Joseph Alois Schumpeter, (1883-1950) an Austrian political economist. At 36 he was the finance minister of Austria and eventually became a professor at Harvard. The main task of his article “Development” was to identify some aspects of novelty which he saw as the source of economic development.
Schumpeter begins his article on development with some caveats. He means to be clear that he is talking from a purely economic perspective. However, he recognizes that this article could be interesting to other fields of knowledge but warns that his insight came from working through the process of economic development. Another caveat is the notion that what “develops” has to maintain some sort of identity during the development – it may be helpful to not require the continuous identity of the object that develops. Two further notions that we need to disavow ourselves of when trying to understand Schumpeter’s concept of development are: 1) a faith in progress and 2) evolutionism.
“Faith in progress implies a positive valuation of changes. Precisely because it implies valuation, it has no right of place in science.” Given that the effects of development can have on people’s lives, it is not too hard to see why some would have this faith and why others would reject it. However, in studying development we must not be for or against development, rather we should recognize it as a phenomenon worthy of examination. Evolutionism in the 1930s was fashionably associated with materialism, the notion that matter is the fundamental substance of nature; that there is no mind, only brain; no soul or spirit only body. Materialism is for Schumpeter a “metaphysic” today it means that it is an ideology, that is a simple and speculative explanation, not rigorous or based on real science. It seems that in the 1930, very much like today, evolution was thrown around without really meaning anything precise.
There are three scientific perspectives through which Schumpeter would like to look at development: 1) the process, 2) the causes, and 3) the goal. Firstly, to scientifically study the development process, we would need to “investigate the concrete characteristics of events at different points in historical time and the differences among these characteristics.” Like a moving image made up of many still images. Secondly, to scientifically study development causes we need to “identify concrete circumstances, in the best case measurable quantities which can be said…to have ‘caused’ the change.” Like a historical narrative explanation. Finally, “apply the notion of goal in a nonmetaphysical way, and for the purposes of a particular science. Teleological aspects…are not taken into consideration….” We must never expect to find the correct end or purpose of development, but for simplicity of expression we can say that the result of development was a goal and in that sense we can speak of the goal of development: reaching the expected results.
Consider a thought experiment bound by time and space: 13th century Florentine painting can be recognized as an “imprinted form” linked in many aspects. The same can be said of 15th century Florentine painting and yet the fact that they are linked by time and space does not account for the differences. “No list of identifiable environmental elements will suffice to clearly determine how a particular change actually took place. Rather, artistic creation…could have turned out in a different way; the process could have reacted differently to influences from the environment…. the appearance of novelty as such always gives rise to indeterminacy.” This indeterminacy is independent of causal connections. Finding causal connections is what we do best and doing so does help overcome certain horrors originating in not knowing. For this reason, causal explanations can easily be abused and accepted as the single true explanation when in fact they are nowhere near true nor the single explanation. Accepting indeterminacy is hard in the face of theory that gives plausible causal explanation and the fact that there is a lot of determination all around any tiny bit of indeterminacy. Schumpeter defines a change or “development” that is causal and therefore determinable within a causal theory’s explanation as properly called an adaptation, not a novelty. Adaptation is predictable given a known change in data or situation.
“Novelty changes the previously considered matter and substitutes it with another one that reacts differently to changes in the data. In a different sense, it might still be possible to interpret [novelty] as an adaptation, but not a passive and determined one. From the perspective of any adaptation-theory, novelty is incomprehensive” Novelty cannot be understood from a causal perspective. Furthermore, searching for causes will not help you find novelty: “the true core of everything that must be accepted as indeterminate in the most profonde sense.” Schumpeter, repeats that novelty is surrounded by determinacy. The environment can cause many changes but not the ones that are caused by novelty. It may be that proximate causes may be necessary but they are not sufficient. The environment may make it possible for a novelty to appear, but it cannot be said to have caused the novelty. “The change transmuting one imprinted form into another one must represent a crack, a jerk, or a leap if the problem that I tried to identify should arise. When starting from the old form, the new one must not be reachable by adaptation in small steps.”
Schumpeter, in his description of deterministic phenomena uses a linear algebra analogy. Given that all deterministic relationships can be expressed in a mathematical formula and given that all the formulas can be put into a matrix that can be manipulated with matrix algebra and be expressed in the form a vector, then all possible variations both forward and backward in time can be calculated. This vector, produced from all the components of the deterministic relationships, Schumpeter calls, the norm of the economy. It represents where the economy has been and where it can go. Continuous changes in the norm of the economy are simply deterministic linear transformations. “a change in the norm can be interpreted as an adaptation of the economy to continuous changes in data. In practical terms we are talking about small changes in data at each time point.” This is how all deterministic systems work and how their past and futures can be calculated. However, “this procedure fails where a leap-like change of the norm occurs.” An example: “A continuous increase in population and wealth immediately explains an equally continuous improvement of roads and an increase of the mail coaches in circulation in a step-wise adapting manner. But add as many mail coaches as you please, you will never get a railroad in that way. This kind of ‘novelty’ constitutes what we here understand as ‘development,’[:]…transition from one norm of the economic system to another norm in such a way that this transition cannot be decomposed into infinitesimal steps. In other words: Steps between which there is no strictly continuous path.” Development is different from growth. Growth is merely more of the same: increases that are continuous, predictable and quantifiable. Development, by contrast is something really new, discontinuous, unpredictable, and not immediately quantifiable.
For Schumpeter, economics is a special science for two reasons. First, because its main measure, price, is expressed directly in numerical terms and thus is exactly precise on its own. All other sciences must impose numbers on to their domains of study to achieve precision. This means that the observer is imposing one more outside schema on the phenomena and increasing the inherent possibility of measurement error. Secondly, because it expresses itself in numerical terms, economics is, in principle, clearer in terms of the distinction between determinate and indeterminate systems. Economics is, therefore, the best science to study the emergence of novelty. The hard sciences like physics and chemistry are precise because they study deterministic systems; the so-called soft sciences, like economics and sociology, have entities from which novelty emerges and which disrupt their deterministic systems.
Novelty, because it is outside the determinability of a deterministic system cannot be studied in a purely and strictly scientific way. Science works best studying a deterministic system. The scientific method relies on a degree of repeatability which is unavailable in instances of novelty. Novelty seems impossible to study with any precision in quantitative terms. We can, however, benefit by studying the deterministic systems that surround novelty and once identified, we can then observe the results of novelty unconfused about what is deterministic adaptation and genuine novelty.
To study novelty, there are five things that can be done in qualitative terms. We can, 1) document and catalog instances, 2) “We can observe and describe the jerks and leaps in detail. [3)] We can estimate their importance to the phenomena of each domain. [4)] we can comprehend the effects and counter-effects they trigger, not only in descriptively but also theoretically…. [5)] We can…identify the entry points of novelty not only in the specific case but also generally, and thus build a theory of the mechanisms that are involved.” Schumpeter, however, did not believe that an adequate study was ever made in his lifetime.