Two Views on Social Stability: An Unsettled Question

Two Views on Social Stability: An Unsettled Question.      

From: The American Journal of Economics and Sociology     |     Date: 10/1/1999             |         Author: BIRNER, JACK;EGE, RAGIP            


Emil Durkheim published The Division of Labour in Society as part of his strategy to create a placefor sociology as a science independent from economics. The bookdescribes how social cohesion and cooperation evolve spontaneously in the course of the process of the division of labour. Friedrich Hayek developed a theory of markets and competition which was later extended into a theory of society, in which spontaneous evolution is a centralelement. The main force behind this process is competition and theevolution of coordination. Both authors address the problem of socialstability. Hayek rejects Durkheim's analysis as constructivistic, but this criticism is unjustified. Further analysis reveals many similarities between the two authors' theories of societal evolution. A striking point of convergence is that Hayek's theory of markets is a network theory, add that sociological network theory is directly inspired by Durkheim's work. The main differences are Hayek's emphasis on the division of knowledge and on coordination as the fundamental stabilizing forces as opposed to Durkheim's stress on the division oflabour and cooperation. The network approach, together with an elaboration of Hayek's psychology, offer perspectives for integrating coordination and cooperation into a unified theory of social stability.

I   Introduction  

   THE QUIP ABOUTECONOMISTS showing that something works and sociologists showing why itdoesn't is usually passed off as just a joke. But like so many jokes,it contains a kernel of truth. A methodologist might point out that itcaptures the idea that sociologists think of themselves as offering amore general theory of society than economists do. Tolerantsociologists leave a place for economic explanations where they areapplicable and true. At the same time they claim their theories to bemore general than economics in that they specify the special conditionsunder which economic explanations are true, while also providing anexplanation of what happens outside the domain of validity ofeconomics. In the Popperian and Polish traditions in the philosophy ofscience this is known as the correspondence principle. [1]
  Establishing a correspondence relation would be an accurate descriptionof the aim of Emil Durkheim when he published his first book in 1893,La division du travail social [The division of labour in society]. Itscontent and method are direct consequences of Durkheim's problemsituation. This is defined by his objective of placing sociology on themap as a scientific discipline in its own right. In order to createthis intellectual space, Durkheim meets sociology's nearest rival,classical political economy, in the doctrine that constitutes itscornerstone, the division of labor. He does so by saying that the mostimportant consequence of the division of labor is not efficiency, butsolidarity. Given the intellectual situation in the "moral sciences" atthe end of last century, this compels Durkheim to define his positionvis-a-vis Adam Smith. According to The Theory of Moral Sentiments, whatmakes a civil society possible is sympathy, the human capability ofimagining the others' position. Sympathy is based on the similarity ofhuman beings. However, the division of labor, which is the subject ofThe Wealth of Nations, presupposes that humans are different from eachother. This is an internal contradiction in Smith's thought, and forDurkheim its solution constitutes the birthright of sociology. However,the fact that he explicitly seeks to create a place for a new socialscience that is independent from economics does not mean that he deniesthat economics has its merits. On the contrary:
   C'est auxeconomistes que revient le merite d'avoir les premiers signale lecaractere spontane de La vie sociale, d'avoir montre que la contraintene peut que la faire devier de sa direction naturelle et que,normalement, elle resu1te, non d'arangements exterieurs et imposes,mais dune libre e1aboration interne (DTS, p. 380). [Credit is due theeconomists for first having seen the spontaneous character of sociallife, and having shown that constraint could only make it deviate fromits natural direction and that, normally, it results not inarrangements which are external and imposed, but in a free internalelaboration. (Durkheim 1964, p. 386)]. [12]
   Despite thisgenerous recognition, Durkheim skillfully maneuvers into the positionof secondary factors the mechanisms that "the economists" (except forSmith and Spencer he is never explicit who they are) think aresufficient to explain social stability and harmony. The whole of BookII of DTS is devoted to a systematic analysis to "the causes andconditions" of the division of labor. By causes Durkheim means thesufficient conditions, and by (secondary) conditions the necessaryconditions. This reveals a rather modern approach to causality. But forDurkheim the main function of the distinction is methodological andstrategic: it serves to define his own position with respect toeconomics. All that he finds of value in economics is relegated to thedomain of necessary conditions, while his own explanatory factorsconstitute the sufficient conditions. Thus, he obtains what he regardsas an incorporation of the economic theory of the division of labor andof the emergence of modem society into his own theory.
   Onthe casual observer modern society leaves an impression of a confusingcomplex of millions of actions of disconnected individuals, each ofwhom is motivated by his own goals, rather than a relativelyharmoniously evolving and stable whole of coordinated behaviors. Thescientific object of Durkheim's investigation in DTS is the explanationof social stability. He is fascinated by the capacity of theindustrialized society of his time to grow without a central organizinginstitution to keep it from falling apart. Durkheim's analysis isinspired by the search for the conditions for the surprising stabilityof society in the face of its apparently anarchic structure.
  Not only is modern society highly stable, it also harbours mechanismsthat enable its members to benefit from its possibilities forself-deployment whose scope and level are unmatched in human history.One of Durkheim's great merits is that he has given expression to thissense of wonder about the fact that a social structure that has notbeen rationally and deliberately organized does not fall apart.
  We find the same sense of wonder and the same fascination at industrialsociety's capacity for self-organization and stability in the work ofthe economist and social theoretician Friedrich Hayek. From the late1920s to the early '40s Hayek's main occupation was to find anexplanation for the lack of coordination and the economic instabilitythat characterizes business cycles. Gradually his interest moved to thegreater question of the stability of society as a whole, and to theproblem of how to preserve the freedom of the individual. His mostimportant publications in this field are The Constitution of Liberty(1960) and Law, Legislation and Liberty (LLL), published in threevolumes in 1973, 1976, and 1979. The similarities between Hayek andDurkheim do not stop here. They both share a more "practical" concern:the intellectual, moral and social crisis that they detect in theircontemporary society. Durkheim speaks of the state of legal and moralanomy in which the economy finds itself (see, for instance, DTS, p.II): conflicts and economic crises, due to the lack of rules withincertain professional groups. The theme that inspires Hayek's work fromThe Road to Serfdom (1944) to his very last book, The Fatal Conceit(1988), is his sense of alarm at the intellectual hubris that makesmodern humans think they can organize the complex processes thatcharacterize the evolution of social institutions according to theirdesires. Hayek fears that this attitude may destroy everything that hasbeen achieved in the domain of freedom of the individual, As to thequestion of how to solve the social problems they analyze, both authorsshare the conviction that a--possibly violent--solution imposed fromthe outside would not work. [3] There are many more similaritiesbetween Durkheim and Hayek, as we hope to make clear.
   Thesame sense of wonder about the stability of society, which is the rootof political economy and sociology, inspired Adam Smith to develop theidea of the invisible hand. Both Durkheim and Hayek declare themselvesto be the intellectual heirs of the Scottish philosopher, although theyemphasize different aspects of his thought. This is closely related tothe different intellectual traditions in which the two authors placethemselves. The tradition to which Durkheim belongs emphasizes the roleof law in society, while Hayek's lineage pays particular attention tothe competitive market. The difference can be formulated as thatbetween cooperation and coordination. Our point of departure in thisarticle is that the similarities and differences between sociologistDurkheim's and economist Hayek's theories of social stability justifyour choosing them as representatives of alternative explanations ofsocial stability. Our goal is to give an answer to the questions ofwhat each has to contribute, how these cont ributions are related, andhow we may use them to make progress in the explanation of socialstability.

[ 本帖最后由 fussfun 于 2007-7-1 19:39 编辑 ]

  II The Division of Labor and Social Cohesion

  DURKHEIM TRIES TO MAKE the characteristics of modem industrial societyclear by contrasting it with earlier, more primitive forms of socialorganization. This is more of a methodological device than a realhistorical analysis. His "segmentary society" is a fiction orconceptual artifact that allows him to define the institutions andmechanisms of modern society, rather than a truthful descriptiveaccount. In Durkheim's conjectural history, the most primitive state ofsociety is the horde. It is characterized by a type of cohesion that isdue to similarity:

   Si l'on essaye de constituer par laspensee le type ideal d'une societe dont la cohesion resulteraitexclusivement des ressemblances, on devra la concevoir comme une masseabsolument homogene dont les parties ne se distingueraient pas les unesdes autres, et par consequent ne seraient pas arrangees entre elles,qui, en un mot, serait depourvue et de toute forme definie et de touteorganisation. Ce serait le vrai protoplasme social, le germe d'ouseraient sortis tous les types sociaux. Nous proposons d'appeler bordel'aggregat ainsi caracterise (DTS, p. 149). [If we try to constructintellectually the ideal type of a society whose cohesion wasexclusively the result of resemblances, we should have to conceive itas an absolutely homogeneous mass whose parts were not distinguishedfrom one another. Consequently they would have no arrangement; in shortit would be devoid of all definite form and all organization. It wouldbe the veritable social protoplasm, the germ whence would arise allsocial types. We propo se to call the aggregate thus characterized,horde. (p. 174)].

   Primitive society is a repetition ofidentical aggregates of hordes. Its structure is, in terms that weborrow from Herbert Simon, maximally redundant. [4] The next step inthe development of society is the clan: a horde that has ceased to beindependent and has become the element of a more extended ("plusetendu") group: segmentary society. Clan chiefs are the only form ofsocial authority in this structure, marking the beginning of adiversification. Still, in these "inferior societies," the only form ofsolidarity is that which derives from similarity. This type of societyis no longer purely hypothetical. Drawing on Fustel de Coulanges,Durkheim gives the examples of Australian aborigines, Indian tribes,etc. In such societies, religion is all-pervasive. Because in a smallsociety everybody is faced with the same conditions of existence, thecollective environment is essentially concrete. Individual experiencesare the same and have as their objects (the same) specific things. Thecollective character is well-defined.

   Durkheim's segmentarysociety is ruled by "droit repressif." Repressive law punishes thoseacts that offend the collective convictions and that are aninfringement on the rules and values that are shared by the communityas a whole. So, repressive law is the rule in those societies in whichcriminal acts need not be explicitly codified. Its rules are firmlyrooted in the collective memory that each individual carries with him.In such a society there is no room for individual variations. Eachsegment is the bearer of the same strong feelings, the same traditionalvalues, and the same social rules the infraction of which justifies theseverest form of punishment.

   Since the segments areautarchic, there is no need (or incentive) for exchange between them.The solidarity in this society is solidarity by likeness ("parsimilitudes"), which Durkheim calls mechanical solidarity. This societyis characterized by "communism":
   Le communisme, en effet,est le produit necessaiore de cette cohesion speciale qui absorbel'individu dans le groupe, la partie dans le tout. La propriete n'esten definitive que l'extension de la personne sur les choses. La donc oula personnalite collective est la seule qui existe, la properieteelle-meme ne peut manquer d'etre collective. Elle ne pourra devenirindividuelle que quand l'individu, se degageant de la masse, seradevenu, lui aussi, un etre personnel et distinct, non pas seulement entant qu'organisme, mais en tant que facteur de la vie sociale (DTS, pp.154-55). [Communism, in effect, is the necessary product of thisspecial cohesion which absorbs the individual in the group, the part inthe whole. Property is definitive only of the extension of the personover things. Where the collective personality is the only one existent,property also must be collective. It will become individual only whenthe individual, disengaging himself from the mass, shall become a beingpersonal and distinct, not only as an organism but also as a factor insocial life. (p. 179)].

    There is no possibility for individual personality to develop in a  society where collective conscience rules supreme.  
  As we observed above, in Durkheim's discussion of the necessaryconditions for the division of labor to progress we find the moretypically economic factors, such as the need for individuals to developtheir individual traits, innovation, and, most fundamentally,rationality. This last factor appears in the form of an explanation ofthe development of rational thought. Individuals become moreindependent from the groups to which they belong when the members of asociety increase in number. When societies become more voluminous,i.e., when their physical and hence their social density increases,this collective conscience changes. It is forced to elevate itselfabove local differences and to cope with space and distance; hence ithas to become more abstract. It is thus that abstract concepts arise(DTS, p. 272). For Durkheim, the more general collective consciencebecomes, the more room it leaves for individual variations (DTS; p.275). Since collective conscience is almost entirely a product of thepast (so, of tradition), its role diminishes as segmentary society isleft behind. Within the extended group there is more individual liberty(DTS, p. 284). When society becomes dispersed over a larger area it hasto become more abstract:

   [e]lle est elle-meme obligee des'elever au-dessus de toute les diversites locales, de dominerdavantage lespace et, par consequent, de devenir plus abstraite. Car iln'y a guere que des choses generales que puissent etre communes a tousces milieux divers. Ce n'est plus tel animal, mais telle espece; tellesource, mais les sources; telle foret, mais la foret in abstracto (DTS,p. 272). [the common conscience "is itself obliged to rise above alllocal diversities, to dominate more space, and consequently to becomemore abstract. For not many general things can be common to all thesedivers environments. It is no longer such animal, but such a species;not this source, but such sources; not this forest, but forest inabstracto" (p. 287)].
    This also explains the increasing rationalization of society:

  Cela seul est rationnel ce qui est universel. Ce qui deroutel'entendement, c'est le particulier et le concret. Nous ne pensons bienque le general, Par consequent plus la conscience commune est prochedes choses particulieres, plus elle en porte exactement l'empreinte,plus aussi elle est inintelligible (DTS, p. 275. [5] [This alone isrational which is universal. What baffles understanding is theparticular and the concrete. Only the general is thought well of.Consequently, the nearer the common conscience is to particular things,the more it bears their imprint, the more unintelligihle it also is.(p. 289-290)].
   Once rational thought has emerged, there isno way of keeping it in check. This is a mixed benefit, as instinctalways has a more compelling force than reason: "Parce qu'elle devientplus rationnelle, la conscience collective devient donc moinsimperative, et, pour cette raison encore, elle gene moins le libredeveloppement des varietes individuelles" (DTS, p. 276). ["Because itbecomes more rational, the collective conscience becomes lessimperative, and for this very reason, it wields less restraint over thefree development of individual varieties." (p. 290-291)]. However, themain force of collective consciousness derives not so much from thefact that it is shared by contemporaries as from the fact that it is aproduct of the past which took a long time to develop.

  L'autorite de la conscience collective est donc faite en grande partiede l'autorite de la tradition. Nous allons voir que celle-ci se diminuenecessairement measure que le type segmentaire s'efface (DTS, p. 277)[The authority of collective conscience is, then, in large partcomposed of the authority of tradition. We shall see that the latternecessarily diminishes as the segmental type is effaced (p. 291)].

  As society becomes more open and mobile, tradition loses its sway. Forinstance, the more advanced a society, the less the aged are held inreverence. It is from the young that change is to be expected, providedthey have dissociated themselves from the erroneous practices of thepast.

   Modern society is characterized "non par unerepetition de segments similaires et homogenes, mais par un systemed'organes differents dont chacun a un role special, et qui sont formeseux-memes de parties differenciees" (157) ["not by a repetition ofsimilar, homogeneous segments, but by a system of different organs eachof which has a special role, and which are themselves formed ofdifferentiated parts" (p. 181)]. Industrial society rests on anadvanced division of labor, a degree of specialization anddifferentiation which is inconceivable in segmentary society. Theelements of which modern society consists no longer exist side by sideas the links of a chain, but show a more complicated, hierarchicpattern that is defined by the logical requirements of their function.They are "coordonnes et subordonnes les uns aux autres autour d'un memeorgane central" (157) ["co-ordinated and subordinated one to anotheraround the same central organ" (p. 181)]. However, individuals are notpassively subjugated to the moral prescriptions of collectiveconscience, as they were in primitive society, where they wereinterchangeable and where mechanical solidarity cemented the socialbonds. The type of cohesion which rules in modern society is organicsolidarity. Durkheim agrees with Spencer that social harmony derivesfrom the division of labor, but criticizes him (in ch. I.VII) forthinking that industrial solidarity is characterized by spontaneity--sothat there is no need for a repressive apparatus--and for thinking thatthe only link among individuals is that of absolutely free exchange andhence competition. But neither is it true that self-interest is theonly basis of social stability (DTS, pp. 180-81). The growing numberand complexity of legal rules that develop as society progresses issufficient to make this clear. Nevertheless, Durkheim observes that theidea of a social contract in the sense of Rousseau is also hard todefend. More specifically, repressive law cedes more and more torestitutive law, which aims no t at punishing infractions of agenerally valid moral code, but serves as a means for redress so as toput things back into the order that they should have been (DTS, p. 79).The situations to which restitutive laws apply are usually veryspecific, so that most members of society remain in complete ignoranceabout these legal rules. As the division of labor advances, so does thetechnicality of the legal rules. The increasing complexity of socialrelationships in modern society has necessitated an ever more complexsystem of laws to manage them. This does not mean that judges candecide arbitrarily; they remain bound by the rules that are generallyaccepted in society.
   Like Comte, Durkheim believes that whatis really spontaneous is society itself. Neither the division of labornor competition can be considered as spontaneous phenomena in the truesense of the word. Admittedly, when the organic division of laboremerges, it solidifies the social bonds. But that does not mean that itcreates them. "Cette interaction en suppose une autre qu'elle remplace"(DTS, p. 262) ["this integration supposes another which it replaces"(p. 278)]. And further on we read:
   Les organismes pluscomplexes se forment par la repetition d'organismes plus simples,semblables entre eux, qui ne se differencient qu'une fois associes. Enun mot l'association et la cooperation sont deux faits distincts, et sile second, quand il est developpe, reagit sur le premier et letransforme, si les societes humaines deviennent de plus en plus desgroupes de cooperateurs, la dualite des deux phenomenes ne s'evanouitpas pour autant (DTS, pp. 262-63). [more complex organisms are formedby the repetition of more simple, similar organisms which aredifferentiated only if once associated. In short, association andco-operation are two distinct facts, and if the second, when developed,reacts on the first and transforms it, if human society become groupsof co-operators, the duality of the two phenomena does not vanish forall that (p. 278-279)].

   Specialization and differentiationtake place on the basis of association. For Durkheim, the social wholealways precedes the individual parts. See for example DTS, p. 264,where he writes: "La vie collective n'est pas nee de la vieindividuelle, mais c'est, au contraire, la seconde qui est nee de lapremiere" ["Collective life is not born from individual life, but itis, on the contrary, the second which is born from the first" (p.279)].

   Comte believes that the division of labor, if pushedtoo far, will result in the disintegration of society. Individuals whospecialize in ever more narrowly defined and abstract tasks will retireinto their own private and solitary universes. Here, the division oflabor becomes a centrifugal force that makes society fall apart intoisolated atoms. A different unifying principle is needed tocounterbalance this disintegrating force. The government or the statemust intervene:

   Il est clair, en effet, que le seul moyenreel d'empecher une telle dispersion consiste a eriger cetteindispensable reaction en une nouvelle fonction speciale, susceptibled'intervenir convenablement dans I'accomplissement habituel de toutesles diverses fonctions de l'economie sociale, pour y rappeler sanscesse Ia pensee de l'ensemble et le sentiment de la solidarite commune(Durkheim quoting Comte, DTS, p. 349). [It is clear, in effect, thatthe only real means of preventing such a dispersion consists in thisindispensable reaction in a new and special function, susceptible offittingly intervening in the habitual accomplishment of all diversefunctions of social economy, so as to recall to them unceasingly thefeeling of unity and the sentiment of common solidarity (p. 358-359)].

  According to Comte, the solidarity that is produced by the division oflabor is more fragile than the cohesion of a society that rests on theprinciple of likeness or homogeneity. For this reason he advocates theconscious intervention of the state in creating solidarity, and, in"Note sur la definition du socialisme" (Durkheim 1893), socialism.Whereas Durkheim shares the idea that there must be rules, he does notshare Comte's pessimism. Durkheim thinks Comte's pessimism is based onhis failure to recognize the real nature and the power of organicsolidarity. The malfunctions and anomalies of modern society are not somuch due to the disappearance of the pervasiveness of the common goalsof segmentary society as to the slowness with which individuals adoptthe rules and regulations that are necessary for a peaceful andharmonious coexistence of functions that have become separated by thedivision of labor. The highly efficient mechanisms that createdcohesion in traditional society have been destroyed by t he progress ofspecialisation in modern society and a new equilibrium has not yet beenfound:

   Des changements profonds se sont produits, et en trespeu de temps, clans la structure de nos societes; elles se sontaffranchies du type segmentaire avec une rapidite et clans desproportions dont on ne trouve pas un exemple dans histoire. Par suite,la morale qui correspond a ce type social a regresse, mais sans quel'autre se developpat assez vite pour remplir le terrain que lapremiere laissait vide de nos consciences (DTS, p. 405) [Profoundchanges have been produced in the structure of our societies in a veryshort time; they have been freed from the segmental type with arapidity and in proportions such as have never before been seen inhistory. Accordingly, the morality which corresponds to this socialtype has regressed, but without another developing quickly enough tofill the ground that the first left vacant in our consciences (p.408)].
   Hence the crises and anomalies to which Durkheimdevotes the third and last book of DTS. Durkheim repeatedly andexplicitly mentions that the system of rules which serve to avoidmalfunctions and anomies in modern society must be the product of aspontaneous process, lest it create instability: "la vie sociale,partout ou elle est normale, est spontanee; et si elle est anormale,elle ne peut pas durer" (180) ["social life, wherever is normal, isspontaneous, and if it is abnormal, it cannot endure" (p. 202-203)].This is a very important point in his thought. He also argues thatrules that do not suit the individuals in their specific situations andwhich are imposed from above create the anomaly of the "division dutravail contrainte." As we have seen, Durkheim opposes the idea thatcompetition alone can lead to a stable institutional framework.
  Competition has a centrifugal effect on society if it occurs in asituation in which there is no social cohesion; only in conditions ofsolidarity does it contribute to social stability. This conclusionserves to further emphasize the difference with the economic theoryaccording to which cooperation emerges as a consequence of competition:
   Puisque la concurrence ne peut pas avoir determine cerapprochement, il faut bien qu'il ait preexiste; il faut que lesindividus entre lesquels la lutte s'engage soient deja solidaires et lesentent, c'est-a-dire appartiennent a la meme societe (DTS, pp.259-60). [Since competition cannot have determined this conciliation,it must have existed before. The individuals among whom the struggle iswaged must already be solidary and feel so. That is to say, they mustbelong to the same society (p. 276)].
   But despite the factthat he thinks he has incorporated economics in his own theory, hegives no description of the institution which, according to Smith,determines the extent of the division of labor: the market. [6] That isunderstandable since even economists rarely offer an analysis of thefunctioning of markets. Among the few exceptions is Friedrich Hayek, toa discussion of whose work we now turn.


  III  Equilibrium, Markets, and Coordination


HAYEK REJECTS THE STANDARD NEOCLASSICAL ANALYSIS of markets because it is static and tautological. Static, because it provides aclassification of various types of markets that are defined in terms ofstatic conditions, such as the number of sellers or purchasers and the shape of the demand and supply curves. Tautological, because it makes anumber of assumptions that deprive its models of empirical content. As far as this latter aspect is concerned, Hayek singles out the assumptions of perfect competition and perfect information. The neoclassical assumption of perfect competition describes a situation in which goods and sellers or producers are completely homogeneous and the price is given. This characterizes the absence of any competition. The assumption of perfect information is one of the conditions of market equilibrium. Here Hayek's criticism is that equilibrium is defined as the state in which all buyers and sellers have perfect information.

  The concept of equilibrium is central to Hayek's alternative explanation of the way in which markets work. [7] The basic unit of analysis is the planning individual. The idea of a plan logically presupposes rime. Hayek defines equilibrium as the correspondence between the expectations on which each individual bases his or her plans and the informational input which serves as feedback, The individual applies the "pure logic of choice" to his or her own preferences and his or her perception of the environment when planning his or her behavior. An economic system is composed of a multitude ofsuch perceiving, planning, and utility-maximizing individuals whointeract and communicate with each other. The system is in equilibriumif the plans of all individuals are compatible with one another.Markets are the social institutions in which individuals exchange goodsand services using their prices as guidelines. The interaction onmarkets creates a communication structure that transmits priceinformation efficiently and rapidly since individuals' fields ofperception are partially overlapping. Competition is a crucial elementin the spreading of information; it consists of a process in whichindividuals actively seek to discover new and so far untriedopportunities. Hayek emphasizes the role of learning and knowledge bysaying that the division of knowledge and its coordination are morefundamental than the division of labor. Markets are social institutionsthat have arisen and survived in an evolutionary process because theyresponded to the need for coordinating dispersed knowledge. They createthe high degree of availability of knowledge to everyone thatcharacterizes developed economies. [8] Their functioning relies on anefficient communication structure.

   The inclusion of thestructure of communication in the analytical framework is very unusualin economics, [9] and has a number of far-reaching consequences. One isthat it highlights the importance of an agent's position for theacquisition of the knowledge which is necessary for the economic systemto show a tendency to an equilibrium: "the relevant knowledge which hemust possess in order that equilibrium may prevail is the knowledgewhich he is bound to acquire in view of the position in which heoriginally is, and the plans which he then makes" (Hayek 1937, p. 53).This introduces the element of what we may call position-constrainedlearning. The passage just quoted continues:

   It is certainlynot all the knowledge which, if he acquired it by accident, would beuseful to him and lead to a change in his plan. We may therefore verywell have a position of equilibrium only because some people have nochance of learning about facts which, if they knew them, would inducethem to alter their plans. Or, in other words, it is only relative tothe knowledge which a person is bound to acquire in the course of theattempt to carry out his original plan that an equilibrium is likely tobe reached (Hayek, 1937, p. 53).

   Another factor that we findhere is path-dependency. Hayek also discusses the amount of knowledgewhich would be needed in a decentralized system in order that it mayreach the same equilibrium that an omniscient dictator would impose as"a sort of optimum position" (Hayek 1937, p. 53) [10]:

   One condition [for the decentralized system equilibrium to coincide withthat of the centralized dictator-economy] would probably be that eachof the alternative uses of any sort of resources is known to the ownerof some such resources actually used for another purpose and that inthis way all the different uses of these resources are connected,either directly or indirectly (ibid.).

    The note to this sentence elaborates:

  That it is not necessary, as one might think, that every possible useof any kind of resources should be known to at least one among theowners of each group of such resources which are used for oneparticular purpose is due to the fact that the alternatives known tothe owners of the resources in particular uses are reflected in theprices of these resources. In this way it may be a sufficientdistribution of knowledge of the alternative uses, m, n, o, . . . y, z,of a commodity, if A, who uses the quantity of these resources in hispossession for m, knows of n, and B, who uses his for n, knows of m,while C, who uses his for o, knows of n, etc., until we get to L, whouses his for z, but knows only of y. [11]

   As we haveobserved, Hayek's criticism of traditional equilibrium theory is thatit is circular: "Correct foresight is ... not, as it has sometimes beenunderstood, a precondition which must exist in order that equilibriummay be arrived at. It is rather the defining characteristic of a stateof equilibrium" (1937, p. 42). [12] This criticism may be reformulatedas follows: The assumption of correct foresight implies thatindividuals have full access to all available knowledge about thefuture. In other words, there are no gaps or other imperfections in theintertemporal communication structure. For their current exchangerelationships the assumption of perfect information implies a similarperfection of the present communication structure. Neither is the casein reality. So, the perfect information assumption relegates thestandard neoclassical 'analysis at the most to the status of a limitingcase, or an idealizing model without empirical content. On the otherhand, markets and competition are considered t o be crucial elements increating an efficient structure of communication. Neoclassical analysishas little or nothing to say about this, Hayek does: "The whole acts asone market, not because any of its members survey the whole field, butbecause their limited individual fields of vision sufficiently overlapso that through many intermediaries the relevant information iscommunicated to all" (1945, p. 86). However, Hayek does not see thatthis defines a research problem rather than a solution (which hepresents in the form of the price system). To mention one problem, thespeed with which the information is transferred obviously matters. Ifthe local information spreads slowly, there may be no tendency towardsequilibrium.

   The acquisition of implicit knowledge andexperience by means of personal contacts is another element that isintroduced with the communication structure. Hayek speaks of "a body ofvery important but unorganized knowledge" (1945, p. 80) which is notscientific. "We need to remember only how much we have to learn in anyoccupation after we have completed our theoretical training, how big apart of our working life we spend learning particular jobs, and howvaluable an asset in all walks of life is knowledge of people, of localconditions, and of special circumstances" (Hayek, 1945, p. 80).

  A great merit of Hayek's analysis of markets is that he has calledattention to the crucial role of an economy's interaction andcommunication structure. In fact, Hayek's explanation is one of theearliest instances of a network analysis of markets, whose explanatoryfactors include connectivity, [13] the strength and frequency ofinteractions, the development of personal relationships, and thetransmission of information. It is one of the curious facts ofintellectual history that until recently Hayek's type of analysis ofmarkets was only taken further by one economist, G. B. Richardson, [14]and that this work, too, failed to produce a research tradition.Network analyses of markets by economists first began to reappear,uninfluenced by either Hayek or Richardson, with the work of AlanKirman and Rob Gilles. [15] Network analysis is a much more flourishingresearch tradition in sociology. The most important applications ofnetwork models to markets and competition can be found in the work ofMark Granovet ter, Harrison White, and Ronald Burt. [16] Although verysimilar in spirit to Hayek's approach, these sociologists do not referto it either. This is apparently an instance of disconnectedintellectual networks. What can be the reason for this lack ofinfluence on economics of network analysis by sociologists andeconomists?

   The hypothesis that seems to arise from ourprevious analysis is the following. Sociological network analysis isdirectly inspired by Durkheim. He introduced an approach that isdifferent from that of economics. Network analysis has inherited thisdifference. This would explain why it has failed to create ananalytical tradition in economics. In order to examine the validity ofthis idea we propose a comparison between Durkheim's social theory inDTS and Hayek's theory of society. But, we will first devote a coupleof paragraphs to showing how Hayek's economic theory developed into atheory of society.

[ 本帖最后由 fussfun 于 2007-7-1 20:02 编辑 ]


  IV Social Institutions as Coordinating Devices


  THE CENTRAL PROBLEM of Hayek's analytical economics is the explanationof the lack of coordination which causes disequilibrium growth. Theproblem arises because individuals only perceive their direct economicenvironment. Prices are part of this environment. The monetary systemhas a dynamics of its own which permits the amount and the division ofmeans of exchange in society to diverge from the barter ratio betweengoods that reflect the conditions of real scarcities. Whenever thisdivergence occurs, the decisions to consume, save and invest, which arenecessarily based on the only prices that can be perceived, viz. moneyprices, are mistaken. When the real conditions have revealed themistakes, it is too late to put things right immediately, and theresult are business cycles (i.e., disequilibrium growth). Like themarket, the monetary system has evolved spontaneously as a solution tothe need to enter into exchange relationships with ever more distanteconomic units. In that sense it is a solution to the problem of socialcomplexity. However, this particular solution has its cost: economicfluctuations.
Here we have in a nutshell the three elements that Hayek  generalized gradually into a theory of society:

    1. the basic problem of an economy (society) is the problem of  coordination;  
  2. individuals have only limited knowledge; one consequence is that theeconomy (society) as a whole appears to them as a phenomenon of such adegree of complexity that individuals by themselves will never be ableto oversee, to understand all the detail of, nor to coordinate all theindividual transactions with others that make up an economy (asociety);
   3. markets (social institutions) are solutions tothe problem of social complexity that have spontaneously developed inan evolutionary process.

   It is with respect to the economicsystem that Hayek first expresses his sense of wonder at the fact thatthe millions of interactions between individuals do not result in totalchaos; instead, what we usually observe is a relatively stable set ofrepeated interactions that occur according to some set of rules thatare shared by most. [17] The facts that human knowledge is so severelylimited and that the spontaneously evolved social institutions haveapparently mastered the complex problem of social coordination leadsHayek to defend non-interventionism and liberalism. He contrasts thiswith the tradition which he calls "constructivism" and which heidentifies with the tradition of Cartesian rationalism and the FrenchSchool of Engineers. He accuses it of seeing only one solution tosocial complexity, conscious intervention and regulation. Prominentrepresentatives of this tradition are Saint Simon and Comte.

  In Hayek's social philosophy, history plays a role that is very similarto that in Durkheim's. He introduces the "tribal society," not as adescription of a real historical past, but as an analytical device tolend contrast to his analysis of what makes modern society work. Healso develops an evolutionary theory of society which culminates in histheory of cultural evolution. It may be considered to be the scientificunderpinning of the contrast between the mechanisms that rule thetribal and the open societies. Hayek develops his evolutionary theoryin, for example, "Notes on the Evolution of Systems of Rules ofConduct" (1967), Law, Legislation and Liberty, and his last book, TheFatal Conceit (FC, 1988). There are three kinds of evolution in humanaffairs: genetic evolution, which produces instincts and instinctivebehavior; the evolution of rational thought; and cultural evolution.Culture occupies an intermediate position between instinct and rationalthought, not only in the course of the development of the species, butalso logically and psychologically. This is Hayek's rudimentaryexplanation of the emergence of rational thought. [18] Instinctivebehavior is sufficient for the coordination of the actions ofindividuals within small primitive groups, the members of which havecommon perceptions and objectives. On the other hand, within thedeveloped and "abstract" society (or, which is the same, the "extendedorder"), which is too complex to be fully understood by the human mind,coordination is ensured by abstract rules that have developedgradually. These rules govern private property, honesty, contracts,exchange, commerce, competition, profit, and the protection of privacy.So, they have very much the same function as solidarity does inDurkheim. Those rules are transferred by tradition, learning, andimitation (FC, p. 12). There is a continuous tension between the rulesgoverning individual behavior and those governing the functioning ofsocial institutions. The formation of abstract systems and institutions of coordination have forced individuals to change theirnatural or instinctive reactions (ibid., p. 13). The fact that theirbehavior is still largely ruled by the instincts of the tribal societyand has not kept pace with the development of the abstract societyexplains why they try to oppose these systems. "Disliking theseconstraints so much, we hardly can be said to have selected them;rather these constraints selected us: they enable us to survive" (p.14). The institutions that emerge are the result of certain individualsstumbling upon solutions to particular problems in a process ofcompetition. Indeed, competition as a process of discovery is part ofevery evolutionary process (p. 19).

   For understandingHayek's development from a technical economist to a philosopher ofsociety, "Individualism: True and False" (ITF, 1945) is an importantarticle. It is also one of the few places where he refers to Durkheim,two reasons for paying attention to it. In ITF Hayek works out theconsequences of the theory of society of the Scottish Enlightenment andits individualist methodology.

   This argument is directedagainst the properly collectivist theories of society which pretend tobe able directly to comprehend social wholes like society, etc., asentities sui generis which exist independently of the individuals whichcompose them (ITF, p. 6).

   This has to be distinguished fromthe so-called individualism of the Cartesian school, which is usuallyreferred to as rationalism. This is why Hayek calls the trueindividualism of the Scottish Enlightenment antirationalism.

  The antirationalistic approach, which regards man not as a highlyrational and intelligent but as a very irrational and fallible being,whose individual errors are corrected only in the course of a socialprocess, and which aims to make the best of a very imperfect material,is probably the most characteristic feature of English individualism(ITF pp. 8-9). [19]

   This insight is due to Mandeville. Themain differences between the pseudo-individualism of the rationalisticor engineering tradition on the one hand and the true individualism ofthe Scots are that "true individualism is the only theory which canclaim to make the formation of spontaneous social productsintelligible," and which "believes ... that, if left free, men willoften achieve more than individual human reason could design orforesee" (pp. 10-11). This has consequences for political philosophy:

  The great concern of the great individualist writers was indeed to finda set of institutions by which man could be induced, by his own choiceand from the motives which determined his ordinary conduct, tocontribute as much as possible to the need of all others... (ITF, pp.12-13).

   Hayek emphasizes the anti-rationalistic character ofthis philosophy, which is a view which in general rates rather low theplace which reason plays in human affairs, which contends that man hasachieved what he has in spite of the fact that he is only partly guidedby reason, and that his individual reason is very limited andimperfect... One might even say that the former is the product of anacute consciousness of the limitations of the individual mind whichinduces an attitude of humility toward the impersonal and anonymoussocial process by which individuals help to create things greater thanthey know... (ITF, p. 8).

    The great discovery of the classical economists is that  
  many of the institutions on which human achievements rest have arisenand are functioning without a designing and directing mind... and thatthe spontaneous collaboration of free men often creates things whichare greater than their individual minds can ever fully comprehend (ITF;p. 7).


  V The Relationship between Sociology and Economics Revisited


  WE HAVE OBSERVED THAT Smith is an intellectual ancestor that Hayek andDurkheim both share. Hayek would not have been happy with this commonheritage. In ITF (as well as in other publications, the most importantof which here is The Counterrevolution of Science) he criticizes"rationalist constructivism," the tradition that he identifies withDescartes and Comte. He presents this tradition as diametricallyopposed to the ideas of the Scottish Enlightenment. The fact thatDurkheim explicitly recognizes his intellectual debt to Comte mayexplain why Hayek condemns him as a constructivist: "It is in theinsistence on social 'solidarity' that the constructivist approach tosociology of Auguste Comte, Emil Durkheim and Leon Duguit shows itselfmost clearly" (LLLII, p. 11, n. 9). Hayek demonstrates a fundamentalmistrust of the idea of cooperation and an almost dogmatic emphasis onthe efficiency of coordination as the binding principle in society.Compare, for instance, the following passage in The Fatal Conceit: [20]

   One revealing remark of how poorly the ordering principle ofthe market is understood is the common notion that 'cooperation isbetter than competition'. Cooperation, like solidarity, presupposes alarge measure of agreement on ends as well as on methods employed intheir pursuit. It makes sense in a small group whose members shareparticular habits, knowledge and beliefs about possibilities. It makeshardly any sense when the problem is to adapt to unknown circumstances;yet it is this adaptation to the unknown on which the coordination ofefforts in the extended order rests (FC, p. 19).

   Hayekrefuses to understand by "cooperation" anything but a conscious,deliberate, and intentional act of solidarity. As such, cooperation canonly take place in the small face-to-face tribal group with its sharedgoals. However, what Durkheim has in mind when he speaks of cooperationis not the behavior of individuals who, one day, decide to get togetherand agree on common objectives. On the contrary. What he means are thevarious forms of interaction and complementarity between the"functions" which emerge with the increasing division of labor insociety. The individuals are no more than the holders or performers ofthese functions, and through their relationships with other individualsthe various specialized social organs will be in touch with otherorgans in society. It would be entirely justified to apply Ferguson'sexpression "the result of human actions but not of human design," oneof Hayek's favourite quotations, to the durkheimian process of theemergence of cooperation. We have already seen that Durkheim's idea ofcooperation is much subtler than Hayek thinks. For instance, Durkheimis very critical of Spencer's thesis that every society consists ofcooperation. To this he opposes Comte's idea that cooperation does notgive rise to society; on the contrary, it presupposes its existence.[21]

   We have taken Hayek as a representative of an economicapproach to social institutions and social stability in whichcoordination is a central element. However, the distance between hisanalysis and Durkheim's, whom we have chosen as a representative of thesociological approach in which cooperation is crucial, seems to existmore in Hayek's subjective opinion than in his analysis. The conclusionappears to be either that Hayek has not read Durkheim very well (or notat all), or that he read him but forgot the argument. The phrase to theextent that Durkheim is a constructivist, which we quoted above, is anote to a passage stating that the Great Society has nothing to do withsolidarity in the "true" sense of conscious unitedness in the pursuitof common goals. Durkheim says the exact opposite. Hayek's injusticewith regard to Durkheim is compounded in LLLIII, where he accusesDurkheim of being the originator of confusing altruistic with moral.Yet Durkheim's strategy for creating a social science independe nt fromeconomics starts with his taking issue with exactly this idea.Ironically, what Hayek presents as criticism is more like a restatementof Durkheim's ideas, couched in Hayek's language. [22]


  VI  Cooperation versus Coordination?

  HOWEVER, THIS IS A MATTER OF INTELLECTUAL HISTORY. Here we are moreinterested in the question of whether there exist fundamentaldifferences between sociology and economics as represented by Durkheimand Hayek. We have seen how according to Durkheim the division of laborin society, accompanied by restitutive law as the observable expressionof a growing social cohesion or solidarity, goes hand in hand with thedevelopment of individual personality. This is only possible in asociety which provides the material and spiritual conditions forindividuals to distinguish themselves from the collective. The form ofsociety that gives individuals the space for developing their ownpersonal characters is a society that has learned, in a manner ofspeaking, not to severely punish those who dare take their distancefrom the collective values. [23] Here, the visions of Durkheim andHayek are very close indeed. Hayek's Great Society is only possible ifthere is a framework that allows individuals to diverge from theprescriptions of accepted morality and learn to pursue their ownindividual objectives, without having to justify them. The marketorder, or catallaxy, constitutes such a framework:
   The GreatSociety arose through the discovery that men can live together in peaceand mutually benefiting each other without agreeing on particular aimswhich they severally pursue. The discovery that by substitutingabstract rules of conduct for obligatory concrete ends made it possibleto extend the order of peace beyond the small groups pursuing the sameends, because it enabled each individual to gain from the skill andknowledge of others whom he need not even know and whose aims could bewholly different from his own (LLLII, p. 109). [24]

   Thecontinuity of the market order which makes this peaceful coexistenceand the stability of the Great Society possible presupposes the beliefby its members in the existence and beneficial effects of spontaneouscoordination mechanisms that have not been consciously planned orprogrammed. At the very least, it presupposes their willingness not totamper with them. Here lies the vulnerability of the market order. LikeDurkheim, Hayek argues that since this social organization form is sucha recent discovery, the individual values are still more adapted to theenvironment of the primitive group. The risk that threatens the marketorder is "the revival of the organizational thinking of the tribe"LLLII, p. 134), where consciously imposed concrete rules were the norm.In order to preserve the catallaxy, its members must learn to renouncethe sort of solidarity that is the result of the adoption of commongoals. Hayek condemns socialism as a reassertion of "that tribal ethicswhose gradual weakening has made an approach to the Great Societypossible" (ibid.).

   According to Durkheim, individualpersonality develops as the division of labor and restitutive lawoccupy an ever more important place in society. According to Hayek, itsdevelopment goes together with the extension of the market andcompetition in an evolutionary process. [25] Durkheim finds the unitythat characterized segmentary society back, in a more stable and moredeeply rooted form, in the cooperation between the many differentfunctions which is the essence of the division of labor and the socialcohesion that results from the recognition of the mutual dependence inmodern society. For Hayek, the coordination of individual actions thatis achieved by markets and competition, together with the "minimaljustice" of the application of abstract rules of behavior aresufficient conditions for the cohesion that allows individuals topursue their personal goals, provided they respect this spontaneousorder.

   So, for Durkheim solidarity comes about throughcooperation, whereas according to Hayek at least a minimal form ofcohesion is the result of coordination.


  VII  Collective versus Individual?

   THIS DOESNOT APPEAR to be the only difference. Hayek's methodology is generallyidentified with individualism and Durkheim's with collectivism. Forexample, his definition of a social fact presupposes the existence of acollective conscience, [26] whereas for Hayek a social fact is based onindividual expectations and perceptions. However, for Hayek rules andindeed rationality are fundamentally social phenomena--this is themessage of, for instance, ITF [27]--and for Durkheim the collectiveconscience must be internalized by the individuals of whom societyconsists; they must feel that they belong to the same society (DTS, p.260). Also, when Durkheim speaks of the mechanical causes and forcesthat make men live more closely together, he mentions consanguinity,sharing the same land, the cult of ancestors, and having in common thesame habits (p. 262). Perhaps with the exception of the first two, allof these factors are mental and attitudinal rather than physical orobjective. These are the factors on the basis of which groups form, andonly after groups have formed does cooperation evolve. So on the matterof the (social) part-whole relationship the two authors are much closerto one another than they appear to be. The point of contact lies inwhat Hayek calls the primacy of the abstract, which is mirrored byDurkheim's explanation of the emergence of abstract thought andrationality.

      VIII  The Division of Knowledge

  AN ASPECT IN WHICH HAYEK seems to differ from Durkheim is in his theoryof the distribution of knowledge and its coordination:

   Farmore important than this moral attitude [of caring for oneself andone's family], which might be regarded as changeable, is an undisputedintellectual fact which nobody can hope to alter and which by itself isa sufficient basis for the conclusions which the individualistphilosophers drew. This is the constitutional limitation of man'sknowledge and interests, the fact that he cannot know more than a tinypart of the whole society and therefore all that can enter into hismotives are the immediate effects which his actions will have in thesphere he knows (ITF, p. 14).

   Is this where Hayek's trueoriginality with respect to Durkheim lies? The next passage creates theimpression that Hayek himself thinks so:
   All the possibledifferences in men's moral attitudes amount to little, so far as theirsignificance, for social organization is concerned, compared with thefact that all man's mind can effectively comprehend are the facts ofthe narrow circle of which he is the center; that, whether he iscompletely selfish or the most perfect altruist, the human needs forwhich he can effectively care are an almost negligible fraction of theneeds of all members of society (ibid.).

   The division ofknowledge as such is not discussed by Durkheim. However, it is obviousthat with increasing specialization there takes place a fragmentationand localization of knowledge. For Durkheim the growth of knowledge,both in the sense of local, possibly tacit knowledge, and in the senseof explicit and scientific knowledge, is a consequence of the divisionof labor. He only needs to explain the initial "jump" in the level ofknowledge which sets the process of diversification of the socialstructure in motion, i.e., the higher intelligence that turns aparticular individual into a leader within the horde. This, however,seems to belong to the domain of necessary conditions that allow theeffects of a greater social density or volume to start creating thedivision of labor. For Hayek, too, the division of labor and thelocalization of knowledge develop together in what would now be calleda process of co-evolution. This does not seem to distinguish him fromDurkheim.

Hayek's originality lies in the way in which he links theexistence of dispersed knowledge and the problem of its coordinationwith the concept of equilibrium, which is the subject of "Economics andKnowledge." This shows a way forward that promises to add somethingsubstantial to Durkheim's analysis without justifying Hayek's harshjudgment (which we may read as an implicit reply to Durkheim's claimthat sociology is more general than economics) "that, however gratefulwe all must be for some of the descriptive work of the sociologists,for which, however, perhaps anthropologists and historians would havebeen equally qualified, there seems to me still to exist no morejustification for a theoretical discipline of sociology than therewould be for a theoretical discipline of naturology apart from thetheoretical disciplines dealing with particular classes of natural orsocial phenomena" (LLLIII, p. 173).


  IX  Psychology

   BEFORE INDICATING HOW this"cognitive" part of Hayek's research program may be developed further,we have to dwell upon a part of his work that has remained relativelyunknown. We mean his theoretical psychology, which, in fact, was hisearliest contribution to science. In 1920, when he was still a student,Hayek wrote an analysis of the mechanism by which the human braintransforms sensory perceptions into knowledge about the world. Usingthe latest results in brain research and the psychology of perception,he constructed a theory that was ahead of its time. So much so thatwhen he published an extended version of the manuscript in 1952, underthe title The Sensory Order. An Inquiry into the Foundations ofTheoretical Psychology, it was as much ignored as Donald Hebb's TheOrganization of Behavior three years earlier, which contains a theorythat is very similar to Hayek's. [28] We have mentioned earlier thatperceptions play a crucial role in Hayek's economics. It is thereforesurprising that he did not us e, or even refer to, his earlierpsychological analysis of human perception. Indeed, like his friend thephilosopher Karl Popper, he explicitly rejects psychological analysisin social science explanations.

   Nothing compels us to takeover this anti-psychologism. Indeed, in his analytical economics Hayekhimself seems less dogmatic than in his methodology. In PriceExpectations, Monetary Disturbances and Malinvestments, he declareshimself "in complete agreement with [Myrdal] when he stresses the greatimportance of this element [expectations] in the further development ofthe theory of industrial fluctuations" (1933, pp. 155-56). This seemsto imply that he did not exclude psychological explanations from thedomain of economics. By a complicated series of transformations in histhought, [29] a number of elements of his psychological theory ended upas ideas central ideas in his methodology and theory of society. Amongthe former are subjectivism, i.e., the principle that the facts of thesocial sciences are the opinions of the agents; the compositive method,according to which all social phenomena have to be reconstructed fromthese social facts; and the idea that an important difference betweenthe social and the natural sciences is that in the former the scientistis equipped with a mind that obeys the same principles as the minds ofhis objects of study. The most important ideas in his social theorythat were inspired by his psychology include that of socialinstitutions as self-organizing systems; the market system as astructure of distributed knowledge; and social institutions ascontaining the implicit knowledge of earlier generations (and hencebeing path-dependent). From the idea that the facts of social sciencesare the products of human minds, together with the idea that no entitycan explain anything that is more complex than itself (another of thephilosophical consequences of Hayek's psychology), follows the core ofHayek's theory about the limits to human understanding of socialphenomena, and hence to the possibility of intervention.

   Wemention these facts about Hayek's psychology for two reasons. First,because Hayek, in a process of intellectual development that is verydifferent from Durkheim's, arrives at a number of conclusions that arevery similar to Durkheim's. Second, because Hayek's work harbours an asyet unexplored potential for improving our understanding of socialprocesses. To this we turn next.


  X  Minds, Networks, and the Reintegration of Sociology and Economics

  WE HAVE SHO ABOVE how Hayek generalized into a theory of socialinstitutions a number of ideas that are central to his economic theory.Curiously enough, he failed to do so with what we may rightfully callthe central element of his economics, the concept of equilibrium. Hayek(like Durkheim) is a moral scientist in the sense of Mill in that heconsiders social institutions as owing their existence to theperceptions, expectations, and ideas of individuals (Durkheim speaks ofthe collective conscience). Hayek defines economic equilibrium as thecompatibility of the perceptions and expectations on which individualsbase their economic plans. To explain social stability, thisequilibrium concept can be extended straightforwardly to include allperceptions and expectations, not just those that concern economicmatters.

Hayek never makes this extension although it seems a naturalway to link his theory of society with, for example, the sociologicaltradition of symbolic interactionism, which has as central concep t thedefinition of the situation. The advantages of this extension of theequilibrium concept become clear when we include the role of thestructure of communication, In order for a set of social institutionsto be stable, it is necessary that the individuals who populate thesocial framework continue to have perceptions and expectations (fromnow on we will use the word ideas) that are sufficiently compatible orcongruous. In order for that to be the case, they need to calibratethese ideas. This can only take place when they can communicate withone another, which presupposes a structure of communication. Theirlocation in that structure and the number, the type, and the intensityof their contacts or ties with other individuals influence thiscalibration process. Including these factors in the analysis allows usto increase its explanatory power and empirical content.

  Above, we have referred to the work of the network sociologist HarrisonWhite. White studies the emergence and stability of different types ofmarket relationships (White 1988, 1993). More specifically, he studiesthe market of products that are purchased by industry. These marketsare established and remain in existence (White speaks of these marketsreproducing themselves) only if the structure of relationships amongsuppliers and purchasers is such that their perception of the situationis sufficiently congruent. Otherwise, a market will disappear. Thus,these markets exist by virtue of the mutual compatibility of theperceptions of the market parties. White limits his analysis to thecongruence of the perception of the cost and utility functions of themarket parties. As in Hayek's case, this seems an unnecessaryrestriction. Indeed, including the perception of more than thesetraditionally economic factors offers the perspective of enriching theanalysis with elements that allow us to include both coope ration andcoordination. [30]

   A further extension of Hayek'sequilibrium concept builds upon his cognitive psychology and his theoryof cultural evolution. It is the idea that mental models and theirevolution have to be included in the explanation of social phenomena.While social institutions are the result of mental models, these mentalmodels have evolved in interaction with a particular institutionalenvironment. This enables us to subject to further examinationDurkheim's claim that cooperation is more fundamental thancoordination. A natural point of departure would be the analysis of theeffects of competition. We remind the reader of Durkheim's comment thatwhether or not competition has the beneficial effects that for instanceHayek ascribes to it (viz. that by discovering knowledge it enhancescoordination) depends on the presence or absence of social cohesion.This is a claim that can be tested empirically. But a more general typeof analysis must address the question of which explanatory factors arefundamental. Economic the ories ultimately invoke rationality. It isnot entirely clear what the fundamental factors in sociologicaltheories are. In the recent tradition of "explanatory sociology," forinstance, utility maximization is taken over from economics to explainsocial phenomena. In his network analysis of competition, Burtemphasizes the social structure as an explanatory factor. [31] Hespeaks of "[c]ausation resid[ing] in the intersection of relations"(1992, p. 192), which he contrasts with the "debilitating alternativeof using [individual] attributes as an ersatz explanation" (ibid., p.193). However, entrepreneurial behavior in Burt's analysis is driven bythe perception to "turn a profit," which seems to presuppose amaximizing principle. Even Durkheim, despite his efforts to maximizehis distance from economics, in the end invokes a maximizationprinciple. He does so when he discusses the division of labour as aresult of the struggle for life: "La division du travail est donc unresultat de la lutte pour la vie .. . (DTS, p. 253)." ["The division oflabor is, then, a result of the struggle for existence...."]. Everyspecialization has as a result the increase or amelioration ofproduction (DTS, p. 255), It is difficult to understand why efficiencywould lead to specialization if its advantages were not noticed by atleast some individuals. So, Durkheim at this point introducesefficiency by the back door so to speak. This, in turn, leads us backto utility: "pour que la vie se maintienne, il faut toujours que Iareparation soit proportionnee a la depense. . ." (p. 255) ["but forlife to be maintained, reparation must be proportionate to theexpenditure" (272)]. In fact, this is the Friedman-Alchian evolutionaryargument for utility maximization (Alchian 1950). This stronglysuggests that cost-benefit considerations are more than necessaryconditions. So it seems that the question about the relationshipbetween economics and sociology that was raised by Durkheim has not yetfound a definitive answer. A recent suggestion that there may be morethan one fundamental drive motivating human behavior deserves furtherinvestigation. It is due to Hermann-Pillath, who distinguishes anegoistic and an altruistic element. [32] This theory seems topresuppose a type of modular structure of the human mind that can alsobe found, for instance, in the work of Jackendoff (Jackendoff 1989).This suggests that the study of mental structures, either in terms ofmodules or models, may be a necessary ingredient for answeringDurkheim's question about the relationship between economics andsociology. Hayek has pointed to the direction in which we may look: astudy of minds connected by networks. Nothing in Durkheim's thought isinconsistent with that. [33]

   (*.) Jack Birner is Professorof Economics at Maastricht University and the Laboratory of CognitiveScience at the University of Trento. His publications include Hayek,Co-ordination and Evolution: His Legacy in Philosophy, Politics,Economics, and the History of Ideas, co-editor with Rudy van Zijp,Routledge, 1994; and "Cambridge Histories True and False," in C.Marcuzzo, L. Pasinetti and A. Roncaglia (eds.), The Economics of JoanRobinson, Routledge, 1996. Ragip Ege is Professor of Economics at BETA,Universite Louis Pasteur, Strasbourg, France. Ege has publishedarticles about Friedrich A. Hayek and Karl Marx in Revue Economique andRevue d'Economie Politique. Recently he has co-authored with R. dosSantos Fereira, "Le Temps et la conception du capitalisme chez Marx,"in the 1998 volume of Revue d'Economique Politique.



    (1.) Compare Popper 1972, ch. 5; Krajewski 1977; Birner 1994.  
  (2.) Except when stated otherwise, in the case of Durkheim pagereferences preceded by DTS are to Durkheim 1893 (1994). References toHayek's three volumes 1973-79 will be given as LLLI, II, or III. Alltranslations from DTS are taken from Durkheim 1964. In the sequel wewill only give the page references.
   (3.) Compare for exampleDurkheim, DTS, p. III: "si le vaincu peut se resigner pour un temps aune subordination qu'il est contraint de subir, il ne la consent pas,et, par consequent, elle ne saurait constituer un equilibre stable. Destreves imposees par la violence ne sont jamais que provisoires et nepacifient pas les esprits" ["if the conquered, for a time, must suffersubordination under compulsion, they do not consent to it, andconsequently this cannot constitute a stable equilibrium. Truces,arrived at after violence, are never anything but provisional, andsatisfy no one"; pp. 2-3]; and Hayek LLLII, p. 136: "the attempt tosecure to each what he is thought to deserve, by imposing upon all asystem of common concrete ends towards which their efforts are directedby authority, as socialism aims to do, would be a retrograde step ...."
    (4.) Simon 1968.  
    (5.) Durkheim adds that this explains why we find primitive  societies so difficult to understand.  
  (6.) Or perhaps because of the fact that Durkheim thought his theory tobe more general. In that case, one may suppose that he found theanalysis of markets in political economy satisfactory.
   (7.)It would be an exaggeration to speak of Hayek's theory of markets,since he never produced a fully-Hedged and coherent explanation.However, his work of the 1930s and '40s contains the most importantelements for such a theory (compare Birner 1999). Hayek's mostimportant publications dealing with markets and competition are Hayek1937, 1945, 1946, 1947, and 1968.
   (8.) A natural extensionof Hayek's ideas on this issue would be to say that this generalavailability of knowledge, together with a constructivist philosophy,has led neoclassical economists to construct their highly idealizedtheory of markets. Indeed, the only economy to which standardneoclassical perfect competition models approximately apply is the nowalmost completely defunct centrally planned socialist or communistsystem.
    (9.) Hayek is the first economist to do SO (1937).  
  (10.) Hayek does not explain what he means by optimum, but the textmakes it clear that it is a situation in which no relevant knowledge isleft unused so that no individual has a motive to change his plan--anindirect way to express a Pareto optimum.
   (11.) CompareDesai (1994) for a discussion of the revolutionary character of Hayek'sposing the problem of the division of knowledge.
   (12.)Compare also Hayek: "The statement that, if people know everything,they are in equilibrium is true simply because that is how we defineequilibrium' (1937: 46).
    (13.) Desai is the only one to notice this aspect of Hayek's  analysis of markets. Cp Desai 1994: 41.  
    (15.) Cp Kirman 1983, 1985, 1991; Gilles 1990 and later published  work.  
    (14.) Cp Richardson 1960 and 1972. Richardson's sources of  inspiration are Hayek and Marshall.  
  (16.) Cp for instance Granovetter 1982, 1985; White 1988; Burt 1992.Among economists, the best known of these is Granovetter, and his work,too, failed to give rise to an economic network tradition.
    (17.) Cp Hayek 1937 and 1945.  
  (18.) Unlike Durkheim, Hayek never formulates a theory of thedevelopment of rational thought. On this compare Birner 1995, 1999.
  (19.) This is very similar to Popper's approach to social science.Watkins has coined the fortunate term "negative utilitarianism" forthis.
   (20.) Given the fact that Hayek did not himselfcomplete the book, it has to be cited with caution. However, what Hayeksays here is consistent with his neglect of the incentives oncooperation in earlier work, a neglect for which he has been criticizedby, for instance, Witt 1994, Shearmur 1994, and Bianchi 1994.
  (21.) "[L]a cooperation, bien loin d'avoir Pu produire Ia societe, ensuppose necessairement le prealable etablissement spontane" (DTS, p.262). ["co-operation, far from having produced society, necessarilysupposes, as preamble, its spontaneous existence."-p. 278].
    (22.) For a more extended discussion of this issue in intellectual  history cp Birner and Ege 1999.  
    (23.) Cp Bianchi 1994.  
    (24.) This can be read as an accurate summary of Durkheim's  theory of the division of labor.  
    (25.) Cp Ege 1995.  
  (26.) Cp Les regles de la methode sociologique. "Est fait social toutemaniere de faire, fixee ou non, susceptible d'exercer sur l'individuune contrainte exterieure; ou bien encore, qui est generale dansl'etendue d'une societe donnee tout en ayant une existence propre,independente de ses manifestations propres" (p. 14, italics deleted).[In our translation: "Social facts are all types of behaviour, whetheror not laid down in rules, that are capable of acting as an externalconstraint on the individual; or alternatively, that are generaleverywhere in a particular society while having an existence of theirown, independent from their specific manifestations"].
   (27.)Cp ITF, p. 15, where Hayek presents the market (order) as a socialtrial-and-error process, where Reason with a capital R exists only byvirtue of many individuals contributing their specific knowledge tosociety in an unplanned manner.
   (28.) Both became stimuli tothe development of neural network models and the re-introduction of thestudy of mental processes into psychology that now dominates research.The publication of Hebb's book almost made Hayek give up the project ofpublishing his own. For a discussion of Hayek's psychology and its(paradoxical) place in the whole of his work (to which the text belowrefers briefly), cp Birner 1999a.
    (29.) For which the reader may want to consult Birner 1999a.  
    (30.) Richardson (1960) is an early attempt to include both these  factors. Cp also Birner 1999.  
  (31.) Burt 1992. This analysis can be considered as an extension and aformalization of a type of competition that was analyzed by Mises andHayek. Cp Birner 1996.
    (32.) The idea is of course much older. One finds it, for instance,  in Menger. Compare Birner 1990.  
  (33.) Hayek's social theory was strongly influenced by hisneural-network model of the mind; cp. Birner 1996 and 1999a. InDurkheim, too, we find references to physiological psychology (ofWundt), though they have a different function and seem to play amarginal role in the development of his thought (DTS, pp. 322-3).Making comparisons with biology is a tradition that goes at least backto Comte and Spencer. However, like Hayek, Durkheim does not succumb tothe temptation of an organicistic theory of society.
    Aichian, A. (1950), 'Uncertainty, Evolution and Economic  Theory," Journal of Political Economy 58: 211-21  
  Bianchi, M. (1994), "Hayek's Spontaneous Order: the 'Correct' versusthe 'Corrigible' Society," in Birner and Van Zijp (eds) (1994).
  Birner, J. (1990), "A roundabout solution to a fundamental problem inMenger's methodology and beyond," History of Political Economy
  __________. (1994). "Idealizations and Theory Development in Economics.Some History and Logic of the Logic of Discovery," in B. Hamminga andN. de Marchi (eds.), Problems of Idealisation in Economics, PoznanStudies
   __________. (1996). "Mind, Market and Society.Network Structures in the Work of F. A. Hayek," Working paper.Computable and Experimental Economics Laboratory, University of Trento.
   __________. (1999). "Making Markets," in S. C. Dow and P. E.Earl (eds.), Economic Organisation and Economic Knowledge: Essays inHonour of Brian Loasby. Edward Elgar
    __________. (1999a) "The Surprising Place of Psychology in the  Work of F. A. Hayek," History of Economic ideas.  
  __________. (1999b) "Hayek, psicologia e economia: elementi per unnuovo programma di ricerca nelle scienze sociali," Nuova Civilta delleMacchine
   Birner,J., and Ege, R. (1999). "The moralconsequences of the functionalism of Durkheim and Hayek," paperpresented at the Hayek conference organized by AHTEA, Paris
  Birner,J. and van Zijp, R. (eds) (1994). Hayek, Co-ordination andEvolution; His Legacy in Philosophy, Politics, Economics and theHistory of Ideas. London: Routledge
    Burt, R. S. (1992). Structural Holes. The Social Structure of  Competition, Cambridge, MA: Harvard University Press  
    Desai, M. (1994). "Equilibrium, Expectations and  Knowledge," in Birner and Van Zijp (eds)  
  Dow, S. C., and Earl, P. E. (eds) (1999) Economic Organisation andEconomic Knowledge: Essays in Honour of Brian Loasby. Edward Elgar
    Durkheim, E. ([1893]) 1978). De la division du travail social.  Paris: Presses Universitaires de France  
    Durkheim, E. (1893a), "Note sur la definition du  socialisme," Revue Philosophique  
    Durkheim, E. (1895/1937), Les regles de la methode sociologique,  Paris, Presses Universitaires de France (1997)  
  Durkheim, E. (1964), The Division of Labour in Society, translated byGeorge Simpson, New York, The Free Press, and London,Collier-Macmillan, 7th Printing, 1969
    Ege, R. (1995). "Emergence du marche concurrentiel et  evolutionnisme chez Hayek," Revue Economique  
  Gilles, R. P. (1990). Core and Equilibria of Socially StructuredEconomies; The Modelling of Social Constraints in Economic Behavior.Ph.D. thesis, University of Tilburg.
   Granovetter, M, (1982)."The Strength of Weak Ties; A Network Theory Revisited," AmericanJournal of Sociology, 78: 1360-80.
   __________. (1985)."Economic Action and Social Structure: The Problem of Embeddedness," inGranovetter and Swedberg (eds) (1992).
    Granovetter, M. and Swedberg, R. (eds). (1992). The Sociology of  Economic Life. Boulder, CO, Westview Press.  
    Hayek, F. A. (1937), "Economics and Knowledge," in Hayek  (1949)  
    Hayek, F. A. (1945), "The Use of Knowledge in Society,"  in Hayek (1949).  
    Hayek, F. A. (1946), "The Meaning of Competition," in  Hayek (1949).  
    Hayek, F. A. (1947), ""Free Enterprise" and  Competitive Order," in Hayek (1949).  
    Hayek, F. A. (1949), Individualism and Economic Order. London:  Routledge & Kegan Paul  
    F. A. Hayek (1952), The Sensory Order, Chicago, Ill.: Chicago  University Press.  
    F. A. Hayek (1955), The Counterrevolution of Science, Glencoe: The  Free Press  
    Hayek, F. A. (1960) The Constitution of Liberty, London, Routledge  & Kegan Paul  
    Hayek, F. A. (1968), "Competition as a Discovery  Procedure," in Hayek (1978).  
    F. A. Hayek, (1969), "The Primacy of the Abstract," in  Hayek (1978)  
  Hayek, F. A. (1973-79) Law, Legislation and Liberty (3 vols), London,Routledge & Kegan Paul; quoted as LLLI, II, or III
  Hayek, F. A. (1978). New Studies in Philosophy, Politics, Economics andthe History of Ideas. London: Routledge & Kegan Paul.
    Hayek, F. A. (1988) The Fatal Conceit, London, Routledge; quoted as  FC  
  Hermann-Pillath, C. (1994). "Evolutionary Rationality, 'HomoEconomicus,' and the Foundations of Social Order," Journal of Socialand Evolutionary Systems 17(1)
    Jackendoff, J. (1989). Consciousness and the Computational Mind.  Cambridge, MA: MIT Press  
    Kirman, A. (1983), "Communication in Markets: A Suggested  Approach," Economics Letters, 12: 101-8.  
    __________. (1985). "Organisation et communication dans les  marches," Economie qppliquee 3: 597-609.  
    __________. (1991). "Information, communication et  coordination," Economie appliquee, XLIV: 105-24.  
    Krajewski, W. (1977). Correspondence principle and growth of  science. Dordrecht: Reidel  
    Popper, K. R. (1972). Objective Knowledge. Routledge  
  Richardson, G. B. (1960). Information and Investment; A Study in theWorking of the Competitive Economy. Oxford: Oxford University Press.
    Richardson, S. B. (1972) "The Organisation of Industry,"  Economic Journal  
    Shearmur, J. (1994), "Hayek and the Case for Markets," in  Birner and Van Zijp  
    Simon, H. (1968). The Sciences of the Artificial. Cambridge, MA:  MIT Press  
    Witt, U. (1994). "The Theory of Societal Evolution:  Hayek's Unfinished Legacy," in Birner and Van Zijp