Commentary on Political Economy

Wednesday 13 April 2016

Praxeology as Language-Game: Mises and Hayek

Nearly all bourgeois economic theory - what is taught around the world - is essentially a language game. The greatness of Mises is to have intuited this: his misery is to have believed that it is a legitimate "theory". What is a language game? Take the bourgeois "theory" of economic life. You have individuals who own "goods". These goods are, by definition, available for "exchange" with other "individuals" who "own" other goods. The "exchange" occurs free of coercion among these "individuals" - this is "the market". "Market prices" are the ratios of "exchange" between these "individuals". Prices are set by "demand and supply" because the "utilities" of "individuals" are "given". Nothing is "produced" or created in this economy: it is a zero-sum game in which there is no "profit" except through inter-temporal exchanges leading to final equilibrium prices.

As you can see, all these categories - individuals, ownership, goods, exchange, market, prices, equilibrium, supply and demand - all these categories are "defined" in such a way that this economic theory constitutes a "game": and the rules of the game - its "language" - make it so that every element or rule of the game is defined consistently with every other element and rule. There is no "escape" from this "prison". A language game is a "false prison" when it is applied to a human reality - because it limits the scope pf human action to the artificial axiomatic rules set by the language game.

Such is bourgeois "economics"...and the world is starting to rebel against this false prison. Watch out!

The ego is not

sharply marked off, its limits are very indefinite and

arbitrarily displaceable. Only by failing to observe this

fact, and by unconsciously narrowing those limits, while

at the same time we enlarge them, arise, in the conflict

of points of view, the metaphysical difficulties met with

in this connexion.

As soon as we have perceived that the supposed unities

"body" and "ego " are only makeshifts, designed for

provisional orientation and for definite practical ends (so

that we may take hold of bodies, protect ourselves against

pain, and so forth), we find ourselves obliged, in many


more advanced scientific investigations, to abandon them as

insufficient and inappropriate. The antithesis between ego

and world, between sensation (appearance) and thing, then

vanishes, and we have simply to deal with the connexion of

the elements [sensations]…

Kant’s critical idealism had sought to preserve the separation of Subject and Object: by treating the Object as an unknowable and impenetrable “reality” – the “thing-in-itself” – Kant had preserved the “autonomy” of the Subject, that is, the ability of the Subject to initiate action in the lifeworld (in keeping with St. Augustine’s dictum that “man was created so that there might be a beginning”, ut initium esset). At the same time, however, this very impenetrability of the Object gave the followers of Kantian philosophy, the neo-Kantians from Hermann Cohen onwards, the theoretical impulse to exclude the Object from the realm of human action altogether – to push “physical reality” or “the thing-in-itself” to the realm of “meta-physics”. In this specific context, Schopenhauer and Mach march abreast with neo-Kantism. Not until Lukacs and Heidegger challenged this extrusion of metaphysics from the sphere of human enquiry and action from very different political and philosophical perspectives did the errors of Machian and neo-Kantian positivism attract the attention of Western thinkers. But by then this insipid formalism had spread through Austrian logical positivism and even the early Wittgenstein to all spheres of thought, including economic theory, as exemplified by the Neoclassical Revolution from Menger to Mises and the rest of the Austrian School. In the words of Mach quoted above, once we treat all human experience as confined to the sphere of “passive” sensations, rather than “active” decision, “the antithesis between ego and world [between appearance and reality] …vanishes”. Again, this is pure Schopenhauer. But it is also pure neo-Kantism – because once the ability of human beings to act in and upon the world is denied as being beyond the reach of “science” – when in fact such a denial is as “metaphysical” as anything could ever be! -, then all that is left for science to do is simply “to classify” the world, to pigeon-hole life into its various “categories”.

For science, nothing can ever be created – everything is transformed. Because Machism obliterates the difference between acting Subject and passive Object in the uniformity of “sensations” or phenomena, even the difference between dream and reality evaporates:

The facts are not to blame for that. In these cases, to

speak of "appearance" may have a practical meaning,

but cannot have a scientific meaning. Similarly, the

question which is often asked, whether the world is real

or whether we merely dream it, is devoid of all scientific

meaning. Even the wildest dream is a fact as much as

any other. If our dreams were more regular, more connected,

more stable, they would also have more practical

importance for us. In our waking hours the relations of

the elements to one another are immensely amplified in

comparison with what they were in our dreams. We

recognize the dream for what it is. When the process

is reversed, the field of psychic vision is narrowed; the

contrast is almost entirely lacking. Where there is no

contrast, the distinction between dream and waking,

between appearance and reality, is quite otiose and

worthless. (Analysis, p.11.)

For Mach as for Schopenhauer and for Neo-Kantians philosophers and Neo-Classical economists later, there is no scientific difference between reality and dream, between reality and appearance, between existing reality and human finality, human action. All that remains are “sensations”, then. And sensations are “facts” by definition. And because sensations are “passive”, then clearly all “facts” are passive entities also. Human beings perceive the world “as they find it”: reality, even social reality, cannot be changed, it can only be trans-formed through new combinations of “elements” already in existence! This conclusion which forms the essence of positivism will be the starting point of Menger’s and Bohm-Bawerk’s Austrian economic theory.

Contrary to its etymological meaning (from the Latin facere, to do, to act), for positivist science “facts” are not what human beings create or initiate but what they are observed doing – as if human agents were robotic mechanisms whose behavior can be ascribed to scientifically-determined causes.

If the Ego can be reduced to the sum of human sensations; if human behavior can be explained by the regularities of these sensations, then clearly the sphere of subjectivity, of creative human action, vanishes. And so does the Ego understood not as an individual and not even as “the Subject”, but even as the ineluctable reality of the primacy of human action. But the disappearance of the Ego and the Subject does not entail the disappearance of the “in-dividual” for Machian science and for neoclassical economics. Individuals continue to count and to be counted by positivist “science”, but their behavior is not taken to be “political action”; instead, it is “measured” nomologically, in pure positivist statistical terms that can yield “predictive regularities” for the purposes of social policy. For the positivist, reality is not what human beings make it but it is purely what individuals actually do; reality is what individuals are observed doing – it is “data”, that is, “givens”. And for the positivist, science is the empirical ability to predict what human beings will do. (Cf. on this, of course, Milton Friedman’s “The Methodology of the Positive Sciences”.) But why do “individuals” act as they do? Why do they engage in one kind of behavior rather than another? How did they evolve to their present state? And how can we interpret “scientifically” what they supposedly “do”? If, indeed, the nomological value of positivist social science is to be able to predict social behavior, then certainly economic science has achieved very little precisely because its predictive power is virtually fallimentary.

Yet, as Robbins pointed out, the aim of economic science is not merely classificatory: it is not purely descriptive or taxonomic; nor is it explicatory. Economic science – like all formalist positivism – does not tell us “why” reality is what it is – because the sphere of ultimate explanation belongs to “meta-physics”. Instead, the aim of economics is to establish a set of rules that can have predictive power over existing “facts”; the aim of economics is purely “functional” – it is quintessentially “caeteris paribus”, what happens to one variable when other variables change assuming that everything else remains the same. Clearly, therefore, this functional formalism must inevitably involve an unbridgeable gap – an apory – between empirical reality as it defines it for its own practical and strategic political purposes, on the one hand, and, on the other hand, the logico-mathematical forms that positivist science develops to describe “scientifically” the functional relations that it presumes to find between “facts” seen as “data”, as “givens”.

And it is this unbridgeable irrational gap (Fichte’s projectio per hiatus irrationale), what Boettke and Leeson (‘Was Mises Right?’) call “this dichotomy” in the passage below, between empirical content and theoretical form that is the bane of all bourgeois science, economics in particular. For Mises, who was nothing if not a thorough theoretician, it was clear that the theoretical form alone capable of maintaining a rigorous logical relation with empirical content in economics had to be a tautology – because only a tautology can deal with economic institutions, which are intrinsically historical, and yet claim to have the status of a “science”, which can apply only to inert objects:

Mises eschewed the traditional analytic/synthetic dichotomy, successfully both revealing the illegitimacy of the positivist approach and defending the empirical relevance of ‘mere tautologies’ in economic science. (p.249)

But this is precisely where Boettke and Leeson are totally wrong: Mises could not perform this ambitious operation “successfully” for the simple reason that “mere tautologies” can only yield a language game: they can never ever have any “empirical relevance” at all, as both Kant and Wittgenstein knew only too well. Kant himself was quite aware of Wittgenstein’s central insight in the Tractatus. In decrying Fichte’s Wissenschaftslehre (science of knowledge) as an impossible oxymoron, Kant commented:

For the pure science of knowledge is nothing more nor less than mere logic, and the principles of logic cannot lead to any material knowledge. Since logic, that is to say, pure logic, abstracts from the content of knowledge the attempt to cull a real object out of logic is a vain effort and therefore a thing that no-one has ever done,(in G. MacDonald Ross, Kant and his influence, p.159).

The problem however is that by classing logico-mathematical statements as synthetic a priori Kant equated them epistemologically with physical equations describing causal chains of events – which do formulate “material knowledge” – that is, they encapsulate it and facilitate it – although they too do not “lead” to it. The crucial and essential difference between physical equations, on one side, and logico-mathematics on the other side is that the former are not tautological because they merely point to an experimental relation or link (loosely called “causal”) between two different sets of phenomena, whereas logico-mathematics purely defines the same concept “in other words”, and therefore is tautological. The “objects” of physical science are inert bodies; the “objects” of logico-mathematics are either abstract concepts or else they are categories that are applied to “agents”. But any category that is applied to “agents” is ipso facto invalid because it is impossible to categorise “scientifically”, that is “objectively”, agents who are, by definition, not objects!

The illustrious authors, keen “to illuminate the real world” with the embarrassment of Misesian drivel, fail to understand that “praxeology” can have applicability to “reality” only as a “language-game” – that is to say, only as a strait-jacket, only as a Procrustean bed into which “reality” must fit if it is to qualify as reality; and not – as they sheepishly overextend themselves – as “analytical narrative”, a “somewhere between” formalism and empiricism. They are wrong because they fail to see that mathematical identities applied to physical events have a categorically different status from when they are applied to social institutions (including economic ones) that are conventional in the sense that they are not immutable but are political and therefore subject to historical change. Only if we reify social political and economic institutions – only if we turn them into “objects” of study - can we hope to apply logical identities to them so that they come to form part of a language-game. But in that case we are performing an impermissible task: we are assuming that these institutions are “immutable natural entities” when in fact they are pro-ducts or “facts” prone to social transformation to which logical identities cannot validly be applied because to do so deprives them of their very content! We would be creating, to use a Wittgensteinian metaphor, “a false prison”! [Cf. discussion in ‘Eq&SocRlty’, quote from Arendt. Myrdal makes similar point.]

‘‘Praxeological reality is not the physical universe,’’ Mises argued,

‘‘but man’s conscious reaction to the given state of this universe. Economics

is not about things and tangible material objects; it is about men, their

meanings and actions. Goods, commodities, and wealth and all the other

notions of conduct are not elements of nature; they are elements of human

meaning and conduct’’ (1949: 92).

The colossal stupidity of this statement is just too much: ‘Assez!’ For what Mises does is effectively “reduce/traduce” these “elements of human meaning and conduct” such as “goods, commodities, and wealth” to precisely what he says they are not, that is, “things and tangible material objects” or “elements of nature”!

Additionally, in contrast to the natural sciences, Mises argued that there

were no constant relationships in human action. As such no universally valid

quantitative laws were possible in the realm of human affairs. Standing

between the claims of methodological monism on the one side and

historicism on the other side, Mises sought to carve out a niche for the

science of human action—one that agreed with the cultural critics of

methodological monism that the human sciences were unique, yet resisted the

implication of these critics that there were no nomological laws possible

in the human realm. Mises’s position was that while the science of human

action (praxeology) was different from the natural sciences for the reasons

enumerated above, it generated nomological laws that had the same

ontological claim on our attention as that of the natural sciences. (P.254)

Unbelievable leap from “language games” (with their tautological “inexorable laws”) to physical reality – how can “purposeful action” (an open-ended multi-verse) have the same “ontological status” let alone “claim” (!) as physical observation (of the inanimate world)? As a minimum, one would expect Schopenhauer’s categorical distinction between ‘Sinne’, “Reiz’ and ‘Motiv’ als Faktoren der Kausalitat (Tchauscheff, p24).

Here the authors clearly fall into a petitio principii in that they presume to know what the “purpose” of a particular “human action” is! This is what Hayek disputed. It is of no avail to invoke the Vichian ‘verum ipsum factum’- a notion that goes back to Aquinas through to Machiavelli and Hobbes to Vico. The “fact” that human beings make their own history and are therefore capable of “reflecting” upon it (Mises’s “introspection”) does not at all mean that human history has the same ontological and epistemological status as “physical events” – for the exact reason that human beings can “change”, “transform” or otherwise “revolutionise” their social relations. This was precisely Hayek’s disagreement with Mises: - that the “introspection” could only facilitate the interpretation of human behavior but could not determine it positively or scientifically because in social studies it is not the social scientist who decides but the “individuals” whom he or she observes extrinsically but whose “minds” he or she cannot know – however much he or she may try “to read” them!

From these categories implied in the axiom of action, Mises contends we

can deduce the pure logic of choice. The theories thus arrived at, because

they represent the elucidation and teasing out of the implications of the

fact that man acts ‘‘are, like those of logic and mathematics, a priori’’

(Mises 1949: 32). If no logical error has been made in the process of

deduction from the axiom of action, the theories arrived at are aprioristically

true and apodictically certain. Their aprioristic quality, however, does not

render them irrelevant to the real world. ‘‘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 as it appears in life and history. Praxeology conveys exact and precise

knowledge of real things’’ (Mises 1949: 39).(p258)

Complete garbage! That is precisely what Mises does not and cannot have: - “knowledge of real things”! Because a “language game” is only as good as the “game” it regulates: the ‘truth’ of its propositions is non-existent. In fact, Mises would not apply praxeology to “real things” (objects like physics), but rather to “human action” which is knowable through introspection and “thought-experiments”. But applying logico-mathematical identities to “human action” can only yield “language games” because, as Mises himself rightly pointed out, human beings do not behave like objects – the “circumstances” of their actions cannot be “reduced” to logico-mathematical identities. To the extent that they can, “human action” has been “reified”, reduced to a “thing” – and “the world” reduced to “the limit of language”. But the world is much more than language can describe or indeed “circum-scribe” or encapsulate!

(Kant and Schopenhauer were aware of the ‘technical’-only validity of logic and mathematics, and of the vital distinction between ‘logical form’ and ‘real content’ – whence the primacy of perception in the drawing up of scientific hypotheses. Of course, this is what the logical positivists were arguing – but methinks they presumed too much, that is, “proof”!)

Mises was mistakenly attributing the status of Kantian synthetic a priori status to what is instead a strictly Vichian foundation, verum ipsum factum:

The social scientist, on the other hand, is in a relatively better position, for

qua man, he is himself the very subject of his study. This fortunate position

allows him to get inside the mind of his subject. Thus, in the social sciences,

the scientist begins with knowledge of the ultimate causes driving his

subject’s behavior. And it is in this sense that the social scientist is in a better

position for the study of his field than the physical scientist in terms of

understanding causation. This fundamental difference between the relationship

of the physical scientist to his subject of inquiry and social scientist to his

subject of study suggests a fundamental difference in the epistemological

status of their insights and implies a methodological dualism in the realm of

science. (p254)

This essential difference between social and natural sciences is highlighted by Hayek (CRS) (as L&B acknowledge in fn.12). But Hayek draws the exact opposite conclusion, namely, that we must refrain from attributing to human agents motives arising from mere observation – the case of “behaviourism” and “scientism” generally – or even through Misesian “introspection”, which may apply to a single individual but not to interacting individuals. (Napoleoni moves this objection to Robbins as did Hayek, but propounds that Pareto optimality solves the problem for Walrasian equilibrium – which it does not because individual utilities are heterogeneous.) Because the “concepts” adopted to describe human behavior already involve “finality” (an implicit purpose), Hayek cautioned against drawing conclusions about the real motives or knowledge behind human use of the “objects” behind them, which can be illegitimately manipulated by “the observer” (the social scientist) who is also using “concepts” in the analysis. - Which is what Mises does, and Hayek wished to avoid by the time he wrote CRS and what led him to develop the concept of “spontaneous order”, precisely to be able “to observe” economic behaviour “from the point of view of individuals, whose motives for decisions are ‘inscrutable’” rather than manipulate it as in the “omniscience” of equilibrium analysis. This is the entire basis of Hayek’s “subjectivism”! This subtle yet essential difference between Mises and Hayek is detected with characteristic brilliance by Festre’ in “Von Mises vs. Hayek”. Leeson and Boettke have clearly missed it – but not Salerno (“Mises and Hayek Dehomogenized”) who foolishly chastises Hayek about it, as does Rothbard in his vituperation against Hayek. Hayek is more in line with (his cousin) Wittgenstein, who was far too subtle to believe that there is any “rational/logical” connection between what humans say and think and what they do. Indeed we may regress to Schopenhauer and the Scholastic qualitates occultae or “ultimate causes” or aims, which he excluded from scientific inference both a priori and a posteriori.

Here, in a nutshell, we have the difference between Hayek’s “homo quaerens” (where equilibrium is an approximate real state, the result of an empirical quest for prices or “price discovery”, and only a “heuristic” device for empirical observation until equilibrium is attained); and Mises’s “homo agens” where equilibrium is a logical language game schema indispensable for the “caeteris paribus Gedenkenexperimenten” of logical analysis but unattainable in fact – indeed, for Mises, profit depends on the impossibility of a competitive market process that leads to equilibrium: for him, competition and equilibrium are antithetical notions for profit to be possible. It may be said that here Hayek adopts Machian methodology and Mises adopts Neo-Kantian aetiology precisely in that “practical” sphere in which Kant could admit only of ethical and not causal relations.

According to Mises our nature as actors—beings who purposefully act—is

known through introspection. Reflection on what it means to be human

reveals that purposeful behavior is our primary and distinguishing feature.

This knowledge is aprioristic. We do not become aware of our uniquely

human characteristic through experience because we cannot, in fact,

‘‘experience’’ without purpose. Thus, ‘‘man does not have the creative

power to imagine categories at variance’’ with the category of action

(Mises 1949: 35). In taking action as the starting point for all of economic

13 Economic laws are deduced from the axiom of action aprioristically with the aid of the ceteris paribus

assumption that enables a sort of controlled mental experiment. And theoretical progress in the human

sciences, according to Mises, occurs by way of these mental experiments. Mises goes as far as to say that

the method of praxeology is the method of imaginary constructions (Mises 1949: 237 – 238).



theory, Mises roots the logic of choice in the broader logic of action he calls

praxeology. (Boettke and Leeson,‘Was Mises Right?’ p.254)

The fatal flaw in Mises’s “praxeology” is that he seeks to turn “human historical institutions” (goods, money, demand, supply, competition, individuals) into “reified” eternal or immutable natural entities subject to controlled experimentation – whilst at the same time protecting his fantastic theory from scientific examination by saying that it is not subject to experimentation because it arises “apodictically a priori from mental introspection”! But Mises cannot have it both ways! Either human institutions (including economic ones) are “reified” and capable of “logico-mathematical analysis” (the pure logic of choice), in which case we are brutally and violently imposing the present economic institutions as an ineluctable “destiny” of humanity; or else we admit that these are “historical institutions” that are not subject to logico-mathematical analysis except in the most brutal and repressive sense of adopting this analysis and its “categories” so as to preserve the established order!

Indeed, Mises’s quite pathetic and ridiculous attempt to attribute the “reification” of human institutions to some absurd “evolutionary process” (a foolhardy attempt undertaken first by his mentor Carl Menger) is the conclusive proof of the colossal stupidity of his entire undertaking!

In examining the a priori nature of these logical categories, Mises (1949,

1978) offers a speculative history as to how they evolved as part of the

human mind. According to Mises the a priori categories evolved along

with humans in a Darwinian fashion. We have the categories of the mind

that we do today precisely because they were best able to impart accurate

information about the real world to us necessary for our survival.

The categories are subject to future evolution as improved variations

enable us to better understand the world or the underlying reality of the

world itself changes. This hypothesized evolutionary process helps

explain the necessary connection of the starting point of action, and the

categories that it implies, to the real world. If they were not connected in

this way to the world, humans possessing them could not have evolved

as they have. There is a mutually interactive process between our minds

and the world, forming a feedback loop between the evolution of our

a priori mental categories that determine the world we experience, and the

reality of the world that conditions our way of thinking and understanding


Of course, this is the “psychologism” that Frege, Husserl and Wittgenstein refuted.

No comments:

Post a Comment