The empirical problem and explanatory strategy of economics

2009-01-07T04:35:00Z

Some of the "gut" of my dissertation was presented at an economics conference in 1996.

What I want to do with my dissertation is present this so that economist understand this is an advance of understanding in economics in and of itself as _economics_ strictly speaking, and I also want to present it so that philosophers understand that is in an advance of understanding in philosophy in and of itself as _philosophy_.

In other words, we are advancing understanding here in a single step in two fields, not in two separate steps, one "economic" and one "philosophical".

This is one of the things that delayed completion of the dissertation -- beyond personal biography.

Many philosophers read this and see economics. Many economists read this and see philosophy. Hayek's view -- similar to my own -- what that many philosophers and philosophers of social science had philosophy and philosophy of social science wrong, and also that many economists had the explanatory strategy of economics wrong, in part because their philosophy and philosophy of social science was essentially a dumbed down version of the philosophy and philosophy of social science that Hayek rejected.

Here's the second part of my 1996 paper:

2. The empirical problem and explanatory strategy of economics.

In Friedrich Hayek's famous 1937 paper "Economics and Knowledge", again in the first two chapters of his classic The Pure Theory of Capital, and throughout the rest of his career Hayek identified the valuation constructions of marginalist economics -- including the intertemporal construction that he had introduced to economics in 1928 -- as pure and tautological logic, incapable of providing causal explanations of resource use coordination and industrial fluctuations in an extended society. Yet Hayek never doubted the importance of the intertemporal valuation construction for economics, or its role in fulfilling the explanatory promise of the discipline. How are we to understand Hayek's position on these matters -- just what use did Hayek find for the intertemporal valuation construction, and what role does this play in his solution to the problem of the logical character and explanatory strategy of economics?

I want to use this question as a fulcrum around which to report on my current research into the economics of Friedrich Hayek. I also want to use it as an anchor around which to address important questions about the function of narrative as both a necessary tool and as a sometimes unsuspected barrier to the advance of our understanding in both science and intellectual history

The choice of Hayek and his problem is an apt one, for Hayek is a recurring figure in the various narrative accounts that contemporary economists have offered to communicate the character and significance of their own efforts. For example, economists from Robert Lucas, Joseph Stiglitz, and James Buchanan to John Roemer, Israel Kirzner, and Sanford Grossman, have found it useful to articulate their own research agendas in relationship to Hayek's classic castings of the question of the relation of (1) valuational constructions and (2) the fact of our dispersed and imperfect knowledge to the core problems of resource use coordination and industrial fluctuations in extended societies. Economists, philosophers, historians, and social theorists who have characterized their own conception of political economy in relation to Hayek's reads like a who's who (if I've somehow forgotten to include your own name on this list, please correct my error and include it here). I might mention, beyond those already named, Kenneth Arrow, P. S. Atiyah, Abram Bergson, Marina Bianchi, Mark Blaug, Peter Berger, Alan Brinkley, E. J. Dionne, Ronald Dworkin, John Eatwell, John Elster, John Galbraith, Pierangelo Garegnani, John Gray, Gottfried Haberler, Frank Hahn, Alvin Hansen, Garrett Hardin, Russell Hardin, Daniel Hausman, R. G. Hawtrey, Robert Heilbroner, John Hicks, Albert Hirschman, Jack Hirshleifer, Geoffrey Hodgson, Kenneth Hoover, Leonid Hurwicz, Terrence Hutchison, Nicholas Kaldor, John Keynes, Israel Kirzner, Arjo Klamer, Tjalling Koopmans, Irving Kristol, Frank Knight, Ludwig Lachmann, Oscar Lange, Charles Larmore, Alex Leijonhufvud, Maurice Mandelbaum, Murray Milgate, Philip Mirowski, Ernest Nagel, Richard Nelson, Otto Neurath, Robert Nozick, Mancur Olson, Michael Polanyi, Karl Popper, Paul Samuelson, Joseph Schumpeter, G. S. L. Shackle, Herbert Simon, Peiro Sraffa, Robert Sugden, Robert Townsend, Edna Ullmann-Margalit, John Watkins, E. Roy Weintraub, Larry White, Oliver Williamson, and Sidney Winter. This list, of course, does not pretend to be complete.

This is a widely varied list of individuals. Although each of the people listed here has his or her own reasons for using Hayek's work in political economy as a point of reference, it is easy to understand why especially economists will continue to do so. In recent years three fundamental research agenda items pressed by Hayek have come to the forefront of important discussions among economists.

(1) In the 1930's and the 1940's Friedrich Hayek identified deep problems with traditional conceptions of the explanatory relevance of the intertemporal relational valuation construction he had introduced into economics in 1928 for the real world of resource use coordination and industrial fluctuations in an extended economy. Since 1960's and increasingly so in every decade since, Hayek's perception of problematic explanatory relevance of the intertemporal equilibrium construction for the problems of resource use coordination and industrial fluctuations has come to be shared by an ever widening circle of economists and philosophers. {It seems appropriate to recognize the important work of our conference keynote speaker Axel Leijonhufvud in this regard, who has never made a secreteof the role Hayek's work played in the evolution of his own thinking on these problems.}

(2) Hayek's "Economics and Knowledge" paper, his Pure Theory, as well as his famous essay "The Use of Knowledge in Society", identified the unavoidably imperfect and dispersed character of our knowledge of alternative production possibilities as at the center of the problems of resource use coordination and industrial fluctuations in an extended society. In recent decades economists has sought to incorporate these facts within the body of economic explanation.

(3) In his "Economics and Knowledge" paper and in later work Hayek identified learning in the social context of changing relative prices and shared rules of just conduct as the core explanatory element and contingent cause in economics. Over the last ten years or so economists have increasingly come to identify learning in a social context as a key agenda item for the explanatory project of economics.

Before I say more about Hayek and his place in contemporary political economy narrative, I should say something about narrative within intellectual inquiry, especially in those realms which supply us with special understanding of things beyond our everyday understanding, such as in geometry, cosmology, chemistry, biology, and economics. The following remarks reflect a least in part what I have learned from Thomas Kuhn, Joseph Rouse, and most especially Larry Wright.

The intelligibility and significance of a scientific practice comes in part for the narrative understanding in which it is situated. In order to understand the increase in knowledge we have attained we need some way to recall our prior lack of understanding, as well as some of the conceivable alternative accounts of the world we need to picture what various deficiencies in our knowledge might look like. One of the difficulties of this aspect of the process in which our understanding advances is that the conceptual picture provided by the articulated narrative history of a discipline is invariably incomplete, misleading, and in some ways a distortion of past efforts and current alternatives. The effects of this can be seen in the heated arguments between paleontologists and population geneticists. Modern Darwinian biology is understood today within a narrative of the controversies between 'group selectionists' and 'gene selectionists', or between 'Ultra-Darwinist' population geneticists and the 'anti-Ultra-Darwinist' paleontologists. The narrative that radical gene selectionists provide of these controversies differs, sometimes significantly, from that provided by the "anti-Ultra-Darwinist" paleontologists, and neither account can be considered representative of the narrative understanding being provided by most of those within either the population geneticist or paleontologist communities. Incomplete, misleading, and distorted narratives abound, yet in providing these narratives, gene selectionists and the 'anti-Ultra-Darwinian' paleontologists are participating in the development of their science, contributing to the richness of its intelligibility and significance. In the process we get a more developed view of the rival alternatives, a richer insight into the range of current agreement, and a deeper sense of the potential for underlying conceptual unification. Historians of theoretical physics in the 19th century document the functioning of a like process in the development of the physical sciences.

Darwinian biology is enriched today by the competing narrative histories that help to flesh out differences between competing explanatory conceptions. Yet narrative accounts of the history and contemporary character of a discipline are not always advantageous contributors to the development of our understanding of things. The narrative of a scientific practice which provides some part of its significance and intelligibility can do more than simply distort conceptual space, it can also block access to potentially more fruitful avenues of understanding, very often even those which have already been explored to some success by others. Narrative accounts of the problem of species in the biological literature since the time of Plato and up to the work of the biologist Michael Ghiselin and the philosopher David Hull almost invariably identified the species concept as a class category, rather than a shifting historical individual or population. (Roots of the Ghiselin-Hull development can be found in the work of Darwin and Ernst Mayr). Ernst Mayr nicely shows in his grand treatise The Growth of Biological Thought how this essentialistic picture of the notion of particular species as universal types with features logically given irrespective of time or place served to shackle conceptual space, making it almost impossible for naturalists and logicians to perceive incipient non-essentialistic views of the species concept, and falsely imply that other alternatives were not available, thus blocking access to nascent evolutionary and selectionist explanations of environmentally apt morphologies and the origin of species, both old and new. To put it in a phrase, when not simply excluded from creative thought by blinding fetters on the imagination, auspicious endeavors to advance our understanding in biology were invariably killed while still in the crib by Plato and the logician's picture of the character of knowledge and thought. The ancient log-jam was finally broken by Charles Darwin's profound recasting of the problems of biology and the character of its explanations in his Origin of Species, but not without a number of missteps along the way in the process of his "one long argument", and not without the need even today to continue sorting out a nest of puzzles, even despite the profound explanatory success of the Darwinian program. All of this, of course, is part of an ongoing process taking place at the "High Table" among paleontologists, philosophers, population geneticists, and micro-biologists.

Part of what makes the rival narrative accounts supplied by paleontologists, micro-biologists, philosophers, and population geneticists so interesting and fruitful is that within these narratives various biologists provide rival accounts of the character and significance of Darwin's classic casting of the content and tasks of evolutionary and functional biology. This gives biologists and philosophers a well-worked out reference point to advance a shared understanding of what biology is all about or what it might be all about. (Niles Eldredge, David Hull, and Ernst Mayr, among others, make it clear that the work of philosophers has been integral to this process and continues to be so.)

My own work in this field serves as a useful introduction to some of the most basic research results I have obtained in my investigations of Hayek's work and its place in the development of contemporary narratives in political economy. In biology the environmentally apt functional features, divisions of labor, and teleological doings of organisms such as we observe in the pumping of the heart, the coordinated operation of the digestive system, and the fleeing of rabbits raises the central explanatory problem for any evolutionary account of the origin of species. As Darwin puts it, ".. such a conclusion [the origin of species by descent], even if well founded, would be unsatisfactory, until it could be shown how the innumerable species inhabiting this world would have been modified, so as to acquire that perfection of structure and coadaptation which most justly excites our admiration." The hypothesis of the origin of species by descent is an alternative to the hypothesis that species are a product of acts of Providential Creation. If the conjecture of the origin of species by descent is accepted as a fact, then within this alternative observational frame of reference the manifest functional features and teleological doings of organisms suddenly become an explanatory problem. The features and activities of organisms present us with patterns and results which look as if they were created or generated by the design and intent of a hand from above operated by the creative intelligence of a super-mind something like our own. But if the populations which exhibit observably apt characteristics are mutable products of history linked together somewhere in the past, and not the product of singular acts of Providential Creation, then the functional and teleological features of the organisms in these populations can no longer themselves be accounted for by this now abandoned process of Providential Creation.

To account for this directly observed teleological phenomena, which now raises a pressing question in the new context of the conjecture of modification by descent, Darwin provides a new bottom-up causal process as a rival to the top-down Providential Creation explanation. This bottom-up process shows how the manifestly observable apt structure and behavior of biological features could be generated to look as if they where invented or constructed by the design of a 'blind goods-maker' or the intent of an 'invisible hand'. Significantly, the bottom-up explanatory elements provided in Darwin's bottom-up causal account are open-ended in the sense that they involve an open-ended disjunction of causes, the physical constitution of which is undisclosable in advance of the unique unfolding of evolutionary history. The reward of reproductive opportunity can be given not only to structurally identical biological expressions, but also to very nearly similar structural constitutions. Ultimately the only common property which a structure must have to be identified as belonging to a functional or teleological category is the shared effect these structures have in producing the replication and persistence of entities sharing the same historical origins and which constitute parts of an evolving historical species, without it ever being possible to make any principled distinction between chance event and selective event in any particular instance. What we have at bottom in this situation is the primacy of our direct perception of the teleological characters of organisms. The physical constitution of particular exemplifications of the adaptive functional features and teleological doings Darwinian theory is designed to explain cannot be provided in advance of the historical unfolding of an historically unique population, which is to say that our direct teleological and functional observation of the features requiring explanation in evolutionary biology cannot be given a replacement in the time and place independent categories of physics and chemistry.

There are four things to note here.

(1) The contingent character of Darwinian explanation is assured by the existence of rival causal explanations.

(2) Darwinian explanation begins with a problem raising pattern in our experience, a problem which arises when we try to make sense of design-resembling phenomena without any top-down ordering hand or intelligence.

(3) The empirical character of Darwinian biology is founded in the first instance upon our direct observation of aptly corresponding features, such as functional appropriateness, division of labor, and teleological direction.

(4) In Darwinian biology the phenomena that fall within what we recognize as a question raising pattern to be explained are open-ended and irreducible to the categories of things whose characteristics would count as theoretical kinds in the physical sciences outside of the contingent unfolding of the unique course of history. That is, the directly observed functional features and teleological doings that ask to be explained in Darwinian biology are categorically autonomous of the kinds that provide explanations in physics. Conversely, the class of explanatory causal elements responsible for the products of evolution is itself open-ended and cannot be physically characterized according to the theoretical kinds physics and chemistry in advance of the unfolding of evolutionary history.

Many attempts before and after Darwin have made to fit the problems of the functional aptness of organisms to their environment and the origin of species within the ancient tradition which demands that our knowledge fit a particular conception of how logic and language gets their significance. The hope of grammarians and other students of language in the 13th century was to develop a linguistics as an `Aristotelian science', a domain of secure knowledge providing us with a body of essences and necessities. This picture of `science' has continued to influence the perception of those investigating the explanation of the functional aptness of organisms and the orgin of species. Individuals under the sway of this ancient picture look for the necessary laws, essential kinds, and cognitive certifiers or confirming crucial tests in adaptive and evolutionary biology. Not finding these, they declare Darwinina biology `not a science', but a `metaphysical research program'. Karl Popper, for example, has interpreted species as having fixed properties, with some necessary direction of development. And finding no such universal necessities, and no crusial tests, Popper at one time concluded that Darwinian biology cannot be a contingent empirical science. After Hayek, Michael Ruse, and others pointed out the false conceptions presupposed by Popper in his account, Popper later changes his judgment about the scientific status of Darwinian biology, but still without managing to fit it very coherently within Popper's wider picture of knowledge, science, language, and explanation.

Let me now give you the picture of economics that I have extracted from Hayek's work on the problems of the logical status and explanatory strategy of economics, placed in the frame of what I have learned from Larry Wright and others. The picture looks something like this.

The repeated pattern in which prices approach costs of production and the intricately coupled network of aptly divided labor and knowledge within an extended society raises a question about the phenomena of the market similar to Charles Darwin's problem of order without design. The problem is raised, as it is in Darwin, when we locate an order in observed events displaying systematic characteristics we identify with intentional design or deliberate production, yet lacking any top-down ordering hand behind that systematicity.

One of the roles of the economist's equilibrium construction, in which costs and prices or values are made to exactly equal each other, is to help us to observe the design-like order in the economy, by pointing to the deep order within change within the extended domain of resource use coordination implied by the repeated pattern in which prices approach costs of production. The observation of a systematic design-like order within the larger extended economy becomes problematic when we realize that the individual understandings which go into this wider social coordination are limited, imperfect, and divided, making our cognitive relation to this order very different from our relation to the posited facts which go into a resource using intentional plan given to the understanding of a single mind. Within the logic of an unflawed resource using plan worked out by a single intelligence, the elements which go into the plan are part of a single understanding, given as a commensurable totality within the scope of the plan. As parts of a well-considered logic, they are perfect and unlimited `given's within the domain of the plan, without room for the sort of incommensurable differences in thought which distinguish different minds, or the open-ended changes in our knowledge which takes place when we begin to see the world in a new way. If the individual understandings that go into making the social order of resource use coordination are necessarily limited, imperfect, divided, and always changing in an open-ended fashion in the way that individual human understandings are limited, divided, imperfect and always changing, then the pattern observed in the market cannot be the product of a (single) human mind and producer. A problem then arises: if we observe a plan-like systematicity in the extended social order of resource use coordination which no individual planner and producer could create, then how does this extended systematic social order arise? And if this systematic order in the social coordination of resource uses shares only some but not all of the logical characteristics of a plan given to the understanding of a single individual, then which of these does it share, and what are its structural features? And if the social order of resource use coordination is not the result of the coherent deliberative plan of a single human being, then what is the cause and underlying process behind this order?

The equilibrium construction, in its dated-goods-through-time relational valuation form, functions to expose the elements which allow us to answer these questions. In the first instance by shear exclusion the intertemporal equilibrium (IE) construction exposes the most plausible causal element which might explain the observed pattern of plan-like order in the extended economy of coordinated resource uses. It does so isolating the universally recognized causes of changes in human understanding that necessarily stands outside the `givens' of any logical construction. In this capacity the IE construction serves as a kind of isolating foil or background exposing template, allowing us to see the world of causes -- changes in understanding -- outside the perfect logic of a resource using plan.

The following are examples of this point in Hayek:

"In distilling from our reasoning about the facts of economic life those parts which are truly a priori, we not only isolate one element of our reasoning as a sort of Pure Logic of Choice, but we also isolate, and emphasize the importance of, another element which has been too much neglected."

".. [economics must] complete the isolation of this branch of logic and restore to its rightful place the investigation of causal processes . . ".

" .. it is these apparently subsidiary hypotheses or assumptions that people do learn from experience, and about how they acquire knowledge, which constitute the empirical content of our propositions about what happens in the real world."

"This kind of causal explanation of the process in time is of course the ultimate goal of all economic analysis, and equilibrium analysis is significant only is so far as it is preparatory to this main task."

"I am far from denying that in our system of equilibrium analysis has a useful function to perform. But when it comes to the point where it misleads some of our leading thinkers into believing that the situation which it describes has direct relevance to the solution of practical problems, it is time that we remember that it does not deal with the social process at all and that it is no more than a useful preliminary to the study of the main problem [the problem of local knowledge and its coordination]." (1945, p. 530)

" . . though the discussion of moral and social problems based on the assumption of perfect knowledge may occasionally be useful as a preliminary exercise in logic, they are of little use in an attempt to explain the real world."

"A 'logic of choice' can say something only about the consequences to be drawn from a set of statements known to some one mind, and in this sense it can account for the behavior of one individual. But . . the step from the logic of choice to an empirical science which tells us anything about what can happen in the real world requires additional knowledge abut the process by which information is transmitted or communicated."

By isolating learning or changes in understanding as a causal element outside of the givens of a logical plan, the IE construction isolates a contingent cause with other possible rivals. The order in the market might be explained by these possible alternative rivals:

The order in the market might be explained by several alternative rivals:

1. Postulate a top-down omnipotent super-mind production master in the image of the individual human planner.

2. Conjecture that learning or changes in understanding in the context of changing relative money prices and stable negative rules of just conduct (such as honesty and property rights) is responsible for the undesigned order of resource use coordination in an extended society, suggested by the repeated pattern in which prices approach costs of production and the extended division of labor.

3. Postulate that we are ant-like creatures or crude robot-like machines who produce a plan-like social order as the result of simple and physically predictable regularities in our behavior.

4. Take for granted unimaginable luck in the completely random `casino' economy (Keynes), or postulate unimaginable luck in the relational valuation structure through time of production (Kaldor).

A second function of the intertemporal equilibrium construction is to identify elements contained in any display of the pattern of resource use coordination over time. One example of this would be the relational contextual dependence between time preference, length to maturity, and output in the valuational relationships between goods in any paper of resource use coordination displaying some degree of concatenation between production and consumption plans. In plainer language, it shows that any social order displaying economic coordination will have within it the systematic relationships between plans, goods, and production processes that economists (in one use of this rather ambiguous word) have labeled `interest'.

Hayek expresses this function of the IE construction when he remarks, "In so far as we analyze individual thought in the social sciences the purpose is not to explain that thought but merely to distinguish the possible types of elements with which we shall have to reckon in the construction of different patterns of social relationships."

Neither of these two elements isolated by the intertemporal equilibrium construction can be reduced to the categories of physics, nor to physical prediction, nor physical symmetries, nor physical conservation principles, nor any of the mathematical functions of physics. The causal element of learning or changes in understanding cannot be reduced to physical predictions or categories, nor can it be reduced to a logic or formal construction. Conversely, unlike the time-invariant significance of the functions, symmetries, conservation principles, and categories of the physical sciences, the valuational relations within a coordination through time of resource uses are only of relational significance within a time and place marked historical unfolding. Furthermore, the conceptual categories and logical relations which characterize the structure of a coordination of resource uses through time cannot be reduced to physical categories, or functions, or causal explanations. As in the domain of geometry, the time-invariant structures of thought in which we all conceive things and relate one thing to another are in a domain distinct from the non-formal world in which we provide contingent causal explanations within contending theoretical frameworks.

An understanding of the open-endedness of learning or changes in understanding, and the predictive and conceptual autonomy of learning and changes in understanding from the explanations and predictions of physics given in contemporary physical categories and structural laws is one of the achievements of twentieth century students of brain, science, and knowledge. As Hayek long ago pointed out, physical science cannot predict or give laws specifying its own advance. The advance of science is conceptually open-ended, and a many-many problem exists blocking any sort of one-to-one reduction of the everyday shared distinctions we depend upon in our observations of environmental conditions to the risky conjectured explanatory categories used in the physical sciences.

The significance of this many-many problem for the theory of knowledge and mind was first worked out by Hayek in his important The Sensory Order, but the importance of this problem is perhaps most economically expressed in Thomas Kuhn, ".. people do not see stimuli; our knowledge of them is highly theoretical and abstract .. much neural processing takes place between the receipt of a stimulus and the awareness of a sensation. Among the few things that we know about it with assurance are: that very different stimuli can produce the same sensations; that the same sensation is in part conditioned by education. Individuals raised in different societies behave on some occasions as though they saw different things. If we were not tempted to identify stimuli one-to-one with sensations, we might recognize that they actually do so." "None of this would be worth saying if Descartes had been right in positing a one-to-one correspondence between stimuli and sensations. But we know that nothing of the sort exists. The perception of a given color can be evoked by an infinite number of differently combined wavelengths. Conversely, a given stimulus can evoke a variety of sensations, the image of a duck in one recipient, the image of a rabbit in another. Nor are responses like these entirely innate. One can learn to discriminate colors or patterns which were indistinguishable prior to training."

There are four things to note here:

(1) The contingent character of the explanatory strategy Hayek provides for economics is assured by the existence of rival causal explanations.

(2) Hayek's explanatory strategy in economics begins with a problem raising pattern observed in our experience, a problem which arises when we try to make sense of design-resembling phenomena without any top-down ordering hand or intelligence.

(3) The empirical character of Hayek's explanatory strategy is founded in the first instance upon our direct observation of aptly corresponding features, such as functional appropriateness, division of labor, and teleological direction.

(4) In Hayek's explanatory picture the phenomena which fall within what we recognize as a question raising pattern to be explained are open-ended and irreducible to the categories of things whose characteristics would count as theoretical kinds in the physical sciences outside of the contingent unfolding of the unique course of history. That is, the directly observed patterned resource using coordinations that ask to be explained in Hayek's explanatory picture are categorically autonomous of the kinds that provide explanations in physics. Conversely, the class of explanatory causal elements responsible for the production of resource use coordination in an extended society is itself open-ended and cannot be physically characterized according to the theoretical kinds physics and chemistry in advance of the unfolding of human history.

Hayek Dissertation Blog

Back and forth with Boettke

The empirical problem and explanatory strategy of economics

The Theme of My Dissertation is Good Explanation