P2: Methodology for (Re)-Reading Keynes

aaeaaqaaaaaaaahhaaaajdq5nmzhzwvllwmxn2utndg0yy05mtg2lwe5ymqxzjhhmji0nqThe first post on Reading Keynes provided an outline of the reasons why this is a good idea. It is clear that economics is broken. We need a new macroeconomics for the 21st century, one which can solve the massive problems which humanity as a whole is facing on political, social, economic, and environmental dimensions. Keynes faced similar problems, and found solutions which guided economic policy in the mid twentieth century. It is always useful to absorb the insights of our predecessors, before trying to build upon them. Such a methodology is essential for the advancement, progress and accumulation of knowledge. Our current stock of human knowledge is based on the collected insights and labors of hundreds of thousands of scholars, accumulated over the centuries. We would return to the stone ages if we were to reject it as being full of contradictions and errors (which it is). Instead, progress occurs by absorbing the past accumulated wisdom, and trying to remove the errors, or add missing insights, building on our heritage, rather than discarding it and starting over from scratch.

Several of the central Keynesian insights into the causes of the Great Depression never made it into the economics textbooks. However, our goal in studying Keynes goes far beyond just the re-discovery of these lost Keynesian insights.   A central goal is to apply and illustrate a radically different methodology for studying economics in particular, and social science in general. This is derived from a study of The Methodology of Polanyi’s Great Transformation. This is an extremely important point, which we proceed to amplify and explain further.

1.       Problems with contemporary economic theory arise from a fundamentally flawed methodology, which is incapable of learning from real world experiences. As Romer said, macroeconomics has gone backwards for the last several decades. This is because the methodology currently in use does not lead to progress and accumulation of knowledge. Very briefly, this is because current methodology is the Axiomatic-Deductive Methodology of Greek Geometry, which was never successful in dealing with natural phenomenon. Instead, what is needed is scientific methodology as practiced and demonstrated by Ibn-ul-Haytham. Unfortunately, logical positivism created massive confusions and misunderstandings regarding scientific methodology, which persist to this day, despite the fact that logical positivism has been rejected.

2.       Why have modern economists adopted and practiced a deeply flawed methodology? This is a complex and tangled tale, but its origins lie in the Battle of the Methodologies (Methodenstreit) in the 1890’s. In this battle, the German Historical School of Schmoller, which advocated a contextual and historical approach lost to the Austrian School of Menger, which advocated a more scientific, mathematical and a-historical methodology. The details and consequences have been explained at length in “How Economics Forgot History” by Geoffrey Hodgson. As a consequence of the economists’ search for scientific laws which are universal invariants, economic theorists have invented an artificial world of maximizing robots without history, culture, institutions, and social norms.The process of economic modelling — learning to think like an economist — involves translating economic problems to this artificial world and then calculating the results. This can be done because all the robotic agents behave in predictable ways, and the environment is sterilized of all particular historical, social, environmental elements. However, most often, economic outcomes in this artificial world bear no resemblance to outcomes in the real world. Mistaking a highly distorted map for the territory, economists are confused when real world phenomena do not match the results of their models.

Some of the key methodological issues which we will try to develop in this re-reading of Keynes are highlighted below:

3.       Theories cannot be separated from their historical context. Thus Keynesian theory can only be understood within its historical context. We cannot understand Keynesian theory as a collection of principles and/or mathematical laws, taken out of context and understood to apply to all economies across time and space. When placed within it historical context, Keynes becomes much easier to understand.

4.       Even more important, theories interact with history. Human being formulate theories in order to try to understand and explain changing social circumstances. When circumstances change rapidly, theories are devised to understand the change. These theories, whether right or wrong, are used to  respond to changes, and thus end up shaping history. From this perspective, it is important to study Keynes, regardless of whether his theories were right or wrong, because economic policies from mid-twentieth century onwards were guided by his views. Thus Keynesian theories have shaped economic history. There is a complex interaction of theories and history, and we cannot understand history without theories, just as we cannot understand theories without their historical context.

5.       Because of the central importance of point 4 above, we provide a simple illustration to clarify it. As described in greater detail by Polanyi, the process of enclosures of common land deprived the masses of access to livelihood and created poverty on a large scale in England. Large numbers of authors described the problems and searched for causes of this phenomena. However, the analysis of Malthus, which blamed the problem on the excessive fertility of the poor, came to dominate. His theories deeply influenced the Poor Laws, and the British response to poverty, and thus millions of lives. Even though Malthusian theories about the arithmetic increase of food and the geometric increase of population were empirically incorrect, we must understand Malthus to understand the economic policies and circumstances of England at that time.

Accordingly to widely accepted methodological principles underlying the development of modern economics, theories are formulated without historical context. In addition, economics is studied in isolation from politics and society. We propose to study Keynesian theories within their historical context. This will substantially enrich our understanding of Keynes. In addition, the historical context includes the political, social, and economic environment, which will allow us to see that economic events cannot be studied in isolation, since all these dimensions of human lives interact with each other. Again our approach goes against a core methodological commitment of modern economics, which insists that economics can be separated from political and social circumstances and studied in isolation.

HOMEPAGE for Re-Reading Keynes. Links to more material on Methodology

  1. Thank U sir for your another masterpiece.The main problem that hinders the progress of knowledge is that, “when one believes that he knows every thing about the problem”. I think “The Black Swan” from Nassim Nicholas Taleb is a good anecdote to realize ones own faults about knowing everything.

  2. David Chester said:

    It is clear that economics is broken. We need a new macroeconomics for the 21st century, one which can solve the massive problems which humanity as a whole is facing on political, social, economic, and environmental dimensions. Keynes faced similar problems, and found solutions which guided economic policy in the mid twentieth century. It is always useful to absorb the insights of our predecessors, before trying to build upon them. Such a methodology is essential for the advancement, progress and accumulation of knowledge.

    And wonder of wonders such a new macroeconomics is already sitting on your doorstep!

    My new book on macroeconomics theory will allow you to explore any change in policy that you may wish to introduce. This book provides a sold analytic numerical method for understanding how our macroeconomics works. It is titled “Consequential Macroeconomics–Rationalizing About How our Social system Works” and I will send an e-copy to those who will ask it of chesterdh@hotmail.com and who tell me why they are interested.

  3. You begin the post with an interesting assertion, “We need a new macroeconomics for the 21 century, one which can solve the massive problems which humanity as a whole is facing on political, social, economic, and environmental dimensions.” If economics is the science that deals with the production, distribution, and consumption of commodities; the branch of knowledge concerned with the production, consumption, and transfer of wealth. In other words, economics and economists study, research, investigate how people produce, distribute, and consume commodities and wealth. These two pursuits are contradictory, so which is the purpose or justification for economics? Non-economists invent ways to make, distribute, and consume commodities and wealth. Then economists study and explain this work. It seems, to me at least that economists propose being both scientist doing the research and the subject of such research. This is related directly to my second concern. The post is filled with mentions of theory (macro theory), all created by economists. Where is the study of the theories of the actors involved in creating economic commodities and wealth? Again, economists seem bent on being both theorist and scientist studying the theories. This is not even bad science. It is no science.

  4. I have mentioned that the project of “Social Science” is misconceived. There is no way to study human beings scientifically, since each life is completely unique. I have mentioned that economics CANNOT be studied in isolation from politics, society, environment, culture etc. Furthermore, as should be clear from context, we are not studying Keynesian theory from the point of view of learning economics, we are studying it as an outsider — in a mata-theoretical ways. How did this theory originate from the crises created by history, and how did the theory itself impact and change history? This complex interaction between the subjective and the objective — “theories” and the object of these theories — is not possible within a scientific framework of neutral detached observation.

    • Let me see if I get what’s you’re saying. You are proposing a program to study the history of a theory and related theories, all centered around the economic theories of Keynes. It is not a scientific proposal. Your main focus is the creation of the subjective and objective through and with the interaction of these theories. And the results of those processes of creation. Is all this correct?

      But your view of science in my view is not correct. Science is not neutral or detached observation. In fact, it’s just the opposite. It’s observation from an ever widening number of stand points. Each new stand point adding to the picture from our observations.

  5. Your first paragraph is correct. I believe that in “social science” the observed and observer interact in essential ways — our attempts to understand our history (theories) impinge directly on our responses to this history, and hence shape history. This means a radically different approach to methodology is required to study human beings and societies.

    Regarding your second paragraph, my understanding of science is what is conventionally understood about methodology of science. I have my own views (different from yours) which I will express later. However, here I am concerned with the fact that economists use “scientific methodology” as they understand it, not as you or I understand it. I have referenced this in one of my articles that a survey shows that most of them are still Logical Positivists — they believe this misunderstanding of scientific methodology. Postivism clearly separates objective and subjective, and posits the scientist as a neutral detached observer, and that their is no interaction between the theorist and the subject matter of the theories.

    • Thanks for the additional clarifications. You and I seem to be on the same path in paragraph one. But if you look closely I think you’ll find that radically different methodology is even more radical than you think. Look at John Law, After Method: Mess in Social Science Research.

      As to paragraph two, I arrived at my view of science by watching scientists. With some exceptions scientists spend their time studying the same concerns from various perspectives (e.g., the climate scientist examines climate via laboratories, field sampling, paleontology, the lives of animals (including humans), astronomy, etc.) and then assessing one another’s work from each of the perspectives. Whatever name you give this process, it is the work of scientists. Philosophers of science may argue how philosophy and science fit together. Scientists do not. According to Richard Feynman the scientific method involves: 1) guessing a new law (unchanging relationship); 2) compute the consequences of the guess; 3) compare the computed consequences to experience (experiment, observations); 4) if the guess disagrees with experience, then it’s wrong; 5) move on to new guesses; 6) if the guess agrees with experience, it is accepted as likely right; 7) as a result the scientist can prove any given guess wrong (within existing consensus in science) but can never prove any guess right (again, within existing consensus in science). This is how scientists spend their lives. Feynman also points out that a vague or unclear guess cannot be compared to experience. The guess must allow the computation of consequences that can be compared to experience. My major argument with Feynman is that he seems to assume the comparison step is simple and straight forward. I’ve observed enough scientists at work to recognize that the debates and struggles over making comparisons can go on for decades and in some instances for centuries.

  6. Your view of scientific method is based on nominalist ideas going back to Bacon, but is deeply mistaken. The correct, realistic, view of scientific methodology has been explained by Manicas at book length. Very briefly, perhaps cryptically, the standard views of science are based on induction (Bacon — Feynmann) deduction (Positivism, currently dominant philosophy of science among economists) — but NEITHER is a correct understanding. Scientific method is based on abduction (Pierce) and transdiction (Manicas) = just as these terms and their meanings are unfamiliar to most, so the true scientific method is unfamiliar to most.

  7. Realism is marginalized for a good reason. It doesn’t work. While they abbreviate, compact, and often rearrange the process Feynman describes, scientists (social and physical) generally stay near to that path. A scientist has an interest (e.g., parental neglect causes juvenile crime, climate change denial is the result of political and religious beliefs, the human eye evolved in stages, changes in human perception result from chemical changes in the brain), states that interest (usually for a funding proposal or academic requirement), and then begins the search for evidence (experience) to support that interest. Observing these scientists shows them looking in every corner with as many tools as they can find and afford for bits and pieces of experience to support their interest. And then working like hell to assemble these into a gestalt that supports their interest and other scientists will accept. The bits and pieces used and the gestalt created for an interest vary with scientist, with time, and with other related experiences. Per realism, the real is just real. Science does not then deal with the real. But with the bits and pieces of experience that can be accessed and the gestalts that can be constructed from these experiences, and tested by other scientists. As I’ve said, the philosophy of science is not science. Scientists utilize abductive reasoning in much of their work. Don’t know if any of them know that’s its name. Later in his life Feynman offered the view of scientific method I prefer. It is whatever works. Everything scientists work with is uncertain, reversible, unclear, reflexive, and difficult to assess – it is complex. If we want to study these subjects the methods must fit the subject. Realism can’t do that. Neither can deduction or induction, or any other approach voiced by philosophers of science. Law’s book offers some options.

  8. Asad, you need to distinguish Roger from Francis Bacon here. Francis’s method was taking things to bits to see how they work, but Hume took the established view back to Roger, distorting his ‘induction’ from agreement between observations to agreement between observers on the significance of observations, thereby letting in conflicting and mischievous motives for agreeing.

    Manicas looks interesting. I haven’t been able to find an explanation of ‘retrodictiction’ as against retroduction, which is the focus of Bhaskar’s Critical Realist philosophy of fundamental science as against the practical adequacy sought in ‘normal’ practical science.

    Looking at your methodology of Ibn-ul-Haytham, I found this, “Greek axioms and logic had led to two rival theories about vision, which had remained deadlocked for 800 years. One way of framing axioms led to the conclusion that light emanated from the eyes and struck the object, while the other led to the reverse conclusion. Ibn-ul-Haitham used observational evidence to definitively settle the matter”.

    Francis Bacon’s method then takes a further step, taking US to bits to see how WE work inside. Given light comes from outside (which Hume denies we can know), what happens inside? Light is demonstrably not communicated from the retina into the memory. What is consistent with what does happen is logically equivalent to what happens in the automatic focussing of a digital camera. We remember the settings of our senses when they are focussed, and restoring the settings tunes in to enough [random] energy to make us dimly see the structure in what was originally present. Whatever that was! I’m now using the analogy of a TV displaying what’s going on in the world. Social science needs to be studying how TV systems work (and how well: c.f. the developments in system design from Baird’s mechanical scanning to analogue and now high definition digital electronics). Not what’s on, as in the probability that the news will be on at six o’clock. Applied to the re-reading of Keynes, I’ll say again, the advance is from steam engine thinking to the beginning of electronic thinking, as in our economic system having not just a built-in boiler thermostat but also a wired-in room thermostat (with the possibility, incidentally, of different rooms each having their own thermostat to eliminate inequalities between the entertainment rooms and the kitchen).

    Ken, I’m afraid, is waffling again. He’s too busy watching television to wonder what goes on behind the screen.

  9. Ken: I am surprised by the comment that “realism does not work”. DO you mean that you do not believe in electrons and gravity?

    Dave Taylor: A fundamental flaw — i dont know where it entered western thought — is the objective/subjective distinction. When I taste a fruit, the chemical properties of the fruit are objecive external stimuli, but my taste buds are personal subjective elements, and then there is the interpretation of stimuli received within the brain. External objective stimuli interact with the hardware in our bodies and the software in our brains to create the world we live in. As Putnam puts it — the objective and subjective are inextricably entangled. The Enlightenment attempted to put facts and reason on a pedestal and rejected the subjective, which led to an extremely defective theory of knowledge, which continues to be extremely harmful to mankind. All of the most precious achievements of human beings, our self-knowledge, spirituality, love, courage, sacrifice, are subjective and removed from the body of “scientific” knowledge.

    • No, Asad, your taste buds are objectively yours, and it is the taste – the interaction between the stimulus and your senses – which is subjectively yours, because our senses differ (as is more obviously seen in people who – like a friend of mine – are colour blind). I don’t agree with Putnam. The objective and subjective are not “inextricably entangled” any more than North-South and East-West are. They form a complex, but it is not complicated. Nor do I agree with what you say about the Enlightenment. Hume effectively did with facts what our government recently did with marriage: redefined the word ‘fact’ for scientific purposes as our government did [gay] ‘marriage’ for legal purposes, and thereby redefined ‘science’, which originally and more generally means knowledge or the acquisition of it.

      Hume’s ‘facts’ are subjective because, he claimed, all we can know is what we have experienced in our consciousness. Kant’s rejoinder about knowing necessary interpretive concepts was on the right lines; not in his day knowing how the senses work (and that things represent themselves) he couldn’t see how Nature has built interpretive concepts into our bodies in much the same way as a computer maker builds at least a loading program into the wiring of his hardware. The interpretation I’ve given you above shows how what we dimly remember is what is objectively there to be seen, even if we don’t detect all of it; and the effect of hallucinatory drugs on the brightness of such internally generated images is evidence for the mode of operation I’ve proposed, which is intelligible even to those who don’t understand the internal physical processes. The fact that an image can be digitised, transmitted as waves and reconstituted on a TV screen is surely clear evidence that a message is not altered by the encoding of it, even though what we see is. One just has to know (by their having become built in) the procedures necessary for interpreting it, as in the right keywords for an encrypted letter. [I’m saying this as a scientist who has been involved with computer hardware and its programming, including development and computer-interpretation of humanly intelligible programming languages, since 1957, when the original computers were still around; and with the physiological psychology of perception since 1964. I’ve been saying more or less what I’m saying about Hume since I first encountered him in a philosophy of science course in 1958].

      Despite Hume, we know “all of the most precious achievements of human beings, our self-knowledge, spirituality, love, courage, sacrifice”, and we know of them by looking at our world in the Francis Bacon way, not by measuring them and agreeing on the measurements. The difference between science and non-science, I argue, is one of motivation. A scientist is looking for new knowledge or understanding of new discoveries, whereas an engineer wants to use his scientific knowledge but may need to become (or seek the aid of) an applied scientist to correct the detail of it.

    • It’s not an either/or situation. Ian Hacking and William James make more sense to me. In the words of Hacking,

      “There is only one way in which my thesis is contrary to a bundle of metaphysical doctrines loosely labeled ‘realist.’ Realists commonly suppose that the ultimate aim or ideal of science is ‘the true theory about the universe.’ I have never believed that even makes sense. …. Our perceived theories and the world fit together so snugly less because we have found out how the world is than because we have tailored each to the other.”

      In the particular world in which we happen to live, scientific inquiry has, as a matter of fact, arrived at a set of particular conclusions, and created an empirical reality to match. There is really no other option since inquiry, scientific or otherwise is conducted always in specific social and cultural contexts. Realist metaphysics fails to pay attention to actual events.

  10. So in different social and cultural contexts, electrons might fail to exist, and gravity might fail to operate? To the best of my knowledge, Hacking started out as a nominalist, but eventually converted to realism — your quote might be from his pre-conversion period. Or he may have flopped again — i havent been following the scene for a while now.

    • In a word, yes. Hacking is an interesting guy. Like most philosophers he loves to hear himself talk. But he makes good points. Hacking, like many others have worked hard to get their heads around the notion of “social construction,” or just “construction in interactions.” Two starting points are fundamental. Humans exist within certain social and cultural contexts. Second, humans interact with one another and nonhuman “things” (experience). From this all human knowledge arises, and all human culture and society. My view of this is close to that of Andrew Pickering. Who says this (in a letter or maybe email) to Hacking. “I would never say that ‘Constructing Quarks’ is about ‘the idea of quarks.’ That may be your take on constructionism re the natural sciences, but it is not mine. My idea is that if one comes at the world in a certain way — your heterogeneous matrix — one can elicit certain phenomena that can be construed as evidence for quarks.” As Hacking notes physicists on the other hand view the quark solution as inevitable. Can this dispute be settled? I don’t see how at this point. To paraphrase William James, I prefer to believe constructionism.

      • If I look, I see what is there (if through smoky glass, darkly). That’s not me constructing what I see, it is m being able to decode what’s there.

      • Dave, you’re welcome in this metaphysical debate. But don’t get too cocky. This is not a debate you can win. You say, “That’s not me constructing what I see, it is me being able to decode what’s there.” Okay, now prove it!

      • Ken challenges me to “prove it”. In three lines? But words are not the proof. The proof is that despite Hume, things happen; and they happen in the same [i.e. equivalent] way in a brain as they do in an electronic communication system; and that I can show, but not in three lines.

      • Dave, you take a 1000 lines or 100 book volumes. But with all that you cannot escape that direct comparisons between theoretical constructs (i.e., gravity is a theoretical construct) and observations is impossible. For every observation or set of observations there are an unlimited number of theories that explain these observations. And when you contend that the scientist must then exercise judgement to select the correct theory for the observations, you negate direct comparison of theory with observations. And it is impossible to confirm that each observation or set of observation is equivalent to every other observation or set of observations. Again, which is the correct observations? No way to figure that out except by the judgement of the the scientist. Thus endeth the lesson.

  11. I think there is every possibility that what you observe today may be different from what one experienced yesterday.

  12. There was no point in continuing this argument with Ken, as he is not willing to concede he could be wrong. However, as Asad is still struggling with the issue at P4 of his discussion, let me suggest anyone interested read Roy Bhaskar’s “A Realist Theory of Science”, in terms of which Ken is perpetuating Hume’s “epistemological fallacy”, presenting the argument in terms of how one acquires knowledge, when the issue is “ontological”: whether there is anything there to know. In terms of my TV analogy, whether the TV set is “constructing” the pictures we see, or whether it is merely displaying those which have been previously transmitted in non-pictorial form. Of course one cannot reliably predict what producers will broadcast, but one can predict the shape, colours, picture quality, refresh-rates etc possible with a given type of TV system. Ken’s committed to the first view; I’m committed to the second, because I know how TVs work.

    At root, Ken’s commitment derives from Hume’s 1740’s axioms, which assume linear Cartesian coordinates and consciousness of what we can see changing: this corresponding to an infinite universe containing pre-existing matter. Mine derives from 1940’s Big Bang Einsteinian axioms in which energy exists and is transmutable into matter, this corresponding to an expanding and evolving universe analogous to the Biblical story of Creation.

  13. At no point, have I contended that I could not be mistaken. But still you show me nothing that addresses the riddles of how humans know anything. Or, why humans assume there is something there to know. I’ve not read Hume in years, but the one thing I carried away from reading his work is that he did not take causality for granted. In my view, causes are created in the interactions of humans and the experiences they perceive. Apart from assuming causes are given by a divine being or through some magic they are revealed to humans, this is the only origin for them I can figure out. If you have another, please share it. As far as scientific axioms go the “expanding and evolving” universe is interesting. Scary for some, supportive for others. But still it remains a theoretical construct. And observations, used by scientists to assess such constructs cannot be directly connected to the constructs or to one another. If you know a way around these problems that doesn’t involve simply assuming them away or skipping over them, please share. My solution is to believe that humans can and do make judgements about how certain experiences are linked to others, as well as how observations either confirm or disconfirm these judgements. I focus much of my work on how these judgements are formed and used. My final words on these issues, unless something new and interesting is offered. Thanks for the recommendation for Bhaskar’s book. Read it. Not impressed. Why should science want to return to Kantianism? Something it’s been trying to escape for over 100 years.

  14. Asad,

    On a small point: My reading of Keynes is that he is not claiming any absolute for ever theories, but only adequate partial descriptions of how things were then, relevant to the main issues at hand. So anyone who lifts a ‘theory’ from way back then and tries to apply it now is not in any meaningful sense following Keynes. Rather, I think we need to look at his overall theorising, including his Treatise, and see how his lines of thinking would work out now.

    This leads us back to consider the significance of the above comments. It seems to me vital to think such abstract thoughts, but then – like Keynes – to bring it all back to our concrete needs. Good luck!

    • Concur with your overall perspective. As human generations come and die reality changes from just that evolution. Each generation creates a new, or several new realities. And part of those realities are theories that explain those realities. Realism is out the window. Relativity is in.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: