The Road to Fallibilism (Part I)
The Methodological Missteps of Friedrich von Hayek, and the Popperian Cure
By John Saudino
3 October 2019
To the Hayekians and Austrian Economists of all schools[1]
1. Introduction:
The fundamental nature of this topic is complex, writing an introduction thereto is daunting, and constructing a systematic explication thereof is herculean, yet do I cast myself “once more unto the breach, dear friends”.
What I intend to do here is twofold: The first and primary task will deal with epistemology/methodology and is essentially heavily grounded in the philosophy of science. This section, which will take up the Lion’s share of the article, will focus on the methodological differences between Hayek and Popper. This task has been split over two articles. A third and final article presents my conclusion, in which I will sketch what I believe to be some of the significant ramifications of these differences for the social sciences, economics in particular, and the real world.
Task one is a systematic and critical, though not antagonistic, analysis of Hayek’s epistemological positions and a reconciling or critique of these in consideration of Karl Popper’s conceptions regarding the growth of scientific knowledge and the problem of demarcation regarding the natural vs. the social sciences.
In completing these tasks I will be focusing on a selection of key texts.
In the epistemological[2]/methodological part I will examine the texts of both philosophers and will take the liberty to also examine the philosophical roots of their positions in an effort to point out the implicit, historically rooted, ontological and epistemological problems behind them and to evaluate to what extent the particular approach either solves, ignores or obscures these problems.
2. Monetary Theory and the Trade Cycle: Epistemological Problems
Hayek’s epistemological approach: a radical rejection of empiricism
Hayek describes his trade cycle theory in his famous book Monetary Theory and the Trade Cycle[3] as the “Additional Credit Theory”, but also examines significant methodological considerations in which he develops an approach to science that is quite different than Popper’s.
I will argue that Hayek’s approach here does indeed amount to a radical rejection of empiricism. This is not something terribly unique for an Austrian economist. Though this tendency is not often described in these stark terms as “a radical rejection of empiricism”, I will argue that this approach is not only so appropriately named but is also problematic in a number of ways.
2.1 Aprioristic/anti-Empiricist tendencies among Austrian Economists
This Hayekian tendency and the weaknesses inherent in it are even more prevalent in Hayek’s Austrian School colleagues. In Mises and Rothbard the approach is an unambiguous adoption of apriorism. In his 1933 work Epistemological Problems in Economics[4] Mises crowns economics as the prince of disciplines, as it were, by giving it the noble title of “the universally valid science of human action”. According to Mises its development by philosophers of the 18th century till now has presented mankind with the great realization of a profound truth of the most vital significance, namely that “…in the social realm too there is something operative which power and force are unable to alter and to which they must adjust themselves if they hope to achieve success, in precisely the same way as they must take into account the laws of nature.”[5]
Without elaborating too much here on the moral implications of this “discovery” of irrefutable laws in the historical context of the rising statist totalitarianisms of Mises’ times, I will just mention, with a nod of respect to this seminal philosopher of liberalism, that the desire to find a set of laws as powerful as those of nature as an irrevocable check on these evils is more than understandable, if not noble in its intent. However, this a priori, or at the very least, conventionalist[6] approach nonetheless has some serious defects regarding its rejection of empiricism as an essential element of the growth of scientific knowledge.
Mises goes on to elucidate his favoring of logical deduction over empirical observation when he describes this science thusly: “in the last analysis, logic and the universally valid science of human action are one and the same”;[7] and later on even more explicitly: “We do not maintain that the science of human action should be made aprioristic, but that it is so already”;[8] and to seal the deal regarding the unassailability of this science: “a proposition of an aprioristic theory can never be refuted by experience.”[9]
Another good illustration of this tendency toward radical apriorism among Austrian Economists of the Misian School is found in an important paper by Murray Rothbard:
If the axioms of praxeology are absolutely valid for human existence [which he, of course, insists they are], then so are the consequents which can logically be deduced from them. Hence, economics, in contrast to physics, can derive absolutely valid substantive truths about the real world by deductive logic.[10]
The last phrase alone seems to present such a radical departure from the principles of science, be it social or natural, that one cannot help but be reminded of the stultifying Platonic biases of medieval scholasticism. If there are “valid substantive truths about the real world” to be discovered by any field that dares to call itself scientific, they will be obtained by a combination of both logic and empirical fact or they will have discovered nothing at all but a rationalist projection into the real world of what is actually only the axiomatic internal system in that particular “scientist’s” mind. Such an approach that so consistently eschews the essential criterion of empiricism cannot help but eventually develop into an arbitrary construct. Although the godlike capacity to discover “substantive truths” i.e. genuine scientific knowledge about the real world by means of mere “deductive logic” may be appealing to philosophers of this tendency, the impossibility of doing so lays bare the Platonic impoverishment of their ontological and epistemic approach.
When one considers Mises’ pronouncements regarding the insurmountable constraints placed on organized human society by his conception of economics, the real world consequences of the inherent dogmatism of his approach become even more poignant. That is to say if one considers the claims made by this “universal science of human action”, it becomes clear that economics plays a key role in not only describing but prescribing what the world is or is to become by means of the sway it holds over political philosophy and economic policy and hence over our daily lives.[11]
This is because in contrast to sociology, anthropology, psychology, etc., economics, once known as political economy, represents a point of interaction with the real world through which social science and philosophy are not mere academic pursuits but rather have a genuinely profound impact on our daily lives. Whereby a social science like sociology describes the world, the science of economics also prescribes. It prescribes the economic policy in our society that creates the world in which we must live work and survive. Economic policy creates our world in that it provides the framework for the juridical superstructure and at the same time governs the substructure embodied in the material conditions of production upon which this superstructure is built. [12] In contrast to the deterministic and historicist approaches taken by many vulgar Marxists to this relationship between superstructure and substructure, that view the superstructure, the entire complex of laws and political ideologies guiding society, as being determined mechanistically by the material conditions of production themselves, it is quite clear to the more perceptive among us that theory in fact takes on a much more active role than this simplistic determinist view would suggest. When one takes a more dialectically materialist approach, i.e. one that takes into account the reciprocal causation acting between the two levels, it becomes obvious that the realm of theory, which is the product of autonomous human thought, has a much more decisive effect on social and historical processes than this simplistic materialist analysis would indicate. It is the fruits of such theories in the form of laws and policy decisions that feed back into the causal nexus to create the framework that governs the economy. In this codetermination process the theoretical social science of economics is both the effect and the cause of our social reality.[13]
Economists like Rothbard and others stemming from the Mises School of Austrian Economics are well known for their radical tendencies toward right-libertarianism or a kind of Anarcho-Capitalism. The problematic nature of this body of thought and its influence on the world, i.e. the effect of economic policy based upon this theory, is something that I cannot elucidate in the scope of this article, though I reckon it is so sufficiently understood by those familiar with it that I can legitimately table the discussion of it for now.
The good news about Hayek is that he does not go nearly as far down the path of outright apriorism at any time in his career as do Mises and Rothbard. Later on in fact, in part because of Popper’s influence, he seems at various times to take a wide variety of methodological positions, which becomes more apparent later in his Counter Revolution of Science. I will also mention as an aside that there are aspects of Hayek’s thought that contrast him to the more fundamentalist opponents of the Welfare State that put him in a different category than Mises and Rothbard. A good example would be his acceptance of an essential reconcilability between the existence of various forms of social insurance and the preservation of a free society.[14]
In spite of his more nuanced approach, Hayek still suffers from some of the rationalistic/anti-empiricist biases of his Austrian colleagues.
Hayek begins Monetary Theory and the Trade Cycle with a chapter called “The Problem of the Trade Cycle”. Far from being an exposition of his particular trade cycle theory, the chapter is actually a profound explication of methodological considerations that reveals a great deal about Hayek’s philosophy of science and his conception of scientific progress. His great treatise expounding his rejection of “scientism”– The Counter Revolution of Science—was not to come out until 1955, and the question of the qualitative distinction that, according to Hayek in that work, obtains between the natural vs. the social sciences I will deal with later on in this article. What is essential here in this first chapter of Monetary Theory and the Trade Cycle is his addressing of genuine questions of scientific progress as such. These questions are related to his conception of the role played in science by theory on the one hand and empiricism or “statistical analysis” on the other.
In a way that is similar yet distinct from Mises, Hayek stresses the primacy of theory over empirical data. The reasons he does this have to do, I will argue, with specific unresolved philosophical problems lingering in his system, problems that relate to what at the time were common yet undoubtedly faulty attempts to deal with the problem of induction. But before we get to that, let us look at his text in some detail to ascertain his points succinctly. Regarding the relationship between theory and practice Hayek writes that empirical input, i.e. statistical analyses: “… can, at best, afford merely a verification of existing theories; they cannot, in themselves, provide new insight into the causes or the necessity of the Trade Cycle.”[15]
The reason he believes this is true has to do with his belief in the broad distinction that must be made between theory and practice or theory and empirical fact. As his use of the concept of “necessity” here and elsewhere in the text indicates, he is still to a certain extent chained to some of the philosophical conundrums left by German Idealism, one of them being the analytic/synthetic dichotomy, or as it is here manifested in a false dichotomy between necessary and contingent truths. I will expand on this problem and its implications later on in the article.
His point in making a radical distinction between theory and practice or theory and statistical analysis here is essentially the idea that empirical observation might tell you what something is but only theory can tell you what it must be, hence theory is the only path to necessity or what Popperians might call certainty. This form of certainty is clearly the one that is formed by non-empirical rationalistic, i.e. apriori argumentation of the kind derived from axiomatic logic, mathematics or geometric proofs, a kind of necessity that is devoid of empirical content and hence in and of itself an inadequate basis for genuine scientific progress. For these reasons and others it will become clear later on that both of these concepts, necessity and certainty, are essentially invalid categories if they are applied to the empirical sciences.
Hayek elaborates on the primacy of theory by quoting another economist named Löwe as follows: “to expect an immediate furtherance of theory from an increase in empirical insight is to misunderstand the logical relationship between theory and empirical research”.[16]
Again the deductive nature of the theory places it on another plain of existence, in an almost classically Platonic way, above the mere world of “cave shadows” represented by statistics. He does not dismiss the importance of empirical input to the extent that Mises and Rothbard do, but he still grants the decisive and central role to deductive logic: “Empirically established relations between various economic phenomena continue to present a problem to theory until the necessity for their interconnections can be demonstrated independently of any statistical evidence.”[17]
Here we see the word necessity again. What he means by this is to say that although one considers empirically established relations, their contribution to the growth of knowledge remains naught until the necessity of the “interconnections”, which in observable phenomena represent mere correlations, can only be demonstrated by deductive logic.
Statistical examination is of very limited value because it does not form the basis of the two requirements necessary for the theory, which are the following:
- The theory must be “deduced from unexceptional logic from the fundamental notions of the theoretical system”;
- It must explain the phenomena observed in trade cycles “with all their peculiarities” and do so “by a purely deductive method.”[18]
There are two elements revealed in these two requirements laid out here by Hayek. The first is the dependence of new theories on the “fundamental notions of the theoretical system” which I will argue engenders a reactionary element by means of its inherent conceptual rigidity. The other is the reiterated notion of basing scientific discovery on “a purely deductive method” which as I have pointed out above and will elaborate later, entails what I have declared to be a radical rejection of empiricism.
2.2 The inherent conceptual rigidity of Hayek’s foundationalism
An example of what I call here conceptual rigidity comes out in Hayek when he explains that the earlier theory based on equilibrium as expressed by static analysis cannot be reconciled with the later, more complex, theory of the trade cycle because: “We cannot superimpose upon the system of fundamental propositions comprised in the theory of equilibrium, a Trade Cycle theory resting on unrelated logical foundations.”
By emphasizing the need for the later theory (T2) Trade Cycle Theory to conform to the “logical foundations” of the earlier theory (T1) Equilibrium Theory rather than beginning to question, test or falsify T1, Hayek is essentially eternalizing the logical assumptions of his earlier axiomatic system in a way that seems to unnecessarily eschew empirical input and retard scientific progress. This is because the new theory must be made distinct from the old one so that it will not be in conflict with its basic assumptions, for as he points out above, T2 must be “deduced from unexceptional logic from the fundamental notions of the theoretical system”. For Popper, as we will see, it is the very essence of the later theory (T2) that it refutes and supersedes T1 because it “has passed tests which T1 has failed to pass”.[19] It follows from this that T2 need not be based on any of the “logical foundations” of T1 to do so.
Hayek, on the other hand, in his insistence on basing later theories on earlier ones seems quite eager to preserve the “logical foundations” of older theories at all costs rather than falsify them as Popper does. This bias comes out elsewhere in his arguments for the fundamental distinction to be made between theory and practice. In arguing for this distinction he contrasts statistics on the one hand with economic theory on the other. He writes that “the perception employed” in statistics cannot be reconciled with that of economic theory because statistical analysis relies on “unrelated logical foundations”. Here he once again in his anti-empiricist bias places the value of logical consistency over testing and fallibility, in that the one field is claimed to exist in a qualitatively different realm even though the objects of observation are very real phenomena in this one very real world. His use of the word “perception” here to represent what can only be interpreted as an ontological concept in this context also represents a kind of psychologism and hence an additional error, in my view. How could it be that the varying perceptions inherent in two legitimate approaches to understanding reality (statistics vs. rational inference) render them so incompatible that one of them, the empirical one in this case, should be all but dropped from consideration? Once again it seems that the tail of theory is waging the dog of practice and the rigid requirement as we see here that newer theories not conflict with or overthrow older theories flies in the face of Popper’s falsification principle that he shows to be essential for science. Taken to an extreme this approach could serve to eternalize the logical foundations of old theories and lead to the conceptual rigidity I refer to. The danger is that the earlier theory may indeed be based on severe and, in consequence of this approach, unchallenged philosophical fallacies.
2.3 A purely Deductive method
The emphasis on theory and deduction reach their zenith in the course of the rest of chapter one in the form of statements like the following: “the observed phenomena will have to be deduced from the theoretical system, independently of empirical detection.”[20]
Here again the desire to break free of the rigors of empiricism is apparent as well as the somewhat bizarre notion that observation requires for its truth content a deductive process conducted “independently of empirical detection”.
He ends section III of the chapter with a plea to use theory consciously in a way that is distinct from that of “the ‘practical man’”:
The ‘practical man’ habitually acts on theories which he does not consciously realize; and in most cases this means that his theories are fallacious. Using a theory consciously, on the other hand, always results in some new attempt to clear up the interrelations which it assumes, and to bring it into harmony with its theoretical assumptions; it results in the pursuit of theory for its own sake.[21]
His reiteration of the primacy of a conscious use of theory as opposed to the intuitive observation on the part of the “practical man” and this final statement suggesting the proper practitioner of science endeavor in the “pursuit of theory for its own sake” both represent a profound anti-empiricist bias that is clear in Hayek and in other Austrian Economists.
This approach, I would argue, forms the basis of the many methodological and practical errors of both classical and Austrian economists. As I will point out later, some of the consequences of this conceptual rigidity account for the stubborn insistence of classicists and Austrians to cling to long outdated theories such as Say’s Law and the quantity theory of money.[22] Another consequence of rigidity or dogmatism is Hayek’s insistence that the trade cycle must be governed by purely endogenous factors, an a priori assumption that led, I will argue, to some major errors of interpretation involved in the ill-conceived attack on Keynesian economics of the 70’s and 80’s.
Hayek’s conceptual rigidity as detailed
above combined with his eschewing of the standards of empirical science lead to
an inevitable tendency on his part and on the part of other Austrian economists
to become unnecessarily doctrinaire in the their approach to economic problems.
There will be more on this in my conclusion.
[1] There is an obvious irony meant with this title in its imitation of Hayek’s The Road to Serfdom of 1944, which itself was a takeoff on Bertrand Russell’s The Proposed Roads to Freedom of 1918. The dedication below is also an echo of Hayek’s “to the Socialists of All Parties”. However, the “destination” as it were of “fallibilism” in my title is intended to be a positive outcome like Russell’s “Freedom” rather than a menacing one like Hayek’s “Serfdom”.
[2] Epistemology here is meant in the common understanding of the word as relating to the theory of knowledge and not in sense Popper meant it in the Conjectures and Refutations as a concept that must be expressly distinguished from methodology.
[3] Monetary Theory and the Trade Cycle, by Friedrich A. Hayek, Augustus M. Kelley Publishers, New York, 1966 (first published in 1929 in German and in 1933 in English)
[4] Epistemological Problems of Economics, by Ludwig von Mises, D. Van Nostrand Company inc., Princeton, NJ, 1960 (first published in German in 1933) p. 12
[5] Ibid p. 12
[6] See “Austrian Economics without extreme Apriorism: Construing the fundamental axiom of Praxeology as Analytic”, by Alexander Linsbichler, Springer Verlag, published online 12 March 2019
[7] Mises, 1960, p. 13
[8] Ibid, p. 16
[9] Ibid, p. 28
[10] “Toward a Reconstruction of Utility and Welfare Economics”, by Murray N. Rothbard first published in On Freedom and Free Enterprise: The economics of Free Enterprise, (Princeton, NJ Van Nostrad, 1956) p. 4
[11] See Saudino, “The Quantity Theory of Money and Classical Economics”, April, 2019
[12] Marx, Preface to the Contribution to the Critique of Political Economy, 1859
[13] Saudino, “The Quantity Theory of Money and Classical Economics“, 2019
[14] F. A. Hayek, The Road to Serfdom, Routledge, 1944, pp. 66-67
[15] Monetary Theory and the Trade Cycle, by Friedrich A. Hayek, Augustus M. Kelley Publishers, New York, 1966 (first published in 1929 in German and in 1933 in English)
[16] Ibid, p. 28
[17] Ibid, pp. 30-31
[18] Ibid, pp. 32-33
[19] Conjectures and Refutations, by Karl Popper, BASIC BOOKS, Publishers NEW YORK LONDON, 1962, p. 234
[20] Hayek, 1966, pp. 37-38
[21] Ibid, p. 40
[22] See Saudino, “The Quantity Theory of Money and Classical Economics”, 2019
