History, Theory, and Statistics from the Perspective of Mach's Writings
- Top of page
- History, Theory, and Statistics from the Perspective of Mach's Writings
- Marx's Economic Interpretation of History as a Key to Schumpeter's Economic and General Sociology
- Discussion and Concluding Remarks
It is useful to briefly consider how Schumpeter arrived at the problems and solutions that he focused on. When it comes to the content of history, theory, and statistics, and the relations among them, the late 19th-century “Methodenstreit” was still heavily affecting economics when Schumpeter started his studies. On one hand, there was the Austrian, Carl Menger, according to whom economics should be heavily focused on just certain aspects of a phenomenon while other aspects could be assumed constant. The outcome was supposed to be a clear theory developed largely by deduction. Parallel to such an approach, but secondary to it, there could be a realistic and empirical orientation where the phenomena are analyzed in all their complexity. Menger put individualism at the theory's core. On the other hand, there was the German, Gustav von Schmoller, according to whom if a general theory were to emerge, it must come out of rich empirical data. Furthermore, Schmoller advocated a more collectivist perspective than Menger (Louzek 2011; Shionoya 2005).6
Schumpeter (1908) developed and explained his position in his first book. His final position is a fine-tuned rather than more radically revised version of his early position. Here is a comment in History of Economic Analysis (Schumpeter  1997), which contains a description of the way that all science evolves and the “rationale” for economic theory and theory in science in general:
Experience teaches us that the phenomena of a given class—economic, biological, mechanical, electrical, and what not—are indeed individual occurrences each of which, as it occurs, reveals peculiarities of its own. But experience also teaches us that these individual occurrences have certain properties or aspects in common and that a tremendous economy of mental effort may be realized if we deal with these properties or aspects, and with the problems they raise, once and for all. For some purposes it is indeed necessary to analyze every individual case of pricing in an individual market, every case of income formation, every individual business cycle, every international transaction, and so on. But even where this is necessary we discover that we are using, in each case, concepts that occur in the analysis of all. Next we discover that all cases, or at least large sets of individual cases, display similar features which, and the implications of which, may be treated for all of them together by means of general schemata of pricing, income formation, cycles, international transactions, and so on. And finally we discover that these schemata are not independent of one another but related, so that there is advantage in ascending to a still higher level of “generalizing abstraction” on which we construct a composite instrument or engine or organon of economic analysis—though not the only one, as we have seen—which functions formally in the same way, whatever the economic problem to which we may turn (*). (Schumpeter  1997: 15–16, italics in the original)
Interpretations could be developed directly, but it is worthwhile addressing the footnote related to the asterisk in Schumpeter's quote first:
The statement above is a brief rendering of E. Mach's doctrine that every (theoretical) science is a device for effecting economy of effort (Denkökonomie). (Schumpeter  1997: 16n4)
These quotes show that Schumpeter continued to cherish Mach's views throughout his career. Mach's original texts (especially  1976; see also  1919,  1898) must be carefully studied if one is to understand Schumpeter's views on questions such as what science is, what scientists do, and how science evolves at the level of an individual scientist.7 Such considerations are intimately related to Schumpeter's views on history, theory, and statistics.
In the above quote, Schumpeter talked about formation of “concepts.” Mach's writings described in detail how “concepts” and “ideas” gradually evolve out of the mass of memorized “sensations” (Mach  1976: Ch. 2). The process is explained in the following quote where Mach describes how numbers and the concept of number may have been first learnt:
Even animals are no doubt driven by vital needs to distinguish small groups of the same fruit for example as to content, and to prefer the larger to the smaller. In the need to refine this ability to distinguish lies the origin of the number concept. The more members can be gathered into a group without losing the overview and the individuality of members, the more we shall value this ability. To start with, children manage groups of 2, 3 or 4. Contiguity in space and time may be helpful in forming the group, while difference of spatio-temporal position may condition the process of distinguishing the members. Thus arise the first idea of number, with or without name according to the influence of the environment. These ideas develop through the senses of sight, touch and hearing, in the last case through attention to rhythm. As we work with ideas of number while the objects change, we are led with the help of number names to the view of a uniform activity of reaction independent of the nature of the object, that is, to the concept of number. (Mach  1976: 241)
The first thing to highlight is Mach's instrumental view of knowledge and knowledge generation. He argued that practical problems have always driven the process of knowledge generation, and that the origins of knowing are evolutionary. This big picture appears repeatedly in Mach's writings (and in this section).
As to the individual-level process in the quote, there is first an inexperienced individual not equipped with the specific concept(s). As more experience is gained and sensations memorized, the individual learns to recognize that different amounts matter. If other members of the community have already learned the lesson, the individual may follow them or be a teacher by showing cases with different amounts of items. In any case, one learns by experiencing multiple cases (Mach  1898: 230). Gradually, the individual starts to recognize further similarities and differences: two apples and two oranges versus three apples and three oranges and so on. Gradually, the individual arrives via intuition at the abstract concept of “number.”8 Sensations and vague ideas get reorganized into concepts (Mach  1976: 93, 99). In the individual situations that have been the basis of the process, various aspects of the situations are increasingly left without attention as the boundaries of the concept become established in the mind of the person (Mach  1919: 9–10).
Schumpeter apparently accepted all of this and even more. Mach explains in the quote above that once a concept is formed, it can be related to names. Similarly, Schumpeter ( 1997: 45) argued that new concepts in scientific work “receive labels or names in order to retain their identity.” Mach provided more detailed insights. He argued that words play at least two functions in the development of thinking: first, words can keep thoughts organized; and second, words are means of forming, adjusting, and coordinating the habits of mind across the population (Mach  1898: 411–412).
In the quote at the beginning of this section and elsewhere (Schumpeter 1908), Schumpeter does not regard concepts as the ultimate aim. As mentioned above, it is possible to ascend to “a still higher level of generalizing abstraction.” Schumpeter talked about the “engine or organon,” and about theories that are constructed out of concepts—whether related to words or not—by linking them. Mach wrote:
Just as it is biologically important to observe how reactions hang together (say the appearance of fruit and its food value), so every science aims at finding constancies of connection and of combination and interdependence of reactions. (Mach  1976: 98)
In this quote, one can recognize the instrumental nature of theories, and realize that they range from laymen's everyday observations to more abstract, qualitative, or quantitative formulations by scientists. Theorizing is not a phenomenon that exists only among scientists, but is a part of human nature.
Regarding the relations among concepts, it is worth noting that Mach was critical of the cause and effect analyses. To Mach ( 1898: 484,  1976: 204) the concept of cause was a sign of a rather primitive attempt to reproduce facts in thought. The ideas of cause and effect are developed instinctively, and “the necessity of the causal connection” is “probably created by our voluntary movements in the world and by the changes which these indirectly produce” (Mach  1919: 485). Instead, in any “profound investigation,” the investigator “must regard the phenomena as dependent on one another” (Mach  1919: 579;  1898: 253–254). In addition, “close analysis almost always reveals that the so-called cause is only one of a whole set of conditions that determine the so-called effect” (Mach  1976: 204).
Following Mach, Schumpeter (1908) was also critical of the concepts of cause and effect. In Business Cycles, Schumpeter (1939: 33n2) noted that there is “a strong argument against using that questionable term [“cause”] at all.” He also argued that in social sciences, in particular, instead of searching for final causes one should focus on interactions between multiple elements (Schumpeter  1991: 410–411). Schumpeter ( 1997: 20) also commented, with admiration, on Newton and his critique of the use of metaphysical concepts like “cause.”
If that is theory, then what is history? Mach did not develop any explicit (philosophical) account of what history and its study are. There are just scattered comments on this subject in his writings. For example, Mach ( 1919: 579–581) said that the basis for key aspects of his theoretical writings on psycho-physics was his work as a teacher and practical instructor—as a physics teacher. In addition, Mach ( 1919) practiced historical research and made use of it in his theorizing. Therefore, practical experience and history appear to be very close relatives.
Similarly, Schumpeter also had scattered abstract thoughts about historical research. In his first book, Schumpeter (1908: 41–42) wrote that both history and theory describe the world, but where history stops at making a catalogue of facts, theory continues to construct schemes that provide a better overview of facts. A few years later, Schumpeter ( 2003: 59) said that a social scientist collects facts (for theory) through daily experiences, history, ethnology, and statistics. Many years later, he regarded history as equal to “historical sense” or “historical experience” and points out that “history” also refers to various fields that have acquired different names like prehistoric reports and ethnology or cultural anthropology (Schumpeter  1997: 13). History also includes “present-day facts” (Schumpeter ( 1997: 12). In brief, the study of history is about recording the sensations and experiences of various situations and about gradually turning them into more abstract and general concepts and theories. The process is very understandable from the Machian perspective. When it comes to the practicalities of gaining such historical sense and experience, the best that Mach and Schumpeter can offer is their own example as historians.
For both Mach and Schumpeter, history is not a mere random collection of sensations and experiences, but the study of history (or fact-finding) co-evolves with theory:
Attempts at conceptualization invite the hunt for further facts and the new facts discovered must themselves be inserted and conceptualized. (Schumpeter  1997: 45)
A historian can do very little without preexisting concepts that frame the observations. A theorist has to maintain contact with the real world by various means, of which the study of history matters the most—when understood in such broad terms as discussed above. What if one insists: Which comes first—theory or history? If one considers Schumpeter's ( 1997: 41, 45) emphasis on “vision” as a “preanalytic cognitive act,” then a-priori theories based on intuition are there first. One must, however, keep in mind Schumpeter's acceptance of the Machian view of the emergence of concepts. From that perspective, sensations and history come first. Schumpeter's ( 1997: 45) term “give-and-take” is most appropriate. History and theory are dependent on one another like symbiotic organisms. In particular, development of one is dependent on development of the other.
How does statistics fit into the picture? It is useful to begin from Mach's views about the role of numbers in everyday thinking as well as in science. Mach (see also  1976: 149–151, 353) wrote about qualitative and quantitative presentations:
At closer range we notice that quantitative investigations are merely a special and rather simple case of qualitative ones. Physics has reached a higher level than for example physiology only because it deals with simpler and easier problems and because these are all more of a kind so that their solutions are more readily given in comprehensive expression. Indeed, description by counting is the simplest imaginable and can be pushed to any degree of fine and accurate discrimination by means of a number system ready to hand without need for any new invention. (Mach  1976: 239–240)
Mach ( 1976: 238–239) claimed that in physics the phenomena could be described quantitatively, but in the study of plants in biology it was difficult to do so, while chemistry was somewhere in the middle. If quantification is not possible, one has to settle for qualitative theory.
Mach treated quantitative theories as the inherent (and as the preferable) direction of the process, for at least two reasons. First, reactions and relations are presented in a highly clear and economical fashion such as the calculation of space elements from time elements in s=1/2gt2 (Mach  1976: 150). Quantitative representation is economical and clear also in the sense that mathematical representation can be taken towards finer elements and beyond the existing qualitative categories without new “nomenclature” (Mach  1976: 150). Second, with quantitative representation, complex interrelationships can be considered with little influence from the observer (Mach  1976: 240).
Schumpeter also had similar viewpoints. First, mathematics pushes economists towards clearer claims and proofs, which contribute to the productivity of the work of economists, and second, in quantitative format, economic theory can become an instrument to politicians and businessmen (Schumpeter  2008: 107). In addition to these two Machian points, Schumpeter recognized two more issues. Third, economics has always been a quantitative subject more than any other field, even physics, because some economic facts like prices would not be there unless expressed numerically (Schumpeter  2005,  2008: 100–101). Fourth, the general public only takes proper notice of arguments when they are made in figures (Schumpeter  2008: 107,  1997: 794).
Once numerical expressions are possible, statistics can emerge, which introduces two functions for theory builders. First, statistics is needed “in order to know precisely what there is to explain,” and secondly, statistics is needed “for explaining things” (Schumpeter  1997: 14). A statistician can cope with numerous cases, but apparently in other respects, the relationship between statistics and theory building is in many ways similar to that of history and theory building. When it comes to the importance of history versus statistics, the former is more important than the latter, because the statistician must ultimately appreciate the historical detail.
Like Mach, Schumpeter did not see the future of science to be entirely quantitative. In an essay aimed for economic historians, Schumpeter ( 2008: 235–236) pointed out that there are important areas where quantification has not been possible and that economists are prone to treat such facts, due to general human weakness, as “nonexistent.” In addition, Schumpeter (1939: 32–33) warned about the dangers of “nonsense induction” and “spurious verification” in statistical work.
The contents of theory, history, and statistics and their interrelations have been clarified above from the perspective of Mach's theory. If “thought experiments” and “physical experiments” (Mach  1976), and the experience gathered from them, are read in those terms, then there is a (meta) theory that describes the means with which any science is carried out.
How about economic sociology? Does Mach's work provide insights into it? Mach focused on the individual-level processes, but some of the points that he has made are highly suggestive. For example, he wrote on the opening page of his magnum opus:
An animal can withstand more intricate and less stable conditions only if it can adapt to a wider range of spatial and temporal surroundings. This requires a farsightedness in space and time, which is met first by more perfect sense organs, and with mounting demands by a development in the life of the imagination. Indeed an organism that possesses memory has wider spatial and temporal surroundings in its mental field of vision than it could reach through its senses. It perceives, as it were, even those regions that adjoin the directly visible, seeing the approach of prey or foe before any sense organ announces them. What guarantees to primitive man a measure of advantage over his animal fellows is doubtless only the strength of his individual memory, which is gradually reinforced by the communicated memory of forebears and tribe. Likewise, what essentially marks progress in civilization is that noticeably wider regions of space and time are drawn within the scope of human attention. With the partial relief that a rising civilization affords, to begin with through division of labor, development of trades and so on, the individual's life is focused on a smaller range of facts and gains in strength, while that of society as a whole does not lose in scope. (Mach  1976: 1)
As mentioned above, Mach ( 1898: 214–235) linked his views directly to Darwin's account of the origin of species.9 He regarded human knowing as an adaptation. Knowing and knowledge have enhanced the life success chances for human beings. Theories are “restrictions that under the guidance of our experience we prescribe to our expectations” (Mach  1976: 351). They can be useful instruments that consist of abstract descriptions made for some purpose(s) in the course of human activity. When it comes to the direction of the evolution, Mach ( 1919) argued that (ideally) knowledge evolves generally, and in science, to simpler forms that are more economical, i.e., fewer, more reliable, and less difficult to use. Mach ( 1976: 1; see also  1898: 190–191) noted that science and scientific knowledge have gradually emerged out of everyday considerations as the division of labor has advanced.
When Schumpeter accepted Mach's “Denkökonomie,” he apparently also approved of Mach's Darwinism. Samuelson (1951: 99, 1970: 62), formerly Schumpeter's student, made notice of Schumpeter's Darwinism. The enthusiasm about Darwin's ideas is recognizable also in the comments Schumpeter ( 1997) made on Darwin and his work. Like Mach, Schumpeter regarded theories as instruments (Shionoya 1990, 1997). Schumpeter also imposed this view onto economics. Ideally, economists would provide policy-makers with something similar to what chemistry, physiology, and biology had provided for medicine (Schumpeter  1997: 1141) and what pure mechanics had provided for machine builders (Schumpeter  1997: 1145). When it comes to the history of sciences, much like Mach, Schumpeter ( 1982: 1052,  1997: 7) defined science as “refined common sense.”
How does this relate to economic sociology? Both Darwin and Mach make it clear that human beings are highly malleable.10 The study of individuals and their behavior is the study of something highly malleable. When one studies collections of individuals, which is what economists and other social scientists do, the malleability can hardly be ignored. While a society may sometimes be treated as a machine with fixed components and relations, in other occasions it needs to be acknowledged that all the parts and their relations change. Schumpeter appears to have regarded economics as largely focused on the former type of a situation. While he found that to be acceptable, much like Menger, he considered it necessary, much like Schmoller and especially Marx, to keep change, and the latter type of situation, firmly in mind. Economic sociology was one step in that direction, but Schumpeter aimed much further.
Marx's Economic Interpretation of History as a Key to Schumpeter's Economic and General Sociology
- Top of page
- History, Theory, and Statistics from the Perspective of Mach's Writings
- Marx's Economic Interpretation of History as a Key to Schumpeter's Economic and General Sociology
- Discussion and Concluding Remarks
From the Machian perspective, the role of all scientists is to develop abstract descriptions of their surroundings that are useful in respect to some particular purpose(s). History and statistics are the means to that end. Schumpeter (1908, 1909, 1912) approached economics from this perspective in his early writings and again in his last book (Schumpeter  1997). He (1908, 1909) regarded the non-evolutionary and static (Walrasian) mainstream economic theory as a description of certain aspects of society that had its specific uses. However, beyond the narrow purposes, he thought that it was not an appropriate description of reality, and that a very different kind of account was needed. For some of those other purposes, Schumpeter introduced his own evolutionary (and somewhat dynamic) economic theory in Theorie der wirtshchatflichen Entwicklung (Schumpeter 1912) and also in the better-known abridged English version (Schumpeter 1934).11 The culmination of Schumpeter's work on economic theory is his massive Business Cycles: A Theoretical, Historical, and Statistical Analysis of the Capitalist Process (Schumpeter 1939), in which he showed how theoretical, historical, and statistical approaches could be applied in parallel.
Economic sociology seems to be missing from Schumpeter's work. This is not completely true. Economic sociology was not an integral part of his analysis, but the analysis in those writings was based on an economic sociology framework. In the subtitle of his 1939 book, the emphasis on “the Capitalist Process” must be noted. Similarly, in his earlier books, Schumpeter (1912: 4, 1934: 5) emphasized that economic theories, including his theory of economic development, only describe certain phenomena within the “commercially organized state, one in which private property, division of labor and free competition prevail.”12 Economic theories are descriptions of the economy within the boundaries of specific “institutions” that constitute capitalism (Schumpeter 1949,  1976,  1997).
The nature of economic sociology and its role with respect to economic analysis can be recognized in the definitions of those two fields. Following Gerhard Colm, Schumpeter (1949: 204,  1997: 21) argued that “economic analysis deals with the questions how people behave at any time and what the economic effects are they produce by so behaving; economic sociology deals with the question how they came to behave as they do.” According to Schumpeter (1949: 203–204), an alternative, if not fully synonymous definition, is that economics consists of interpretation and description of “mechanisms that play within any given set of those institutions,” but interpretation and description of those institutions belongs to the field of “economic sociology.” It is noteworthy that he made no mention of theory, history, and statistics. The discussion is at a different level. It is not about techniques within a science, but rather about two different sciences, where economics rests on economic sociology.
As further evidence, one may consider Schumpeter's definition of “institutions”:
By “institutions” we mean in this course all the patterns of behavior into which individuals must fit under penalty of encountering organized resistance, and not only legal institutions (such as property or the contract) and the agencies for their production and enhancement. (Schumpeter  1991: 438)
Institutions are behavioral channels that human beings constitute and that guide the behavior of each individual, directly or via the conduct of others, into certain directions as certain patterns of behavior are sanctioned by some (or almost all) agents. In other words, if economics contains appropriate descriptions of capitalist conduct (at all), such capitalist conduct exists because people collectively make each other behave in that capitalist ways. Schumpeter (1908,  1997) regarded it as acceptable that economics was built around “methodological” individualism, but did not regard individualism as appropriate in sociology (Schumpeter 1908: 92,  1997: 888). They have two different takes on society, and they are two different sciences.
The importance of economic sociology to economics is obvious considering the possibility that institutions change. Therefore, while Schumpeter approved of the limited focus of economists in their economic theories, he did not regard it as appropriate for economists to be ignorant of economic sociology. Indeed, he saw it as a serious flaw in the economists' practice that they paid so little attention to institutions and to the possibility that institutions change. He wrote about the 19th-century thinkers and about the emergence of mainstream economics:
Many writers, primarily the English theorists—such as Ricardo, James Mill, and Senior—did not bother to specify the details of the institutional frame they visualized, but took them for granted. (Schumpeter  1997: 544)
The theorists he mentioned, and J. S. Mill, who was more explicit about the institutional framework, had to some extent acknowledged the historical relativity of the social institutions, but they believed in the permanence of the capitalist order and excluded the possibility of change of those institutions from their inquiries (Schumpeter  1997: 544). Marshall had also simply accepted them (Schumpeter 1941: 244). In early 20th-century economics:
Its institutional framework was left practically untouched, that is to say, it was left in the shape in which it had been thrown by the English “classics” and in particular by J. S. Mill. (Schumpeter  1997: 886)
Of the 19th-century theorists, J. S. Mill had in Schumpeter's ( 1997: 544) view given attention to institutional change, but still held that “men could, should, and would change capitalist institutions through a rational perception of what he considered to be their defects.”
Parallel to the ignorance, there was a strong argument that institutional change was not only possible, but that it was actually happening. To “Marx alone” belonged the idea that “capitalist order is only a historical phase and bound to develop, by virtue of its own inherent logic, into something else” (Schumpeter  1997: 544). Within such perspective, it is crucial to pay attention to the big picture. At the very least, it is necessary to consider what the institutions have been in the past when a theory was established in economics and what they are at the time of the analysis. Making economists aware of the necessity of institutional analysis was clearly one of Schumpeter's objectives. Economists had to acknowledge institutions and institutional change in order to secure the connectedness of their theories to the real world—if they were doing science rather than just intellectual exercises or fictional writing.
Whether the writings of Marx were Schumpeter's primary source of insights into the possibility that economic conduct could fundamentally change is an issue that deserves further consideration elsewhere. Shionoya (1997), for example, emphasized the role of Schmoller's writings. Writings of other thinkers may have played their roles as well. In any case, Marx was not the only one to elaborate the institutional relativity of economic laws. Furthermore, the reasoning behind Schumpeter's call for economic sociology is clear enough even without considering Marx's writings.
It becomes necessary to consider Marx's writings when Schumpeter's attention to economic sociology is seen as a part of a more ambitious project. Marx did not merely notice the possibility of ubiquitous social change, but he had sketched a theory to describe the process: the economic interpretation of history. Schumpeter (1912: 98,  1993: 100, 1949,  1976: 10,  1997: 438–442) paid attention to it throughout his career. While he was not entirely satisfied with Marx's original version, he highly appreciated it. As a young researcher, Schumpeter ( 1993: 100) regarded the economic interpretation of history as possibly the greatest achievement in the sciences focused on human societies. Towards the end of his life, Schumpeter ( 1976: 10) wrote that it is doubtlessly one of the greatest individual achievements of sociology. Schumpeter ( 1991: 341) also noted that “innumerable facts can be invoked to its verification.”
What is the economic interpretation of history and why did it interest Schumpeter so much? Due space limitations, Marx's writings cannot be fully analyzed; therefore, we will limit the discussion to a mere reference to the key texts and a very brief summary of the main ideas. Schumpeter (1949: 204–205) refers to the famous Preface of the Critique of Political Economy (Marx  1904), but points out, following Engels, that Marx may have gotten the idea many years earlier, and that the frame is implicit also in the Communist Manifesto. Schumpeter's interpretation follows faithfully the famous Preface. In brief, Schumpeter ( 1997: 439) called the economic interpretation of history a hypothetical description of the way the economy or productive activities contribute to certain kinds of social relations of production, and of the ways that such arrangements support specific kinds of cultural manifestations. There is also an important auxiliary claim. Schumpeter (1912: 98,  2008: 166–167) applauded Marx for the discovery of the fact that productive arrangements evolve continuously. As Schumpeter developed his famous theory of economic development, he developed support to Marx's views. The combination of the views that economic theories describe conduct within capitalism, Schumpeter's economic analysis, and Marx's economic interpretation of history inevitably lead to the view that economists' descriptions will not remain in sync with the institutional reality.
Furthermore, the economic interpretation of history is not necessarily just about capitalist institutions. Marx did not necessarily just describe the economically relevant institutions, but also their evolution, and indeed, the evolution of society at large. Marx kept building a grand (sociological) theory of history (Schumpeter  1993,  1997). Schumpeter's economic sociology appears to have been a part of such larger (potentially general) sociological theory.
There are very clear indications that Schumpeter cherished such an ambitious vision and kept working towards its realization. The following quote shows his enthusiasm as he (1915: 54) acknowledged that many thinkers like Ibn Khaldun, Vico, and Adam Smith had attempted to build such a comprehensive theory. He wrote about Smith's considerations:
In 1785 A. Smith wrote to a friend that he had the idea of writing a “philosophical”—what can be here understood as analytical—history of mankind and its sciences and arts. A universal theory of humans and society and all changes of life (forms) of which history narrates! What a plan—surely never as doable as now, indeed, this insight is as comforting as the one that the joys of youth were just foolishness. Also Turgot has planned similarly and such plans buzzed then in the air … Not as dreams of dilettantes or as speculations, but as concrete intentions of strong, prudent minds. What a challenge, what a legacy! (Schumpeter, 1915: 55, author's own translation)
A sketch of such a big theory, built out of the same key ingredients as Schumpeter's own theory of economic evolution, was already included in Schumpeter's (1912: Ch. 7) first book. It was left out of the later editions, but Schumpeter did not forget the idea. For example, Schumpeter wrote that in his own economic theory there are elements that could be used in “what may be called the theory of cultural evolution” (Schumpeter 1934: xi) and in “a much larger theory which applies to changes in all spheres of social life, science and art included” (Schumpeter 1939: 97). A few years later, he said in an interview that his life's work had always been in “comprehensive sociology” (Schumpeter 1944, quoted in Swedberg 1995: 531; and in Shionoya 1997: 308). I suspect that13 Schumpeter was silent about the early sketch, because he became increasingly convinced of the usefulness of Marx's theory as the starting point.14
An attempt to reconstruct the big theory is well beyond the scope of this article, but Schumpeter's definition of (general) “sociology” provides some further insights into it and functions as further support to Schumpeter's continued work towards the big theory. In Schumpeter's ( 1997: 25–26) view, (general) sociology was “… the general analysis of social phenomena such as society, group, class, group relations, leadership and the like.” Schumpeter kept developing his views on such topics throughout his career and he also wrote about topics like “imperialism,” “state,” and “ideology.” There is considerable overlap between his work and that of Marx. His views on those areas often differed considerably from Marx's views, but it is possible to see steps towards a theory that considerably resembles Marx's grand vision.
Schumpeter was not satisfied with the state of such theories. While he recognized that economists could use sociological descriptions, he hesitated to promote them openly. Schumpeter promoted history as a way to rectify the economists' ignorance of institutions. In the following quote, Schumpeter lists reasons why he, if he were a young student of economics, would focus on history prior to theory and statistics:
First, the subject matter of economics is essentially a unique process in historic time. Nobody can hope to understand the economic phenomena of any, including the present, epoch who has not an adequate command of historical facts and an adequate amount of historical sense or of what may be described as historical experience. Second, the historical report cannot be purely economic but must inevitably reflect also “institutional” facts that are not purely economic: therefore it affords the best method for understanding how economic and non-economic facts are related to one another and how the various social sciences should be related to one another.[*] Third, it is, I believe, the fact that most of the fundamental errors currently committed in economic analysis are due to lack of historical experience more often than to any other shortcoming of the economist's equipment. History must of course be understood to include fields that have acquired different names as a consequence of specialization, such as prehistoric reports and ethnology (anthropology). (Schumpeter  1997: 12–13)
The study of history secures that the potential uniqueness of the institutions gets recognized. The footnote related to the asterisk in the above quote appears to address economic and general sociology:
Owing to the unreliability of “theories” on this subject, I personally believe the study of history to be not only the best but the only method for this purpose. (Schumpeter  1997: 13n4)
Until proper sociological theories emerge, history remains the best substitute. Besides, the study of history could produce such sociological insights. Ultimately, there could be abstract sociological theories and research programs such as the ones that emerged in natural sciences and in economics, and that could be described in the Machian terms discussed in the previous section.
Discussion and Concluding Remarks
- Top of page
- History, Theory, and Statistics from the Perspective of Mach's Writings
- Marx's Economic Interpretation of History as a Key to Schumpeter's Economic and General Sociology
- Discussion and Concluding Remarks
The discussion that is placed under the title “techniques” in the History of Economic Analysis seems to contain at least two kinds of messages. First, there is the view of scientific practice in any field of science. Theory, history, and statistics can be found or can be expected to emerge in every field of science even if the naming of those approaches differs across sciences. While Schumpeter drew insights from many sources, the logic of his thinking can be unlocked with Mach's writings. The kind of approaches that Mach had found in well-established sciences can also be followed in other sciences. In many respects, such an approach is also natural. Still as a mature researcher, Schumpeter ( 1997: 16, 34) explicitly pointed out that the differences between natural and social sciences are limited to the limited opportunities for social scientists to do experiments and to the possibility of social scientists to access the meanings of actions.
Second, while Schumpeter promoted a largely unitary view about scientific aims and methodology, he also wrote:
But it is true that “economic laws” are much less stable than are the “laws” of any physical science, that they work out differently in different institutional conditions, and that neglect of this fact has been responsible for many an aberration. (Schumpeter  1997: 34)
As ways of avoiding such aberrations, Schumpeter urged economists to study history and economic sociology. If that were to be done, there would be awareness of the conditions in which the economic theories can be expected to provide broadly useful limitations to expectations. His hidden agenda seems to have been the development of a sociological theory that would describe the evolution of the conditions in which the economic mechanisms described in economic theories are expected to function.
These points have relevance in at least two respects. First, the article provided clarity into the discussion placed under the title “techniques” in History of Economic Analysis and into Schumpeter's views about the practices that are actually followed and preferred across sciences. In that sense, the techniques of theory, history, and statistics are one thing. It is important to recognize that Schumpeter applied a similar approach that had been followed in natural sciences to social sciences and that he did it with great success. So far, surprisingly little attention has been given to Mach's texts as a potential source of insights into Schumpeter's work. From our perspective, it is crucial to continue the discussion on that topic in order to understand Schumpeter's research practice, and as a result, to expose some of his key accomplishments.
Economic and general sociology are another thing. Economic sociology was placed confusingly under “techniques” in History of Economic Analysis. Its role has remained unclear. For example, Shionoya (1991: 200) proposed that in Schumpeter's research program, economic sociology is a bridge between history and theory. Shionoya (1997: 51) also proposed that Schumpeter was going to build his analysis of institutions on neoclassical economics. I strongly disagree. Economic sociology is in Schumpeter's program a part of general sociology and outside of the boundaries of economics. Even though it has not been possible to enter reconstructive work, the acknowledgment of institutions by Schumpeter and his definition of it, already demonstrate that his sociology was not based on the same individualistic foundations as neoclassical economics. His advocacy of Marx, whose starting point was the critique of the attempt to base social theory on rational individuals (Schumpeter  1997: 438), also supports my point. The arguments by Swedberg (1995) and Dahms (1995) that Schumpeter did not try to build a grand sociological theory appear to be equally problematic. How far Schumpeter got with the project remains a relevant question, but his aim and his vision were always there.
When it comes to the content and relations of the three techniques and the content of economic and general sociology, this article has hopefully increased our understanding of Schumpeter's work. In this respect, this article may serve the purposes of researchers interested in Schumpeter's work and in intellectual history involving Schumpeter in one way or the other.
Second, a better understanding of Schumpeter's ideas about these topics may help improve modern research practice in economics and other social sciences. There has been a lot of controversy over the last 100–150 years about the question of whether social scientists can follow the same approaches as natural scientists. Schumpeter provided one solution to these debates. He argued that the topics of social sciences should be approached much like natural scientists have approached their topics. This point may easily be regarded as scientism—as copying of approaches from other sciences. That need not be the case. As Schumpeter ( 1997: 16–17) emphasized, the similarity of approaches may simply be due to the scientific approaches having been first established in the natural sciences. Schumpeter's perspective encourages one to ask whether in many social sciences there has been too much emphasis on the differences, and whether such a viewpoint has impeded progress in social sciences? Schumpeter's views may also support increased cooperation between natural and social sciences.
At the same time, Schumpeter made clear certain peculiarities about social sciences. Above all, he acknowledged that unlike many natural sciences, social science researchers can be certain that the subject matter keeps on changing. Hence, if one wishes to build laws like those in some sciences studying nature, it has to be acknowledged that even if those laws would be somewhat successful at one time, in one place, and among specific people, different laws may exist later on, in different places, and among different people. Attention to the context is vital. This message was aimed especially to the economists who had started to ignore history and sociology. Schumpeter still presents the challenging question to economists: Is sufficient attention given to the context in terms of history and economic sociology? A related question is: Should it not be taken into account that the descriptions (of the past) may become prescriptions (of the future) and change the world? To elaborate, if economics describes mechanisms within capitalism, is the promotion of such worldviews already not present as policies promoting capitalism and specific interests? Developers of curricula, researchers, and users of researchers' advice ought to consider such questions.
Finally, Schumpeter's vision of general sociology that underlies his comments on economic sociology and his views on the historical relativity of economic laws may deserve further attention. As Schumpeter recognized, there have been many attempts to develop a comprehensive sociology to describe social change and history's evolution. Before Schumpeter's time, Marx had made the most advanced attempt, which Schumpeter took as his starting point. During Marx's lifetime, many new ideas that Marx could not yet exploit had been developed. The work of Darwin is perhaps the most important example. Also the few comments of Marx on it and the work of Mach deserve acknowledgment. They were all sources of inspiration to Schumpeter. Still, Schumpeter did not get there. Since his time, further advances have been made on various fronts. The time may have come again to consider whether Schumpeter's goal ought to be revived. Schumpeter and his key sources may have a lot to offer in terms of achieving it.