The Mengers versus Mises on matters methodological

Abstract

The paper argues for three points. The first purpose of the paper is to show that Carl Menger would have rejected Ludwig von Mises’ methodological apriorism. Second, I argue that Carl Menger was a pluralist about the methods of theoretical economics and that Mises was rather less of a pluralist, if not altogether a monist, about the legitimate method(s) of economic theorising. Finally, I try to establish the broad consistency of Menger’s pluralism with the tolerant methodological attitude of his son, the mathematician, logician, and philosopher of science, Karl Menger.

Keywords:

• Carl Menger
• Ludwig von Mises
• Karl Menger
• apriorism
• Austrian economics
• economic methodology

JEL codes:

• B13
• B25
• B31
• B41
• B53

Acknowledgements

An early version of the paper was presented at the 15^th Biennial Conference of the International Network for Economic Method (INEM) at Arizona State University, November 12–14, 2021, and (via Zoom) at the 91^st Annual Conference of the Southern Economics Association, Houston, Texas, November 20–21, 2021, and the conference on “Carl Menger: One Century Later. Originalities and Modernities” at Nice, France, November 24–26, 2021. I wish to thank the organizers and attendees of these sessions. I would also like to thank two anonymous referees whose comments and suggestions served to improve the paper. Finally, I wish to thank the editors of this symposium and of European Journal of the History of Economic Thought. Any remaining errors or omissions are my own.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 Murray Rothbard (1976; italics added) describes Mises’ methodology as “The Methodology of Austrian Economics.” Robert Nozick (1977, 361) explicitly associates apriorism with “[o]ne branch of Austrian theorists (Mises, Rothbard),” but the title of his paper implicitly identifies it with “Austrian Methodology.” Mark Blaug (1980, 91–92) associates Misesian apriorism with “Modern Austrians…a small group of latter-day Austrian economists…numbering among its adherents such names as Murray Rothbard, Israel Kirzner, and Ludwig Lachmann[.]” Terence Hutchison (1981, 214) also seems to associate Misesian apriorism with “modern Austrians,” with the exception of “Hayek II” and possibly Kirzner (against Hutchison’s posit of a Misesian “Hayek I,” see Scheall [2015], and against Hutchison’s posit of a Popperian “Hayek II,” see Caldwell [1992]). Bruce Caldwell (1982) identifies Misesian apriorism with “Austrian methodology.” David Gordon ([1994] 2020) argues that “The Philosophical Origins of Austrian Economics” culminate in an Aristotelian version of Misesian apriorism. Hans-Herman Hoppe ([1995] 2007; italics added) identifies Misesian apriorism with “The Austrian Method.” Peter Leeson and Peter Boettke (2006, 247–248; italics in the original) argue that “to the Austrian economists who trained with Mises during his New York University period (1944–1969), like Murray Rothbard, adherence to methodological apriorism is the distinguishing characteristic of the Austrian school, and alternative methodological positions are interpreted as undermining Mises’ strong claim about the nature of economic reasoning.” More recently, Gabriel Zanotti and Nicolás Cachanosky (2015) identify “Austrian epistemology” with Fritz Machlup’s (1955) failed attempt (see Scheall [2017b]) to align Mises’ epistemology with a moderate kind of proto-Lakatosianism. The subtitle of Alexander Linsbichler’s (2017) Was Ludwig von Mises a Conventionalist? A New Analysis of the Epistemology of the Austrian School of Economics implicitly identifies Mises’ epistemology (whatever it was) with that of the entire Austrian School.

2 The comparative fame and respect that Hayek, relative to Mises, is broadly accorded outside the Austrian School is the reverse of their respective standings within the School, especially in the praxeological variant associated with Murray Rothbard and several of Mises’ other American acolytes, who treat Mises as something of a demi-god and Hayek as decidedly mortal.

3 As a preliminary point in support of this thesis, it is worth noting that Mises (2013, 141), in his Notes and Recollections, argued that Carl Menger was “too much under the sway of John Stuart Mill’s empiricism to carry his own point of view to its full logical consequences.” Without taking a stance on whether and to what extent Menger was influenced specifically by Mill’s empiricism (as opposed to, say, a more Scottish, Humean-Smithian, empiricism), I am happy to note Mises’ likely agreement with my first thesis. Many thanks to an anonymous reviewer for encouraging me to mention this point.

4 Criticisms of Mises’ apriorism as extreme and defenses of it as moderate are surveyed in Scheall (2017b). Linsbichler (2021) and Lipski (2021) are two more recent contributions to the literature.

5 I lay the blame for this confusion primarily at the feet of Mises’ Austrian defenders, who have tended to ignore the substance of his critics’ objections (Scheall 2017b). Critics of Mises’ apriorism have offered relatively clear objections that have not been adequately addressed by his defenders, who have tended to declare victory for their interpretation of Mises as a moderate apriorist on grounds irrelevant to these objections.

6 I have argued that the Greater Certainty thesis is a non-sequitur (Scheall 2017b). It does not follow from the fact (if it is a fact) that knowledge of the action axiom is “within” us, as Mises ([1933] 2003, 137) put it, that this knowledge is any more certain than knowledge acquired from without. It may be easier to introspect, but this does not per se make knowledge acquired through introspection more epistemically secure than observation of the external world.

7 One of the difficulties involved in trying to understand his apriorism lies in the fact that Mises ([1949] 1998, 68) explicitly defended fallibilism – the thesis that all beliefs are open to revision, that no beliefs are infallible – about every belief, apparently, other than belief in the action axiom. This has led some commentators (e.g., Leeson and Boettke [2006, 258]) to claim that Mises was a fallibilist full stop and that his fallibilism extended even to the action axiom. But, given Mises’ explicit assertion of apodictic certainty about the action axiom, this is not a tenable position. Mises’ fallibilism did not extend to belief in the action axiom. Mises’ fallibilism is thus rather unusual in the history of ideas, given that most fallibilists are fallibilists about all beliefs. It is difficult to tell a coherent story that unifies all of Mises’ epistemological and methodological musings.

8 It is interesting to note, given the argument that follows here, that Carl Menger may have been more extreme than Mises along the extent dimension. Many thanks to Alexander Linsbichler for suggesting this point to me. However, given that the extent dimension does not figure in the many criticisms of Mises’ apriorism, that Menger may have been more extreme along the extent dimension is immaterial to the question of their methodological affinities. Menger and Mises may have both been extreme apriorists, but they were extreme in different respects, and it is only the epistemological respect in which Mises was an extreme apriorist that is significant for the relevant literature.

9 The qualifier “methodological” is meant to distinguish versions of apriorism as manifested in scientific research from more general epistemological apriorisms. A particular version of methodological apriorism provides reasons (or a justification) for exempting particular propositions from empirical testing. An epistemological apriorism provides reasons why we can possess relevant knowledge without (i.e., “prior to”) some kind of experience, typically, without experience of the external world. Part of what makes Mises’ methodological position unique is that he connects apriorism about the action axiom to an epistemology according to which we can allegedly possess knowledge of the action axiom sans experience. That is, Mises argues that we can exempt the action axiom from testing because (somehow) we already know it to be true. However, to be a methodological apriorist, it is by no means necessary to be an epistemological apriorist. I discuss several versions of methodological-but-not-epistemological apriorism in “What is Extreme about Mises’ Extreme Apriorism?” (Scheall 2017b). Many thanks to an anonymous reviewer for encouraging me to clarify this distinction.

10 I do not intend to engage in a comprehensive analysis of Carl Menger’s methodological thought or practice. My concerns are narrower in scope. There are a number of unresolved issues in Mengerian historiography that I will not touch upon (and some, frankly, that I would not touch with a ten-foot pole). For example, I will not venture an opinion concerning the fraught question of the extent to which Menger was an Aristotelian. Neither will I evaluate either the significance of the Methodenstreit for the history of economic thought or the correctness of Menger’s methodological arguments. I will also have little to say about the methodological arguments of Menger’s adversaries in the Methodenstreit, the German Historical School. Such issues are far from my purposes in the present essay.

11 Menger acknowledged that there were more than two orientations of theoretical research in economics. The exact and the realistic-empirical orientations were merely the two “most important for our science” (C. Menger [1883] 1985, 55). Indeed, the realistic-empirical orientation itself breaks down into (at least) two sub-orientations: the historical and the statistical.

12 Given the standard Austrian concern for the realisticness of assumptions, it might also be noted that, on this interpretation, the assumption is realistic – persons are self-interested – and, to the extent persons are not in error or motivated by considerations other than self-interest, also true.

13 Of course, it is the fact that introspection is not per se more reliable than sensory observation that makes the Greater Certainty thesis a non-sequitur.

14 I have argued that Mises occasionally failed to appreciate the significance of various developments in philosophy and science for his methodology (Scheall 2017a, 2017b). Perhaps this was just another instance of Mises insisting on the truth of a proposition that was already passé at the time he put it forward.

15 Related to this point, see C. Menger ([1883] 1985, 215, fn. 145):

It is a peculiarity of the exact social sciences that exact research in the realm of the phenomena of human activity starts with the assumption of a definite volitional orientation of the active subjects. This does not, however, establish an essential distinction between exact research in nature and exact social research, for the former starts with presuppositions [e.g., concerning the effects of gravity and other forces] which exhibit a formal analogy to the one under discussion here.

16 It might be more precise to say that each orientation implies an array of methods more or less appropriate to the orientation.

17 An anonymous reviewer helpfully noted that the action axiom is part of Mises’ praxeological theory and this means that the extreme interpretation also makes Mises a monist about economic theory. A theory inconsistent with the (true) action axiom is necessarily false.

18 Compare this with some sentences taken from the last paragraph of the ninth and final appendix of Carl Menger’s ([1883] 1985, 237) Untersuchungen: “As if the worth of a science were dependent on its object, and not rather on the importance, depth, and originality of the results of their investigations!”

19 Karl Menger’s mother, Hermine (“Mina”) Andermann, was born Jewish, but converted to Catholicism in 1893 (Scheall and Schumacher 2018). After Karl emigrated to the U.S. in March 1938, administrators at the University of Vienna prompted an investigation by the Sippenamt (“kinship bureau”) of his racial background, an investigation that ended with Menger declared a Mischling or “half-breed” (Sigmund 2017, 356).

20 Classifications between sciences and disciplines are “historically evolved” (C. Menger [1884] 2020, 477).

21 Also see C. Menger ([1884] 2020, 491): the sciences differ “by the tasks they have to solve,” tasks that were determined by the unique goals of the respective sciences.

22 Mayer would later gain eternal infamy in Austrian economic circles for his quick expulsion of Jewish members from the Nationalökonomische Gesellschaft in the same week as the Nazi Anschluss.

23 It should be noted that, whatever he might have learned, Mises never modified his methodology in light of Menger’s (or, to my knowledge, anyone else’s) criticisms.

24 Compare with the passage quoted above from Carl Menger’s abandoned Critique of Metaphysics and of the so-called Pure Reason from the Empiristic Point of View.

25 Karl Menger cites Böhm-Bawerk (1924) here, though the prefatory discussion to the paper as published in Menger’s Selected Papers, quoted in the text above, indicates that Mises’ contention that some laws of economics could be logically proved without reference to experience was the original inspiration for the paper.