COGNITION AND ECONOMICS PART 10 pptx

Posner John Dewey and Friedrich Hayek, despite their differences in field, nationality,generation, and politics, have a number of things in common1 and amongthese is the derivation of a c

GROUND OF POLITICAL THEORY

IN PLATO, POPPER, DEWEY, AND

Richard A Posner

John Dewey and Friedrich Hayek, despite their differences in field, nationality,generation, and politics, have a number of things in common1 and amongthese is the derivation of a comprehensive social, political, and economictheory from a theory about the structure or operation of the human brain,what I am calling ‘‘cognitive theory.’’ In the case of Dewey, a philosopher, onewould be inclined to substitute ‘‘epistemology’’ for ‘‘cognitive theory’’2and inthe case of Hayek, who dabbled in biology, one might substitute ‘‘cognitivescience.’’ But the similarity is considerable, and so the choice of a single termfor both is warranted My particular focus is on Hayek’s cognitive theory, withDewey used mainly as a stalking horse I begin by situating the Deweyan andHayekian projects in the broader philosophical tradition of deriving politicalfrom cognitive theory I end with a brief discussion of the bearing of Hayek’scognitive theory on several current issues in social science

$

This is the expanded text of a talk given at the Third Annual Symposium on the Foundations

of the Behavioral Sciences – entitled ‘‘Hayek, Dewey and Embodied Cognition: Experience, Beliefs and Rules’’ – sponsored by the Behavioral Research Council of the American Institute

of Economic Research and held on July 18–20, 2003.

Cognition and Economics

Advances in Austrian Economics, Volume 9, 253–273

Copyright r 2007 by Elsevier Ltd.

All rights of reproduction in any form reserved

ISSN: 1529-2134/doi:10.1016/S1529-2134(06)09010-7

253

1 PLATO AND POPPER

This project of derivation that I have just described may seem strange, but isnot In this as in many respects Plato set the fashion for the millennia tocome The ideal state sketched in the Republic is not only an analogy to thesoul (though it is that too); it is an implication of Plato’s conception ofhuman mental capacity, a conception that is ontological as well as episte-mological It was Plato who, according to Aristotle, first separated auniversal (i.e., a concept) from particulars (i.e., a concept’s physicalembodiments or expressions) There are a multitude of chairs, very differ-ent in size, shape, color, and design, yet there is also a concept of the ‘‘chair,’’

in which all the physical chairs participate The concept has no physical bodyand therefore in a sense exists outside time and space – it is immaterial andeternal But Plato believed, reasonably as it seems to me, that it is real It isreal in the same way that a line or circle in Euclidean geometry is real eventhough it is not identical to any physical line or circle and cannot be – theEuclidean line has only one dimension, and the Euclidean circle only two,and there are no one-or two-dimensional objects in the physical world(although electrons are dimensionless), as far as we know

But Plato was a metaphysical realist in a stronger sense than that ofsomeone who merely believes that concepts are real though not ‘‘embod-ied,’’ that is, not located in space, as physical objects are He was an ‘‘ab-solute’’ realist That is, he believed that what we would call concepts and heForms or Ideas existed independently of any knowing mind and any phys-ical example He believed that there was a Heaven of Forms in which thesearchetypes resided, as it were It is unclear and unimportant whether hethought one would find there the archetypes of artifacts – the archetypicalchair or horse, for example, to which actual chairs and horses and circles inour world were mere approximations His emphasis was on abstract no-tions, such as the Just, the Good, the Equal, and the Large, and thus therewas, for example, the Just as well as behaviors that were just And, crucially,

he believed that the Forms could be discovered by a human being of sonable intelligence and philosophical aptitude, provided he was rigorouslytrained in philosophical reasoning Philosophy was the royal road to moraland political truth and goodness

rea-The leap from numbers, as metaphysical ‘‘reals,’’ to normative concepts,

as metaphysical ‘‘reals,’’ was highly questionable, because while every petent person believes that two plus two equals four, suggesting that twoplus two ‘‘really’’ does equal four, competent people often disagree on nor-mative moral and political questions, suggesting that there is no reality out

com-there to force convergence But the leap was crucial from a political point because it enabled Plato to claim that, through philosophy, what wastrue and good in politics could be ascertained, just as in mathematics Fromhere it was only a small step for Plato to advocate that, ideally at least – for

stand-he was aware of tstand-he practical obstacles3– philosophers should rule Untilthey did rule, the state would never be well governed Philosophy alonecould distinguish truth from opinion, and it is natural to suppose that thehandful of persons who know with certainty what is the right course inpolitics should be the rulers of the state, rather than that ignorant hoi polloishould be the rulers The latter have access not to truth but only to opinionand so are radically less competent The ideal state as sketched in the Re-public is anti-democratic, indeed totalitarian, though disinterestedly so be-cause the philosophers do not have a lust for power

The fiercest, though not the fairest (Hallowell, 1965, p 273), enemy ofPlato’s cognitive-political theory was the philosopher of science Karl Popper(1902–1994) (Popper, 1963, vol 1;Lessnoff, 1980, p 99) The ferocity derives

in part from the fact that Popper too had a cognitive theory from which hederived a political theory – what he called the ‘‘open society’’ and corresponds

to modern notions of liberal democracy But it was a cognitive theory that wasthe opposite of Plato’s, so it is no surprise that it generated a political theorythat was the opposite of Plato’s political theory Popper was a fallibilist Hebelieved that we can know error but never truth.4His model of the reasoningprocess was scientific inquiry For him the scientific method consisted of pro-posing empirically testable hypotheses and then testing them If the observedresults of a test were inconsistent with the hypothesis being tested, the hy-pothesis was falsified and the scientist would have learned that he was barking

up the wrong tree But if the observed results were consistent with the pothesis, this did not prove that the hypothesis was true A future experimentmight yield discrepant results, and not just because of a defect in the testingmethod Newton’s universal law of gravitation passed empirical tests for cen-turies, but eventually was determined to be erroneous because contradicted byobservations designed to test Einstein’s general theory of relativity

hy-The example of Newton’s law is an important one in saving Popper from

a charge of epistemological nihilism Although eventually proved false,Newton’s law yielded and continues to yield useful knowledge, because itwas less false than the theories it displaced A hypothesis that withstandsrepeated empirical tests demonstrates by this that it has sufficient reliability

to warrant action in accordance with it even though the possibility that itwill someday be overthrown cannot be negated, and, in some cases, such asthat of Newton’s law, even after it has been overthrown

If Plato’s capital-T ‘‘truth’’ is not attainable, if instead intellectualprogress depends on generating and testing empirically rich hypotheses, thevesting of intellectual and political authority in philosophers makes nosense Since scientists do not enjoy seeing their hypotheses refuted, scientificprogress, depends on the creation of a competitive community dedicated tofreedom of inquiry Since, moreover, it is far more difficult to test hypoth-eses concerning the moral or political Good than it is to test scientific hy-potheses, there is no basis for giving a monopoly of moral and politicalcontrol to a body of intellectual specialists Better to have a diversity ofinquirers, or what I have elsewhere called ‘‘moral enterpreneurs’’ (Posner,

1999, pp, 7, 8), in the hope that some of their moral and political insightswill survive, for a time anyway, the challenge to moral guidance and po-litical governance posed by the actual conditions, personalities, and events

of particular societies

2 DEWEY

John Dewey (1859–1952), like Popper though earlier, opposed Plato’s nitive and political theories, proposing his own cognitive theory and thepolitical theory that seemed to him to flow from it I have written aboutDewey’s cognitive and political theories at some length elsewhere (Posner,

cog-2003, Chapter 3), so I shall be brief here Remember that for Plato lectual inquiry is a search for the antecedently real – for that which existsindependently of human cognition The universe, including mathematicaland even moral and political entities or concepts, is basically a passiveobject waiting to be discovered by human beings using the methods of exactreasoning The quest for its secrets is seen by Plato, at least as understood

intel-by Dewey, to be both lofty and lonely, conducted intel-by virtuosi The ification ‘‘at least as understood by Dewey’’ may be important, because theSocratic method employed in the dialogues presupposes a plurality ofinquirers rather than a lone scientist in a lab On the other hand, thephilosopher kings of the Republic are not envisaged as deliberating inthat fashion

qual-Dewey questioned the emphasis that the approach taken by Plato andPlato’s successors placed both on truth and, as Dewey saw it, on the in-dividual Scientific and other inquiry, he argued, actually is oriented towardthe cooperative acquisition of useful knowledge by whatever tools lie tohand, including imagination, common sense, know-how, and intuition ‘‘Aproposition may be said to be verified if it serves as a useful guide to futureconduct’’ (MacGilvray, 1999, pp 542, 545) True knowledge thus includes

tacit (‘‘how to’’) knowledge as well as the articulate knowledge acquired byformal reasoning and systematic empirical methods, for both are useful.Dewey rejected what he called ‘‘the wholly at large view of truth whichcharacterises the absolutists’’ (Dewey, 1906, pp 293, 305 n 1; emphasis inoriginal), by whom he meant Plato and Plato’s followers in the mainstreamWestern philosophical tradition Dewey said that ‘‘it would be a great gainfor logic and epistemology, if we would always translate the noun ‘truth’back into the adjective ‘true,’ and this back into the adverb ‘truly,’’’ for

‘‘truth means truths, that is, specific verifications’’ (pp 305–306; again,emphasis in original) There is no way of knowing when one has found ‘‘thetruth’’ because one cannot step outside the universe and observe the cor-respondence between it and one’s descriptions All that people are able to

do, and all that most of them are interested in doing, is getting better controlover their environment, enlarging their horizons, and enriching and im-proving their lives The knowledge required for these endeavors is collective,being produced by the cooperative efforts of diverse inquirers – intelligence

is distributed throughout the community rather than concentrated in an elite– and validated by the community’s judgment of its utility

We might call this theory of knowledge ‘‘epistemic democracy.’’ No elitehas a monopoly of truth In fact, knowledge of truth is always just out ofreach, making ‘‘truth’’ a regulatory, an orienting, ideal (Not that Dewey,any more than Popper, is a skeptic; what they doubt is not that our claims oftruth ever correspond to an external reality, but that we can ever be certain

of that correspondence.) Thus, to Dewey, the proposal in Plato’s Republic ofrule by an elite possessed of ‘‘a comprehensive rational understanding ofeternal reality and truth, to be nurtured by a rigorous and extended highereducation in all the mathematical sciences from arithmetic to astronomy’’(Schofield, 2000, pp 190, 224) was quixotic

Dewey’s skepticism about the truth claims of conventional epistemology

is a version of fallibilism quite close to Popper’s Fallibilists ascribe thesuccess of science in producing useful knowledge to lack of certitude and aresulting insistence that beliefs be tested against evidence – in Popper’sversion of fallibilism that hypotheses be confronted by data and abandoned

if the data are inconsistent with them Dewey dubbed his own approach

‘‘experimentalism’’ and the word aptly conveys the tenor of his thought aswell as the resemblance to Popper’s cognitive theory, though Popper did notemphasize the communal character of knowledge and inquiry

Dewey is more the ‘‘psychologist,’’ emphasizing the structure of humanreasoning; Popper is more the ‘‘sociologist,’’ generalizing from his under-standing of the practices of the scientific community Dewey commended the

temper that, impatient with convention and the accustomed ways of doingthings – the sediment of habit – insists on trying now this, now that, in acreatively restless search for better means The search yields, as a byproduct,better ends as well One might take up ballet in order to improve one’sposture and discover that one loved the ballet for its beauty; a means wouldhave become an end Popper had made a similar point even more strongly.

He thought that a great weakness of utopian social engineering, of makingover society as the Marxists, for example, wanted to do, besides the inor-dinate demands it made on knowledge, was that in the process of workingtoward a distant end the social engineer would discover and embrace newends and discover that the means he would have to employ to achieve themwould carry him away from rather than toward the distant end

As Dewey explained under the rubric of ‘‘interactionism,’’ our beliefs are

a product not of pure thought but of the interplay of thought and action.When airlines were deregulated, consumers did not ‘‘know’’ what kind ofairline service they wanted; they learned what they wanted by experiencewith the various new services that the airlines, freed from the dead hand ofregulation, offered A central planner could not have designed the optimalconfiguration of a deregulated airline industry, because the essential infor-mation concerning consumer demands simply did not exist before the de-regulated services were offered, just as the person who took up ballet toimprove posture could not know beforehand that the pleasure of balletwould become an end in itself

If experts do not have the lock on knowledge that Plato thought they had,the epistemic basis for authoritarian rule by philosophers (or theologians, orMarxists, or other utopian social engineers, or others who claim the right torule on the basis of superior wisdom or knowledge) is removed, along withany basis for the censorship of moral and political ideas on the ground thatthey are false and any legitimating ground for a fixed and durable politicalhierarchy Dewey’s philosophical project of overturning Platonic epistemol-ogy provides support for making democracy the default rule of politicalgovernance in the same way that Platonic epistemology provides support forthe authoritarian political system described in the Republic Dewey like Pop-per turned Plato on his head by accepting the linkage between knowledge andpolitics but arguing that knowledge is democratic5and so should politics be:

Democracy for Dewey is a good form of political organisation because it is the priate political modelling of a more general form of human interaction which has both epistemological and valuative advantages, and which finds its best realisation in a free scientific community devoted to experimental research Just as such a research commu- nity is trying to invent theories that will allow us to deal with our environment in a

appro-satisfactory way, so a good human society would be one that was a kind of experimental community devoted to trying to discover worthwhile and satisfying ways of living (Geuss, 2001, pp 124–125)

This is not a compelling argument From the fact that Plato was wrong tothink rule by philosophers desirable it does not follow that democracy is thebest form of government; for it is not the only alternative to rule by phi-losophers There is an impermissible leap in the Deweyan claim that ‘‘in-telligence is present most distinctively not in the contemplative life of theleisured elite, but in the workaday practicality of the masses’’ (MacGilvray,

1999, p 551) The ‘‘leisured elite’’ may well lack the kind of intelligencenecessary for good politics, but it does not follow that the kind of practicalintelligence possessed by ordinary people for navigating their personal andprofessional lives equips them to govern or make political judgments Po-litical democracy has to be defended on its own merits rather than by ref-erence to its consonance with sound cognitive theory We shall see that frompsychological premises much like Dewey’s, Hayek reasoned not that thepeople should rule, but that the scope of government – the domain of dem-ocratic rule – should be contracted Given the history of democracy, and itsvirtual disappearance from the political scene between the fourth century

B.C and the eighteenth century A.D., it is doubtful that ‘‘the workadaypracticality of the masses’’ is always and everywhere a superior basis forpolitical governance to the knowledge possessed by an elite

Dewey wrote extensively on law and public policy in an effort to defendpolitical democracy and a socialistic organization of the economy But thesewritings, while some of them, especially those on law, are excellent, owe little

to his cognitive theory (Posner, 2003, Chapter 3) The social significance ofthat theory seems to have been largely negative, in undermining the Platonicapproach to political governance Yet not completely negative, for Dewey insharp contrast to Hayek was a socialist, that is, one who believed in col-lective action as the answer to society’s social and economic problems; and

he may have thought that since thinking was a collective activity, solutions

to social and economic problems should be collective as well But the clusion does not follow from the premise

con-Dewey’s cognitive theory that I have been describing is Darwinian in thesense of being derived from reflection on the type of mental operations thatwould be adaptive in the environment of early man As Richard Rortyexplains, speaking of pragmatists, such as Dewey:

Pragmatists are committed to taking Darwin seriously They grant that human beings are unique in the animal kingdom in having language, but they urge that language be understood as a tool rather than as a picture A species’ gradual development of

language is as readily explicable in Darwinian terms as its gradual development of spears

or pots, but it is harder to explain how a species could have acquired the ability to represent the universe – especially the universe as it really is (as opposed to how it is usefully described, relative to the particular needs of that species) (Routledge Encyclo- pedia of Philosophy, 1998, vol 7; s.v ‘‘Pragmatism,’’ emphasis in original)

That is, language, and mental operations more broadly, are a tool forcoping with the environment rather than a method for establishing thestrong sense of truth (‘‘the universe as it really is’’) that Plato thoughtattainable by philosophers

3 HAYEK

Dewey’s cognitive theory is abstract in the sense that, while characterizinghuman mental processes in a certain way (what I am calling epistemic de-mocracy) and attributing the characterization to selection pressures, it doesnot specify the mechanism by which the brain operates; and that is equallytrue of the cognitive theories of Plato and Popper One hesitates to call thesephilosophers ‘‘cognitive scientists.’’ Yet that appellation just might be at-tached to Hayek, to whose cognitive theory I now turn

On a superficial analysis, Hayek’s cognitive theory consists of just twopropositions, and both are empirical (inductive or observational) ratherthan theory-based The first is that human knowledge is so widely distrib-uted throughout the human population that no single person or agency(‘‘central planner’’) could acquire enough of it to allocate the society’s re-sources efficiently This is a result of the division of labor, which has beencarried to an extreme in modern society and greatly increases efficiency byenabling specialization, but which has brought about, as an unavoidablebyproduct of specialization, a narrowing (along with a deepening) of theknowledge possessed by any single individual He knows more about less

So far there is close convergence between Dewey and Hayek, both phasizing the radical dispersion of knowledge across persons under theconditions of modernity But they quickly diverge because Hayek, unlikeDewey (who was not an economist, as Hayek was), saw that the price systemwas a method, probably the best method and certainly a better method thancentral planning, of aggregating this dispersed knowledge An individualmay realize that a particular input that he needs in his business is likely tobecome scarce, so he buys up a large quantity and stores it His action forces

em-up the price of the input, and the higher price increases the costs of otherusers and leads them to raise their prices Price thus operates as a method(Hayek would say the method) by which private information is diffusedthroughout the entire market.6

The second proposition of Hayek’s cognitive theory is that private orlocal information (the sort of unsystematized information possessed by anindividual and illustrated in the preceding paragraph, as distinct from in-formation that is codified in general principles stated in books or articlesand thus is readily accessible) is impounded not only in price but also inrules.7A firm adopts a new practice – it might be a new method of com-pensating its employees The firm might have hit on the new practice byaccident or by hunch rather than by explicit cost–benefit analysis or otherconscious reflection on how best to fit means to ends Suppose the practice,whatever brought it about, results in lower costs and higher profits for thefirm That is important information to which the firm is likely to respond bycodifying the practice as a rule The difference between a rule and a standard(‘‘profit maximization,’’ for example) that requires fresh analysis in everycase is that a rule singles out one or a small number of facts to be outcomedetermining The person applying the rule does not have to know its pur-pose, or the net benefits of applying it in a particular case; all he has to know

is whether the fact that triggers application of the rule is present or absent.Eventually the reason for the rule may be forgotten (and there may havebeen no reason, or at least no articulable reason), yet this may not matter;continued adherence to the rule will be a way of exploiting the informationimpounded in the rule without need for thought beyond what is necessary todetermine whether the condition for the application of the rule is satisfied.Complying with a rule, like responding to a change in price, can be a method

of utilizing knowledge without actually possessing knowledge

I said that the practice might have been hit upon by accident or by hunch.The first possibility would represent the operation of trial and error, thePopperian method of acquiring useful knowledge The second belongs to thedomain of ‘‘tacit knowledge.’’8 People know how to do many things thatthey cannot ‘‘package’’ as transferable knowledge A person who knowshow to ride a bicycle cannot convey his know-how in words to anotherperson in a way that a person who knows how to bake a cake can conveythat know-how just by handing the other person a detailed recipe Hayeksensibly believed that a great deal of knowledge, including a great deal of theknowledge utilized in business, is nontransferable just as knowledge of rid-ing a bike is (He believed this in part because he thought a complex systemcould be understood by only a more complex system, and hence the mindcould never fully understand itself I don’t know whether that makes anysense.) This belief strengthened his challenge to the feasibility of centralplanning If much of the knowledge that is scattered across the millions ofeconomically active persons in a society is tacit, then no matter how great

the intellectual capacity of the central planner is, he (or it, if we speak morerealistically of an agency rather than of an individual) will not be able toobtain all the information needed because much of it will be in uncommu-nicable form – uncommunicable except by pricing The price system is thealternative to central planning, and to the extent that knowledge is tacit itwould be a superior alternative even if the central planner had an unlimitedcapacity to absorb and analyze data.

Hayek – this is his most important contribution to economics, and itcomes directly from his cognitive theory – thus offered a new rationale forthe price (or market) system The old rationale, the rationale associated withclassical or neoclassical economics (as distinct from ‘‘Austrian’’ economics,the economics of Hayek, his predecessors Karl Menger and Ludwig vonMises, and his relatively few successors), was that the price system was themethod of overcoming self-interest, of turning private vice into public vir-tue, of leading selfish individuals by an invisible hand to serve the publicinterest People on this view are rational not only in the sense of beingcommitted to means-end reasoning but also in the sense of having a clearperception of the end and of the costs and benefits of the alternative means

to that end However, they are too self-interested, too deficient in altruism,

to be trusted to use their knowledge for social ends and so the trick is toinduce them to behave in a way that will maximize social welfare as a whole;and the price system is the trick Hayek did not think the main problem ofwhich the price system was the solution was self-interest and a resultingshortage of altruism; he thought that people needed the price system inorder to overcome the deficits in their knowledge

The power of that insight was that it denied the feasibility of centralplanning, which at the time that Hayek began propounding his thesis,namely in the 1930s, the depression decade, was considered by most econ-omists a viable and by many a more efficient method of allocating resourcesthan the price system Hayek was one of the first to see that, and to explaincogently why, this was incorrect

Hayek’s second proposition, about the importance of rules as a method fordealing with the knowledge deficits that gave rise to his defense of the pricesystem, belongs to the last part of his career, beginning in the 1960s.9Whereasthe first proposition undergirds Hayek’s defense of free markets, the second isthe key to his political and legal theory With knowledge dispersed and much

of it tacit, there is no way in which a central authority, such as a legislature or

a court, can gather and marshal the knowledge necessary for sensible sions on issues of law or policy The dispersed and tacit knowledge will,however, be found aggregated in rules that grow out of the practices of the