Taking Hayek Seriously

Darwin, Galileo, Mayr, Copernicus, Edelman, Newton, and Hayek all rethought the very fundamentals of their sciences, from the problems and explanatory strategies of their disciplines, to the logical status and conceptual role of the elements of their activities. Yet among these, only Hayek is attacked by professors for being a “philosopher” — and attacked as not being a “scientist” — for having done so.  And only the work of Hayek is constantly smeared by the professors as “philosophy” and not “science”.

Yet Hayek is hardly the only economist who has worried about the explanatory fundamentals of economics and the conceptual status and role of the elements of their explanatory practices. Indeed, economists are recurently agitated about their explanatory failures and the self-evident conceptual hand-waving they do when attempting to justify their cognitive practices to themselves, other economics, and to the public at large.  The list of figures in economics who have done this begins with many of the leading figures of the discipline today and extend to almost every leading economist across the history of the discipline.  For example, former IMF economist Oliver Blanchard has been attempting to defend his explanatory program in macroeconomics and his vision of the conceptual role of his preferred math models here, here, here, here, and here. Esther Duflo used her Ely lecture at the AEA to argue for conceiving economics not as science like physics, but as a field of blue-color engineering like plumbing.  Harvard economist Dani Rodrik has written a book trying to explain to economists and the public what economists are doing and what they are in fact justified in doing with all of the math modeling and statistical work.  World Bank economist Paul Romer has launched an ongoing attack on the dominant explanatory practices current among many of his follow economists here, here, and here, and has offered a defense of his own explanatory practices and conceptual preferences here, here and here.  The response to Romer has been almost endless, with responses and replies coming from across the discipline of economics, for example here, here, here, here, here, here, here, and here.

There a some big issues involved here, almost none of them any different from the ones that induced and inspired Hayek to rethink the explanatory problems and strategies of economics and the conceptual role of its various logical elements.  What I want to do here is lay out how economics constantly generates these problems, how Hayek has successfully resolved these problems, and why and how the natural of mathematical and logic, combined with our conceptual traditions in economics and our philosophical tradition, have blocked economists from finding their way out of conceptual fly bottle that sciences like Darwin’s explanatory science of biology have managed to escape.  In the way will will resolve some linguistic issues concerning different uses we can make for central but conceptually brought words “philosophy”, “empirical”, “science”, “causal”, “explanation”, etc.

Larry Wright shows us how everyday reasoning and argument differs substantively from those points in our practices where we are advised to take time to deliberate and rethink what it is we are doing and what we might do instead. There can be all sorts of things which inspire us to deliberate and rethink things.  Thomas Kuhn has written an important book and a number of essays on the topic as it arises in science. If you read closely the work of Darwin, Newton, Galileo, Edelman, Mayr, and Hayek etc you will find that this rethinking often requires a rethinking of the background tradition held among men of learning about what it is to do science or create knowledge.  Newton, Darwin, Galileo, Mayr, Edelman and Hayek all discuss how their explanatory efforts defy and redefine both traditional and contemporary norms of what it is to do science and produce knowledge. Is science suppose to be demonstrative and not hypothetical, is science suppose to be induced from facts and justified by direct evidence, does it work with and support to Platonic kinds and Aristotelean forms, conceptions and categories? Newton, Darwin, Mayr, Galileo, Edelman and Hayek were prepared to defy all of these demands for what science and knowledge must be.  What we will find in economics is a new set of dogmas, developed in the wake of Kant and the neo-Kantians, which continue to ensnare the minds of economists.  Hayek, we will show, suggests a way out, one similar in important way to that taken by Darwin, and one which was seen already by Adam Smith, Carl Menger and in various ways by other giants of economic science.

But for now lets get back to simpler issues and more concrete examples, using Esther Duflo Ely lecture and economist Beatrice Cherrier’s discussion of that lecture as firm and easily understandable bone to chew upon. Cherrier points out that economists have repeatedly sought to conceive their science on the picture of what they imagine to be the practices found in other fields, from physics (here, here, and here), morals and art, to medicine , dentistry, and plumbing.  What Duflo is working to do, it to encourage here peers to move from what is today the dominant physical picture of economics, to a more practical, hands-on conception of economics as the everyday, craftsman-level engineering of a plumber, water-system technician or pipe-fitter.  So what are these two models and where do they come from?

Significantly, in large measure, these models of “physics” and “plumbing” do not come from the actual practices of physics or plumbing, they come from what economists imagine to be the practices of physics and plumbing, shaped by their training in elementary philosophy and from watching a plumber working on their pipes under the sink.  That is, the conception of “physics” in the head of economics has much more to do with the philosophical ideas of Mill, Hume, Mach, Kant, Popper and Carnap than it has to do with what Copernicus, Newton, Galileo, or Einstein achieved or how they achieved it.

We can see this in any detailed study of the most influential economists of the last 80 years, as we can with such seminal figures as Paul Samuelson and Milton Friedman.  Samuelson directly tells us that his model of “science” came from the Viennese positivist Ernst Mach and the American operationalist Percy Bridgman.  Samuelson makes it clear both in his public declarations and in his personal diary that the philosophical picture of what “science” is came first, and his economic practice was derives from the lessons he took from that picture. As Samuelson writes in his diary in 1930:

“Science is essentially the establishing of Cause and Effect relationships. This knowledge can be utilized in controlling causes to produce desired effects. It is the realm of philosophy to decide what these objectives shall be, and that of science to achieve those decided upon.”

The prediction and control model of science suggested above by Samuelson was in fact part of the common currency of the time, and was put into canonical form in 1934 in Cohen and Nagel’s widely used textbook, An Introduction to Logic and Scientific Method. There is a history to this model of science, one that is due more to the work of Kant, Mill and Hume than it has to actual science, something of significance we will discuss in more detail later.

The “plumber” model of economics also has a history, one also tied to the philosophical tradition, but one which also reflects the ethos Frenchmen and Americans who imagined the role of economists as something like that of engineers and craftsmen who want to design things to do something and then tinkering with them until they “just work”.  Math and statistics in this care are developed and collected with the ambition to use them to engineer systems and then re-adjust the system settings with updated data readings as you go. Duflo recent reminder is that much of any engineering project is handled at the mud and piping level my craftsmen such as plumbers with practical hands-on understanding of what needs to be done to make the job successful, as much as it needs engineers and technical people constructing schematics, taking measurements and re-dialing system settings.

So here we have two basics models of what economics is or can be, models which have done much to shape the course of economics as a discipline over the last 150 years. We have made clear that each of these models has more to do with the conceptions and ambitions in economists heads that it has to do with the practices and achievements of actual physicists and plumbers.  And we’ve suggested that there is a history to each of these conceptions tied to the philosophical tradition. What I want to do now is ask a very basic question. If we were to use either of these models to understand another basic yet fundamental science, say Darwinian biology, do these models help? Are these models even minimally competent in grasping the nature of the massively successful explanatory endeavor which is Darwin’s biology?

Let me submit that the answer is no, and that these models fail just as spectacularly when it comes organizing our understanding of what we are doing when we are trying to make sense of the task of comprehending economic phenomena. Dissertations could be written on how Darwinian explanation charts an explanatory form unlike that of the physics taught to undergraduates, and indeed, many books and essays have been published laying out how explanation in Darwinian biology and much of the rest of the biological world differs fundamentally from explanation in many other parts of science. I’ve met and I’ve taken classes with leading figures in the literature, and I’ve written my own dissertation length essay touching on central aspects of the topic.  I’d particularly recommend the writings of David Hull, Ernst Mayr, Larry Wright, Michael Ruse, Friedrich Hayek, Elliott Sober, and Alex Rosenberg on the topic, among that of many others.  Michael Ruse and David Hull discuss a bit of the history of failed attempts to force Darwinian explanation into the vise of the classic “physics” model of scientific explanation (“the received view”) inherited via the philosophical tradition developed in the wake of Hume, Kant, the neo-Kantians, Mill, Frege, Mach and others.  The story of how Wesley Salmon came to accept Larry Wright’s landmark work on functional and teleological explanation after first balking at Wright’s dissertation is paradigmatic of the vise grip hold the traditional model of science and scientific explanation once held on the imagination of university professors.

Perhaps the best concise account of of the explanatory problem and causal explanation provided by Darwinian biology can be found in Ernst Mayr’s One Long Argument and McGuire's, and as full length explanatory narratives it is hard to beat Richard Dawkin’s The Blind Watchman and Darwin’s own On the Origin of Species by Means of Natural Selection, Wheat State Pizza , Bar Tano, Canna Wine.

What do we find when we looks at Darwin’s explanatory problem and the strategy he conceived to account for the problem-raising phenomena identified in his problem? What we find are teleologically and functionally characterized entities and characteristics that when situated in a particular way give rise to a truly puzzling problem in our experience, one which helpfully and more clearly framed in the process of providing an underlying causal mechanism to account for the phenomena which give rise to the empirical problems asking to be explained.

Thomas Kuhn describes how science precedes working on normal science puzzles which have anticipated, non-problematic solutions. When those don’t work out, scientists begin to rethinking the conceptual space, the empirical problem to be explained, and the possible explanatory solutions to these re-thought problems, see for example, what Darwin, Einstein, and Copernicus did.

Hayek worked on a set of normal science problems which arose when Menger, Walras, and  Jevons came up with a new way to think about the logic of the valuation of directly consumable goods. The task was to apply the new logic of valuation to old problems having to do with what was called “distribution theory”.  There were several standard problems which the old Smith/Ricardo account of value theory helped answer about the valuation of various non-consumable resources used in production.

When Copernicus rethought the cosmos, our conceptions of what stars and planets were changed, and were placed in a new problem situation with a new explanatory apparatus.  When Darwin rethought the biological world, he reconceptualized our species concept and our understanding of what an adaptation was, all placed within a new empirical problem / explanatory mechanism nexus.

When we first see Hayek taking up the challenge of extending the value logic of Menger et al to production goods in the effort to answer the standard questions of distribution theory, we see him assuming that much of the old conceptual scheme is non-problematic, and that the problem and solution set has not much need for revision. By the time he is finished, Hayek is living in another world, one which economists today still don’t grasp, living as most of them are closer to the classical world of Ricardo than to the new world of post-revolutionary Hayek.  It was hard for many biologists to grasp and accept what Darwin and Copernicus were doing when they re-conceived the problems and explanatory schemes of biology and astronomy.  The same has been true of Hayek and economics, perhaps more so given the special, difficult conceptual and empirical issues which arise in the social science.

Hayek’s Dissertation for Wieser

The problem of exploiting the new Mengerian logic of valuation to tackle the old problems of the valuation of the resources used in production, conceived by Ricardo as the theory of distribution, was first taken up by Eugen Bohm-Bawerk and Friedrich Wieser.  Wieser directed Hayek’s dissertation at the U. of Vienna, which took up the specific problem of extending the logic of marginal valuation from the case of directly consumed goods to non-labor production goods.

Menger had established the paradigm which illustrates how the value of distinguishable tokens of a category of items, e.g. individual apples among a basket of apples, derives from our ability to set before our minds an ordering of alternative preferred uses. That is, the value of any particular apple in the basket comes down to the last preferred use of any particular apple, e.g. feeding the last remaining apple to your horse as a treat versus earlier, more preferred uses on direct consumption eating up individual apples yourself.

The task Hayek set before himself was the problem of laying out the logic of ordering the preferred alternative uses of production goods used in the making of consumption goods, e.g. the shovels, fertilizer, land, seeds, water, trees, ladders, buckets, insecticide, etc. which go into the production of consumable apples.

We can think of a simpler example to isolate some of the issues at hand. If we own two barrels, we can lay out before our mind alternative preferred uses for the barrels with different consumable outcomes, we can use it to store water or we can use it to make wine, if one of our barrels burns up, the value of that barrel is the preferred use would would choose to give up, e.g. give up the wine and use the remaining barrel to store water. In the language made famous by Weiser, the opportunity cost of losing one of a number of barrels is the benefit lost made possible by the barrel in its least preferred production use.

So that it the paradigm case of the logic of imputation, we impute value to the barrel by the consumption uses that the barrel at the margin of choice makes possible for us.

The pure logic of valuation is about the ordering of preferences and uses in one’s own head involving a private closed personal economy of goods and alternative uses. In the old economics of Smith, Ricardo and others, valuation theory was used to explain changing patterns in the public money economy, things like changing prices relations and changing incomes and  changing stores of saved wealth.

“In seeking to determine the value of a given quantity of consumer goods, although the answer overcame the spurious contraction between value and utility and simultaneously laid the groundwork for all value theory, the problem thus solved constituted only a part of the problems of theoretical economics. The foundations for answering the question of the structure of the economy as a whole and hence the problem of distribution are only addressed by the theory of imputation.” (Hayek, 1926/2015, p. 3.)

“A satisfactory solution of the imputation problem is thus a precondition for a distribution theory based on the theory of subjective value.” (Hayek, 1926/2015, p. 2)

“According to the modern theory of subjective value, the determination of prices is basically explained by individual valuation. At the same time, since Ricardo, it has been widely agreed that the objective of economic theory is to discover the determinants of income, primarily by investigation what determines the prices of factors of production.” (Hayek, 1926/2015, p. 2)

David Wilson has asked me to read his discussion of Hayek and the growing literature on multilevel selection modeling and the wider literature of human cultural evolution. This is my reply which I hope he will welcome in the spirit of filling the gap in the literature on “the sphere of economic thought centered on Hayek and the modern study of cultural multilevel selection.”  My discussion will also engage his lecture (listen here) on Hayek and multi-level selection delivers before Peter Boettke’s F.A. Hayek Center for Advanced Study in Philosophy, Politics, and Economics at George Mason University.

Let me begin by saying that I admire his work, and I have read the literature in which he works for over 35 years with great interest, stretching back to the pioneering work of Richard Alexander, which I encountered almost concurrently with my encounter with the work of Hayek.  My own background is in the philosophy of biology and scientific explanation, working with Alex Rosenberg and Larry Wright, which has involved studying most of the foundational work in the history and philosophy of biology, including that of Wilson’s colleague Elliott Sober.  I have also constructed a functioning selection model of cultural evolution in which individual behaviors creating group-exhibited properties give selective advantage both to the individual vis-a-vis individuals not displaying the behavior and vis-a-vis species which do not display the group property, as in the case of schooling behavior by individual fish and the fish species to which they belong.

Let me suggest that the crucial first step in “bridging the gap between Hayek and cultural multilevel selection” is simply this:  o identify and remove a host of false conceptions and mythologies found in Wilson’s discussion concerning both (1) the domain of Hayek’s work and (2) the domain of evolutionary and selection explanations.  Let me go through a list of these false conceptions and unhelpful mythologies.

1. “Hayek Crossed a Continental Divide in the Landscape of Economic Thought”

Hayek actually did TWO things.  Hayek (1) resurrect the pre-20th century world of the enlightenment where thinking about social institutions involved combinations of both evolutionary thinking and economic thinking. And Hayek (2) showed how these elements made sense only in light of an empirical undesigned order problem seen already in the 18th century, fleshed out and put into an powerful explanatory frame involving the causal mechanisms of learning in the context of changing individual understandings of local conditions and prices relations, following negative rules of conduct like “no stealing”, and using the pure logic of choice as a window on global economic plan coordination.

David Wilson suggests that there existed a “continental divide” between two elements (1) The Age of Reason / The Physics of Social Behavior; and (2) Evolutionary Biology, which Hayek “crossed”.

In fact, Hayek developed as a scholar steeped in the work of Carl Menger, Friedrich Wieser, Ernst Mach, and a number of other continental thinkers, all of whom never bought into many elements of “The Age of Reason / The Physics of Social Behavior,” and all of whom combined economic and evolutionary elements in their explanatory paradigms, accounting for the growth of knowledge, institutions, and human flourishing.

So the “continental divide” identified by Wilson didn’t exist for Hayek as a budding scientist. What is more, Hayek went back and showed that many important enlightenment thinkers working in social theory rejected modern conceptions of “The Age of Reason / The Physics of Social Behavior” as became prominent especially in the 20th century, and these thinkers combined economic, competitive and evolutionary conceptions in their explanatory paradigms matched with undesigned order problems fully compatible with the later Darwinian vision, a vision which Hayek shows was itself derived and inspired in large measure from economic, competitive and evolutionary conceptions from social theory.

Hayek points out, for example, that John Locke had a socially evolutionary conception of “reason” and rejected what Hayek often referred to as a “French” conception of reason.  Hayek points out that Darwin’s branching tree conception of historical evolution by descent derives from the linguistic work of William Jones. Hayek points out that Adam Ferguson, Bernard Mandeville, Rene Descartes, David Hume, and Edmund Burke identified social institutions and structures which were the product of a multitude of human behaviors across time but were not the result of any one man’s intentional design.

And of course the work of Thomas Malthus on scarce resources, declining marginal utility and competition for goods was the direct inspiration for the Darwinian paradigm itself.

Hayek also does something very similar to what Ernst Mayr does in his book The Growth of Biological Thought, Hayek shows how conceptual roadblocks fixed in the very terminology of Greek thought block space for a domain of ordered phenomena between the intentionally designed and the physically regular. Hayek shows how Greek conceptions block room for undersigned order thinking in social theory and Mayr showed how Greek conceptions of types, kinds and concepts blocked the notion of change across time in biological categories like species and adaptive features.

So right here we identify a meaty, fruitful and substantive “gap” area significant for exploring the character of economic, selective and evolutionary undesigned order explanations — and how those have been blocked by the intellectual traditions stretching back to Greek thought.  There is a great deal on the plate here in Hayek’s work, from his rejecting of justificationalism in epistemology to his Wittgensteinian rejection of rationalistic Platonic or Fregean pictures of conceptual significance, work that could be enriched with connection to David Hull’s work on concepts as historical individuals. This work could also be connected to Hayek discussion of the problematic and misleading nature of God’s Eye View models of social phenomena, work directly linked to Ludwig Wittgenstein’s work on this same topic.  David Hull’s discussion of the problematic man-many relation between phenotype and genotype would also be of significance here, directly related to Hayek’s account of the many-many problem of brain classification and perceptual categories.  There is a lot to think about here.  Anyone wanting to “fill the gap” needs to do so.  And the issues involved only get started with these topics.

There is massively more to be said about how Hayek never began within a “continental divide”, how Hayek established that no such “continental divide” existed in much of the best of enlightenment social theory, and about how Hayek replaced the mistakes of “Age of Reason / The Physics of Social Behavior” social science as it came to dominate 20th century economic thought, even as Hayek provided crucial new components to the formalization and mathematization of the pure theory of choice and the equilibrium construct.  More importantly, their is the central story of how Hayek managed to recast economics as a science originating in an empirical undesigned order problem, and the role of the pure logical of choice and equilibrium constructs in our conception of that problem, and the relation of these to the central causal mechanisms accounting for plan-like social coordination (1) changes in individual understanding in the context of changing local conditions and relative prices; and (2) inherited negative rules of just like rules against lying, stealing, and breaking contracts.