Portal de Periódicos da CAPES

The Battle of Methods in Economics. The Classical Methodenstreit—Menger vs. Schmoller

2011; Wiley; Volume: 70; Issue: 2 Linguagem: Inglês

10.1111/j.1536-7150.2011.00780.x

1536-7150

Marek Loužek,

Economic Theory and Policy

Should an economic theory be built abstractly, deductively and hypothetically, regardless of empirical reality; or is economics condemned to be restricted to the dry collection of empirical facts and their compilation into dull statistics? Does exact knowledge mean getting rid of all empirical evidence, or conversely, making a radical move towards empiricism? Is methodological individualism, as opposed to methodological collectivism, indispensable for economics or vice versa? The above questions are not at all new in economic theory; they have appeared as early as the end of the 19th century in what is referred to as Methodenstreit between Carl Menger and Gustav Schmoller. It is useful to take a look at this dispute in greater detail because it is regarded as one of the most significant methodological disputes in the history of economics. In some sense, it had anticipated the methodological disputes in contemporary economics. As early as the last third of the 19th century, German economists argued that Austrian economic theories were unfounded because they were non-empirical, non-realistic and hence unusable, while Austrians criticized the Germans stating that their historical method only covered a mere description of the existing phenomena and failed to seek generally valid economic laws and relations. Who was more accurate? In the English language literature, the Methodenstreit is usually described as a dispute between the abstract-deductive method represented by Menger on one side and the empirical-inductive method represented by Schmoller on the other side. This description is supported, in various versions, by Schumpeter (1972: 814), Seligman (1962: 274), Lekachman (1959: 249), Newman (1952: 195), Landreth (1976: 275), Ingram (1967: 235) and Haney (1949: 550). The first person who brought this concept to the Western literature was Böhm-Bawerk (1890: 244–271). John Neville Keynes (1891), the father of the famous John Maynard Keynes, also played an important role in the dissemination of the standard Anglo-Saxon interpretation. The seeds of the future terminology had already been present with one of the participants of the dispute, Schmoller. He primarily understood his dispute with Menger as the struggle between induction and deduction (Schmoller 1883). Although the definition of the first battle of methods as a dispute between the abstract-deductive method and the empirical-inductive method is attractive and quite widespread, one can also find some faults in it. These faults lead some authors—especially German and Austrian writers—to arrive at the idea that the traditional concept of Methodenstreit is too reductive and needs to be broadened—Kerschagl (1948: 29–33), Ritzel (1950: 118–124), Albert (1962: 142). The related arguments are as follows. Firstly, Methodenstreit cannot be reduced to a dispute between the abstract-deductive and empirical-inductive method because the key polemic between Gustav Schmoller and Carl Menger also included some other points of controversy, including the issue of the justifiability of exact economic laws. This issue is related to the first dispute to a certain degree, but not completely. Furthermore, the third important aspect of the Methodenstreit is relatively separate—it is known as the dispute over methodological individualism versus collectivism in economics and social sciences. Secondly, the Methodenstreit cannot be reduced to a dispute between an abstract-deductive and empiric-inductive method for the reason that this terminology itself is an interpretation. This interpretation is very close to the terminology of one of the participants of the dispute, Schmoller; however, Menger never used his terminology. Instead, he spoke about the exact and empirical-realistic orientation of scientific research (die exakte und die realistisch-empirische Richtung der Forschung). Thirdly, the Methodenstreit can hardly be reduced to a personal dispute between Menger and Schmoller. The polemic of Menger and Schmoller just touches the surfaceof a dispute that is far deeper and more fundamental. Some authors expressed the opinion that Schmoller did not enrich the discussion with anything original and that the truly dignified opponent of Menger in the Methodenstreit had been Wilhelm Roscher, a representative of the older German historical school (Milford 1995). In this article, we will explain the classic Methodenstreit or the first battle of methods as a methodological dispute that took place in the 19th century between the Austrian school of economics led by Menger and the German historical school led by Roscher originally and later by Schmoller. This article will distinguish three main issues of discussion: 1) a dispute about the justifiability of deductive or inductive methods in economics, 2) a dispute over the existence of exact laws in economics, 3) a dispute over methodological individualism versus collectivism in social sciences. During the first battle of methods in Germany, in the 19th century, economics was not wholly established as a completely autonomous sovereign discipline, equally self-assured as, for example, history, philosophy or natural sciences (Häuser 1988). Universities did not have an economics department; national economy issues were usually included in history, and economic statistics were the only focus that was taught. Even in England and Scotland during the 19th century, political economics was becoming a discipline of its own very slowly. Long after Adam Smith, economists continued to work in a variety of departments (most commonly in the departments of history or moral sciences). The first provable economic professorship in England came into being at East Indian College in Haileybury in 1805. It was called a Chair for History and Economics and its holder was Thomas R. Malthus (Moore 1999). At European universities, there were only classical sciences such as theology, law, philosophy and, later on, medicine and natural sciences. The new economists were emerging from virtually all these disciplines, perhaps with the exception of medicine and partly theology (however, Galiani, Turgot and Marshall studied theology and Thomas R. Malthus worked as a pastor before taking up the professorship of economics and history). Economics was most strongly influenced by philosophy, jurisprudence as well as natural sciences. The English branch of economics became increasingly autonomous primarily according to the example set by natural sciences, while in Germany, economics split off from mainly history and law. The Methodenstreit can be compared to a labor pain accompanying the birth of separate economic theories, which were distinguished from history in Germany (Hodgson 2001). The three disciplines that stood at the birth of economics were far more strictly separated at the German universities than at the English universities. Similarly regarding the division of scientific disciplines, there was virtually no difference between Austrian science and German science but there was quite a substantial chasm between the continental universities and the English schools. According to Häuser (1988), the different organizational structure of the universities was the main reason why no similar dispute, in comparison with the Methodenstreit in the German cultural area, broke out in England. As opposed to students at universities of the continental type, English students were concurrently studying various disciplines; interdisciplinarity was much more developed in Great Britain than in Central Europe. For example, Adam Smith studied mathematics, classical philology and philosophy in Glasgow and Oxford. Malthus was educated in mathematics, natural sciences and linguistics. Keynes, in addition to economics, also studied history, politics, mathematics and arts. As opposed to this, the German and Austrian universities separated the individual disciplines into isolated faculties and students only had a limited possibility of obtaining broader education and having contact with other disciplines. This institutional separation resulted in more competition among the discipline-specific departments and the drive for prestige more often tempted each department to highlight its own merits and, conversely, to question the merits of others. Thus, separation of theoretical economics from its maternal sciences—philosophy and law—was significantly more painful in Germany and Austria and was accompanied by harder clashes than in England. When the German economists succeeded in winning their new department, they soon after closed themselves in their offices and the contact with their colleagues remained minimal. It was not only economics but also natural sciences that took a very long time before they were emancipated from their powerful predecessor disciplines. With its philosophical roots, German economics is, of course, not exceptional: names such as Locke, Hume, Smith or Mill prove that English economics also had significant philosophical predecessors. The law-related roots of economics can also be traced very clearly in France and Italy. Names such as Montesquieu, Bodin, Verri and Beccaria symbolize not only famous law authorities but also the seeds of the new economic science in France and Italy. In spite of this, the law tradition remains typical of Germany and its influence can hardly be compared with other states. Cameralistics was essentially a German-Austrian version of mercantilism. The two basic features of Cameralism, which German economics retained in the form of the historical school throughout the 19th century, is the close interrelation with law and the immediate relation to political power. In the last third of the 19th century, the Austrian school started to split off from this traditional German trend. Austrians tried to get beyond the shadow of the historical method and wanted to create a real theory. Germans interpreted this departure from the fixed line of historicism and law as a betrayal. The first battle of methods, one of the most significant methodological disputes in the history of economics, broke out. The most significant theoretical economic work of Menger is deemed to be the Principles of Economics ([1871] 1981). At the time of publication of the Principles, the school known as the younger historical school dominated in German economics. The scholars belonging to this group were usually oriented towards practical political issues. They demanded a historical-descriptive method for economics and, in their opinion, scientific knowledge was to be postulated in such a way so that it could be quickly used in economic and social policy (Oakley 1997). Therefore, Menger's exact and individualist methodology had a low chance of success in the German community of economists. However, the object of mistrust on the part of German historians was not the theory of subjective value, as it is sometimes claimed. The theory of subjective value had a long-standing tradition in Germany, supported by such names as Hufeland, Roscher, Knies, Rau and Hildebrand (Streissler 1990: 31–68). It was Menger's explicit requirement for methodological individualism that prompted ferocious resistance among German historians. Methodological individualism requires that social science should base the explanations of social phenomena on individual preferences. Methodological individualism, along with the theory of subjective value, was last presented by Hufeland and Menger continued exactly in this line of thought. Virtually all German economists after Hufeland advocated the principle of methodological collectivism and that was the main reason why they disagreed with Menger (Milford 1990). Although Menger was passionately convinced of the accuracy of his theory, he started to slightly revise his work as early as 1873, that is, merely two years after the first publication of Principles. He left his economic theory unchanged; however, he decided to support the economic teachings with more accurate philosophical, gnoseological and methodological reflections. The methodological investigations, which had only been included in the preface in the first edition of Principles, were to permeate and accompany the whole work in the second edition (Alter 1982). Since he was unable to reconcile himself with the fact that his economic theory had not been met with a proper response in Germany, he energetically settled his score with German national economics in an article titled "The Split of German National Economists" in 1873. However, the culmination of his theoretical break-up from German economics takes place in his key methodological work Investigations into the Method of the Social Sciences, with Special Reference to Economics ([1883] 1985). Menger (1985) tries to define economics as a theoretical science. The key problem he deals with is distinguishing between historical and theoretical sciences dealing with society and the economy. Menger holds the opinion that the distinguishing criterion is that historical sciences are concerned with the analysis of specific, individual phenomena, while theoretical sciences analyze that which is general and common to the phenomena (or certain groups of phenomena). Purchase, sale, supply, demand, price, capital or interest rates are typical phenomenal forms of economy. Regular decrease of price resulting from increase of supply, decrease of interest rates resulting from increase of savings are typical relations or correlations between the phenomena that we can see in the national economy. The task of economics as a theoretical science is to analyze the general essence of these phenomena and the general relations among them (Hutchison 1992). Thus, distinguishing economics as a theoretical science from history does not, of course, mean to say that historians should not be concerned with the national economy. In Menger's opinion, the same phenomenon can be explained both in a historical way and a theoretical way. An increase in revenues from land or a decrease in interest on capital can be explained historically as a result of specific individual circumstances. However, these phenomena can also be interpreted (no less valuably) in a theoretical way—as a consequence of exact economic laws. Both approaches are justified. Theoretical economics can consist only in the exposition of the general nature and the general connection of the laws of economic phenomena, but by no means, for instance, in the exposition of the nature and connection of individual phenomena of economy, such as in historical presentations, or in practical rules for the economic activity of people. The theory of economy must in no case be confused with the historical sciences of economy, or with the practical ones. (Menger (1985: 51) In Menger's opinion, theoretical research can take two different paths: A scientist tries to grasp the full empirical reality, the totality of phenomena in their entire complexity, and then sorts the sum of the real phenomena into certain phenomenal forms and subsequently presents patterns or coexistences of these phenomena and phenomenal forms as he experienced them in practice. This method of work led to the establishment of what is known as a realistic-empirical orientation of theoretical research. A scientist focuses on a certain specific aspect of phenomena or phenomenal forms, while assuming that all the other aspects and circumstances are constant. Thus, he/she examines the general relations of the subsequence and coexistence of phenomena (or certain aspects of phenomena), while assuming that the other components of the empirical reality are constant. This method led to establishment of what is known as exact orientation of theoretical research. Both the exact and the realistic orientation of theoretical research are therefore justified. Both are means for understanding, predicting, and controlling economic phenomena, and to these aims each of them contributes in its own way. Anybody who denies the justification and usefulness of the exact or the realistic orientation is comparable to a natural scientist who one-sidedly values physiology highly, perhaps under the pretext that chemistry and physics are based on abstractions, and would deny the justification of the latter or their justification as means for the understanding of organic structures. Otherwise, he/she resembles a physicist or chemist who would deny to physiology the character of a science because its laws are for the most part only "empirical" (Menger 1985: 64–65). In order to specifically show the fundamental difference between exact and empirical analyses, Menger uses the law of demand as an example. If a layman were to formulate the law of demand, he would probably say that if the need for a commodity increases, then the demand will increase and consequently the market price will increase. However, such a formulation is in fact insufficient. The law of demand can be formulated in two different manners: exactly and empirically. The exact law of demand says: If the need for a commodity increases, then under otherwise identical conditions its market price will always increase. As opposed to that, the empirical law of demand states: If the need for a commodity increases, then its market price will usually increase. The key difference between these two formulations lies with the fact that while the empirical law of demand is based on experience, the exact law is based on the empirical framework, but at the same time also adds other non-empirical (ideal) assumptions. Of these model assumptions, the most important one is the thesis that other conditions of the examined relation are unchanged (the ceteris paribus condition). However, these assumptions have an important consequence: while the empirical law of demand is valid only in relative terms, for example, "normally,""usually," according to the observed information, the exact law is valid in absolute terms and does not permit any exceptions. This difference between the exact and empirical laws is absolutely fundamental and also has a key importance for theoretical economics. While the empirical law of demand is only valid in some cases, the exact law of demand is valid—under the defined assumptions—absolutely and inevitably, and it cannot be refuted by pointing out an empirical fact that is in conflict with it. The law that the increased need for an item results in an increase of prices, and indeed that a definite measure of the increase of need also results in an increase in prices determined according to its measure, is not true—is unempirical, when tested by reality in its full complexity. What else does this say, that the results of exact research do not find their criterion in experience? (Menger 1985: 71–72) The goal of exact theoretical science is to find the simplest elements of social or natural phenomena and then, by putting together their mutual relations, correlations and interplays, to explain more complicated phenomena. In Menger's opinion, this approach is also justified in a situation where the simplest elements, at which the exact analysis arrived, are of non-empirical nature. In Menger's opinion, this means that the social sciences have a great advantage over the natural sciences. The ultimate elements to which the exact theoretical interpretation of natural phenomena must be reduced are "atoms" and "forces". Neither is of empirical nature. One cannot imagine "atoms" at all, and natural forces only by a representation, and by these we merely understand unknown causes of real motions. From this there arise ultimately quite extraordinary difficulties for the exact interpretation of natural phenomena. It is otherwise in exact social sciences. Here the human individuals and their efforts, the final elements of our analysis, are of empirical nature, and thus exact theoretical social sciences have a great advantage over the exact natural sciences. (Menger 1985: 142) In Menger's opinion, the exact method of theoretical research in the field of the organic world does not deny the unity of natural organisms at all. Rather it searches for the origin and functions of these individual formations and tries to explain how the real units function. According to Menger, for this purpose, it is always necessary to use the methodological individualism (as opposed to the empirical-realistic method of scientific research). Menger demonstrates an exact analysis on the origin and emergence of money and a number of other social institutions (Zuidema 1988). It is with Gustav Schmoller that the Methodenstreit is most frequently associated (in addition to Carl Menger). Schmoller wrote a review of Menger's book in a journal called Jahrbuch für Gesetzgebung, Verwaltung und Volkswirtschaft (Schmoller [1883], 1888) in 1883 and the review was not very commending. As a contrast, Schmoller used Dilthey's work "Introduction to Human Sciences," which he, conversely, praised highly (Dilthey 1883). The passionate discussion that broke out in response to this review is labeled as the first battle of methods (Hutchison 1988). Schmoller's review is important in several respects. Firstly: Schmoller was the first one who interpreted Menger's exact and empirical orientation of scientific research in the terms of abstract-deductive and empirical-inductive method. "The two paths, which Menger sees in economics, are called realistic-empirical and exact by him. By this, he means that which is usually referred to as inductive and deductive procedures" (Schmoller 1888: 280).1 The new label was gradually accepted in economics and has persisted in standard interpretations of the Methodenstreit until today. More importantly, Schmoller's not quite self-evident opinion that the difference between an abstract-deductive method and an empirical-inductive method is not fundamental because both methods are integral parts of economic analysis was also accepted. The division into two orientations of scientific research, on which Menger bases his reflections, undoubtedly has certain justification . . . but this contract must be understood as a chasm that can be overcome. The individualist science or—a descriptive science provides the material for the general theory. This material gets more complete as all the important aspects, transformations, causes and effects are better described. (Schmoller 1888: 278) In Schmoller's opinion, the deductive and inductive methods are not in conflict with each other; rather, they supplement each other. Every consistent descriptive work that involves collecting empirical or historical data is then a prerequisite for obtaining general pieces of knowledge. First, we must gather rich empirical material, and only then can we build a general theory. If we create complex deductive theories regardless of the reality, they become mere daydreams. When a presumption of the pursuit of one's own interest had been used as a seeming constant for price examinations at one time, it was done to explain the simplest market processes. But it is a mistake to make this a rule for future research or examination of more complex national economic processes. In any case, one must always be aware, if one uses this procedure, that science based on hypotheses will always only provide hypothetical forecasts. However, such forecasts are only seemingly exact. (Schmoller 1888: 281) Schmoller criticizes Menger for restricting theoretical economics merely to a discipline dealing with value, price and the essence of money. Menger does not need to examine other aspects of social and economic life. However, in Schmoller's opinion, the undesired result is a questionable theory that is sheer speculation and is suspended in mid-air (Peukert 2001). Schmoller stated in a ridiculing manner, "Menger says that who wants laws must abstract. We answer that in the end, all our thinking and cognition is based on abstraction. However, the point is to abstract in the correct manner so that our abstractions result in scientific truths and not schematic phantoms or visionary escapades as is, regrettably, often the case" (Schmoller 1888: 283). Schmoller admits that Menger's criticism of the German historical schools was valid in certain specific cases. Nevertheless, he emphasizes that it is not possible to draw broad conclusions about an entire scientific discipline from a few isolated blunders. The conclusions contained in the second book of Investigations are correct in many respects. One could also partly agree with the list of sins of the historical school,but does this completely cancel the justifiability and merits of the entire scientific discipline? Menger is absolutely incapable of understanding the fundamental causes and merits of the historical school because he lacks the authority to do so. The historical school represents a return to the scientific grasp of reality instead of vague abstractions lacking the desired connection to reality. (Schmoller 1888: 287) To Schmoller's harsh criticism, Menger responded by the pamphlet "Errors of Historicism in German Economics" (Menger [1884] 1970: 1–99). However, this polemic treatise was not a very strong one because the substantive essence of the dispute got lost among the caustic attacks against Schmoller, and key arguments were also missing. Of course, such arguments did exist, but Menger simply did not have enough time for them (Schumpeter 1972: 814). The standard of argumentation of the founder of the Austrian school of economics is indicated by a sharp quote. "I would like to point out all the possible and impossible distortions of my words by Schmoller. There is no doubt that Schmoller has achieved a true mastery in his personal and vulgar style. Yes, after all, it is the only mastery that can be attributed to this man with a brilliant knowledge of the German language" (Menger 1970: 16). In the "Errors of Historicism," Menger does not present any new arguments; he only more or less repeats the ideas that had already been expressed in the Investigations. He insists that it is useful to distinguish between exact and realistic-empirical methods of scientific research; he insists on a strict separation of history from theoretical social science, and criticizes the methodological collectivism of Schmoller. Menger even provoked Schmoller by sending him his pamphlet for review directly to Jahrbuch für Gesetzgebung, Verwaltung und Volkswirtschaft. Schmoller flared up and immediately sent the pamphlet back to Menger with an insulting cover letter, the text of which he published in the Jahrbuch at the same time: "It is impossible not only to read but even to review something so outrageous" (Bloch 1940: 432). Why did such a sharp controversy take place between Menger and Schmoller? Why did both participants of the dispute insult each other in spite of the fact that both of them were openly building on the work of Wilhelm Roscher? Was their dispute only an unfortunate misunderstanding resulting from the need for scientific prestige or was it really inevitable? Was the Methodenstreit only a methodological dispute or were different philosophical roots also reflected in it? There are a number of interpretations of the Methodenstreit in literature. The standard economic interpretations are presented by Eugen von Böhm-Bawerk (1890: 244–271), J. N. Keynes (1891: 5–31) and J. Schumpeter (1972: 814–815). These authors claim that the first battle of methods had been overdone and unnecessary. The standard interpretations express that there are no large programmatic differences between Menger and Schmoller that would justify such an enormous intensity of the dispute. Neo-Austrian interpretations, presented by Ludwig von Mises (1933: 68) and/or F. A. von Hayek (1930: 393–420) usually emphasize the deep meaning of Methodenstreit and tend to take sides with Menger. The specific feature of Neo-Austrians is that instead of methodological pluralism officially advocated by both Menger and Schmoller, they demand methodological monism. Thus, the situation that prevailed in the middle of the 19th century has been reversed to a considerable degree. Consensualist interpretations are close to the standard economic interpretations in the sense that they seek compromise between the clear-cut Austrian and German positions. However, unlike standard interpretations, they recognize the noetic value of the dispute. Karel Engliš, a Czech economist (1992), sees the Methodenstreit through the eyes of his teleological method (explaining social phenomena through their purpose); Gerhard Ritzel (1950: 89–129), a Neo-Historian, tends to be more of a fan of Schmoller but also recognizes Menger's merits. Philosophical interpretations are focused on the broader intellectual context of the dispute or they try to address some of the deeper epistemological issues. The most significant philosophical interpretations of the Methodenstreit include S. Bostaph (1978), R. Hansen (1968) and U. Mäki (1990). Most authors emphasize the Aristotle tradition, on which Menger's background was based, and the Neo-Kant tradition, with which Schmoller was raised. The standard economic interpretations are right in saying that the Methodenstreit was a dispute that was overdone to a certain degree. The statement that the dispute was unnecessary is somewhat less acceptable. The Methodenstreit ultimately revealed certain hidden assumptions on both sides and compelled both camps to clearly formulate their stances. The Methodenstreit cleared the air significantly, and it is highly unlikely that the first battle of methods could have been avoided. The analysis of the philosophical context shows that the dispute reached far deeper than just pure methodology. In the Methodenstreit, two different philosophical traditions—the Austrian and the German—were also reflected. Carl Menger, as the founder of the Austrian school of economics, was firmly tied to the philosophical