argument stems from his treatment ofeconomic history as being a subset of economics, in stark contrast to Mises's position that
There is economics and there is economic history. The two must never be confused.
For example, Horwitz points to several instances of Austrian-influenced scholars working in economic history (which of course is empirical) as evidence that "modern" Austrian economists resort to empiricism a great deal (as if there were ever a time when Austrian scholars did not use empirical evidence in their efforts to study economic history).
In his discussion of the history of thought, Horwitz does make one rather good point:
Despite the impression that one might get from reading some Austrians, Mises's term "praxeology" was not intended to be a "method" for economists.
I agree with this. Also, I brought this up with David Gordon at Mises University, and he agrees too. "Praxeology" itself should not be confused with the distinctive "praxeological method." Praxeology is a science with a distinctive method; it is not itself a method or methodology. Moreover, it is not anapproach to economics; it is a science that includes economics.
But unfortunately, right after accurately clarifying what Mises did not intend with the term "praxeology," he gets entirely wrong what he claims that Mises did intend:
Instead, that term, which has roots in the Greek for "action," described a field of study. That field comprised all the "sciences of human action."
This is simply incorrect. Mises wrote, "There are two main branches of the sciences of human action: praxeology and history." For Mises, praxeology is one of the sciences of human action, and history isanother science of human action. Praxeology does not include "all" of the sciences of human action.
Horwitz then writes,
So what today we would call economics, political science, anthropology, and sociology would all fall in this group.
This too is incorrect. Of those, only economics is classed by Mises as a branch of praxeology. Mises wrote,
History is the collection and systematic arrangement of all the data of experience concerning human action.… It is on the one hand general history and on the other hand the history of various narrower fields. There is the history of political and military action, of ideas and philosophy, of economic activities, of technology, of literature, art, and science, of religion, of mores and customs, and of many other realms of human life. There is ethnology and anthropology, as far as they are not a part of biology, and there is psychology as far as it is neither physiology nor epistemology nor philosophy. There is linguistics as far as it is neither logic nor the physiology of speech. 
And then in a footnote:
Economic history, descriptive economics, and economic statistics are, of course, history.The term sociology is used in two different meanings. Descriptive sociology deals with thosehistorical phenomena of human action which are not viewed in descriptive economics; it overlaps to some extent the field claimed by ethnology and anthropology. General sociology, on the other hand, approaches historical experience from a more nearly universal point of view than that of the other branches of history. History proper, for instance, deals with people or with a certain geographical area. Max Weber in his main treatise deals with the town in general, i.e., with the whole historical experience concerning towns without any limitation to historical periods, geographical areas, or individual peoples, nations, races, and civilizations.
Horwitz also writes,
Some Austrians argue as if one can deduce all of economics in one's armchair, but Mises was pretty clear that this core of economics was fairly limited.
This, again, is wrong. Mises wrote that "All the concepts and theorems of praxeology are implied in the category of human action." Not just a "limited core," but all of praxeology, which includes all of economics, is "implied in the category of human action." This would explain why, according to Percy Greaves, one of Mises's closest friends,
Mises used to say that all a good economist needed was some sound ideas, writing materials, an armchair, and a waste basket.
Horwitz tries to support his characterization of Mises's position:
He points out that even the notion that labor is unpleasant is not part of that core, but rather an auxiliary assumption we make based on observation. So too is the existence of things like money. When the economist goes to analyze the world, the core toolkit that comes only from reflection on action is a rather small set of basic propositions. Most of the interesting work in economics is institutionally contingent. For example, even if we recognize the importance of being able to engage in economic calculation, our ability to do so effectively depends upon the set of institutions in the economy under analysis. Moving from what Carl Menger called "exact laws" or pure theory, to applied theory means we must include the human beliefs and social institutions of the empirical world.
This would seem to imply that, for example, the theory of indirect exchange is not part of the "core toolkit" of pure theory but is rather applied theory. This, of course, is incorrect. Horwitz's error is to think that, if you introduce assumptions to your theorizing because of experience, that makes the result "empirical" or "applied theory." But that is not the case. For the truth of a theory, it makes no difference why it was constructed the way it was (that is to say, why it includes the assumptions it does). A self-contained theorem is true or false based on its logical structure, regardless of whether the assumptions are introduced because of experience, because of pure fancy, or for any other reason.
The theory of indirect exchange is pure, aprioristic theory, not applied theory. We choose to formulate it, because we observe that indirect exchange is part of the world around us. We take the effort to formulate it, because we want to later apply it to economic history. But the theory itself is true independent of experience. Whether it is applicable to any given experience is another question. It only becomes "applied theory" when it is applied to particular historical episodes.
But the end of science is to know reality. It is not mental gymnastics or a logical pastime. Therefore praxeology restricts its inquiries to the study of acting under those conditions and presuppositions which are given in reality. It studies acting under unrealized and unrealizable conditions only from two points of view. It deals with states of affairs which, although not real in the present and past world, could possibly become real at some future date. And it examines unreal and unrealizable conditions if such an inquiry is needed for a satisfactory grasp of what is going on under the conditions present in reality.
However, this reference to experience does not impair the aprioristic character of praxeology and economics. Experience merely directs our curiosity toward certain problems and diverts it from other problems. It tells us what we should explore, but it does not tell us how we could proceed in our search for knowledge.
Comment from Joseph Salerno
Not only does Danny's excellent article straighten out the profoundly confused essay by Steve Horwitz on Mises's approach to economic theory; it also highlights a strange phenomenon that is unique to modern Austrian economics as an intellectual movement. One would be hard-pressed to find another eminent economist besides Ludwig von Mises who has regularly had positions attributed to him that blatantly contradict positions that he has clearly, emphatically, and repeatedly expressed in his writings. (Karl Marx does not count because he wrote in an obscure and convoluted style and expressed contradictory positions on various issues when it suited him throughout his career.)
It is certainly true that Friedman's, Samuelson's, Keynes's, Ricardo's, etc., economic doctrines have been subject to alternative interpretations and debates over their precise meaning. But rarely does it occur that someone attributes to these economists views that are flagrantly at odds with what they have plainly written, as Danny has shown Horwitz has done with Mises. I do not mean to imply that Steve deliberately or consistently misrepresents Mises's views or that he is the only one to have grossly misrepresented them on occasion. In fact in the last several decades, views have been imputed to Mises — especially regarding the nature of entrepreneurship, the essence of the problem with socialist calculation, and the optimality of "monetary equilibrium" — that he has clearly rejected in his writings.
One can only speculate on the sociological reasons for this peculiar episode in intellectual history.
 Emphasis added throughout.
Originally published at Mises.org.