Reprinted from Mises Brazil
Armed with this model, Richard von Mises argues for the unity of method (Ch. 17). Instead of criticizing the methodological dualism of his elder brother, he prefers to expose and criticize the dualism in Dilthey and Rickert, whose views were dominant in Germany. According to his concept of dualism, the natural sciences strives for simplification in order to generalize knowledge. Where as, the social sciences seeks an understanding particular to the study subjects. The first, thereby intends to explain everything in terms of an atomistic physics. While the second, using more realism, rejects the reduction of mental phenomena to mechanistic explanations. Richard dismisses these differences, noting that the population dynamics in Malthus would be a case of social knowledge generalized by science, and that a natural science such as geology is also interested in unique phenomenon of our planet’s history. Moreover, the distinction between mental and physical phenomena would not generate significant methodological differences, since we learn about mental states from sensory observation, reading, listening or observing the actions of others. Finally, physics could not be reduced to an outdated atomism.
The author minimizes the importance of other differences, such as the impossibility of controlled experiments (in astronomy for example, one cannot control celestial bodies) or the difference between positive and normative spheres (engineering would be a form of normative physics). Only one difference concerns the author: the existence of propositions influenced by subjective views within the social sciences. For a positivist of course, this would rob the scientific character of a theory, since such a pre-popperian doctrine would still believe that good science only begins with objective sensorial observations. Recognizing certain particularities of each discipline, Mises (p. 213) reaffirms his view that in any area of study the scientific method should be the same, a progression from observation to theoretical generalization.
Convinced about the validity of that thesis, Richard examines the area of ??expertise of his brother: the social sciences. There, we can note the contrast between the admiration the author has for the sociology of Auguste Comte and the skepticism with which he evaluates economic theory, ancient and modern. In praising the old project of building a positivist social science from an empiricist methodology, Mises ventures the guess (p. 258) that it would be in the area of sociology, where we would see the greatest scientific progress in the future.
As for the economy, Richard compares the performance of classical mechanics with certain economic theories. Such as, what he calls “The theory of spontaneity of the economy and the doctrine of the necessary and sufficient function of egotism” (p. 249). Although classical mechanics still explains much of the physical phenomena, the second, although advocated by many authors, has not been proven through observation. Furthermore, the author argues (p. 249, 364) for the simultaneous existence of competing theories and the existence of economic problems in societies as examples of weaknesses of the discipline. The diagnosis of this situation is again attributed to the use of inaccurate language, which in turn is associated to the lack of empirical observations (p. 250).
The empiricism suggested slips into the thesis of the historicist since: “… all theory is but a systematic description of earlier observations, while experience demonstrates that with the course of economic and technological development all premises, including the psychological dispositions of men, change” (p. 364). Therefore, theoretical results such as those suggested by economists would not be sustainable. The author believes that the existence of the army or public justice, which would exclude the pursuit of private profits refutes the axiom “that only ‘sound egotism’ provides a useful incentive for actions that are desirable in the interest of all…” (p. 364).
In assessing the modern theory, he believes that the main problem exists primarily in the demand (ie, how individuals evaluate specific types of goods) and that the answer to this question of the concept of marginal utility would be a vague statement about reality (p. 251), and would not be adequate to explain the economic problem in its entirety.
Continuing on the modern theory, Richard von Mises attributes to economists the belief in an objective scale of measurement, with values established by a system of prices. Böhm-Bawerk in particular, is seen as an author proud of his ability to compare current and future assets. According to the author, this would not be valid if the technology or the institutions changed (p. 257).
As to the question of mathematics in economics, although skeptical of its potential himself, Richard believes that it’s not even worth evaluating the opinion of those who think such a method would be contrary to the essence of the discipline (his brother?). Because this position would imply that the sum or subtraction of prices could be illegitimate (p. 251).
These naive observations reveal a lack of familiarity with the economic theory. Which lead us to ask, whether Richard actually read the economic writings of his older brother, whose contributions are never actually mentioned. Furthermore, this lack of direct reference does not make the contrast between their views less clear. Let’s consider then, the views of Ludwig von Mises.
The opinions of Ludwig on the methodology of the social sciences, might rather be listed as another chapter in a long tradition of methodological defense of economics from the attacks made by empiricist critics. In fact, economics has always dealt with these criticisms, made continuously ??since the nineteenth century by the German historical school, until the rise of econophysics of the twenty-first century.
One of the main authors in this tradition of defending the economic theory is none other than John Stuart Mill, one of the patron saints of empiricism (whom was also an economist). Mill opposes the empiricist methodology, subsequently defended by this author, as being appropriate for the physical sciences, whereas the a priorimethod would be appropriate for the social sciences. The first would not apply within the latter camp because of the impossibility of conducting controlled experiments, coupled with the large number of factors that can affect human behavior. Ultimately, the crucial difference lies in the greater complexity of the social phenomena. Aside from the fact that astronomy cannot conduct controlled experiments, it is possible to construct models with few variables that generate highly accurate predictions. Whereas the simplest of economic models can involve a multitude of variables and disturbing causes that prevent isolating phenomena satisfactorly.
For Mill, since the a posteriori way is blocked, economics could proceed in the same way as geometry, starting from true fundamental assumptions about human behavior and deducing from them valid results. The author, in this case, starts from the assumption of Homo economicus – the impetus for the accumulation of material wealth – without denying absolutely the existence of other factors operating simultaneously. Theorems derived from the assumption that only one isolated factor occurs, would be empirically observed if it were not for the simultaneous confluence of a myriad of disturbing factors. The presence of these disturbing factors, when referring to concrete phenomena, would only allow us to speak in terms of tendency laws. For example, David Ricardo’s law of comparative advantage could never be confirmed or refuted empirically, given the impossibility of finding two countries absolutely identical except for their foreign trade policy.
Carl Menger, founder of the Austrian School of Economics, uses the same argument when he condemns the confusion between that which he calls the empirical-realist research orientations and exact research in the economic theory. The latter, starting from a priori principles about the actions of individuals, would deal exclusively with the effects of economic forces working alone. While the first, when considering the reality in all its complexity, would never be able to establish exact correlations between concrete phenomena. Menger never tires of reminding us that attempts to replace “atomism” and the simplifications used in economic theory with any new methodology based on observation of reality as a whole (as the German historical school wanted) has proved itself unable to generate any kind of results.
Ludwig von Mises, one of the heirs to the Mengerian tradition of the twentieth century, also contributed to the defense of the economic theory from the positivist attacks. His thesis on methodology can be found in several of his works. The last book published during his lifetime, is however an actual critique on positivism in economics, and why we chose this work to contrast his views from his brother’s. Just as Richard criticizes the economic theory from the positivist perspective without making references to Ludwig, the latter in turn criticizes the positivist design of unified science without making references to his brother. Nevertheless, the disagreement between them is revealed in each argument.
Right from the preface of his criticism, Ludwig complains that natural scientists (among others, his brother?) should first study the economic theories before critiquing them. This study would reveal that the positivist outline for the social science, called panphysicalism by the author, is doomed to failure by its inability to deal with a fundamental element of these disciplines: the intention of human action. Although the evolution of the natural sciences, as described by the positivists, has indeed been marked by the purge of animistic notions, borrowing methods from these sciences would result in a setback in the development of the human sciences: “The physicist may laugh today at the doctrine that interpreted certain phenomena as the effect of a horror vacui. But he fails to realize that the postulates of panphysicalism are no less ridiculous. If one eliminates any reference to judgments of value, it is impossible to say anything about the actions of man…” (Mises, L. 1978, p. 39)
For Mises, a method that allows only observational protocols would impede the understanding of human behavior. Unlike the thoughts of his brother, Ludwig did not believe that we learn about others only through the senses: the social sciences would be essentially teleological (pp. 7, 37) and explanations of social phenomena must necessarily involve references to the intentions of the agents, even if they are not explicitly stated. The failure of behaviorism — the attempt to replace mental categories of psychology not directly observable by the allegedly more scientific method (that considers only observable stimuli and responses) — is presented by Mises (pp. 41, 121) as an example of the impossibility of eliminating the concept of the purpose for human action from social theories.
Not only does the subjective nature of facts relevant to the social sciences conspire against the positivist claims, but attempts to quantify and measure social phenomena were unable to reveal regularities that could inspire satisfactory theories on the subject. Mises (pp. 26, 57, 63) never tires of repeating that in the sphere of human action there are no constants such as those found in physics. Echoing Mill, Mises (pp. 43, 76) stresses the point that experience in the social fields is always relative to complex phenomena, and empirical verifications or refutations of social theories are not possible. Thus the coexistence of competing explanations of the same social phenomena, which so annoyed Richard, could not be solved only by an appeal for more observations, or purged of their metaphysical concepts and ideological prejudices.
Aside from stating that there is no positivist alternative to the methods practiced in the social sciences, Ludwig believes the attack by this metaphysical doctrine (p. 117) on the social sciences could actually have pernicious consequences. For Mises (pp. 40, 123-4), advocates of positivism — old and new — would not be known for any contributions of what they have done for the science, but for what they want to prohibit. Its defenders would be effectively advocates for intolerance and dogmatism. There is thus, nothing more emblematic about the differences between Ludwig and Richard von Mises than their opinions on Comte: dogmatic and intellectually sterile for the first and a model of good science for the second. The transformation of positivism in the hands of Comte into a dogmatic religious sect, something viewed accidental by Richard (p.360), is seen as the fruit of dogmatism inherent to positivism, according to Hayek (the economist that succeeded Ludwig in the development of the Austrian School of Economics).
Having reviewed Ludwig’s opinion on the positivist program for the social sciences, let’s consider his views on the appropriate method for them, referred to as the sciences of human action. Mises divides these into two branches: theory and history. The latter uses the results of all theoretical sciences to study particular sequences of events in time. For this to be carried out, one historically uses the concept of “understanding” (Verstehen) the effort to make the actions of certain agents intelligible in terms of its purpose and plans.
Social theories however, economics in particular, treat regularities applicable to all cases. Like Mill, Mises defends the thesis that economic theory use the a priori method, obtaining results through deductive reasoning (p. 21), starting from basic postulates. Unlike Mill, Mises does not adopt the fundamental principle of the restrictive assumption of Homo economicus. For him, contrary to what his brother believes, economics requires only the assumption that people have unfulfilled purposes, regardless of its nature. Given this generalization, the historicist thesis (defended by Richard and today by critics that have not read the theory of Mill or Mises) that economic theory would depend on the false assumption of egoism, or the impulse to accumulate wealth, loses its meaning.
According to this Misesian conception, there would be no strictly economic subjects, but only economic aspects in which for every kind of activity it’s necessary to choose between alternative purposes, regardless of whether the activity is artistic, spiritual, sports or of any other nature. Thus the theoretical social sciences for Ludwig, consist of what he calls praxeology — examination of the logical consequences of the fundamental postulate that states people act when they imagine more satisfactory situations, means to achieve them, barriers to their adoption and uncertainty of outcome.
For Mises, every economic theorem is derived ultimately from the postulate of human action (p. 45). This assumption in turn, consists of true knowledge about the world, derived solely from reason. In Kantian terms, it would be an example of a priori synthetic knowledge, precisely the category whose existence is denied by logical positivism. For Mises (p. 18), the aprioristic character of the postulate of action is derived from the fact that the idea of ??purposeful action is part of the structure of the human mind: we cannot even imagine the veracity of its denial. Thus, theoretical knowledge in economics would always be true and the only thing that historical observation could do is determine if a certain economic theorem is applicable or not to the existing conditions in a certain concrete situation.
Although the brothers’ books address countless theses on various disciplines, we can summarize their opposing views on the method of the social sciences through some labels. For Ludwig, the social sciences are marked by methodological subjectivism, methodological individualism, deductivism and apriorism. Richard on the other hand, we could label inductivist and a posteriori. Which in turn, indirectly imply methodological objectivism, methodological collectivism and historicism. That is to say, without explicitly defending these principles, the logic of Richard’s arguments takes him in these directions.
Having estabilished the contrast between the theses of the two brothers, we will conclude with a discussion of their importance to the development of modern economic science. This discussion can be summarized as a conflict between the desired and the possible: from the thirties of the twentieth century, the evolution of economics was undoubtedly inspired by the positivist program, but the problems inherent to this doctrine limited the progress of this program.
The first point to note shows that, although quite influential among scientists, positivism was progressively phased out of the philosophical discussion. Thus, today almost no one believes that science starts from impersonal observations on which inductive generalizations are made: most scientists actually advocate some kind of hypothetical-deductive model. Ludwig von Mises however, in defense of the methodological dualism, assumes that positivism is descriptive of the methods of the natural sciences: this would be based on observation (p.54) and subject to empirical verification. In turn, Hayek prefers to attack what he calls scientism, defined as the defense for the social sciences from what is erroneously considered the method employed by the natural sciences.
In fact, modern economists employ theories that start from axioms and use inference to get their results. The clamor for rejection of hypotheses not based on observation, especially those related to rationality and profit and utility maximization, has always been the flag of heterodox groups and laypersons, as modern econophysicists.
Nevertheless, positivist ideals marked the modern economics: the saying of Lord Kevin that science is measurement is perhaps the most cited by economists. In fact, much of the development of the discipline in the last hundred years has been guided by the search for models that can be tested empirically and the formulation of working hypotheses that use variables which can be measured in principle (as in the theory of revealed preference by Samuelson). At this point, no doubt the profession moved away from reviews of Ludwig von Mises, who skeptical about the possibility of finding constants and confirming or refuting economic theories, considers econometrics as mere recent economic history (p. 64). Samuelson made famous the statement that he “trembled for the reputation” of economics when he read the methodological reviews of Menger, Mises, Robbins and Knight. The latter actually represent orthodoxy regarding methodological issues prior to the positivist revolution.
This orthodoxy, given the complexity of social phenomenon, adopted hypotheses with a high degree of generality so that theoretical regularities could be established. Thus, for example, every non-reflexive action was considered rational to Mises. For economists committed to the ideal of testability, such a definition would be inappropriate, approaching mere tautologies. Thus, broad definitions have been replaced by more concrete hypotheses and established theoretical concepts were reinterpreted to be more operational.
Thus, the search for testable models in a field marked by complex phenomena, whose complexity conspires against this search yielded an interesting methodological dilemma: if on a more general level being right is seen as unscientific, to adopt specific hypotheses about complex phenomena makes these hypotheses automatically refuted, so the alternative is to be wrong, but with correct method! This dilemma explains the predominance of the modern methodological approach known as instrumentalism: no matter the realism of assumptions, but its predictive power. This reconciles the demand for operational hypotheses with the unpleasant realization that these tend to be easily refuted, maintaining the positivist illusion that theories would be compared by their capacity to predict phenomena.
The realization that theories whose predictions failed were not discarded, so that the instrumentalist criteria functions primarily as rhetorical strategy for the rejection of competing explanations, coupled with lack of theoretical progress derived from the adoption of a more empirical economics and the progressive discredit of positivist philosophy, made ??the contempt for the methodological opinions of Ludwig von Mises were reversed and their opinions reconsidered. After all, according to his brother Richard, logical positivism was not just a metaphysical doctrine to the extent that the success or failure of the positivist project could be verified empirically. Perhaps there is some truth in that statement.
BARTLEY III, W.W. & Radnitzky, G., (eds.) Evolutionary Epistemology, Rationality and the Sociology of Science.La Salle: Open Court, 1987.
CALDWELL, B.J. Beyond Positivism: Economic Methodology in the Twentieth Century. London: George Allen & Unwin, 1982.
DETTERING, R. The Criterion of “Connectibility”, 1953.
FRIEDMAN, M. “The Methodology of Positive Economics” in Essays in Positive Economics, Chicago: University of Chicago Press, 1966.
HAYEK, F.A. The Counter-Revolution of Science: Studies in the Abuse of Reason. Indianapolis: The Free Press, 1979 .
HAYEK, F.A. “The Theory of Complex Phenomena”, in HAYEK, F. A. Studies in Philosophy, Politics and Economics.London: Routledge, 1967.
KIRZNER, I.M. The Economic Point of View: An Essay in the History of Economic Thought. Kansas City: Sheed & Ward , 1976.
MENGER, C. Investigations into the Method of Social Sciences. Grove City: Libertarian Press, 1996 .
MILL, J. S. “Da Definição de Economia Política e do Método de Investigação Próprio a Ela”. In: Os Pensadores. São Paulo: Abril Cultural, 1984.
MISES, L. The Ultimate Foundation of Economic Science: An Essay on Method. New York: Foundation for Economic Education, 1978 .
MISES, L. Ação Humana. São Paulo: Instituto Mises Brasil, 2011.
MISES, R. Positivism: A Study in Human Understanding. New York: Dover Publications, 1951. Translated from Kleines Lehrbuch des Positivismus, Hage: von Stockum, 1939.
NEURATH, O. CARNAP, R. & HAHN, H. “The Scientific Conception of the World: The Vienna Circle ” In: NEURATH, M. & COHEN, R.S. (eds.) O. Neurath: Empiricism and Sociology, Chapter 9, 1973.
SCHORSKE, C. E. Fin-de-Siècle Vienna: Politics and Culture. New York: Vintage Books, 1981.
SKOUSEN, M. The Making of Modern Economics. New York: ME Sharpe, 2001.
ROTHBARD, M. Ludwig von Mises: Scholar, Creator, Hero. Auburn: L v Mises Institute, 1988.
 Menger, 1996.
 Mises, L, 1978.
 In economics, parallel failure is illustrated by the theory of revealed preference, which sought to replace the theory of consumer choice and its subjective concepts of preferences and utility, with another based solely on the choices, prices and observable incomes.
 Hayek, 1979, second part.
 Mises, 2011.
 Kirzner, 1976.
 Hayek, 1976, in his penetrating analysis of positivism in social science, summarizes the contrast similarly. For this author, economics would methodologically be subjectivist, individualist and theoretical, while scientism implies the opposite positions: objectivism, collectivism and historicism.
 The author seems reluctant on this point. Sometimes adopting verificationism, and other times he attacks the falsificationism. In the text under discussion, however, the author states: “the positivist principle of verifiability as modified by Popper is undisputed as an epistemological principle of the natural sciences” (p. 119-120).
 Hayek, 1979.
 “When you can measure what you are speaking about, and express it in numbers, you know something about it, when you cannot express it in numbers, your knowledge is of a meager and unsatisfactory kind; it may be the beginning of knowledge, but you have scarcely, in your thoughts advanced to the stage of science.”
 Caldwell, 1982, p. 118
 Hayek, 1967, states that although we must make theories as falsifiable as possible, the greater the complexity of the subject matter, inevitably the lower the degree of falsification of theories about these phenomena. This would be a price to pay for the study of such phenomena.
 The text of Friedman (1966) is representative of this methodological approach.