Published as: Zaman, A (2016) “Positivist Misconceptions: An Obstacle to Understanding Pluralism” International Journal of Pluralism in Economics Education, Vol 7, No. 1, pp 93-96.(IJPEE Version) SSRN Version http://ssrn.com/abstract=3370707 – More articles on Logical Positivism and its disastrous impact on economics & human knowledge.
At the heart of the problems we face in the world today is a disastrously wrong theory of knowledge. The vast number of errors within this theory of knowledge have been deeply imbibed by the populace, and have become part of the background assumptions under which we operate. These assumptions are not even open to examination and challenge, because they are buried underground as foundations for our thought structures. I would like to examine some of these misconceptions and explain why these create insurmountable obstacles in the path to our quest for the truth.
A Central Enlightenment Myth: Facts and Logic are sufficient to lead us to the truth.
As a corollary, our opinions do not matter. Human beings, their thoughts, anxieties, concerns, as well as personalities, are all irrelevant in the quest for truth. Subjective elements cannot and should not be part of scientific theories. Also, science, scientific knowledge is all there is – genuine knowledge is impossible outside the realm of what is objective. In opposition to this idea, I would like to put forth the following:
The facts by themselves are never enough to lead to the truth.
Just one among many reasons for this is because causality is not observable, and causality is an essential component of the truth we are seeking. Given a sequence of events – for instance a revolution – we can only guess at causes, because there are multiple factors all of which happened simultaneously, and we cannot isolate the cause or causes from irrelevant observable factors. At the same time, understanding of reality requires isolation of the cause.
Many different ways of analysis all lead to the same conclusion. To learn about causality requires an analysis of “what might have been” – what would have occurred if a particular factor had been changed. This is unobservable by definition. According to standard positivist theories of meaning, such analyses are meaningless, because all references to unobservables are inherently meaningless. Note that this conflicts violently with our intuition that is it perfectly meaningful to contemplate what might have happened if Napoleon had won at Waterloo, and there can be sensible analyses as well as wrong analyses about this hypothetical unobservable. It is this type of analysis about unobservables that leads us to clues about causality, which always remain intelligent guesses, and can never be translated into certainties.
Although positivist theories have been proven wrong, many misconceptions generated by them continue to remain embedded in our thought processes. There are three positivist conclusions from the argument that causality is unobservable and hence meaningless, all of which are seriously wrong:
- Myth: Causality does not matter for our analysis, since it is unobservable. As I show in my paper on “Methodological Mistakes and Econometric Consequences,” this is what happened in econometrics as discussions of causality faded from the literature (before making a recent comeback). Truth: Causality is all important for understanding.
- Myth: Knowledge about causality is impossible, since it is outside the realm of the observable. Truth: In our daily lives, we work with a huge number of unobservables using guesswork. This is unavoidable. Learning to make good guesses about how other people think and feel is one of core requirements of our existence as social beings. However, internal states of others are inherently and permanently unobservable to us.
- Myth: We can replace causality by an observable, and this will suffice for our purposes; for example, Granger Causality uses observable predictability to replace causality. This device is similar to the replacement of unobservable preferences by observable choices – equally fallacious. For a detailed discussion of this fallacy, see Zaman (2012). Truth: Unobservables, like electrons, gravity, etc. cannot be parsed out of language, and be replaced by observable manifestations. This is not immediately obvious, and it was when all attempts to do so failed that it was realized by philosophers (but not by economists) that positivism was wrong.
This is just ONE illustration of a general failure of positivism, that it cannot deal with unobservables. In fact the vast proportion of human knowledge is based on unobservables. Even given the impossible, namely complete information about historical events, we would not be able to isolate causal relationships, and could only guess at them. This is because human motivations are unobservable but are vital drivers of history. Furthermore, to understand history we need to know not only what actually happened, but what might have happened. With a limited collection of recorded historical facts, we cannot even approach the absolute unique objective truth that is the goal of positivism. Even within the physical sciences, it is now well established that there is a fundamental ‘under-determination’ of theories by facts; see Stanford (2016) for a detailed discussion. Just as we can fit any number of curves to a finite collection of points, so we can fit any number of theories to any finite (or even infinite) collection of observations.
ANOTHER positivist misconception is to think that this (under-determination) does not matter. This is because the only job of theory is to explain the observations. All theories which fit the data (which save the phenomena) are equally good. There are many ways to prove that this is wrong – all theories may fit equally well till now, but only one of them may work when an unexpected event occurs tomorrow. To which the positivist will respond that if there are no observable differences between the theories, then it is beyond the reach of human possibility to assess which theory is better and which is worse. BUT THE FACT IS that we do have knowledge of unobservables, and use this knowledge all the time; indeed, we cannot get along without it. It is unobservable possibility of a successful revolution which drives struggles for freedom from oppression. But our knowledge about unobservables is NOT certain and can never be certain. It is this quest for the impossible, the search for CERTAIN knowledge that lies at the heart of the difficulties with positivism.
ALL THIS IS OF VITAL IMPORTANCE IN UNDERSTANDING the complete failure of economic theory today.
“Social Science” is an attempt to understand the human experience. This is necessarily historically contingent – that is, it is an attempt to understand a particular historical experience of a particular group of people. This leads to two conclusions which are highly unpleasant to positivists:
- All theories are theories about some particular historical experience of some particular group. While some aspects of these theories MAY generalize outside of the historical context, we can never know this in advance from a prior considerations. This will always be an empirical matter. Try out a theory developed elsewhere, and see how will it fits in a different historical context. We cannot make logical deductions to arrive at such judgments. And in general, it is safe to assume that theories WILL NOT generalize outside of their historical context. Universal invariant social science theories [hence SCIENTIFIC theories] are limited in number and scope and power. Such theories exist, but they will not offer much assistance in understanding issues which are heavily influenced by local historical, cultural, and geographical context.
- Theories are never determined by the observable facts. Especially in face of newly emergent situations (which keep occurring regularly in human history), people invent theories to explain them and to be able to respond to them. Since facts can only be interpreted within the context of a theory, it is inevitable and necessary for us to invent theories. We choose theories from among a menu of available past theories, suitably modified, as well as using the powers of our imagination to invent new theories.
This is why theories MUST be considered within their historical context, and cannot be understood outside of this context. That something as obvious as this needs to be stated and clarified shows the powerful grip of the “Deification of Science” that was the topic of my paper in an earlier issue of IJPEE (Zaman 2015). Similarly, the first two chapters of Appleby et. al. (2011) shows how the Enlightenment generated powerful myths about the absolute truth and validity of science, scientific law, and scientists, which have had a firmly grip on the public imagination until recently.
According to positivist views theories are either true or false – no historical context is needed to understand theories. Scientific theories are objective, invariant across time and place, and hold whether or not there are any observers. The misconception that economic theories are scientific suggest that economic theories hold equally in the eighteenth century as in the twenty first, among the Aztecs as well as Communist China, and even hold if there are no humans around to observe them! Even though this is absurd, a recent survey by Hands (2009) shows that economists continue to subscribe to this positivist conception of economic theory; this is why to them pluralism seems obviously wrong – there is only room for one truth.
Once theories are seen as products of historical circumstances, and localized by time, place, and culture, pluralism follows as a natural consequence. Different societies have different norms, and different ways of arranging social, political and economic affairs. For example, the Cherokee Constitution of 1839 states: “The lands of the Cherokee Nation shall remain common property”. In a society where land is common property, and nature provides adequately for all, laws of economics were substantially different from those of the neighboring colonial states. Similarly, as Polanyi (1957 , Chapter 4) discusses, non-market societies had vastly different arrays of arrangements for distribution of goods to members.
The narrow perspective that there is only one true economic theory substantially distorts our understanding of the history of economic thought. Instead of viewing mercantilism as a response to the demands created by a transition from self-sufficient societies to excess producing societies, we think of it as a primitive and mistaken understanding of economics. Taking theories out of their historical contexts prevents us from understanding the economic interests of different groups and how theories were fashioned in the context of struggles between competing interests. The pluralist perspective views the evolution and change of theories as appropriate responses to changing economic circumstances, with each theory being suitable and appropriate within its historical context. This allow for a far deeper understanding of theories as compared with the non-pluralist view that earlier theories were a sequence of mistakes by incompetent theorists, which were sequentially corrected and improved to arrive at perfection.
Appleby, Joyce; Hunt, Lynn; Jacob, Margaret (2011). Telling the Truth about History (Norton Paperback)
Hands, D.W. (2009) ‘Philosophy of Economics’ in Gabbay, D., Thagard, P. and Woods, J. (Eds.), Handbook of the Philosophy of Science, Elsevier, Oxford.
Polanyi,K. (1944 ) The Great Transformation, Beacon Press, Boston.
Stanford, Kyle, “Underdetermination of Scientific Theory”, The Stanford Encyclopedia of Philosophy (Spring 2016 Edition), Edward N. Zalta (ed.), forthcoming URL = <http://plato.stanford.edu/archives/spr2016/entries/scientific-underdetermination/>.
Zaman, A; (2012) “Methodological Mistakes and Econometric Consequences,” International Econometric Review, Sep. 2012, Vol. 4, Issue 2, p.99-122.
Zaman, A. (2015) ‘Deification of Science and its Disastrous Consequences,’ International Journal of Pluralism and Economics Education, Vol. 6, No. 2, pp. 181-197