The Social Sciences and the A Priori

The Austrian school of economics exhibits a unique methodological feature that sharply contrasts with other schools in the social sciences: methodological apriorism. Praxeology, the science of human action as defined by Ludwig von Mises (1998, 1962), is an axiomatic corpus of knowledge that is obtained by a priori reflection instead of empirical investigation; its consequences follow logically by deduction. Mises (1998, 64) says, “The scope of praxeology is the explication of the category of human action. All that is needed for the deduction of all praxeological theorems is knowledge of the essence of human action. It is a knowledge that is our own because we are men.”

Two aspects of methodological apriorism are striking: the axioms are gained a priori, and praxeological knowledge follows from them only deductively. Some have interpreted these aspects as implying that economics is, according to Mises, a completely a priori science without any room for empirical testability or world-grounded auxiliary hypotheses. This sharply contrasts with the standards and ideals for scientific knowledge defended by the dominating neopositivist philosophy and methodology of science since the 1940s. According to this mainstream and following the standards that made natural sciences so successful, scientific knowledge ought to be guided by empirical evidence and testability. This means that the conditions for the justification of social scientific knowledge ought to be a posteriori, following the paradigm of the natural sciences. This has cast doubts on the scientific value of Austrian economics for assessing and explaining economic and social phenomena.

This paper argues that social sciences based on methodological apriorism have scientific value despite the numerous criticisms. We show this in the particular case of Austrian economics, a social science with explanatory value and a priori foundations in a substantial sense. Our argument relies on what we call “normative apriorism,” following the Pittsburgh school of philosophy (Sellars 1962; McDowell 1994; Brandom 1994). The basic idea involves the normative nature of scientific knowledge when it concerns social phenomena—that is, phenomena brought about by the complex interactions of persons. Such normative knowledge is the bedrock of the social sciences, Austrian economics among them, and is a priori in the sense that it is a knowledge that must be presupposed to conceive social phenomena as such and as distinct from natural phenomena. We thus emphasize the importance of transcendental arguments, in the Kantian sense, for the social sciences. In conclusion, we hold that normative apriorism can offer the epistemological grounds to defend the claim that the social sciences can be scientific and a priori.

PROBLEM

In his classical 1953 book Essays in Positive Economics, Milton Friedman rescues John Neville Keynes’s (1891, 34–35) distinction among a positive science (which concerns what is), a normative or regulative science (which concerns what ought to be), and an art (which concerns ends). According to Friedman, following the elder Keynes, it is of the utmost importance not to confuse these different kinds of knowledge and to recognize that (political) economics belongs to the positive sciences. Beyond the distinction between normative and positive economics, to which we will come back shortly, the epistemologically relevant aspect of the very idea of a positive economics is that there exists a corpus of systematic knowledge concerning social phenomena (economic phenomena) that shares a basis with the natural sciences.

But what is positive economics? As Friedman (1953, 146) describes it, it concerns what is, independently of any ethical or moral judgment or policy. Its ultimate goal is “the development of a ‘theory’ or ‘hypothesis’ that yields valid and meaningful (i.e., not truistic) predictions about phenomena not yet observed. Such a theory is, in general, a complex intermixture of two elements. In part, it is a ‘language’ designed to promote ‘systematic and organized methods of reasoning.’ In part, it is a body of substantive hypotheses designed to abstract essential features of complex reality.”

Hence, the main task of positive economics is to provide a system of generalizations that generate predictions and that enter as premises into scientific explanations. More importantly, the standard for judging their quality is experience. Friedman (1953, 146) says, “[A system of knowledge’s] performance is to be judged by the precision, scope, and conformity with experience of the prediction it yields. In short, positive economics is, or can be, an ‘objective’ science, in precisely the same sense as any of the physical sciences.” Although it is acknowledged that the complexities and uncertainties of economics are much greater because of the human element, the difference from the natural sciences is one of degree and not fundamental.

Contrary to this, normative economics chiefly concerns what ought to be done—that is, policies implemented by state agents to improve the well-being of the population. Normative economics, it is argued, should guide policymakers to implement policies on well-grounded positive knowledge delivered by the explanations and predictions of positive economics. While positive economics can be objective when pursuing empirically grounded knowledge, normative economics is not objective since it concerns ethical statements.

Bryan Caplan (1999) correctly points out that positive economics is the epistemological foundation of neoclassical economics, which is one of the paradigms in contemporary economics. The philosophical context in which neoclassical economics emerged explains this very well. The philosophy of science environment of the 1940s and 1950s was dominated by some elements of the so-called neopositivist philosophy of science and some elements of Karl Popper’s philosophy of science (in particular, falsificationism). This is not the place to dig into the details of this complex and well-studied current (Uebel 2007; Creath 2022), but three pillars are worth mentioning:

• the analytic-synthetic distinction (logical positivism)

• empirical verificationism

• Popperian falsificationism

According to the first, logical positivism had thoroughly adopted the analytic-synthetic distinction. It implies that any cognitively meaningful statement must be either analytic (definitionally true but empirically empty) or synthetic (contingently true but empirically tested). This implies that if the social sciences, and economics in particular, deliver scientific knowledge that is not definitionally true but empirically meaningful, then it must be a posteriori and not a priori. Empirical verificationism, in its Carnapian version (Carnap 1928), states that the possibility of empirical verification is a criterion of the meaningfulness of empirical propositions. Thus, if the social sciences deliver empirical propositions that are meaningful, then they can be verified through observation. Finally, Popperian falsificationism (Popper 1959) establishes that scientific hypotheses can only be held provisionally and must be submitted to steady empirical testability, which rejects untestable a priori truths from the scientific enterprise.

All these ideas were very strong in the epistemological environment of the middle of the twentieth century and had visible influences in the different sciences, natural and social. It is, then, easy to see why the Austrian school of economics in particular was marginalized by the main currents of thought in economics. The issue is largely methodological: if the Austrian school implies methodological apriorism and this is deemed unscientific, then the Austrian school is unscientific and, thus, untenable. In other words, Austrian economics was resisted and eventually marginalized because it failed to meet almost every scientific criterion of the prevailing epistemological paradigm. Mark Blaug (1980, 81), for instance, says that “in the 1920s, Mises made important contributions . . . but his later writings on the foundations of economic science are so idiosyncratic and dogmatically stated that we can only wonder that they have been taken seriously by anyone.” Indeed, methodological apriorism has been criticized on the basis of its being unintelligible, dogmatic, and unscientific (see Caldwell 1984 and also Caplan 1999; for a reply and a defense of Austrian methodology, see Hülsmann 1999). The scientific character of the Austrian school seems to be threatened by the lack of room for empirical testability. Nonetheless, the argument is not only against Austrian economics but is more general: any social science that claims to endorse methodological apriorism is to be rejected as unscientific. The moral, then, is that methodological apriorism is incompatible with the social sciences qua sciences tout court.

Contrary to all this, Mises (1998, 34) very clearly stated that “the fundamental logical relations are not subject to proof or disproof. Every attempt to prove them must presuppose their validity. . . . They are ultimate unanalyzable categories.” And these relations, according to him, underlie not only Austrian economics but any science of human action more generally. Murray Rothbard (1957), in turn, defended an extreme form of apriorism. Even though neoclassical economics emerged from the marginalist revolution, as did Austrian economics, it followed the positivist, prediction-oriented way of doing science, allowing and promoting mathematical models, idealizations, empirical testability, atemporal analysis of economic processes, etc. (see Hutchison 1938; Samuelson 1947; Friedman 1953).

So, this is the epistemological problem that the social sciences face when seen as fundamentally a priori, as many notable libertarians in the Austrian school tradition have argued. We show the argument against Austrian economics, but it can be generalized for any a priori social science:

1. Austrian economics implies methodological apriorism.

2. The hegemonic neopositivist and Popperian philosophy of science implies that methodological apriorism is untenable (false).

3. The hegemonic neopositivist and Popperian philosophy of science is the correct epistemological attitude toward science.

4. Methodological apriorism, then, is untenable (follows from 3 and 2).

5. Therefore, Austrian economics is untenable (follows from 4 and 1).

SOLUTIONS

The debate between different epistemological paradigms in economics—Austrian economics on the one side and neoclassical economics on the other—is thus primarily methodological. Austrian economists face a methodological problem when confronted with the neoclassical paradigm and the prevailing philosophy of science. Hence, the success of Austrian economics mostly depends on how the epistemological problem mentioned above is solved. Defenders of the Austrian school of economics can take at least three different paths to alleviate this tension. Each path rejects at least one of the premises in the methodological problem. One path is to deflate (or simply reject) the link that goes from Austrian economics to methodological apriorism (premise 1). The second one is to reject the neopositivist and Popperian philosophy of science as the right epistemological background for judging what is science and what is not (premise 3). The last one is to offer a more sophisticated reading of the a priori, which reinterprets the problem differently. Let us go down each of these paths.

The first solution states that Austrian economics does not entail methodological apriorism. This is the path followed by Friedrich von Hayek, who rejects, for instance, that entrepreneurs’ knowledge can be given a priori. Agents’ knowledge within markets is neither a priori nor perfect; agents need to learn from mistakes and under varying circumstances; therefore, many principles of human behavior cannot be known a priori, because they are part of the learning process in markets (Hayek 1948, chaps. 2, 4). This opens up a non-Misesian way to do Austrian economics, which is, at least methodologically, a bit closer to neoclassical economics. Thus, even though methodological apriorism can indeed be deemed as unscientific and untenable, this does not affect Austrian economics, which does not necessarily imply methodological apriorism.

This solution does not succeed, because it misses the point of the criticism. Hayek is quite right in pointing out that within markets, agents only have partial knowledge, and then, in an important sense, there are regularities and patterns that they discover and learn over time. But the point here is whether there is a priori knowledge not of the regularities and patterns that are discovered in markets but of the epistemic foundations for having such knowledge. Indeed, Hayek himself seems to presuppose that such a priori knowledge is necessary to identify regularities and patterns as genuinely social (that is, human-related) rather than natural.

The second path consists in rejecting the naïve philosophy of science that logical positivism, verificationist empiricism, and Popperian falsificationism have established. Indeed, the so-called new philosophy of science, pioneered by Thomas Kuhn, Imre Lakatos, and Paul Feyerabend (Kuhn 1962; Lakatos 1978; Feyerabend 1975), among others, moved away from logical positivism and strict falsificationism by rejecting the analytic-synthetic distinction, verificationism, and the idea that hypotheses can be empirically tested (and thereby rejected) in isolation. This would not only offer a more realistic view of how actual sciences work but also lay out a less hostile epistemological environment for Austrian economics and the social sciences in general. Indeed, some have attempted to view Austrian methodology (and methodological apriorism, in particular) under the light of Lakatos’s philosophy of science, which would allegedly be less hostile to a priori statements (see Machlup 1955; Zanotti and Cachanosky 2015).

This is of course correct, but it also misses the point and overlooks what is special about the social sciences and Austrian economics in particular. It can be argued that all sciences, natural and social, involve some a priori statements that are rarely, if ever, rejected. Even particle physics is guided by some principles that are not tested empirically but guide empirical research. They can well be regarded also as a priori in a relevant sense. This is very much true, but we believe that the a priori elements in the natural sciences are of a different nature with respect to the a priori elements in the social sciences, as we will argue in the next section. And this reply to the epistemological problem is insufficient to highlight them.

Finally, the last path is to offer a more sophisticated reading of the a priori. Certainly, the positivist analytic-synthetic distinction is at risk of delivering a naïve concept of a priori that can, and must, be revised. Murray Rothbard, for instance, maintains that principles of praxeology are “broadly empirical” (1957), even though he defends “extreme apriorism” in the vein of Mises. Under an Aristotelian and neo-Thomist framework, Rothbard argues that the right reading of Mises’s apriorism is that praxeological principles reflect “laws of reality” based on common human experience and that have become self-evident (obvious). And, in that specific sense, praxeological axioms can be said to be a priori. An alternative way is to debunk the analytic-synthetic distinction and pave the way to synthetic a priori statements in a neo-Kantian framework. According to some, this is Mises’s own path.

It is true that the analytic-synthetic distinction conveys a naïve concept of the a priori, but an Aristotelian and neo-Thomist framework, like that defended by Rothbard, does not help to clarify it. For instance, the scope and limits of his proposal are unclear. For instance, Aristotelianism is at odds with the contemporary view of nature according to the natural sciences—that is, disenchanted nature. Is Aristotelianism also supposed to apply there? We think that it is important to preserve the scientific view of nature as “disenchanted” and to accommodate the social sciences without rejecting it. We do not see how an Aristotelian and neo-Thomist framework can help here. Also, the ideas of “broadly empirical” and “laws of reality” are obscure. For instance, the foundations of any scientific law are the observed regularities, but these regularities must have different grounds, whether they are social or natural; Aristotelianism does not help to draw such a distinction either. And to regard broadly empirical as a synonym of a priori is, at best, a categorical mistake.

In what follows, we propose a fourth path: normative apriorism. It can be regarded as a subvariety of the third path that stresses the distinction between descriptive and normative statements but starts from modern science.

NORMATIVE APRIORISM

By proposing normative apriorism, we aim to supply social scientists, and Austrian economists in particular, with a more solid epistemological and methodological foundation to face the epistemological problem mentioned in the previous section. Normative apriorism may then serve as the foundation of methodological apriorism. So, in the case of economics, it assumes that Austrian economics entails methodological apriorism (premise 1 is accepted). It also rejects the basis of the neopositivist and Popperian philosophy of science but by rejecting the analytic-synthetic distinction. The proposal mainly relies on a sophisticated reading of the a priori that rejects both the Hayekian deflationary and the Rothbardian-Aristotelian views. It builds on the Kantian-Misesian approach and relies on the normative turn that the Pittsburgh school of philosophy promoted (Sellars 1962; McDowell 1994; Brandom 1994).

First of all, it is important to disentangle some concepts that have been historically associated. The a priori–a posteriori distinction, on the one hand, mainly concerns whether some propositions can be known independently of experience. Hence, it mainly concerns an epistemic distinction with respect to the source of knowledge. The analytic-synthetic distinction, on the other hand, is a semantic distinction to distinguish between propositions that are uniquely true by virtue of words’ meanings and propositions that are true because they relate, somehow, to reality. Immanuel Kant, in his Critique of Pure Reason (1781), famously argued that it is possible to get the four combinations, reserving a special place for synthetic a priori propositions. This is a first step to dismantling the strict positivist analytic-synthetic distinction—it is possible to have a priori propositions that are neither empirically empty nor trivial. Therefore, normative apriorism rejects one of the pillars of the neopositivist philosophy of science on which neoclassical economics and positivist economics rely.

There is, however, another distinction that is important to draw. It has been assumed that a priori propositions also hold true necessarily, while a posteriori propositions are only true contingently. Yet, as Saul Kripke noted (1981), while the a priori–a posteriori distinction is epistemic, the necessary-contingent distinction is metaphysical. Kripke ingeniously showed how a priori and a posteriori propositions can be either necessarily or contingently true. Scientific discoveries, for instance, deliver a posteriori necessary truths, while definitional conventions are a priori contingent truths. These distinctions open widely the realm of the types of a priori propositions for social scientists and Austrian economists to offer a more sophisticated reading of the a priori:

1. Contingent propositions known a priori

2. Necessary propositions known a priori, which in turn could be

   (a) analytic

   (b) synthetic

Which kind of a priori propositions do the axioms of Austrian economics refer to? Our view is that when Austrian scholars consider the axioms (or theorems) of praxeology to be a priori, they use the term in its three senses. For instance, the fundamental axiom (the axiom of action) is a synthetic necessary proposition that is known a priori in the sense that it gives the normative conditions to identify phenomena as the subject matter of praxeology. However, some of its derivations are analytic because they follow from the definition of action.

The second point on which normative apriorism relies is a sharp distinction between descriptive and normative statements. This distinction concerns the sort of relation that is established with the subject matter. Natural sciences need to deprive their subject matter of all traits of subjectivity. This is what John McDowell (1994, lesson 4) calls “the disenchanted conception of nature,” which has made natural science possible as an objective science. This is also the idea behind Friedman’s view of positivist economics and its objectivity. The idea is that natural sciences qua descriptive sciences work on the presupposition that their subject matter is composed of events happening according to some regularities, subjected only to external laws of nature and cause–effect relations. The idea of objectivity, then, works as a heuristic principle that guides scientific research by depriving nature of any subjective-normative traits as much as possible. Knowledge is objective in this descriptive, disenchanted way since it aims to reach some regulative ideal of objectivity; this is what allows natural scientists to identify scientific phenomena as such.

This does not hold true for the social sciences (or the humanities), however. While the subject matter of natural-descriptive sciences is objective, disenchanted nature, the subject matter of the social sciences (or humanities in general) is inherently subjective. Human phenomena are meaningful and purposeful, and the results of persons’ actions cannot be understood independently of their motives and aims—that is, they are normative. This concept might cause some confusion, so it is important to clarify it. In general, normativity is associated with ethical judgments in which it is prescribed that one ought to do such and such. However, normativity comprehends a more varied class of subjective-evaluative phenomena in which intentions, epistemic criteria of correctness, purposeful behavior, moral and ethical rules, etc., are also taken into account. For instance, when it is argued that logic is normative, it means that if individuals want to reason correctly, they ought to follow some rules; otherwise, they are not reasoning correctly. When referring to epistemic norms, we refer to rules that govern what we ought to do or think to form true belief and avoid error.

When we say that the basis of the social sciences (economics among them) is normative, we mean that the basis comprehends a realm of subjective-evaluative phenomena where rule-following actions are essential to capture the phenomenon as genuinely social. When it is said, from a normative stance, that individuals act purposefully, it is not meant that individuals act following their instincts or following some inner mechanic process; it is meant that individuals follow their reason and can form in themselves aims, goals, and rules and act accordingly; they act from their reason and free will, which is the origin of any normative claim in ethics, semantics, epistemology, or the social sciences.^[1]

And this stance is what distinguishes the social and the natural sciences: the relationship between social scientists and their subject matter cannot be the same as it is for natural scientists, because these two kinds of subject matter are radically different in nature. This suggests a sort of methodological dualism since there are at least two irreducible ways to do science. Normative apriorism, we submit, lies at the core of the social sciences, including praxeology, because it makes clear that the subject matter is of a radically different nature: it is human action in its normative aspect—that is, the rules that individuals follow in action and thought, which cannot be grasped externally by mere observation. This opens a schism between nature as investigated by natural science, on the one hand, and norms, on the other. That’s why two epistemological paradigms are confronted: those that want economics to be objective presuppose that economics is about descriptions of regularities that must be deprived of any subjective-evaluative traits—that is, regularities that do not involve individuals’ purposeful actions as such; and those who reject this paradigm (as Austrian economists do) and presuppose that there is an irreducible normative-subjective basis that is the primary matter of investigation of economics.

The Pittsburgh school and the normative turn in philosophy have exploited this schism to resist the naturalization of several fields in philosophy (morality, language, aesthetics, social knowledge, etc.). By naturalization, we refer to the attitude that genuine knowledge must imitate the aims, methods, and principles of the natural sciences. Praxeology is, in this sense, deeply antinaturalist. And the exploitation of the normative-natural schism can also be of some help to it. To begin, social relations do not merely display a series of regularities that can be grasped externally but are normative in the sense that they convey aims, purposes, meanings, criteria of correctness, etc. But this implies that these relations can be neither grasped nor identified as social-normative relations from outside but only from within—that is, praxeological knowledge in this normative construal requires that social phenomena can only be grasped and identified from a normative relation with the subject matter. This is the basis of Hayek’s (1952, 27) example of the difference between an inert rock and an artifact made of stone: without assuming, from within, a normative relation that implies meaning, aims, and purposes and that it is possible to understand them, the distinction would be impossible and, with it, the social sciences; for instance, archaeology would then collapse into geology. So, if we want to understand the phenomenon of toolmaking, for example, we need to presuppose some normative basis; otherwise, the social phenomenon as such is unintelligible.

The nature-norms schism also introduces another element. Social phenomena as normative phenomena do not have their origin in causal laws, randomness, or miracles. To understand social phenomena as normative phenomena implies that they rather have their origins in the actions of free and rational persons. Kant (2002, 85) writes in the Prolegomena to Any Future Metaphysics, “If an appearance is given to us, we are still completely free as to how we want to judge things from it.”

In other words, we are free to position ourselves with respect to everything that is given to our minds, be they sense impressions, desires, or needs. That is why we are free in thinking and acting. Freedom of thought and freedom of the will are one and the same freedom: if a person has the freedom to form judgments about the world, then she also has the freedom to position herself with respect to her own stance in the world; this is the freedom to form goals and intentions to act. Intentions presuppose judgments and thus concepts. They are not mere reactions to desires. Intentions are judgments about how the world should be, combined with attempts to bring about the corresponding changes in the world through physical movements.

Natural sciences must deny that the origin of a regularity they describe is some agent’s will. McDowell (1995, § 3) shows the contrast by asking us to imagine what it would take for a wolf to think and act rationally:

A rational wolf would be able to let his mind roam over possibilities of behaviour other than what comes naturally to wolves. . . . [This] reflects a deep connection between reason and freedom: we cannot make sense of a creature’s acquiring reason unless it has genuinely alternative possibilities of action, over which its thought can play. . . . A possessor of logos cannot be just a knower, but must be an agent too; and we cannot make sense of logos as manifesting itself in agency without seeing it as selecting between options, rather than simply going along with what is going to happen anyway. This is to represent freedom of action as inextricably connected with a freedom that is essential to conceptual thought.

McDowell’s example shows two things: First, descriptions about wolves as a species (a scientific description) require depriving wolves of any rationality in any normative knowledge they may have—that is, “what comes naturally to wolves,” which is informed by biology. Second, such descriptions, however, fail to capture the social-normative phenomena that rational agents produce and engage in. These phenomena are founded on the assumption that agents can act rationally and freely and behave differently from what comes naturally to them. To put it differently, scientific descriptions of wolves cannot justify anything about what rational wolves do. To understand what rational wolves do (or what persons do), it is necessary to establish a normative relation to the subject matter, a relation that recognizes the normative-subjective nature of the phenomena produced by rational and free agents.

Indeed, freedom, reason, and normativity form an inseparable trio. For thoughts and actions—and only for these—one can demand and cite reasons. Events in nature happen according to certain regularities (laws of nature). One cannot demand reasons for them. The same applies to the behavior of animals: if a cat plays with a mouse before killing and eating it, one cannot demand reasons and thus accountability for the cat’s behavior. By contrast, humans can position themselves with respect to their sense impressions, desires, needs, instincts, etc.; they are therefore free in their thoughts and actions and hence accessible to reasons and thus accountability. Normativity entails that if a person positions herself with respect to her sense impressions, desires, needs, instincts, etc., then the question arises concerning what the person ought to think and do—that is, what is right in thinking and acting and what is not.

It is here where normative apriorism lies. This normative knowledge cannot be obtained by experience; it is not a conclusion from having observed regularities in social phenomena. It is an assumption that must be subscribed to in order to be able to investigate social-normative phenomena. In philosophy, these arguments are called “transcendental arguments.” Unlike arguments that refer to transcendent objects that are supposed to exist beyond space and time, a transcendental argument refers back to the subject itself—namely, to the conditions of the possibility for the person to recognize objects in the world and to change the world through actions. These are conditions whose negation would consist in a performative contradiction: the performance—the assertion—of the negation of these conditions would itself be an act that makes use of them.

Descartes famously sets out a transcendental argument in the Meditations on First Philosophy: it is impossible to deny that I think; for if I deny that I think, then the performance (act) of denying is itself an instance of thinking. It follows that I exist as long as I think. This is a transcendental argument that establishes the existence of something—one’s thinking—without depending on empirical observations and without being able to be confirmed or refuted by empirical investigations. A science that works with transcendental arguments is therefore not empirical but an a priori science. Its propositions cannot be verified or falsified by empirical methods. However, logical errors—errors of reasoning—can occur in the formulation of a transcendental argument.

In the same sense, freedom and normativity in thinking cannot be denied without committing a performative contradiction. The act of contesting would itself be an act that claims freedom in forming a judgment (instead of being, for instance, mere noisy sounds) and that makes a claim to validity. The same applies to actions: one cannot deny freedom in forming intentions to act without the act of denial itself being an exercise of that freedom. This act is an action—at least a speech action—for which it makes sense to demand reasons. Consequently, it is not mere behavior but the exercise of the person’s freedom to position herself in relation to sense impressions, needs, desires, etc.

Hence, apart from the natural sciences, there is another kind of science that is built upon transcendental arguments and that can grasp thoughts and intentions to act. Social sciences relying on a normative-subjective basis require a transcendental argument to comprehend social phenomena. In turn, this strengthens the previously mentioned methodological dualism: there is a kind of knowledge that methodologically requires a priori knowledge of normativity. In contrast, there is a different kind of knowledge that methodologically requires depriving the world of normative-subjective traits. This difference is significant since it concerns the foundations of what makes social and natural sciences possible. Praxeology is a genuine science in the former sense, and its foundations are the a priori normative nature of its subject matter—that is, persons acting and thereby adopting normative attitudes toward each other.

To make this point stronger, it is important to stress why the concept of a person (or even human action) cannot be an empirical description of behavior but is rather the expression of a normative attitude toward the being in question. If social relations are relations in which persons engage, then we must have such a normative attitude toward them. Wilfrid Sellars (1962, § 7) aptly writes, “To think of a featherless biped as a person is to think of it as a being with which one is bound up in a network of rights and duties. From this point of view, the irreducibility of the personal is the irreducibility of the ‘ought’ to the ‘is.’ . . . To think thoughts of this kind is not to classify or explain but to rehearse an intention.”

A person, or an acting human, can only be recognized and known on the basis of normativity. Against this background, normative apriorism is about the fundamentals of social knowledge that mainly concerns the interactions of persons. These fundamentals, to repeat, cannot be known empirically; they are not anything humans come to learn by engaging in social relations. They are rather a priori, transcendental assumptions that are necessary to have knowledge of social relations—knowledge that is normative in essence. These assumptions are a priori because they are acquired independently of experience.

All this takes us back to the epistemological foundations of the social sciences in general and Austrian economics in particular. It is not unscientific to assert that knowledge of social relations can be obtained a priori. This is indeed the basis of social sciences under normative apriorism. It is, then, a mistake to require the social sciences to work under the assumptions of natural science, because their foundations are different. Neoclassical economics, by assuming the epistemology of positivist economics, goes astray—it cannot identify normative-subjective relationships (as economic relations are) on the basis of pure experience or by merely observing regularities from outside. Only by acknowledging that normativity is different from disenchanted nature and taking normative attitudes toward society as its subject matter can social sciences discover the normative-subjective nature of social relations. The social sciences can make significant progress on an a priori basis because their subject matter—human action—is normative and not merely descriptive. The fundamental axiom of praxeology, for instance, that human beings act purposefully—that is, that they “engage in conscious action towards chosen goals” (Rothbard 1957)—can only be a priori because it is necessary to consider genuine human action and not merely blind mechanical reactions or random behavior. The fundamental axioms and some of their consequences are genuine synthetic a priori propositions, very much in a Kantian vein. They can only be obtained a priori, but they inform us about persons and refer to how persons think and act in the actual world. Nonetheless, contrary to Kant’s synthetic a priori propositions, they are not formal conditions for having knowledge of objects but normative conditions for identifying social phenomena as different from natural phenomena. They refer, from within, to the normative nature of human behavior. This does not imply that everything that can be known of society and persons is a priori (something that Mises [1962, 68] himself acknowledged) but that their foundations are. In this sense, the social sciences must be a priori if they seek to refer to social phenomena at all.

As we argued in this paper, apriorism with respect to the social sciences is linked with normativity. Indeed, the path from Mises to Rothbard in Austrian economics takes a normative turn. Rothbard (1982, xlvi) places Mises’s praxeology within the framework of a normative theory of freedom. He clearly states this normative turn in the preface to The Ethics of Liberty: “I at no time believed that value-free analysis or economics or utilitarianism (the standard social philosophy of economics) can ever suffice to establish the case for liberty. Economics can help supply much of the data for a libertarian position, but it cannot establish that political philosophy itself. Political judgments are necessarily value judgments, political philosophy is therefore necessarily ethical, and hence a positive ethical system must be set forth to establish the case for individual liberty.”

However, we still have to go one step further. Rothbard tied apriorism to an Aristotelian and neo-Thomist framework instead of to the Kantian one. As we argued previously, Aristotelianism cannot shed light on the nature of the a priori in the social sciences. In our view, the case for individual liberty is in the last resort established by a transcendental argument that shows that freedom cannot be denied without committing a performative contradiction. Indeed, following Rothbard’s normative turn, Hans-Hermann Hoppe (1989, chap. 7) provides the link to the transcendental argument for freedom and emphasizes this link in his most recent publications (notably, Hoppe 2023). Our argument is that this is the appropriate setting to vindicate Austrian economics as a science based on normative apriorism.