Shackle and Other Readings on Uncertainty

I’m doing additional work on knowledge at the moment, and found this wonderful little reading list on Shackle and others. I tend not to look at the world as certainty and uncertainty because I think certainty is a chimera. (That is, it’s a label we use to describe some belief that is sufficiently calculative to tip a willingness to take action. Or, as an associate of mine says, “information is what is necessary to cause a decision.” Calculative exhaustion arises where risk is sufficiently less than reward.)

I am left wondering whether I (we) have a better understanding of time and how it can be derived from certain kinds of data that contain theoretical periodicity, rather than collecting data according to an arbitrary periodicity, than do other people working on this problem.

I have to re-read the general theory. I just didn’t see his point the first time. I think it may be a language problem. I have a very hard time with people who think excessively in statefulness.

This is, fundamentally, what bothers me about the Anglo-analytical tradition in philosophy as well: there is simply something wrong there in the use of time. I just can’t quite get my arms around it.;jsessionid=B6E1E18127EE82763A7213ED65302873?Filename=Published/EmeraldFullTextArticle/Pdf/1600020202_ref.html

Alan Coddington, “Creaking Semaphore and beyond: A Consideration of Shackle’s ‘Epistemics and Economics.'” The British Journal of the Philosophy of Science, 26 (2), June 1975.

G. L. S. Shackle, “The Romantic Mountain and the Classic Lake: Alan Coddington’s Keynesian Economics” reprinted in Shackle, Business, Time and Thought: Selected Papers of G. L. S. Shackle.”

G. L. S. Shackle, “Sir John Hicks’s ‘IS-LM’: an explanation’; a Comment,” reprinted in Shackle, Business, Time and Thought.

Alan Coddington, Keynesian Economics: The Search for First Principles. George Allen & Unwin, 1983.

Backhouse, R. A History of Modern Economic Analysis, Blackwell, Oxford, 1985.

Barreto, U. The Entrepreneur in Microeconomic Theory. Routledge: London, 1989.

Baumol, W. (1968), “Entrepreneurship in economic theory”, American Economic Review, Papers and Proceedings, Vol. 58 No. 2, pp. 64-71.

Bausor, R. (1986), “Time and equilibrium”, in Mirowski, P. (Eds), The Reconstruction of Economic Theory, Kluwer-Nijhoff, Boston, MA.

Buchanan, J. (1982), “The domain of subjective economics: between predictive science and moral philosophy”, in Kirzner, I. (Eds), Method, Process and Austrian Economics, Lexington Books, Toronto.

Carter, C., Meredith, G., Shackle, G. (Eds) (1962), Uncertainty and Business Decisions, Liverpool University Press, Liverpool.

Casson, M. (1982), The Entrepreneur: An Economic Theory, Robertson, Oxford.

Casson, M. (1987), “Entrepreneur”, in Eatwell, J., Milgate, M., Newman, P. (Eds), The New Palgrave: A Dictionary of Economics, Macmillan, London.

Dobb, M. (1953), “Entrepreneur”, in Seligman, E., Johnson, A. (Eds), Encyclopaedia of the Social Sciences, Macmillan, New York, NY.

Drucker, P. (1985), Innovation and Entrepreneurship, Heinemann, London.

Ford, J. (1983), Choice, Expectation and Uncertainty, Robertson, Oxford.

Ford, J. (1990), “Shackle’s theory of decision making under uncertainty: a brief exposition and critical assessment”, in Frowen, S. (Eds), Unknowledge and Choice in Economics, Macmillan, London.

Gilder, G. (1986), The Spirit of Enterprise, Penguin, Harmondsworth.

Hebert, R., Link, A. (1988), The Entrepreneur.

Hebert, R., Link, A. (1989), “In search of the meaning of entrepreneurship”, Small Business Economics, Vol. 1 pp.39-49.

Julien, P. (1989), “The entrepreneur and economic theory”, International Small Business Journal, Vol. 7 No.3, pp.29-38.

Kirzner, I. (1973), Competition and Entrepreneurship, Chicago University Press, Chicago, IL.

Kirzner, I. (1992), The Meaning of the Market Process, Routledge, London.

Kregel, J. (1990), “Imagination, exchange and business enterprise in Smith and Shackle”, in Frowen, S. (Eds), Unknowledge and Choice in Economics, Macmillan, London.

Loasby, B. (1989), The Mind and Method of the Economist, Edward Elgar, Aldershot.

Loasby, B. (1990), “The use of scenarios in business planning”, in Frowen, S. (Eds), Unknowledge and Choice in Economics, Macmillan, London.

Meredith, G. (1962), “Methodological considerations in the study of human anticipations”, in Carter, C., Meredith, G., Shackle, G. (Eds), Uncertainty and Business Decisions, Liverpool University Press, Liverpool.

O’Driscoll, G. Jr, Rizzo, M. (1985), The Economics of Time and Ignorance, Blackwell, Oxford.

Philips, L. (1988), The Economics of Imperfect Information, Cambridge University Press, Cambridge.

Reekie, W. (1984), Markets, Entrepreneurs and Liberty: An Austrian View of Capitalism, Wheatsheaf, Brighton.

Ricketts, M. (1987), The Economics of Business Enterprise.

Shackle, G. (1943), “The expectational dynamics of the individual”, Economica, May, pp.99-129.

Shackle, G. (1955), Uncertainty in Economics, Cambridge University Press, Cambridge.

Shackle, G. (1958), Time in Economics, North Holland, Amsterdam.

Shackle, G. (1962), “Concluding comment”, in Carter, C., Meredith, G., Shackle, G. (Eds), Uncertainty and Business Decisions, Liverpool University Press, Liverpool.

Shackle, G. (1965), A Scheme of Economic Theory, Cambridge University Press, Cambridge.

Shackle, G. (1966), The Nature of Economic Thought, Cambridge University Press, Cambridge.

Shackle, G. (1968a), Economics for Pleasure, Cambridge University Press, Cambridge.

Shackle, G. (1968b), “Policy, poetry and success”, in Shackle, G. (Eds), On the Nature of Business Success, Liverpool University Press, Liverpool.

Shackle, G. (1973), An Economic Querist, Cambridge University Press, Cambridge.

Shackle, G. (1979), Imagination and the Nature of Choice, Edinburgh University Press, Edinburgh.

Shackle, G. (1984), “To cope with time”, in Stephen, F. (Eds), Firms, Organization and Labour, Macmillan, London.

Shackle, G. (1988), Business, Time and Thought, Macmillan, London.

*Shackle, G. (1969), Decision, Order and Time in Human Affairs, Cambridge University Press, Cambridge.

*Shackle, G. (1970), Expectation, Enterprise and Profit, George Allen and Unwin, London.

*Shackle, G. (19??), Keynesian kaleidics;: The evolution of a general political economy.

*Shackle, G. (19??), Epistemics and Economics: A Critique of Economic Doctrines.

Shand, A. (1984), The Capitalist Alternative, Wheatsheaf, Brighton.


The Best Interpreter of Keynes, April 8, 2008
By James F. Mueller “Lachmanniac”
This book is composed of five lectures on Keynes’ economic program. Shackle is a brilliant economist who writes extremely well. Many economists disagree over what Keynes really meant, but very broadly we can divide them up into neoclassical keynesians and postkeynesians, with Shackle belonging to the latter category. But his interpretation of Keynes’ work is interesting because he goes back to Keynes’ early work “The Treatise on Money” to argue that the clearest articulation of his radical message can be found there. Most post-Keynesians argue that it was only with “The General Theory” that Keynes broke away from “neo-classical” economics (and the assumptions of full employment and neutral money) and incompletely at that.

Shackle concentrates on two important themes of Keynes’ work: uncertainty and liquidity preference. The entire body of Keynes’ work can be reduced to these two concepts. Investment, Shackle argues, is determined by the expectations of entrepreneurs which is vulnerable to unpredictable streams of “bad news” causing them to withdraw from the field and leaving resources unemployed. This is made possible by uncertainty. Liquidity preference can also explain the existence of money and the need to hold onto it in the face of uncertainty.

In this book, Shackle provides a lot of fascinating quotes from Keynes. The message, however, is stated very simply in the beginning and becomes repetitive by the third and fourth lectures. While this does serve to reinforce the message behind Keynes’ writings, I think Shackle could have done a great deal more in this book by elaborating on these fundamental points.

For anyone interested in the “radical interpretation” of Keynes’ economic program, Shackle is the best source to consult (along with Paul Davidson and Hyman Minsky).


James F. Mueller “Lachmanniac”
Location: St. Louis, MO USA

Michael Emmett Brady says:

I have to disagree. Shackle was not Keynes’s best interpreter by a long shot. Shackle defined uncertainty as complete ignorance. He was extremely clear on this point. There is either uncertainty or certainty. There is nothing in between. The same position has been argued for 40 years by Shackle’s disciple, Paul Davidson and practically all of the Post Keynesians and Post Autistic economists. The Post Keynesians and Austrians are in agreement on this point. Keynes, on the other hand, rejected this view out of hand in both the A Treatise on Probability (TP 1921;pp.310-315) and in his Feb., 1937 Quarterly Journal of Economics article, titled “The General Theory of Employment.” Keynes viewed uncertainty as having different degrees or grades. You could have complete certainty, slight uncertainty, mild uncertainty, moderate uncertainty, severe uncertainty, acute uncertainty and then total or complete uncertainty (ignorance). Keynes restricts the applicability of complete ignorance to the investment decision in long lived capital goods alone.

The best interpretation of Keynes’s GT is Meltzer’s 1989 book. The best advance made on Keynes’s approach in the TP is in D Ellsberg’s “Risk, Ambiguity, and Decision” (2001; 1962 dissertation). It is tragic that Ellsberg waited 40 years to publish his work.

James F. Mueller says:
Mr. Brady,

Thank you for your thoughtful response. You are absolute correct in writing that:

“Keynes viewed uncertainty as having different degrees or grades. You could have complete certainty, mild uncertainty, moderate uncertainty, severe uncertainty, acute uncertainty and then total or complete uncertainty (ignorance). Keynes restricts the applicability of complete ignorance to the investment decision in long lived capital goods alone.”

Shackle quoted as much from Keynes in several of his books (Keynesian Kaleidics, Scheme of Economic Theory and Years of High Theory). But the real project for Shackle always was to develop more fully Keynes’ ideas, which he believed to be muddled and incomplete. Shackle is probably the most competent champion of Keynes’ 1937 QJE article and the only post-Keynesian to ever speak favorably of his 1930 book. But more importantly, Shackle was convinced that mainstream economic theory was essentially a theory of “an orderly and reasonable world, … a world where what we intend is what will happen.”

Shackle, it must be remembered, believed that Keynes only hinted at a theory of “disorder.” No admirer of Shackle’s work will ever suggest that he thought Keynes had everything right. For example, Shackle thought that Keynes was never really able to distinguish ex ante and ex post language as well as Myrdal, and believed that Keynes’ ch. 7 in his 1936 book was mistaken in its analysis of a comparison of equilibrium positions.

Let me say that when I write that Shackle is Keynes’ best interpreter, I mean to suggest that Shackle is the most interesting (and extreme) interpreter of Keynes’ work. I also think that Shackle took Keynes’ ideas very seriously and tried to study them “in isolation.” Harrod, Kaldor, Robinson, Minsky, etc., on the other hand, all had their own agendas, and co-opted for the most part Keynes’ analysis into their project of building long-term growth models influenced initially by Kalecki and Marx. Shackle is the only one willing to take the implications of uncertainty seriously by arguing that such a fact goes “beyond the reach of legislation or improvements of organization and technology.” In other words, I think Shackle is the only economist to consistently follow the implications of radical uncertainty. And the fact that Keynes’ did not believe in this is no strike against Shackle’s efforts. Shackle recognized Keynes as an important thinker in the “economics of disorder,” but never meant to suggest that the theory of complete ignorance begins and ends with Keynes’ work.

Michael Emmett Brady says:
OK, we agree. There are major differences between Keynes and Shackle. However, you must understand that Paul Davidson assumes that Keynes’s and Shackle ‘s views are the same. The Post Keynesians DO NOT accept the concept of uncertainty coming in different grades or gradations. They fall back on Shackle’s own words which I am paraphrasing – Uncertainty is the opposite of certainty. UNCERTAINTY MEANS YOU ARE NOT CERTAIN. There is nothing in between certainty and uncertainty.

This, of course, leads to complete nihilism. The Post Keynesian school is doomed intellectually because it does not have a solid foundation to deal with uncertainty as a range. Radical uncertainty only has import in decisions involving innovation/long run capital investment. Any attempt to put it at the center of decision making leads to intellectual chaos. Perhaps you will get a chance to read Ellsberg’s 2001 book on “Risk, Ambiguity, and Decision”. He also takes Keynes’s ideas on uncertainty very seriously, although he did overlook Keynes’s technical analysis in the TP.

I believe that the main difference between us is that I do not think Keynes’s ideas and analysis “… were muddled and incomplete”.

James F. Mueller says:
Mr. Brady,

I would like to just make one comment. When describing Keynes’ general outlook, Shackle attached great importance to a particular passage in the QJE 1937 article. Immediately following Keynes’ articulation of uncertainty (which every post-Keynesian knows by heart), Keynes proceeds by writing:

“Nevertheless, the necessity for action and for decision compels us as practical men to overlook this awkward fact and to behave exactly as we should if we had behind us a good Benthamite calculation of a series of prospective advantages and disadvantages, each multiplied by its appropriate probability, waiting to be summed.”

Keynes then gives three examples of ways in which we cope with this “awkward fact” and concludes by writing:

“Not a practical theory of the future based on these three principles has certain marked characteristics. In particular, being based on so flimsy a foundation, it is subject to sudden and violent changes. The practice of calmness and immobility, of certainty and security, suddenly breaks down. New fears and hopes will, without warning, take charge of human conduct.”

…..This is the point on which Shackle developed his remarkable theory. Any pretense to knowledge or certainty is merely a convention, as Keynes put it, or an illusion, as Shackle aptly wrote. Moreoever, Keynes suggested in his 1937 article that the more one presumes knowledge and certainty, the more likely his expectations and plans are going to be violently disappointed by the chain of certain events. Here is Shackle:

“The deliberate self-deception of business in supposing its investment decisions to be founded on knowledge and to be rationally justifiable; the insecurity of its faith in its own judgments, which the awareness of this self-deception engenders; the paralysis of decision and enterprise which can result when the structure of pretended knowledge is violently overthrown by events; this central core of the General Theory is to be found in chapter 12…”

There is no question that Keynes spoke about uncertainty in terms of grades and gradations. But Keynes also wrote that the pretense of “complete certainty”, as you put it in your first comment, is simply an illusion or convention people devise in order to preclude them from falling into a state of paralysis upon the acceptance of the fact that the future is fundamentally unknowable. Moreoever, as Keynes writes, the stronger this sense of certainty is, the more likely things are going to turn sour.

I think this is what “operationalizes” Shackle’s research program. Shackle explicitly acknowledged the point that Keynes spoke of uncertainty “as having different degrees or grades.” But I think Shackle would add that the ‘pole’ of complete certainty is merely a convention that is created by men desperately seeking to behave rationally. It is important to remember that Keynes was first of all concerned with uncertainty and ignorance and only then, through the back door, introduced the concepts of rationality and ‘mild’ and ‘moderate’ uncertainty in order to illustrate the various ways in which people try to cope with an inescapably uncertain world.

Michael Emmett Brady says:
You really need to carefully work your way through, at least, chapters 15, 17, and 26 of the TP. Chapters 15 and 17 are the technical counterparts of the interval estimate approach to probability that Keynes introduces in chapter 3 of the TP and chapter 26 is the counterpart to chapter 6, where Keynes introduces the weight of the evidence (argument). It is in the TP that Keynes provided his overall technical apparatus that underlies his discussions of uncertainty in the GT. You realize that neither Shackle nor Davidson ever read the TP? Post Keynesians can barely grope with chapter 3 alone. Instead, they commit a major error in trying to interpret the GT through what they think is going on in the Ramsey-Keynes debate. Ramsey, in fact, based his two reviews completely on a misinterpretation of the words “non numerical probability ” as contained in chapter 3.

We can use Keynes’s index to measure the weight of the evidence, w, from p.314 of the TP. w measures the knowledge base upon which the probability estimates are estimated. w ranges between 0 (ignorance) and 1 (complete certainty of knowledge). Shackle’s system can only deal with w=0. Keynes’s system deals with the full range – w=0, 0<1 and w=1. This is not a convention. Keynes presents a system for calculating lower and upper limits for probabilities. He refers to these as “non numerical”. Shackle is not generalizing Keynes at all. He is presenting a special case of Keynes’s analysis, although, never having read the TP, he did not realize this. Keynes does not introduce his grading of uncertainty through the backdoor. He assumed that a reader of the GT would realize that weight and uncertainty are inversely related.

Yes. I agree with Shackle that the core of the GT is in chapter 12. What Shackle overlooked is that chapter 12 of the GT is an application of chapter 6 and 26 of the TP.

Note that without any technical apparatus to deal with decision making under uncertainty you are essentially doomed to end up with a nihilist position. Neoclassical economists and Bayesians have been knocking down this position with regularity for 60 years. Unless Post Keynesians can show technically how uncertainty, as opposed to risk, leads to involuntary unemployment they will always come up a loser against a technically trained neoclassical or Bayesian in any exchange about unemployment problems.

Your approach appears to be a standard Post Keynesian handling – you give some quotes from the GT to start. Unfortunately, this is usually where the Post Keynesian ends also. Post Keynesians are unable to advance beyond this point.

James F. Mueller says:
Mr. Brady,

Thanks again for the instructive comments. You are correct in pointing out that both Davidson and Shackle made little to no mention of Keynes’ Treatise on Probability (Minsky, however, did successfully connect Keynes’ notion of uncertainty in the General Theory to his TP). It is also true that post-Keynesians have for the most part written about uncertainty with too much ease and, as a consequence, applied it unscrupulously. Davidson’s Money and the Real World chapters 2 and 6, for example, consists mainly of assertions supported by occasional references to Shackle, Knight and Keynes. Although I do think Shackle is worlds apart from other post-Keynesians when it comes to uncertainty, you are correct in pointing out that Shackle understood uncertainty as w=0 (weight of the evidence covering only complete ignorance).

However, the real question centers on the nature of a “knowledge base” and the extent to which it can be used to discover the “weight of evidence” in formulating probability estimates. In other words, what does it mean to be in possession of knowledge? From where does this knowledge arise, and according to what criteria can it be considered as such? These are questions that I do not think Keynes’ system can answer. Mathematics is, after all, essentially a system of formalized, unambiguous, mechanical processes for manipulating symbols on pieces of paper. They can in no way illuminate the essence of what is trying to be explained or understood.

Shackle understood knowledge and meaningful action as being “originative” and “spontaneous.” He contrasted this understanding of “knowledge” with more mechanistic models containing mutually determining equations. Your interpretation of Keynes’ system is clearly “pretty” and “polite”, and made for “a well-paneled board room and a nicely regulated market”, but it must be remembered that these “techniques” are, as Keynes put it, “liable to collapse.”

I cannot help but think that your understanding of a “knowledge base” is nothing more than:

“a tacit, instinctive convention to believe in the rightness and reassertive power of things as they have known them, broadly, to be.”

Shackle, following Keynes, would argue that knowledge (w=1) is illusory. The fact that our expectations can collapse so suddenly and without warning is proof enough of the precariousness of presumed certainty. Now this statement is indeed nihilistic. But in order to meaningfully challenge it, I think you will have to explain in great detail what knowledge really means and consists of. Does it consist of past experience? And if so, can this be useful for action which necessarily takes place in time? Or is knowledge instead whatever produces practical results? And so on…

But I do think you have provided a great service to post-Keynesian scholars. You are extremely well-read and have developed a unique interpretation of Keynes’ work. I would just remind you that a lot of this discussion is centered around a host of unsettled answers which many philosophers are still grappling with today — the nature of human rationality, the existence of knowledge, etc.

Michael Emmett Brady says:
You will find your answers about Keynes’s views on how knowledge is obtained in Part III of the TP. This is the heart of the TP. Keynes argues, convincingly, that knowledge comes about by means of induction and analogy. Cognitive psychologists over the last 40 years have arrived at Keynes’s conclusions about “pattern recognition” and “intuition” without realizing that all of their conclusions are all in Part III of the TP.

I know of NO Post Keynesian economist who has carefully read the TP. Post Keynesians “read” (actually they browse these chapters) chapters 3, 4, and 6 of the TP and that it is. There is NO Post Keynesian economist who has any understanding of Keynes’s interval estimate approach to probability, which is being, unknowingly, duplicated by “modern” decision theorists working with Choquet capacities, such as Gilboa and Schmiedler.

A fundamental problem in connecting Keynes and Shackle is that you do not seem to realize that Shackle rejects the TP approach in toto. Shackle rejects Keynes’s logical theory of probability, in fact all theories of probability, and he rejects Keynes’s theory of evidential weight. I am assuming that you realize that Shackle also rejects induction and analogical comparisons because of his belief that all moments of time are unique. Shackle defines “knowledge” as certainty. “Unknowledge” or uncertainty means that there is no knowledge. Keynes, on the other hand, links his discussions in chapter 12 (chapter 17) of the GT about uncertainty to his discussions of weight in chapter 6 and the much more powerful discussions in chapter 26. There is always some kind of knowledge for Keynes, although it may be scanty, little, vague, conflicting,”…our knowledge of the future is fluctuating, vague, and uncertain…”, etc. The only time in print that Keynes said we have NO knowledge (we simply do not know) is one time in the Feb., 1937 QJE article. You, like Shackle, rely on only one small piece of evidence upon which to base your conclusions. Where does Keynes ever make this kind of statement again, in comparison to Shackle, who made the argument literally thousand of times in his published work ?

In my opinion,you face the following two choices. You can follow Keynes or you can follow Shackle (Davidson). Perhaps you can finesse this problem. However, it looks to me that Post Keynesians are trapped by their emphasis on uncertainty as total ignorance. How can their policies be shown to be superior to neoclassical ones? Shackle’s answer is that they can’t and that the free market (and ONLY the free market) will spontaneously generate a solution over time. This is the Austrian answer. I’m not sure that this is the answer that Post Keynesians are expecting.


Leave a Reply