Markets and Discourses #5
Serialized treatise about Prosperity and Politics after the Libertarian Revolution
The following text is embedded within the treatise 1st section on Philosophy (see the post Markets and Discourses #2 List of Contents).
1.8. Praxeology after the Linguistic-Pragmatic Turn
Thesis: Misesian praxeology - the logical science of choice - understood as orienting framework for Economic and Sociological Theory formation must be refounded as Habermasian reconstructive science on the basis of the insights of the linguistic-pragmatic turn.
The premise here is that Mises’ concept and theory of ‘praxeology’ remains a unique and valuable contribution to the philosophical self-reflection of political economy and economic science today. However, in terms of its intellectual background assumptions, Mises praxeology is steeped and indeed confined within the philosophy of the early 20th Century. This fact compromises its standing and hampers its progress. The ambition of this chapter is recuperative: the clarification and reinterpretation of praxeology’s status and sources of validity in light of the philosophical insights we might point to with the phrase ‘linguistic-pragmatic turn’. These insights gained ground in philosophy only during the 2nd half of the 20th century after Mises had already elaborated his conception of praxeology. The linguistic-pragmatic turn builds on the insight that all thinking depends on language and emphasizes that language evolved in social life contexts. This implies that concepts must be distinguished and analysed as tools of social coordination in practical contexts rather than as “mental” or “logical” entities. The thesis here is that praxeology can be re-interpreted as a linguistic-pragmatic analysis and systematization of our modern, inherently discursive “lifeworld”, understood as everyday discourse-practices, in particular within practical economic contexts.
Here is how Juergen Habermas sumarizes the radical philosophical reorientation achieved by the linguistic-pragmatic turn: “Pragmatism and hermeneutics oust the traditional notion of the solitary subject that confronts objects and becomes reflective only by turning itself into an object. In its place they put an idea of cognition that is mediated by language and linked to action. Moreover, they emphasize the web of eveyday life and communication surrounding “our” cognitive achievements. The latter are intrinsically intersubjective and cooperative. It is unimportant just how this web is conceptualized, whether as “form of life,” “lifeworld,” “practice,” “linguistically mediated interaction,” a “language game,” “convention,” “cultural background,” “tradition,” “effective history,” or what have you. The important thing is that these common sense ideas, though they may function quite differently, attain a status that used to be reserved for the basic concepts of epistemology.”[i] Habermas further suggests that “Pragmatism and hermeneutics have joined forces … by attributing epistemic authority to the community of those who cooperate and speak with one another.”[ii]
The insights of the linguistic-pragmatic turn are fundamental and their dissemination within philosophy has been pervasive and decisive. Within recent theoretical philosophy it seems that the linguistic turn has been reversed. However, this does not mean that its fundamental insights have been reversed or invalidated. Rather these insights, insights this chapter and this whle treatise builds upon, are now generally taken for granted, and this implies that philosophical work has moved on. The initial implication that was drawn from the underlying insight into the pragmatic determination of meaning and the role of language in all thinking and action, namely that philosophy should focus on, or even restrict itself to describing and analysing everyday language use as a method of conceptual critique of “idle” or “metaphysical” philosophical language, or as a method of clarifying and solving conceptual problems ran out of steam, as too narrow a practice to pick up the grand mantle of what has been philosophy. A more substantively oriented philosophy, even an unabashed metaphysics returned. However, this does not in any way put the historical advances and insights of linguistic-pragmatic turn in doubt and their relevance and importance for the social sciences in particular.
Ludwig von Mises’ conception of economic science as praxeology offers a uniquely insightful counterpoint to most current mainstream conceptions of economics as empirical science. Praxeology is conceived as an a priori science investigating the logic of human action. Mises compares this enterprise to geometry and considers that its final result should take the form of a deductive system.
Although Mises’ conception seems odd today, it was not at odds with the mainstream economics of his time. As Hans-Herrmann Hoppe points out, “Mises did not wish to prescribe what economists should be doing as opposed to what they actually were doing. Rather he saw his achievement as a philosopher of economics in systematizing, and in making explicit what economics really was, and how it had implicitly been conceived by almost everyone calling himself an economist.” This rationally reconstructing rather than prescriptive approach is the mature, sensible attitude that has been adopted by all latter day philosophy of science. The philosopher is observing first of all how the scientist operate and achieve their successes. However, this attitude is not altogether uncritical. Rational reconstruction starts with rather than rejects the achievements of a science, but in systematizing and explicating these achievements it also critically evolves and refines them. Rational reconstruction operates from the more encompassing, self-conscious vantage point of an explicit philosophical reflection that is able to embed the reconstructed science in a more comprehensive scheme of the sciences, of knowledge, even of the human condition and enterprises as such. This encompassing systematisation is the source of philosophy’s guiding insights.
This chapter is adopting the same reconstructive attitude to Mises project of praxeology, i.e. it proposes and sketches the project of a philosophical reconstruction of praxeology, and by implication of (parts of) economic science, from the more evolved, more encompassing perspective of contemporary philosophy.
The thesis here is that from within contemporary philosophy, after the linguistic-pragmatic turn, praxeology can be critically reconstructed as conceptual systematization of our modern, inherently discursive “lifeworld”, understood as everyday discourse-practices, in particular within practical economic contexts. The thesis implies that praxeology can be understood as a version, an economy focused version, of a more general and indeed by now pervasive form of philosophical reflection that was first initiated by Kant’s Critiques. With and since Kant, this type of philosophical reflection, the reflection on the constituting conditions of a competency’s possibility, has been termed “transcendental”, still a well-established and useful ‘terminus technicus’ of philosophical discourse.
Transcendental reflection in philosophy rejects Cartesian doubt and instead starts with a doubtless fact of life, namely with a competency taken for granted and confirmed in practice. In the case of Kant’s Critique of Pure Reason, it is the competency of scientific knowledge as instantiated by the compelling success of Newtonian physics cashed out in survival competence via theory-led engineering.
This type of philosophical reflection is trying to formalize the most pervasive and thus most fundamental underlying world-revealing categories that always already form the basic enabling conceptual horizon of our shared (science-based) lives. Transcendental reflection thus retrieves an endlessly corroborated survival competence or “know how” that is thereby transformed into explicit and thus rhetorically criticisable “knowledge”. The phrase “rhetorically criticisable” is used here because any explicit statement can rhetorically be disputed and negated. However, in the case of a priori transcendental “truths” such negations or denials lead to a performative contradiction, namely denying what is always already presupposed in life and in communication. The demonstration of the a priori validity of these transcendental truths consists then precisely in displaying this contradiction.
The type of statements in question are axioms or truisms like “Events have causes and effects”, or “Men act (mostly, if not always) with intentions in mind.” Such statements just make explicit what is always already presupposed, i.e. they are “valid” a priori. These statements are literally indisputable, their negation makes no sense. As Mises has put it: Their negation is unthinkable. Wittgenstein’s way to put it is perhaps less mystifying: Such axioms represent the “grammar” or rules of application of the concepts involved.
The impossibility to think without these most basic categories is perhaps less certain than commonly presumed. It seems possible to imagine alternate terms that run counter to our usual ways of attributing agency, intention, purpose etc. to explain behaviour, including our own behaviours. Luhmann treats the concepts of action and experience as alternative attribution possibilities that should not be naturalized. Did I do this or did this happen to me? Within certain scientific discourses, namely cognitive science, alternative conceptualisations of agency, intentionality, cognition, knowledge and intelligence emerge. Within this discourse our common sense categories of self-description are referred to as “folk psychology” and the spectre of their eventual substitution, even within everyday discourse, is posited by the so called ‘eliminative materialism’ of Paul Churchland[iii]. The point here is not to argue that Churchland’s project is viable but just to historisize the idea of what is or is not unthinkable.
As Habermas reflects “discursive practice puts into operation the network of inferential relations built into the vocabulary of a language”[iv] and this network of inferences is itself woven and continuously rewoven by discursive practices within cooperative life processes. A reflective a priori science like praxeology then traces these inferences and tries to systematize, order and comprehensively display this network. The self-conception of Mises’ praxeology is deficient in the sense that it lacks awareness of the contingent practice dependency and evolving open-endedness of the conceptual system he traces and systematizes. Following Kant, he locates and fixes this system in the “structure of the human mind” and like Kant treats it – without stating this explicitly - as if it was given as an eternal cosmos of ideas. Indisputably, the network of shared basic praxeological concepts is - at any time – present, as competency, within each competently communicating and acting adult participant, and in this sense, with individually varying degrees, embedded within various well-socialized human minds. Also, the basic praxeological concepts Mises reconstructs are indeed very widely shared across times and peoples. But, as will be explored in more detail below, Mises’ Kantian-Platonic misconception of his own enterprise leads him astray in ways similar to Kant, not only with respect to naïve ahistorical universality claims, but also with respect to the form he believes his systematisation must take as deductive system, the sharp demarcation he believes that must exist between a priori necessary and empirically contingent inferences, and the presumption of a sealed completeness of the system. All these deficiencies which now compromise the credibility and fruitfulness of Misesian praxeology can be remedied via its reconceptualization on the basis of the insights and methodological innovations elaborated since the linguistic-pragmatic turn.
Divers authors like Dilthey, Peirce, Durkheim, Weber, Cassierer, Husserl, Heidegger, Schutz, Apel, Habermas, Strawson, Putnam and Brandom a.o. have worked towards overcoming the otherworldly flavour of Kant’s initiation of transcendentalism. While Kant’s transcendentalism is formulated in the terms of a philosophy on consciousness, the above string of authors started to make attempts to reformulate Kant’s basic insights about the categorical framing of all experience in terms of culturally evolved frames in the context of social life.
Kant took the faculties of the ‘transcendental subject’ as an unexplained basic datum, and so did Mises when he talks about ‘the structure of the human mind’. Mises formulations remained steeped in Kant’s mentalistic conception of the a priori categories, despite the fact that he – as Hoppe emphasizes - grounds them in human action. For Mises action is first of all a mental act, involving intention. Talk about “mental acts” as well as talk about concepts as the unexplained “content” of these mental acts is no longer philosophically respectable. Such talk must be reconstructed in linguistic-pragmatic categories, lest they remain otherworldly mystifications.
Starting tentatively with the above mentioned string of authors, and definitely since the linguistic and pragmatic turn of philosophers like the late Wittgenstein, Whorf, Quine, Sellars, Apel, Kamlah, Lorenzen and Habermas a.o. we can explain the a priori categories and axioms as deep, historically evolved linguistic sedimentations of the competences of the surviving cultures or “life forms”, to use the late Wittgenstein’s key concept, or “life worlds”, the respective phrase used by the late Husserl as well as by Schutz and Habermas.
As indicated above, the thesis here is that praxeology can be reinterpreted as a linguistic-pragmatic analysis and systematization of our everyday discourse-practices within practical economic contexts. With Habermas we could also speak of a ‘rational reconstruction’ of the ‘formal-pragmatic’ conceptual and social conditions of our modern, inherently discursive ‘lifeworld’.
The a priori character of praxeology, and thereby of at least a part of economic teaching, is, (at least in accordance with Habermas weak version a priorism quoted below) beyond dispute. The question here is what we mean and imply by this?
What is first of all implied, and beyond doubt, is that the (in this context) oft-cited postulates of praxeological economics, like the law of diminishing utility, the law of returns, the law of supply and demand, or the quantity theory of money, are not to be subjected to empirical testing. This however, also applies to many basic statements of mechanics. Indeed any scientific venture builds on founding categories and premises which are never questioned within the particular scientific enterprise.
Murray Rothbard’s 1956 statement displays the flawed, dogmatic conception of praxeology when he contrasts praxeology with the empirical sciences: “On the other hand, economics, or praxeology, has full and complete knowledge of its original and basic axioms. These are the axioms implicit in the very existence of human action, and they are absolutely valid so long as human beings exist. But if the axioms of praxeology are absolutely valid for human existence, 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.”[v]
However, a priori conceptual structures and ‘axioms’ are not unique to the science of human action. Every science or scientific project builds on constitutive conceptual structures that are not questioned within the science.
There are also more general basic categories and premises – like the category of substance/accidence, or cause/effect– that not only underlie all sciences as a priori premises, but also underlie our everyday world orientation.
Here we can already identify a necessary clarification: The a priori status and infallibility of the tenets of praxeology should not be sharply contrasted with the supposedly fully empirical natural sciences per se, and does not rest, as Mises claimed, on the special privileged introspective access we have to these particular truth as truth about intentional human action. Rather, the idea of introspective access should be reinterpreted and generalized as reflective access to all fundamental conceptual structures that constitute an always already linguistically mediated lifeworld, including fundamental discourse practices of basic numeracy and geometry.
Mises himself compared praxeology to geometry, but did not explain what undergirds their a priori validity in both cases, namely the constitutive, world-disclosing conceptual structures that we can reflectively retrieve from our everyday discursive practices. (The phrase ‘discursive practice’ denotes here no more than linguistically supported social practice. This does not necessarily imply the more evolved practice of reflective discourse in the sense of debating controversial validity claims or even theoretical discourse.)
We might make another fruitful analogy: While the ‘logic of action’ that is reflectively made explicit by praxeology is more general than the grammar of any particular language, praxeology is “a priori” in the way linguistics or grammar is a priori, i.e. conducted from the arm chair using the grammarian’s ability to retrieve and reflect his language competency as testing ground when trying to make it explicit as grammar or system of linguistic rules. Certain meta-grammars or general theories of grammar offer perhaps an even closer analogy, e.g. Michael Halliday’s ‘functional discourse grammar’. An even closer comparison could be made to Michael Bratman’s ‘planning theory of intention’, a recent innovative and influential philosophical account of action that should probably be classified as praxeology. Further, there are good reasons to classify a large part of the expansive research programme of game theory as a variant of praxeology. Game theory is a formal theoretical science of strategic calculation and action within which the construction of deductive systems, in comparison to Rothbard’s system, has been pushed much further in the direction of an explicit, rigorous axiomatization that meets the modern logical standards of formalisation. However, the attitude prevailing here is, or should be, not a matter of finding irrefutable, “a priori” principles of strategic action that can be relied upon to predict human behaviour and the success or failure in establishing cooperation. Rather, these systems are theoretical models that must be confronted with empirical evidence, that is, historical or experimental evidence.
Praxeology in general, including the praxeology of Mises and Rothbard, might be understood as doing something rather similar to Heidegger’s ‘analytic of Dasein’ in ‘Being and Time’, except praxeology does not only reflect on the basic conceptual structures of ordinary everyday life but also reaches out to reconstruct from there the more elaborate conceptualisations of a sophisticated, modern business life and indeed of the science of economics.
Moreover, Mises is not only underpinning this science but is trying to critically systematize and further elaborate the economic science and the theory-led business thinking of his time. As we shall see below, my interpretation of praxeology as transcendental, linguistic-pragmatic analysis and reconstruction works well to the extent that praxeology is reflecting very fundamental and general conceptual structures of everyday life and basic economic life. This interpretation of praxeology shows why it’s a priori character becomes tenuous and problematic as soon as praxeology makes claims to cover less well established business discourses and calculations and even claims to encompass all of economic science.
Categories and “axioms” are found embedded or implied in all of our everyday speech and practice. We might refer (with Husserl) to these practices as constituting our culturally and often cross-culturally shared lifeworld, which is reflected in our shared language. Lifeworld and language evolved (and continue to evolve) together.
Evolutionary processes are survival processes and to that extent deliver successful structures, inclusive of successful linguistic-conceptual structures. If we build on the pragmatist insight that truth means success, we start to see not only that we are locked in with certain language-imposed and thus unquestionable axioms but that this fact of life is a functional feature rather than a bug.
So my thesis here is that the most basic, nearly “tautological” tenets of praxeology, i.e. that man acts according to his preferences, in pursuit of his ends and chooses means according to his beliefs etc. emerge from the reflective distillation of our lifeworld’s basic conceptual system, and that therefore the praxeologist’s work should be interpreted in close analogy to the linguistic-pragmatic analyses we find in the work of the ‘ordinary language’ school of philosophy, following the example of the late Wittgenstein.
The validity of these most basic of praxeology’s tenets’ is thus rooted in the practical robustness of these basic lifeworld-language structures. These “conceptual” structures have been so thoroughly corroborated that we should not expect to be able to radically innovate here and they are so deeply rooted that we often cannot even imagine alternatives to these.
However, the Kantian understanding of these in fact culturally evolved conceptual structures, which Mises still displays when he says that the principles of praxeology are rooted in the structure of the human mind, must be updated in line with the linguistic-pragmatic turn. The Kant-Mises conception unduly eternalizes the praxeological categories which instead must be historisized, a process that Hegel initiated and thinkers like Durkheim continued without Hegel’s mystification. Habermas very clear about this. In an early text from 1968, Knowledge and Human Interest, concerning the fundamental categories and principles of science, he states that these “cannot be either logically deduced or empirically demonstrated”, and that they are “neither arbitrary nor compelling”. Habermas further elaborates: “They prove appropriate or inappropriate. For their criterion is the metalogical necessity of interests that we can neither prescribe nor represent, but with which we must instead come to terms. … The achievements of the transcendental subject have their basis in the natural history of the human species.”[vi] Habermas also emphasizes that cultural evolution evolved the cognitive capacities and conceptual resources further.
Habermas postulates “reconstructive science” as a sui generis type of science that is in many respects different from the usual empirical sciences, but not altogether a priori, nor apodictic. Habermas counts his own research programme of “universal pragmatics” that aims at reconstructing the universal validity basis of communicative speech amongst these reconstructive sciences. Mises praxeology can and should be reconceptualised, reformed and continued in accordance with this conception of a reconstructive science. In general, according to Habermas, a reconstructive science “transforms a practically mastered pretheoretical knowledge (know-how) of competent subjects into an objective and explicit knowledge (know-that).”[vii] Habermas gives the following sciences as examples of reconstructive science: linguistics, logic, the theory of science, and ethics. These sciences start with practices that produce symbolic objects that are given and understood as well-formed and valid. “Thus syntactic theory, propositional logic, the theory of science and ethics start with syntactically well-formed sentences, correctly fashioned propositions, well-corroborated theories, and morally unobjectionable resolutions of norm conflicts, in order to reconstruct the rules according to which these formations can be formed.”[viii] Habermas stresses here that reconstruction always concerns “universal” competencies. “To the extent that universal validity claims (the grammaticality of sentences, the consistency of propositions, the truth of hypotheses, the rightness of norms of actions) underlie intuitive evaluations, reconstructions relate to pretheoretical knowledge of a general sort, to universal capabilities (original emphasis), and not merely to particular competencies of individual groups, … a general cognitive, linguistic or interactive competence.”[ix]
With a view to Mises’ praxeology we might add here reference to the universal competency of rational-purposive human action, with economic actions as paradigmatic examples. However, the universality of the supposedly universal competencies is not absolute, as we shall see. Habermas is pondering whether the method of rational reconstruction of these reconstructive sciences should still be referred to as “transcendental analysis”. He hesitates, as a Kantian demonstration of the a priori validity of the constitutive categories in question via “transcendental deduction” is no longer credible. “The strong a priorism of Kantian philosophy gives way to a weaker version. From now on transcendental investigations must rely on the competence of knowing subjects …in order trhen to analyse the material with a view to finding general and necessary categorical presuppositions. Every reconstruction of a basic conceptual system of possible experience has to be regarded as hypothetical proposal that can be tested against new experiences. As long as the assertion of its necessity and universality has not been refuted, we term “transcendental” the conceptual structure recurring in all coherent experiences. In this weaker version, the claim that this structure can be demonstrated a priori is dropped. … We can no longer exclude the possibility that the basic conceptual structure has developed phylogenetically and arises anew in every normal ontogenesis, in a process that can be analysed empirically. We cannot even exclude the possibility that an a priori of experience that is relativized in this sense is valid only for specific, admittedly deep-seated, behavioural systems.”[x]
The Spectre of Polylogism, Historicism, and Relativism
At the very beginning of ‘Human Action’ Mises is eager to ward off the spectre of polylogism, historicism, and relativism: “Marxism asserts that … every social class has a logic of its own. This polylogism was later taught in various other forms also. Historicism asserts that the logical structure of human thought and action is liable to change in the course of historical evolution.”
According to this definition of ‘historicism’ this treatise endorses a historicist (polylogist) reformulation of praxeology, albeit without thereby endorsing relativism.
Mises believed that such a historicism implies relativism and would fatally undermine all legitimate scientific pretensions of economics. While the example of the Prussian (German) Historical School’s rejection of universal economic laws might suggest this, this conclusion is not a necessary one. We can unreservedly subscribe and commit to modern Western logic and rationality - a rationality that made science-based technology and the rational modern economy first of all possible - while admitting that this was a rather unique historical achievement, rather than a universal biological endowment of the species homo sapiens.
Modern critical-rational logic evolved within a specific tradition and expanded language community. However, this is not a self-contained language community but has become the open ended, self-transcending, infinite, cosmopolitan language community of our contemporary world society. This inherent open-endedness and self-transcendence implies that parochial encapsulation is being avoided. The fact that there exists only one such global discursive community, which includes all science and thus Peirce's ‘infinite community of scientists’, without any serious rival, implies that a disorienting or contradictory relativism is not on the cards here.
The rapid global universalisation of this historic achievement of modern rational logic can be explained by its pragmatic economic success. This success is also what merits our commitment to both modern natural science and to modern economic rationality, including the explicit statement of this ‘logic of action’ as offered by praxeology and economic science. However, there is no need to set Mises’ formalisation of this modern logic of action absolute, nor the evolved form of this logic itself as it currently prevails in advanced societies.
I agree with Hans-Herrmann Hoppe’s statement that “our mental categories have to be understood as ultimately grounded in categories of action”. If we take the phrase ‘categories of action’ very broadly this formulation comes close to state the general insight and attitude of philosophical pragmatism promoted in this linguistic-pragmatic reconstruction of praxeology. In particular Hoppe is right to emphasise that the general principle of causality that underlies all natural sciences is a category of action, indeed an indispensable category of action, and we might add: just as the economic principles of praxeology, so is the principle of causality an a priori principle. As Kantian, Mises would have acknowledged this. However, neither Mises, nor Hoppe who – unlike Mises – is explicit in using Kantian terminology to describe the laws of praxeology as ‘synthetic a priori truth’ in parallel to the principle of causality, see here occasion to wonder if their sharp separation between the natural sciences and economics cannot be questioned on the ground that natural sciences like (theoretical) physics are also based on a priori principles. Here too, a lot of work is the non-empirical work of developing elaborate theoretical models with only indirect, tenuous connections with empirical observations. Is the principle of the conservation of energy/matter really treated like an empirical hypothesis or finding? Since Quine’s critique the distinction of analytic versus synthetic has rightly become suspect. In both everyday life and in science what is definitional versus what is open to revision due to new experience is rarely explicit. Neither our everyday language games nor the terminologies of our scientific theories operate on the basis of explicit, fully formalized systems of definitions where this distinction between analytic and synthetic statements would be readily applicable.
Mises’ overemphasis of the contrast was indeed based on taking the false conception of the natural sciences promulgated by the logical positivist/empiricist of his time for granted. The same applies to Rothbard. The contrast evaporates when we look at both physics and economics as comprising a whole spectrum of propositions from a priori categories and principles or axioms, via fundamental definitions and a largely unquestioned set of paradigmatic core theorems all the way to explorative hypotheses at the frontier of empirical research. The suggestion here is that if we take a later, further evolved philosophical account of the natural sciences, like e.g. Imre Lakatos’ ‘Methodology of Scientific Research Programmes’, as basis of the comparison, the sharpness of the divide will disappear. This insight has already long since been available to Hayek via Popper and explains via Hayek never embraced Mises’ praxeology.
David Gordon, a faithful adherent of Mises and Rothbard, emphasizes the character of praxeology as a deductive system: “One starts with a self-evident axiom (“man acts”) and with the aid of a few subsidiary postulates, deduces the entire science of human action.” However, he insists that ‘self-evident’ here does not appeal to a psychological experience of certainty, since this would imply that the statement’s evidence would depend on something else, namely the psychological experience. This bit of clever sophistry reveals that ‘self-evidence’ cannot be explicated within the traditional philosophical framework of Mises’ praxeology. It remains a mystery, an inexplicable datum. However, within the linguistic-pragmatic framework offered here the self-evident is simply explicated as that which is unquestionably taken for granted within the language community as community of practice, i.e. “self-evidence” becomes a rather unproblematic, down to earth concept, which also does not rely on the subjective experience of certainty, although it might be accompanied by such a subjective sense.
Gordon seems to have come across interpretations of praxeology that seem in some aspects somewhat similar to the linguistic-pragmatic interpretation offered here and he thinks these can be dismissed simply as another variant of what he calls the “psychological fallacy”. After emphasizing that self-evidence has nothing to do with psychological states he writes: “The point is important because contemporary hermeneuticists sometimes maintain that the self-evident axioms of praxeology are really principles accepted by a particular community. This approach is just a variant of the psychological fallacy we have already considered.”
This is not helpful. Hermeneutics is not psychology. But apart from this, the mere insistence on absolute truth or evidence universally accessible to all rational beings etc. is just this, a mere insistence that cannot be redeemed scientifically, an insistence moreover that operates with an outmoded epistemology. The problem is that Mises, Gordon and Rothbard operate with a platonic epistemology and thus metaphysics. They cannot conceptualize the communicative processes of science, nor therefore their own cognitive-communicative processes, from within their worldview and conceptual universe. They operate from within an unacknowledged dualism that presumes that de jure questions can be unambiguously and categorically kept apart from genetic questions. That platonic dualism is what underlies the rejection of what since Frege and Husserl had been castigated as “psychologism” in the study of logic and science. Against this stand more recent attempts to naturalise truth, logic, science etc. as dependent on linguistically mediated social processes. While the principle insights of such a naturalist unification, the naturalisation of human thinking, was not worked out yet during Mises' formative years, there is no such excuse available to Gordon in the age of artificial intelligence, and after 50 years of philosoohical advances after the linguistic-pragmatic turn.
Going back to a more concrete discussion of Gordon’s proposition: Self-evidence is a rather misleading term here anyway. “Self-evidence” is indeed an oxymoron or paradoxical phrase, as the phrase aborts the language game of giving evidence. The request for evidence is asking for something that supports the proposition is question.
The “truth” of these truisms is not a question of evidence, rather the question of truth is never coming up at all among participants of the respective discursive practices, subjectively or inter-subjectively. At the level of philosophical reflection, these questions are simple questions of fact about these discursive practices: are the respective praxeological principles at work as taken for granted suppositions or not. It is possible to come up with empirical evidence gathering on this count, and we can use our own intuitions as proxy here also, just as linguists rely on their own language competency as access to or proxy for the rules that govern the community’s language use. This is an explicit insight and tenet within linguistic philosophy, where, as for instance Timothy Williamson confirms when he asserts that “Linguistic … philosophers treat intuitions .. as the deliverances of linguistic or conceptual competence.”[xi] To be sure, the introspective armchair work of the grammarians generating grammatically correct sample sentences, or appraising the grammatical correctness of sample sentences, is nowadays augmented and confirmed by computationally supported work with recorded language corpuses.
With respect to “self-evidence” or “intuition” we can, as Gordon in line with the philosophical tradition from Kant, via Husserl, to Frege does, distinguish psychological questions of cognitive capacity, process and experience from logical questions of validity. However, this treatise posits in contrast to Kant, Husserl and Frege, that validity, in the final analysis, must also be naturalized, via an historical and pragmatist appraisal. Logical validity, in a first step, relies on a given logic system, whether explicitly formalized or implicitly institutionalized. However, in a second step, to avoid relativism, we need to rely on a successful real life track record of this system, and try to rationally and comparatively reconstruct this pragmatic success. This treatise thus proposes a two-tier notion of logical vality in analogy to R.M. Hare’s two-tier notion of moral validity. Therefore, as it were, ‘rule utilitarianism’ can be as much applied to logic as to morality, albeit with the additional complication that circularity cannot be avoided, but only unfolded.
When it comes to the psychological questions of cognitive capacity, process and experience, we need to dispel the mystery of self-evidence and intuition, and instead inquire about the cognitive mechanisms and their capacity and robustness in employing the logical system in its various real world applications. What is the intuitive grasp of the self-evident? Surely not Husserl’s mysterious, unexplained platonic “Wesensschau”, seeing of essences. The underlying process is much more down to earth. We can learn here from Herbert Simons explication of the mystery of intuitive thinking. He says intuition is “nothing more and nothing less than recognition”[xii]. Herbert Simon argued this in the context of investigating the cognitive processes of chess players. He found that chess grand masters recognize chess constellations and instantly are aware of forward options. Grand masters have command over up to 100,000 constellations. Intuitive inferences or the recognition of self-evident truths have this character. We recognize constellations of concepts and that means ultimately constellations of words or word substitution classes. Intuition is recognition, somewhat akin to face recognition. We do not know how we do it, no deliberate steps are involved. But the capacity is obviously based on a memory structure that has been build up over a long period. Albert Einstein had the right intuition when he observed that “intuition is nothing but the outcome of earlier intellectual experience.”[xiii] Such insights are useful, and more generally, questions of cognitive processes, and related questions of cognitive capacities are relevant to economic science, and this includes also forms of corporate cognition, i.e. accounting, analysis and calculation.
The question might be raised here whether this pragmatist interpretation leads to a problematic relativism – or as Mises would say ‘polylogism’ - that undermines praxeology’s universalising claims and indeed the universalizing claims of economic science? The answer is not a simple yes or no, because the “particular community” in question is modernity’s inherently self-transcending, universalising world community, or world society. In this sense there is no danger of relativism, as the commitment to modern, enlightened, science-based civilisation is not relativized here, although it cannot be assumed that these discursive practices are human universals, as there have been older, inferior, less viable practices. We can admit that there are different “logics” without assuming that they are all equally valid or valuable, i.e. without relativistic conclusions. We can admit to the fact of polylogism - in the sense that different logics operate factually - without adhering to the fallacy of multi-culturalism that assumes all cultures and their respective logics are equally valid, or deserve equal respect. Validity must here interpreted and appraised pragmatically, in terms of economic success. Not all cultures are equally successful, i.e. empowering and life enhancing.
While we are thus should be confident and unambiguous rather than relativizing in our support for the rationality encoded in Mises’ concept and axiom of action, we cannot set Mises’ praxeology absolute for all eternity. We must allow at least for the possibility that our discursive practices and thus its underlying praxeological principles evolve further. The language game of asking and giving reasons for actions is neither fixed, nor uniform.
In various fields of business, professional work, and research, random mutations are integrated into strategies of innovation. The role of chance discoveries in the progress of science and technology is long since proverbial without systematic acknowledgement on the part of epistemology. Even today the notion of random pursuits rings anti-thetical to notions of strategic conduct or rationality. Mutations, i.e. experimental discoveries, here are theoretically directed mutations, not utterly random. However, even very strategically planned experiments or innovations rarely involve the anticipations of all consequences.
In the history and philosophy of science , as well as in management theory and economics, a new notion of rationality crystallises. Groping, the incorporation of random play and a margin of exploratory investment, are now seen to be necessary ingredients of any strategy aimed at innovation. Sophisticated arguments which challenge the most basic truisms of praxeology and undermine the traditional antithesis of (random) play and (strategic) plan are developed in J.G.March and J.P.Olsen’s "Ambiguity and Choice in Organizations". From the vantage point of management this critique suspends engrained 'certainties' about the logic and rationality of planned, purposeful action. The whole edifice of Western rationality is shaken in this text with no explicit philosophical ambitions.
Within most of the Western world individuals and organisations see themselves as making rational choices. This concept assumes the pre-existence of purpose and poses the consistency of conduct. Those ideas, deeply embedded in modern society, are made the explicit axioms of decision theory. "It is fundamental to those theories that thinking should precede action; that action should serve a purpose; that purpose should be defined in terms of a consistent set of pre-existenting goals; and that choice should be based on a consistent theory of the relation between action and its consequences. Every tool of ... management science, operations research or decision theory ... (and) the entire structure of micro-economic theory builds on the assumption that there exists a well-defined stable, and consistent preference ordering." (March&Olsen).
What decision theory and the ideology of choice can only regard as deficiency - the reality of "the fluidity and ambiguity of objectives"(M.&O.) - needs to be 'redeemed' within a new and more complex understanding of innovative rationality. The whole economy of rationality involving the network of concepts like freedom, coercion, identity and progress will have to be deconstructed/reconstructed. "Goals are thrust upon the intelligent man. We ask that he act in the name of goals. We ask that he keeps his goals consistent."(M&O) Intentionality is seen to be the defining moment of human consciousness, in its individual as well as collective existence. March & Olsen do not indulge in an abstract negation of goal-oriented rationality, rather they propose its sublation into "more complicated forms of consistency". On this basis a more complex rationality can be elaborated which allows for the temporary relaxation of the demand for predefined purposes to engender and enable procedures for the discovery or construction of new values and goals. The current reality of shifting goals seems to force us to "choose now in terms of the unknown set of values we will have at some future time. ... This violates severely our sense of temporal order."(M.&O.) Such a "choice" is, according to praxeology and rational choice theory, utterly non-sensical.
For example, in architecture and other design disciplines, some avant-garde design-processes have precisely this warped time structure: 'choose' now, 'motivate' later. The design process is being systematically purged of any preconceived intention and replaced by an ever extending series of (initially) arbitrary formal moves: formal transformations as form-generating aleatoric processes. Such a process or "method" involves the radical suspension of everything usually associated with "design" as deliberate purpose-lead activity, directed to solve well-defined problems according to known and explicit criteria. Progress can no longer be monitored as the systematic accumulation of solutions on that basis. Instead of such step by step accountable conduct, initially unaccountable graphic proliferation is the order of the day. Freedom and progress are here mediated through coercion in the sense of the designer's (temporary) submission to the arbitrary determination of the graphic or algorithmic process. "Coercion is not necessarily an assault on individual autonomy. It can be a device for stimulating individuality."(M.&O.) In the aleatoric design method the formal process is running ahead and a meaning (programme) is read into it a posteriori, allowing for an innovative re-alignment of both new form and new function. The aleatoric play is an instrument of intelligence, not its negation or substitute. As in biological evolution, the necessary condition for the ability to harness chance for the purposes of innovation is reproduction, i.e. the ability to reproduce an initially unintended and uncontrolled effect. What initially was play, eventually becomes method. "Playfulness is the deliberate, temporary relaxation of rules in order to explore the possibilities of alternative rules. When we are playful we challenge the necessity of consistency. In effect, we announce - in advance - our rejection of usual objections to behaviour that does not fit the standard model of intelligence. Playfulness allows experimentation. At the same time, it acknowledges reason. It accepts that at one point ... it will be integrated into the structure of intelligence." In this context March & Olsen arrive at the Derridian insight about the temporal logic of becoming: "Planning in organisations has many virtues, but a plan can often be more effective as an interpretation of past decisions than as a program for future ones. ... In an organisation that wants to continue to develop new objectives, a manager needs to be relatively tolerant of the idea that he will discover the meaning of yesterday's action in the experiences and interpretations of today."(M.&O.)
This excursus on March and Olsen’s reflections on productive ambiguity are not meant to refute the basic theorems of praxeology. They remain valid, largely unquestioned, taken for granted, but not absolutely so. They evolved historically, and will evolve further. Indeed, the language game of asking and giving reasons for actions is neither fixed, nor uniform, it is evolving.
Similarly, it seems that the logic (and ontology) that underlies physics has evolved since Newton’s mechanics, via Einstein’s theory of relativity, and further via quantum physics with Heisenberg's uncertainty principle, and continues to evolve via ventures like string theory, and chaos theory etc.
In this situation logic - both economic logic and scientific logic - seems to be well advised to heed Rudolph Carnap’s famously provocative ‘Principle of Tolerance’ that recommends logicians to replace prohibitions with the simultaneous construction and “investigation of language-forms of different kinds” . Carnap wants to “cast the ship of logic from the terra firma of the classical forms” to explore “the boundless ocean of unlimited possibilities” and believes that this venture is hampered by the striving after ‘correctness’. Various non-aristotelian logics have since been elaborated.
Networks rather than Hierarchical Deductive Systems
Murray Rothbard follows Mises in identifying economic science with a priori praxeology, which in turn is conceived as a deductive system where economic laws are deduced from the fundamental axiom of action. Rothbard’s opus magnum ’Man, Economy and State’ is trying to make this deductive form of economic science explicit. On the first pages of this treatise we read: “All human beings act by virtue of their existence and their nature as human beings. We could not conceive of human beings who do not act purposefully, who have no ends in view that they desire and attempt to attain. … It is this fundamental truth—this axiom of human action—that forms the key to our study. The entire realm of praxeology and its best developed subdivision, economics, is based on an analysis of the necessary logical implications of this concept.”
Rothbard’s treatise starts with truisms like this: “The purpose of a man's act is his end; the desire to achieve this end is the man's motive for instituting the action.” Concepts like action, purpose, end, mean, motive, want, desire, preference, benefit, intention, choice, alternative, production, utility, cost, ownership, right of use, exchange, money, price, credit, debt etc. are interdependent, albeit without a logically inherent hierarchy. Some of these terms belong to the sphere of ‘economically oriented’ action and interaction. Others are more fundamental and transcend all specialized spheres of social life. However, within a given discursive practice all terms are equally and indispensably implied.
What we find here is a network of many interdependent terms that together constitute the structure of a particular discursive practice. Wittgenstein’s philosophy of language games homes in on the very same insight. A particular language game or network of terms and “axioms” or “truisms” sustains a particular form of life. This network displays no natural hierarchy which would privilege a reconstruction as a deductive system which picks out some terms and some key tenets as basic and axiomatic while then deriving the rest in a sequence of deductive derivation.
We might start at any point within the network, start with any of the terms to trace the dependencies with all the other terms in the network. This is in fact what Mises and Rothbard are doing: tracing dependencies, starting with the presumably fundamental concept of action and the axiom ‘man acts’.
They might have as well started with any of the other terms: preference, end, means, profit, cost etc. In practice we indeed often “start” with means, or rather with potential resources at hand, and then explore their potential uses and purposes. This might even be the dominant case of entrepreneurial action, as presumed in effectuation theory , according to which entrepreneurs will determine goals according to the resources in their possession.
A network of terms emerges through language use in recurring life situations where these terms are combined and follow each other in the questions, descriptions, guiding suggestions and commands of practical economic life. The terms refer to different aspects of an integral process of cooperation that is continuously and faithfully reproduced due to its material accomplishments. However, it also evolves, if only very gradually.
Because these terms refer to aspects of an integral, recurring process, truisms, i.e. general statements can be extracted. These truisms display the way the terms are connected and depend on each other to do their work.
A person which expresses the intention to do something can always be asked about his/her purpose, then questioned about the appropriateness of his/her intended action as a means towards the stated purpose, then queried about how this purpose fits into a wider set of purposes, or he/she can be confronted with the preferences this purpose seems to indicate, or be invited to reflect on the concrete costs of the intended action, or be confronted with the opportunity costs of the action etc. It is these communications and their concatenations that have established the network of terms and their attendant “trusims”, and that continue to reproduce these terms and truisms as long as this language game continues to facilitate material reproduction.
In each language game, or conceptual network of interdependencies, the involved concepts “require” each other. What does it mean that one concept implies another? That we - when we close our eyes and think - see these implications, is a mystery rather than an explanation. Our mind seems to perform some mysterious immaterial operations in the otherworldly realm of the platonic heaven of concepts or meanings. We must try to overcome such misleading pictures.
What does that it mean in concrete, worldly terms to “draw an inference “, or to “see an implication”? And how are implications first established? And what are concepts or meanings anyway?
Meanings are the regular traffic or use patterns of words, i.e. of inert sounds, which we can handle by our lower more animal-like conditioned cognitive processes, a handling that thereby generates the emergent order of our higher, conceptual capacities. But these thinking capacities are in this way language-dependent. We cannot do much thinking with our bare brains.
So that the various terms of an institutionalized language game or discursive practice require or imply each other, means that in the practical discourses of the type that historically evolved the terms in question, participants have the legitimate expectation that if one of the terms is posited, then questions about the other terms are in order and deserve answers, i.e. if somebody declares an intention to do X, then the question of the purpose of X is in order and should have an answer.
So, on the face of it, the “logical or conceptual requirement” is, first of all, a social requirement of a particular institutionalized type of communicative interaction. Implication derives from real life concatenation delivering practical functionality that spurns the reproduction and confirmation of these concatenations which thereby firm up into (nearly) “necessary” implications. As critical philosophers we can then further reflect that all this makes sense in terms of survival and prosperity. Those who have been questioned about their purposes and then induced to query their means perform better than those who just follow ritualized patterns of action. The “conceptual requirement” can thus be rationally reconstructed and validated via a priori reflection. While the conceptual truism cannot be falsified, it is, in some sense, an “empirical” fact, namely a linguistic and social fact. These are transcendental facts of our modern lives.
This insight into the source or cause of the conceptual network and its necessary links leads us to a somewhat different heuristics for a reflective a priori economics, namely we must realize that the links are not quite as apodictic as Mises and Rothbard would want to make them. The various concepts and a priori theorems that Mises and Rothbard rightly presume to be implied within the concept and axiom of action follow from it and each other only with near necessity, not with absolute necessity.
Taking the axiomatic system of geometry as his model, inclusive of its supposed Kantian introspective self-evidence, Mises and Rothbard set the implicative connections absolute, allowing for no exception. Theorems are treated as analytic, i.e. as if of logical necessity. They become tautologies. Mises is indeed explicit about this: “Aprioristic reasoning is purely conceptual and deductive. It cannot produce anything else but tautologies and analytic judgments.”
For instance every action reveals preferences, by logical necessity, and in turn what you have done must have been what you have preferred and what you at that moment considered to be more valuable. Mises, for instance, in ‘Human Action’, talks about “the fundamental praxeological insight that men prefer what satisfies them more to what satisfies them less and that they value things on the basis of their utility”.
The terms are tied up so tightly that their meanings collapse into a single fact. Preferring something now implies that we value it and that it satisfies us and that it delivers utility.
Wittgenstein would ask: What do we gain by this total assimilation of terms, which becomes an elimination of terms, with the net result of impoverishing our language. In our natural language preference, utility, value and satisfaction are surely linked but certainly not identical. Linking here means, as elaborated above, that talk about preference, allows us to ask about utility, and utility allows us to ask about value, and about satisfaction. These terms form a network, they call for each other, “require” each other, to some extent, but not absolutely. Science, for the sake of rigour and clarity, will have to give up some of the nuanced richness of natural language. When it sharpens and streamlines its terms, it loses track of some of the intuitive competencies that are embedded in the natural language operating in the domain in question. This cleansing process must be understood as such and it should be understood that it is an intervention that can lose more than it gains. When Rothbard works with the concept of demonstrated preferences , implying that the preferences are tautologically demonstrated in actions, he in fact eliminates the very concept of preferences as an independent element in an otherwise more complex (but less tractable) account.
Does this, our modern everyday base-economic network of terms and truisms have a determinate hierarchical structure that would allow it to be cast into an axiomatic system? Are the dependencies asymmetric? Are some of those terms more fundamental than others? Are purposes not more fundamental than means and actions more fundamental than costs? If so, is action more fundamental than purpose? What can “fundamental” possibly mean here? None of these terms can be “fundamental” in the sense that they could be free standing, i.e. exist without the other terms. Action is (almost) always a means to an end, and implies choice, and choices imply alternatives and thus costs etc. We can always run up and down a whole chain and in various orders, and this weaves the network.
Let us consider two ways in which we might think of and motivate a hierarchical ordering of the terms of the network of mutual dependencies.
First, there might be a “hierarchy” here with respect to how individual actors typically think through their plans: They might reflect their purposes before they settle on the means to satisfy them. The reflection on the means might start with listing a set of alternatives which are thereafter compared in terms of monetary cost or in terms of time costs or degree of effort. So there seems to be sequential logic at play here, i.e. a hierarchy in terms of what typically comes first within the unfolding of a decision making process for particular practical projects. So is ‘purpose’ rather than ‘action’ therefore the fundamental concept then? But what if purposes are not given but chosen? Is choice then most fundamental? And are choices not based on preferences? Or are we not rather coming up with potential purposes when we are confronted with resources that thereby become means that lead us to their purposes? Surely, that happens also quite often.
Therefore, perhaps, purposes and choices are not the most fundamental categories after all. Just doing things might be enough sometimes. For instance, there are life routines where we do things because they are done without explicit purposes, and without the deliberate choice of means. The means might have no alternative uses and therefore opportunity costs are never considered. To ask your friend at breakfast “Why are you drinking coffee?” makes no sense and does not pose a legitimate claim to an answer.
In any case, even if there is a typical sequence of decision making in human action this does not imply that there is a deductive hierarchy of concepts here with fundamental vs derivative concepts. The network is rather “circular” in the sense that all terms presuppose and “require” each other and lead onto each other in operative discourses of planning and questioning. This does not necessarily prevent philosophers or social scientists or economists to systematize this network by casting it into the form of an axiomatic system, which, by the way, contra Rothbard, cannot be spun from a single axiom. Mises was aware of this as he talks about axioms in the plural form.
There is a second sense and motivation according to which we might distinguish more “fundamental” categories from less fundamental ones, namely depending how relatively wide spread or pervasive the use of the categories in question are.
Wide spread, everyday life situations and routines are queried less deeply than more rarefied and specialised situations which are drawing more categories into the discourse in which they are embedded. Thus the full panoply of societal situations and language games overlap in a subset of categories, the “fundamental” categories. Language games are situationally bounded. In any particular type of social situation only some, not all, categories or questions are pertinent. So some categories are only relatively rarely required or brought into play. They are therefore, in this sense, less fundamental.
Both explications, the explication of ‘logically fundamental’ (deductive hierarchy) as typical time sequence, and its explication as situationally wide spread, i.e. as pervasiveness, refer to simple, ascertainable (empirical) facts of language use. Such a reduction of logic to language use is likely to seem as absurd to an economist as it must have seemed to a traditional logician, because we do not seem to settle question of logic by way of observing language use. Neither do we settle simple grammar questions like this. But again, how do we settle them? By “thinking”, or by using our “intuition”. Again, such phrases need to be problematized, and then demystified. What we activate by our intuitions is our prior conditioning. De jure, reduces to de facto, but to a pragmatically corroborated de facto.
There is also a related historical dimension here that could be drawn on to motivate the ordering of the categories. In older societies most actions were institutionalized via rigid ritualization, without many moving parts that could be rearranged and repurposed. These actions were thus not guided by a discursive game of asking and giving reasons. These were not yet discursive life-forms like our modern life-form. At least to some extent therefore, some of the supposedly universal, necessary categories of Mises’ praxeology were not yet at play. That we as modern observers feel we can always back project and attribute these categories to premodern or even prehistorical societies does not mean that they were in fact at play or required. These were more primitive forms of life with a more restricted language or conceptual apparatus.
The most widespread categories are probably also historically the oldest, the deepest. We might surmise therefore that these two criteria – pervasiveness and historical depth - probably home in on roughly the same ordering of categories.
A similar point of historical evolution can be made by comparing various economies or variously sophisticated economic agents. Farmers, for instance, started very late with capital accounting inclusive of land values. Only in the context of developed financial markets does it make sense for entrepreneurial capitalists to use the concept of opportunity costs to constrain and justify their productive investments with reference to an always available risk free rate of return, and total shareholder value maximising is more involved than profit maximising, as the latter is subsumed in the former.
Categories that are more wide-spread across more language games are probably also older.
However, we should note that the categories evolve their meaning - which according to the insight of structualism is determined by their position/role within the system or network of terms and moves that make up the respective discursive practice at any particular historical stage. The meanings of individual words evolve, i.e. change, with the evolving system. This is what Marx emphasized and called dialectic progression. This insight must be heeded, and stands against the ahistorical conception of Mises’ praxeology.
The Historical Evolution of the Categories as guide to their Systematic Ordering
There is thus another sense in which the intuition of a “hierarchy” of concepts, in terms of “fundamental vs derived” or “primary vs secondary” etc., might be redeemed, namely in the sense of historical evolution.
While the currently prevalent lifeworld of language-supported human action – which in its basic pattern and primary categories is familiar to us since the culture and discourse of ancient Greece – delivers a network or system of co-dependent terms and truisms we still subscribe to, there might have been older and simpler life-worlds with their respectively simpler language games that we might try to retrace. In the evolution of socio-economic life from simpler to more complex forms, the conceptual network also evolved. New terms were added and worked into the system, thereby elaborating it into a new and different system. These elaborations affect many terms in the system, so that many terms acquire enriched and thereby indeed different meanings, even if the words themselves remain. “Meaning” should, with Saussure, always be understood as system-dependent, and with Wittgenstein, always be understood pragmatically as “use”, i.e. as use in combination with other terms, and as use within an increasingly complex practical life pattern. Therefore, language systems with different degrees of elaboration are strictly speaking incommensurable, and with this all or most terms within it are effectively non-translatable, the more so, the more the overall complexity of the respective languages and their corresponding life forms diverge. However, despite this caveat, we can draw lineages and analogies between different languages and their attendant life-worlds that allow us to see conceptual lineages, often supported by word lineages that thus allow us to establish approximate correspondences. On this basis we can start to distinguish in our current conceptual network “primary” concepts that have counterparts in much older language-life formations from “secondary” and then “tertiary” concepts that appeared later on the scene.
In Marx Capital the logical and historico-genetic derivation of the economic categories is going hand in hand. The more fundamental categories are shared by all economic systems, including the most archaic ones.
New concepts emerge at each stage when we trace the stages of socio-economic evolution from a gift economy to an economy based on regular barter, to an economy using money, but as yet without employment or wage relations, to an economy with wage relations, labour markets, capital markets etc. However, the old/primary terms change meaning/use/import as the system becomes more elaborate, and old truisms either change meaning or are substituted by new truisms. “Land” and “labour” function differently after they have been commodified. Money acquires a new meaning in a system that no longer constrains money to facilitate commodity exchange but where it can be invested via credit with interest and where it’s possible default use in risk free, interest-bearing bond investments feeds back as opportunity cost to the appraisal of all its uses. The very concept of “money”, and with it the connected concepts like “credit”, “debt”, “payment”, “savings” etc., evolves further when it becomes fiat money backed by legal tender laws, giving governments the power to monetize their debt, with the attendant risk of inflation, foreign exchange risks, forex markets etc. etc.
This genetic or historical tracing of the concepts of economic action can thus motivate their logical ordering. This genetic ordering of concepts might then also become a guide to the way a systematized logical ordering of the current network is build up and sequentially presented as a theoretical edifice, should such a comprehensive representation be attempted. This is what Karl Marx tried to do in his three volume treatise ‘Das Kapital’. The logical elaboration of his theoretical edifice roughly mirrors the genealogy of the categories. The successive elaboration and thus transformation of these concepts is what Marx calls “the dialectic”, indeed a useful epistemological reflection concept. Mises and Rothbard, to some extent, proceed in similar fashion in their respective treatises. However, their epistemology does not reflect this.
Mises illustrates the a priori character of economics via the claim that “in the concept of money all the theorems of monetary theory are already implied” . The risk here is that such statements (mis)lead us to an essentialist, ahistorical, Platonist conception. The linguistic-pragmatic conception of a priori truth, in contrast, refers us back to historically evolved and evolving categories. Money in the 12th century was different from what money has become in the 21st century. Of course in retrospect we can see that key functions and questions remain but what evolved could not have been predicted via deduction from the 12th century concept. The concept of money and monetary theory evolve with the monetary system, which itself is not wholly independent of the prevailing monetary theory.
Praxeology’s Relation with Economic Science
Mises thought that all of economic science can be encompassed within praxeology. This might have still been plausible with respect to pre-war economic treatises, i.e. in the context when Mises was writing, but the massive expansion of the discipline of economics since WWII can no longer be thus contained and characterized. Economic research and science have evolved into a field preoccupied with increasingly sophisticated mathematical modelling efforts that, however obliquely, are directed towards the explanation and prediction of increasingly abundant empirical economic data.
Mises opus magnum ‘Human Action’ was published in 1949 in English. The book was based on a reworked German-language predecessor published in 1940. At that time Mises’ account of economics as a deductive rather than empirical science was still a reasonable characterisation. In 1951 the Russian-American economist Wassily Leontief, one of the pioneers among those who made efforts to establish economics as mathematically supported empirical science, characterized the mainstream methodological approach in economics as follows: “Physics, applying the methods of inductive reasoning from quantitatively observed events, has moved on to entirely new premises. The science of economics, in contrast, remains largely a deductive system resting upon a static set of premises, most of which were familiar to Mill.”[xiv] He cites Walras, Pareto and Fisher as early pioneers of the new outlook and “effort to develop quantitative methods”[xv] that started to emerge in the post-war era. “Yet such methods have so far failed to find favour with the majority of professional economist.”[xvi] Leontief admits that the results of the new methods have so far not been “significantly superior to those achieved by the traditional procedure”[xvii]. He also recognized that this new quantitative approach, as well as his own specific method of quantitative input-output analysis, “had to await the modern high-speed computing machine as well as the present propensity of government and private agencies to accumulate mountains of data”[xviii].
Thus Ludwig von Mises’, in contrast to those of his 21st century followers who still insist today that economics must be a purely deductive science, can be excused that his attempt at a principled reconstructive rationalisation of the economists’ methods of his time do no longer capture or encompass the state of science of economics in the 21st century. Both the availability of computational power and the availability of data expanded explosively since Mises’ had published Human Action, spawning an equivalent expansion of economic science.
Mises identification of economics with praxeology had a historical empirical justification in the sense that economics, in particular marginalist economics which emerged in 1870, was still a relative immature research programme, a science in its initial formation, and as such sill focussed mainly on its internal logical elaboration. This, according to Imre Lakatos, is typical for the early stages of many research programmes. The abstract general logic of marginalist economics was developed first, building up from 1st principles and assumptions that defined the research programme. There was very little empirical work. Only gradually could more concrete complex phenomena be modelled and real economic processes be measured through data series that also became available only later. So in Mises formative years as economist, economics was indeed largely a deductive system building, as Mises’ concept of praxeology reflects.
The later development of a much more concretely applicable and applied economics does not render praxeology irrelevant, but it implies that praxeology can surely no longer claim to encompass economics. Instead it at best delivers a foundational conceptual skeleton for economic theory construction with heuristic implications for research. What must be rejected is the attempt by latter day followers of Mises to the deny the value and rationality of those parts of economics that exceed the bounds of praxeology by venturing beyond the mere systematization of fundamental a priori principles of human (economic) action. The distinction between theoretical and applied economics, albeit somewhat artificial, might help here. Praxeology is explicitly confined to pure theory, on a most general level, without engaging with empirical data. What we might call applied economics – inevitably involving empirical data, as well as particular popular preferences, expectations and institutional arrangements etc. - was classified by Mises under a separate term, namely ‘thymology’, itself seen by him as a branch of history. Here is the kind of reasoning that Mises classifies under the rubric ‘thymology’, separated out from economic science which he conceived as a branch of the a priori science of praxeology.
“We assume that, by and large, the future conduct of people will, other things being equal, not deviate without special reason from their past conduct, because we assume that what determined their past conduct will also determine their future conduct. However different we may know ourselves to be from other people, we try to guess how they will react to changes in their environment. Out of what we know about a man’s past behaviour, we construct a scheme about what we call his character. We assume that this character will not change if no special reasons interfere, and, going a step farther, we even try to foretell how definite changes in conditions will affect his reactions.”[xix]
It is here, in applied economics, which requires an a posteriori, empirical input, that economic theorizing and modelling becomes useful. What Mises has in mind here as a posteriori basis is not so much hard empirical statistics but an understanding of prevailing dispositions, preferences and motivations.
A clear example where Mises a priori praxeology goes too far with respect to what can be achieved by a supposed a priori deduction: The pure time preference theory of interest.
“Understanding does not deal with the praxeological side of human action. It refers to value judgments and the choice of ends and of means on the part of our fellow men. It refers not to the field of praxeology and economics, but to the field of history. It is a thymological category. The concept of a human character is a thymological concept. Its concrete content in each instance is derived from historical experience.”[xx]
The separation of this kind of applied economics from what Mises calls economic science seems rather too artificial and obviously stands against common usage. Why is Mises proceeding in this way? This alternate classification is motivated by Mises’ attempt to precisely delineate and separate out the a priori conceptual contributions from the contributions of a posteriori experience. This attempt to achieve self-reflective conceptual clarity in terms of a logic of science is rather admirable. Mises attempts to distil a fundamental set of underlying and systematically connected theoretical premises, pure economics, in distinction to applied economics that then gives accounts of historical economic episodes via the further introduction of supplementary hypotheses. The distinction between a stable theoretical hard core and a periphery of malleable hypotheses that become the subject of empirical testing, makes good sense, and is indeed a distinction that has been elaborated in the philosophy of science. However, this distinction cannot be set absolute, and these two parts, hard paradigmatic core and soft hypothetical periphery, are parts of a single science, known by a single name. Theoretical and experimental physics together form physics. Mises terminology with regards to economics is construed so that only purely theoretical economics is economics, as if only theoretical physics was physics. This makes little sense. The theoretical core cannot be freestanding. The point of a science is explanation and prediction but only supplementary a posteriori hypotheses, in Mises terms thymology, can fit out economic propositions enough to become sufficiently content-rich to make concrete predictions that thereby also become empirically testable. The talk of a priori versus a posteriori as a way to distinguish fundamental general theoretical premises from concretizing supplementary propositions, points to an important distinction that is always at play in science. However, this distinction cannot be set absolute. General presupposed principles and disposable hypotheses together derive testable predictions. If the test fails, it is usually the disposable hypothesis that is falsified. However, the doubt might also, or instead, spread to the supposedly a priori premises or principles. This option is rarely exercised but cannot be excluded altogether. The great breakthroughs in science have often been associated with such paradigm shifts.
The specific logical premises Mises works with are rather outdated. Despite the fundamental controversy Mises engaged in with the social science approach of Vienna Circle Logical Positivism, his approach is in line with the more general underlying logic of Logical Positivism which was the frontier of the philosophy of science at the time Mises formed his basic methodological premises. However, within the more recent philosophy of science, since Quine’s famous essay “Two Dogmas of Empiricism”[xxi], this attempt to maintain such a sharp delineation and separation is seen to be impossible, unfruitful and misleading. Quine is undermining the crucial distinction of a priori and a posteriori through its complement distinction of analytic versus synthetic. These distinctions are not rendered utterly bankrupt and useless by Quine’s insight, but the classification of scientific concepts they imply is rarely explicit, nor fixed in any given scientific discipline or even given scientific theory. Rather they are evolving. A corollary of this is what has been termed ‘epistemological holism’ or ‘sematic holism’, implying that that no theoretical statement can mean something definite in isolation or can be tested in isolation but only as embedded in a background of other hypotheses, ultimately involving our whole web-of-belief. There is an ineradicable ambiguity (and thus wiggle room) with respect to which part of this web is to be blamed and revised in the face of a failed scientific prediction or test. This state of affairs reflects the reality of science as a living, progressing practical project. This insight Quine’s is taken on here. It does not altogether invalidate praxeology but leads, together with many other general philosophical insights, to its linguistic-pragmatic re-interpretation.
The new interpretation of praxeology offered here makes clear why praxeology cannot, or can no longer, claim to encompass, comprise or delimit the totality of economics. That this is so is much clearer now, with respect to contemporary economics, than it was when Mises wrote Human Action.
If with Mises we compare the (a priori) character of praxeology to geometry, then we should extend the analogy also to the (limited) scope of these theoretical systems vis-à-vis the substantive, practice guiding sciences they support but not encompass: just as geometry underpins but not encompasses physics, so does praxeology underpin but not encompass economics.
The insight promulgated here, namely that the possibility and validity of praxeology is based on its faithful tracing and displaying the always already presupposed network of terms and truths of our basic practical (economic) lifeworld, gains a critical edge when praxeology starts to venture and cross the threshold that so far had been invisible to itself, from the tracing and reconstruction of the lifeworld’s wisdom, to the reconstruction of classical economics, and further to the elaborate constructions of its own that are no longer backed up by our basic, pre-scientific lifeworld-language.
Consider here for instance the definition of preferences as ‘demonstrated preferences’, or for instance the time preference theory of interest. What status does this theory have? It is meant to be unquestionable due to its deduction from the concept of action. At the same time it is meant to have explanatory and predictive power over all economic phenomena, especially in relation to saving and investment behaviours. How plausible is this? Or rather, how helpful is such deductive knowledge when it predicts events only in “apodictic” terms which might no longer be the terms of everyday common sense or the terms by which the more complex specialized practical economic sphere of saving/investing understands itself?
Mises and Rothbard operate with an unquestioned, naïve conception of logic and inference. For them all the steps, the tracing of basic given language games, the reconstruction of the well-established, widely shared conceptions of classical economics, and finally their own more idiosyncratic conceptual proposals are all equally conceived as “deductive inferences”. But this way of talking about deduction and inference is naïve, based on intellectually bankrupt Platonism or mentalism. What is necessary therefore to redeem praxeology is to reinterpret and reconstruct it in terms that are compatible with our contemporary philosophical horizon that has evolved via the discourse of linguistic-pragmatist philosophy who’s insight we can no longer afford to fall behind.
Although the distinction between our pre-scientific everyday practical discourse on the one hand and the scientific discourse of economics is an important one to make, it is clear that this distinction cannot not be sharply drawn and fixed because the concepts of everyday commerce are necessarily picked up by economic science. Also, some scientific elaborations, as well as new scientific terms, and the economic calculations they suggest or afford, might in due time, migrate back into everyday usage, as e.g. happened with certain asset-pricing models that were first formulated in a specialist economic science discourse.
The terms of everyday practical discourse are never quite fixed, and instead retain a certain plasticity that gives actors sufficient wiggle room to navigate and argue their cases. Within the more complex and elaborate zones of the economic lifeworld we find new “truisms”, but they seem less stable as terms and practices evolve faster here. Also, the pragmatic coping and survival advantage of innovative analytic vocabularies and techniques, and their possible scope and generalisability, needs time to show itself. Mises’ and Hayek’s insight remains an important reminder here: The complexity of the economic system does not allow for any definitive falsification of an economic doctrine, nor for any definitive arbitration between competing approaches via supposedly decisive historical experiences. Orientation via economic science remains an unending quest, to use a phrase from Karl Popper.
What would be the point of formalizing the network of fundamental economic categories into an axiomatic system?
Sciences always try to tighten their network of basic terms, and then try to build out with precision and elaborateness from this crisp base. The purpose is to discover consequences which are not immediately obvious. But how reliable are these discoveries? This tightening always runs the risk of losing nuances, blunting the subtleness of the organically evolved network. The elaboration outward has its own additional risks, especially if it tries to venture out towards unheard of levels of generality. Caution is called for here. For instance, the principles of value maximizing economic conduct that seem axiomatic within our usual business relations, might not extend into the extreme ends of the wealth/poverty scale.
The insistence on tautological truths does not help us very much in these cases when the respective, sharply defined praxeological statements might very well be upheld irrespective of the fact that their common sense understanding is breaking down in these cases.
Although “deduced” or rather constructed on a seemingly firm base, these constructions cannot maintain the very unquestionable certainty of their base axioms. They become all the more “hypothetical” the more original they are as logical constructions. They offer themselves as a potentially enhanced apparatus for the further rationalisation of the lifeworld. Whether they deliver remains to be seen, depends on the practical success of the techniques these conceptual elaborations suggest, like for instance the new forms of accounting and economic calculation these theories might sponsor. We cannot take this for granted just due to a “logically tight” sequence of inferences.
The theoretical-hermeneutic reconstruction of the lifeworld - with systematizing and then guiding intent – is to be distinguished from a theoretical explanation that elaborates its own, novel, potentially counterintuitive, alien theoretical apparatus, an apparatus that might even be incommensurable relative to the language games it describes. The theoretical-hermeneutic reconstruction in contrast must maintain strong semantic and pragmatic continuities with the lifeworld’s own self-descriptions. While scientific explanation is offered as if from a detached, outside observer, hermeneutic reconstruction is evolving the internal perspective of participants.
This distinction is ideal-typical and sets up the ends of a spectrum rather than a strict dichotomy, i.e. we can distinguish degrees of conceptual continuity here. Even the most radically original and anti-common sense scientific theory will still share a basic lifeworld horizon with its societal subjects. After all, science is a societal venture too, and like its social subject domain, relying on ordinary language too, even if it also creates its own conceptual system, perhaps even inclusive of its own unique graphic symbolism.
The same applies to the conceptual constructs and mathematical techniques of economic science. These eventually too connect up with basic lifeworld categories. However, we might presume that a social science that operates with the concept of ‘understanding’ rather than just crunching statistical dependencies – and these too must in the final analysis connect up with everyday practical concepts – must stay relatively close to the participants’ conceptions, at least as long as these remain operational in the interim, i.e. until the systematized elaborations and improvements have been disseminated into general practice. Reconstruction and reform can go hand in hand, but only in small increments, akin to all evolutionary processes.
When Mises’ and Rothbard presume that they can sit in their study room and, from a firm self-evident axiomatic base, spin outward a whole system of economic concepts and truths, they operate with a naïve, unproblematized and indeed magical concept of logic. However, within a linguistic-pragmatic philosophy, which is ultimately consistent with a materialist evolutionary world view, the traditional concept of logic has become problematic and can no longer be taken for granted unquestioned. As we have explained below, meaningful implications are woven via word concatenations that participate in weaving increasingly complex, adaptive and thereby increasingly robust and prosperous life forms.
We should remember Wittgenstein’s insight that rule following is not a trivial or certain matter. When we try to define a rule in words or writing we are presuming to be able to fix an input-output algorithm over an open ended input set. However, definitions break down or become ambiguous when we are confronted with new unexpected input conditions which make the application of the rule uncertain. This is relevant when we define general concepts and supposedly logical-analytic statements associated with them. With respect to genuinely new phenomena it might become irredeemably ambiguous which of our currently established concepts should subsume the phenomenon. The expansion of the term to the new unexpected case or set of cases expands the very “meaning” of the term. Such expansions are analogical rather than logical. This will empirically show up via a temporary divergence in usage which might lead back into a convergence towards one of the usages that emerged at the time when the novel phenomena had to be absorbed into the language. These facts of language evolution are captured in the thesis that all of language is based on analogy or metaphor. As a new usage consolidates and become routinized the sense that it started out via an originally metaphorical or analogical usage gets lost. These reflections also indicate that the attempt to cleanse philosophical logic from ‘psychologism’ or ‘behaviourism’, i.e. from empirical inference behaviour, are counterproductive.
Various concepts of logic and their application must prove themselves in their respective practical domains. The pragmatist stance loops back into the frontline of our discourse again and again. The pragmatic consequences of each step in the build-up of a logic probe this logic even as the logic itself must be employed in the probing. The pragmatist stance fuels the evolution of a discourse practice, not via deduction, but via mutation, selection and eventual (temporary) retention.
[i] Jürgen Habermas, Moral Consciousness and Communicative Action, Wiley, p. 9
[ii] Jürgen Habermas, Moral Consciousness and Communicative Action, p. 19
[iii] Paul Churchland
[iv] Jürgen Habermas, Truth and Justification, orig. German 1999, Polity Press, Cambridge 2003, p.137
[v] Murray N. Rothbard, Toward a Reconstruction of Utility and Welfare Economics (1956), reprinted in: Murray Rothbard, Economic Controversies, Ludwig von Mises Institute, Auburn 2011
[vi] Jürgen Habermas, Knowledge and Human Interests, orig. German 1968, English 1972, Polity Press, Cambridge 1987, p. 308
[vii] Jürgen Habermas, On the Pragmatics of Communication, Polity Press, Cambridge 1999, p.35
[viii] Jürgen Habermas, On the Pragmatics of Communication, Polity Press, Cambridge 1999, p.34
[ix] Jürgen Habermas, On the Pragmatics of Communication, Polity Press, Cambridge 1999, p.34-35
[x] Jürgen Habermas, On the Pragmatics of Communication, Polity Press, Cambridge 1999, p.43
[xi] Timothy Williamson, The Philosophy of Philosophy, Blackwell Publishing, Oxford 2007, p. 3
[xii] Herbert A. Simon & William G. Chase, Skill in Chess, published by The Society of the Sigma Xi, 1973
[xiii] Albert Einstein, Letter to Dr. H. L. Gordon (May 3, 1949 - AEA 58-217) as quoted in Einstein: His Life and Universe (2007) by Walter Isaacson.
[xiv] Wassily Leontief, Input-Output Economics, 2nd edition, Oxford University Press, Oxford 1986, p.3
[xv] Wassily Leontief, Input-Output Economics, 2nd edition, Oxford University Press, Oxford 1986, p.3
[xvi] Wassily Leontief, Input-Output Economics, 2nd edition, Oxford University Press, Oxford 1986, p.3
[xvii] Wassily Leontief, Input-Output Economics, 2nd edition, Oxford University Press, Oxford 1986, p.3
[xviii] Wassily Leontief, Input-Output Economics, 2nd edition, Oxford University Press, Oxford 1986, p.4
[xix] Luwig von Mises, The Ultimate Foundation of Economic Science, Ludwig von Mises Institute, 1962
[xx] Luwig von Mises, The Ultimate Foundation of Economic Science, Ludwig von Mises Institute, 1962
[xxi] Willard Van Orman Quine, Two Dogmas of Empiricism. The Philosophical Review, 60 (1951), pp. 20–43