SCIENCE AND MARKET AS ADAPTIVE CLASSIFYING SYSTEMS Thomas J. McQuade 1. INTRODUCTION As the cognitive sciences – particularly neuroscience, cognitive psychology, and a rejuvenated artificial intelligence movement that has largely abandoned the model of the mind as a formal machine – have seen major development over the past quarter-century, it is inevitable that the findings thrown up by this ‘cognitive revolution’ should be examined for their relevance to the un- derstanding of economic behavior. This ongoing examination has tended to emphasize those characteristics of human cognitive capabilities that call into question the descriptive adequacy of the rational-choice model, focusing on departures from individual rationality that may have economic consequences at the market level. 1 Such a move may be the obvious one for an economist confronted with this interdisciplinary challenge, but it is not the only one. The new insights into the functioning of the brain can also be deployed in the understanding of complex systems in general – and of specific social ar- rangements in particular – and that is the direction taken here. By critically examining the systemic similarities and differences between the social ar- rangements of science and market, the aim is to show how a complex systems approach, inspired by developments in cognitive psychology but applying these at the level of the system rather than of the individual, can provide a new and useful way of understanding social systems. Cognition and Economics Advances in Austrian Economics, Volume 9, 51–86 Copyright r 2007 by Elsevier Ltd. All rights of reproduction in any form reserved ISSN: 1529-2134/doi:10.1016/S1529-2134(06)09003-X 51 First, an important poi nt of nomenclature. The term ‘science’ is used here to refer to the complex of people and institutions that make up the knowl- edge-generating activities of a scientific community rather than, as might be more common, the knowled ge itself that is generated within that social system. Similarly, ‘market’ refers to the complex of people and institutions that make up a community of buyers and sellers in a money economy. Not included under ‘market’, however, is activity that takes place within firms, and not included under ‘science’ is the activity of academic teaching, both of which would be described by the theory developed here as social arrange- ments with their own distinct institutional fram eworks. 2 In the context of those definitions, sc ience i s not a m a rket, 3 but sci ence and market are instances in the social domain of a general class of systems which are characterized here as ‘adaptive classifying systems’. Now, even setting aside for a moment the puzz le as to what e xac tly is an adaptive classifying system, this is certainly not the standard view in the economics of s cience. The basic working hypothesis that the activities of scientists in the production of scientific knowledge can be understood in ma rket ter ms, dep loying s ome va r- iation of th e me thod o f optim ization under constraint, has been conventional wisdom ever since the pioneering forays of Nelson (1959) and Arrow (1962). See, for illustration, the recent collection Science Bought and Sold – a rather telling tit le – edited by Mirowski and Sent (2002). It is true that the crude characterization of science as ‘the marketplace of ideas’ is not a feature of more modern work, and cer tainly wr iters in the ‘new ec onomics of sc ience’ such as Dasgupta and David (1994) hardly mention markets at all except in the context of technological development (which they sharply differentiate from science). In their concentration on individual incentives played out in contexts of im p erfect information, they incorporate obs e rvations of th e actual incentives encountered by scientists, and bring to bear some sophisticated economic tools in analyzing this new domain. 4 But, in characterizing the interactions as exchanges and invoking a benchmark of efficiency that if it is to be me a ningful at all mu st assume a complex o f goods product i on and ex- change in an idealized competitive environment, they are effectively charac- terizing the n ew domai n as a type of market wit hout ac tually usi ng t he w o rd. But while it is perhaps inevitable that economists will tend, like the man with a hammer who sees every problem as a nail, 5 to take every social arrangement they contemplate to be a market of some sort, this is not necessarily the most helpful approach. The obvious downside is that, by forcing our model of science into conformity with that of market, we will have to downplay the differences, and if these differences include phenom- ena that are important to the operation of science then we impoverish our THOMAS J. MCQUADE52 ability to understand science as a social system in its own right. So the first order of business in this paper will be to examine the similarities and differ- ences between the social arrangements of market and science, and to illus- trate that, despite real similarities, the differences are indeed significant enough for one to be very dubious of the wisdom of treating science as a species of market – even as an ‘imperfect’ market. A major thrust of this paper is, however, to develop a theory, not of what science is not but of what it is. 6 To that end, the defining characteristics of an ‘adaptive classifying system’ are described (a theoretical construct that, interestingly enough, may first have made its appearance in a fairly obscure book of Hayek’s – The Sensory Order), and it is shown how market and science can be represented as different implementations of this more general concept. The question as to what could possibly be the benefit of adopting such an approach – an approach which seems, at first sight, to embody a departure from methodological individualism uncharacteristic of an econ- omist – is raised and discussed, and applications are described that illustrate its ability to shed light on scienc e and market phenomena that seem not to be well handled by more standard approaches. 2. SIMILARITIES AND DIFFERENCES BETWEEN SCIENCE AND MARKET There is no question that one can easily point to similarities between the social domains of market and science. Both are populated by self-interested people – self-interested, that is, in the sense of forming (and acting based on) subjective appraisals of the costs and benefits of their actions and plans, so that behavior at the margin is sensitive to incentives. In both, the people involved are constrained by scarcities of resources, by the inability to do many things at once, and by the cognitive limitations of their brains in the context of a complex environment. 7 There are repeated, institutionalized interactions between the participants; interactions which can be quite indirect and often involve complete strangers but are essential to the indi- vidual pursuit of happiness. And, in both, specialization, competition, and entrepreneurial or calculated risk-taking behavior are major forces in the operation and growth of the social network. When one looks at the overall structure of these two social arrangements, one sees no central locus of control in either case, and yet there is voluntary participation, driven by the positive feedback from the subjectively perceived benefits of participation, and general adherence to the rules of interaction. In cases in which the rules Science and Market as Adaptive Classifying Systems 53 are violated, both incorporate negative feedback processes that keep defec- tion at a tolerably low level. 8 Yet, despite all these similarities, the differences are stark – and funda- mental. 9 Most obviously, in the domain of science, there are no market prices. And, since market prices are an emergent phenomenon of the market system, their absence in science points to deep dissimilarities in the processes of interaction through which, in the case of markets, market prices are formed. The relevant institutions of interaction are in fact different, and they are different for a good reason: the content of the interactions (the goods, services, and money in the case of markets and the published articles and citations in the case of science) have very little in common. Published scientific articles, and whatever it is about their content that may be citable, are not regarded as property; the publication process necessitates interaction not with those who may cite the article but with an editor assisted by referees and theref ore to call the publication–citation sequence an ‘ex- change’ is to take great liberties with the meaning of the word; and the acceptance of articles for publication may be based on appraised signifi- cance or interest or the author’s reputation or connections, but not on expected profitability. In short, the major form of interaction in science does not involve property, does not involve exchange, and does not involve eco- nomic calculation. Nonetheless, one might be tempted to say that, after all, whether or not science is a market is really a matter of definition. One can define key terms such as ‘market’, ‘exchange’, ‘price’, ‘payment’, ‘investment’, ‘capital’, ‘property’, ‘product’, ‘efficiency’ – and even ‘economics’ – in a manner sufficiently general to encompass the phenomena observed in both domains . This has been the general device used by a long line of authors, including: Nelson (1959) and Arrow (1962), who treat science as a knowledge-pro- duction process rendered suboptimal by the character of knowledge as a public good; Radnitzky (1986), who applies cost-benefit analysis to the beh- avior of scientific researchers; Diamond (1988), who formalizes a ‘rational scientist’ as a constrained utility-maximizer whose utility function includes aesthetic attributes of theories; Dasgupta and David (1994), who offer a more sophisticated analysis that recognizes the institutional peculiarities of science but who posit information asymmetries and principal–agent issues as inefficiencies in the subsidized production of knowledge (which is taken to be an exchangeable commodity); and Walstad (2001, 2002), who seeks to transfer Austrian insights directly from market to science. It is a common ass ertion (not only in the economic literature ci ted above, but also more generally) that any situation in which there is an THOMAS J. MCQUADE54 observable acknowledgement of value (including the scientific institution of publication and citation) can also be considered an exchange. 10 Thus, for example, we have the idea of there being two types of market – a ‘traditional market’ in which goods are exchanged and monetary market prices are formed, and a ‘scientific m arket ’ in w hich articles are published and sometim es cited and no m arket prices, monetary or barter, are formed. 11 Ignoring the details of price formation in markets and the dis- connection between the acts of publishing an d citing in science, the gen- eralization functions by i dentifying articles as products w hich are offered for sale and citations as payments in exchange for the use of those articles, or, more precisely, for the use of the knowledge or information therein. The s imilarities already noted between science and market (in particular, the fact that both involve interactions between self-interested participants) make this a plausible move, and it comes with the analytical convenience of the ability to transfer wholesale concepts whose meanings a nd usefulness were established in the context of markets. But, for all the attractiveness of the move, the returns to date have been slight 12 and, with the notable exception of providing some impetus to the growing realization that sci- entists are as self-interested as anyone else, the main result seems to have been the contention that there are ‘inefficiencies’ in the arrangements of science – a conclusion that prompts some economists, including Arrow (1962) and Dasgupta and David (1994), to pronounce ‘market failure’ and call for (further) government subsidy or intervention. 13 In any case, it is not good enough to say that the matter is only one of definition, for definitions, and the analogies they cement, are not without consequence. Firm definitions are obviously necessary for clear exposition, but, often subtly and unobtrusively, they create a path dependence, illumi- nating some directions of inquiry while foreclosing others. The definition of science as a type of market, for example, compels one to look for the analog of marketable goods, and most authors on the topic argue that, despite the obvious problem of quantification, it is found in the knowledge or infor- mation content of the scientific publication. But this cannot be so. If there is ‘knowledge’ in a published article, it would have to be the author’s individual knowledge, but individual knowledge, being the current classificatory capa- bility of an individual brain, cannot possibly be a thing separate from the individual involved. If there is ‘knowledge codified as information’ as Da- sgupta and David (1994) describe it, with information regarded as a signal subjectively appraised, one is out of that frying pan but has fallen into a nearby fire – the need to account for, as a separate process, the emergence of the corpus of current scientific knowledge as a classification distinct from the Science and Market as Adaptive Classifying Systems 55 knowledge of the individual scientists who are parties to the information transfer. This codified individual knowledge, the purported analog of a market good, is not the final product, and it is a gross error to conflate it with scientific knowledge. Scientific transactions are not simply a matter of ‘the- ory choice’ – as portrayed by Brock and Durlauf (1999) – in which the chosen bits of individual knowledge add up to the current state of scientific knowledge. Scientists use aspects of each other’s work, modifying, adapting, criticizing, reinterpreting, and perhaps (from the point of view of the original author) misinterpreting it as they develop their own work. Repeated appli- cations of this process are observed to tend to result in a commonly accepted conception (at least within particular schools, but sometimes in whole dis- ciplines, and especially in those disciplines where empirical reproducibility is considered to be significant), 14 and this transformed conception, this tacit agreement as to the classification of phenomena in the subject domain (tem- porary and mutable though it may be) is what we call scientific knowledge. If one wanted to look for an analog of scientific knowledge in the market domain, the spectrum of current market prices for goods and services would be a reasonable candidate since both are emergent attributes of their re- spective systems, both have counterparts in the individual transactions of the system with which they are often confused, 15 and both provide information to the participants of the system that is vitally useful for the pursuit of their individual ends. But the definition of science as a type of market and the concomitant need to identify the ‘science goods’ that are ‘exchanged’ in this market have led economists of science in another direction altogether. Science is not a market, but science and market, as social systems, have much in common. They are related fraternally rather than filially. To clarify the nature of that relationship, we introduce the parental figure, the adaptive classifying system. 3. ADAPTIVE CLASSIFYING SYSTEMS The sort of ‘system’ we are focusing on is a mutable network of interacting components that is sufficiently stable to be an identifiable entity, distin- guishable from its environment. The system is open to its environment, so that it can be affected by (and can itself affect) that environment, and so the drawing of the system-environment boundary has an element of theoretical convenience to it. Changes in the environment can threaten the integrity of the system and entities in the environment can, once incorpo rated into the system, strengthen its integrity and promote its growth. The fact that the THOMAS J. MCQUADE56 system can and does change in a manner consistent with the maintenance of its structural cohesion as a result of interaction with the environment is the reason for the adjective ‘adaptive’. The way in which this adaptation is effected is particularly interesting – the system, in the course of its expe- rience of the environment, gradually modifies its internal struc ture (realized, at its most basic level, as connections of varying degrees of permanence between its components) in such a way that it builds up, inherent in the structure, a repertoire of useful classes of response that can be deployed as appropriate depending on the environmental situation. In the most general sense, then, the system is continually engaged in ‘classifying’ the phenomena in its environment. 16 Hence the term ‘adaptive classifying system’. 17 One might think that, given this list of unlikely sounding criteria, adaptive classifying systems, if they ex isted at all, would be hard to find. They are not mechanical systems, since the connections between the components are variable and impermanent and the components themselves are changed by their interactions. They are not simple biological systems, like a bacterium, for example, whose adaptive capability depends directly on negative or positive feedback evaluated against a built-in preference rather than on the cumulative modification of the interactions between its components. They are not sim ply ensembles that adapt as a result of their environment’s se- lection pressure in promoting the existence and reproduction of some com- ponents and discouraging others, like a species or an immune system (although their components may, indeed, be subject to such selection). 18 Nevertheless, they are not rare. Systems as diverse as ant or termite colonies – see Hofstadter (1979, pp. 310–336), Tullock (1994), and Sun (2002) – and the brains of higher animals do fit the description. And so, we claim, do various systems of human social interaction, including markets and science. Suggestions that economies and other social arrangements can be under- stood as examples of ‘complex adaptive systems’ have been made many times before in complexity theory – see, for example, Kauffman (1993, pp. 395–402) – and sociology – see, for example, Buckley (1998). It is also, of course, implici t in Hayek – see, for example, Hayek (1967, pp. 66–81), where he asserts that ‘there is no reason why a polycentric order in which each element is guided only by rules and receives no orders from a center should not be capable of bringing about as complex and apparently as ‘purposive’ an adaptation to circumstances as could be produced [in a more hierarchi- cally organized system]’. Kauffman (1993, pp 173–235; 1995, pp. 71–92), discussing ‘the twin sources of order’, makes the important point that, in biological systems, natural selection is not the only source of order, for the tendency in certain systems to self-organization is, in a sense, ‘order for Science and Market as Adaptive Classifying Systems 57 free’. By this he means that the formation of that sort of order is ‘spon- taneous’, a designation that will resonate with those familiar with the work of Hayek – see, for example, Hayek (1973) and Boehm (1994). But systems involving some form of self-organization are members of a very large and diverse set, and so the commonalities are likely to be of such a general nature as to provide very little assistance in understanding particular social systems. So, while endorsing Kauffman’s insight, we prefer to considerably narrow the field so that more concrete things can be said about the structure of the systems of interest. Our purpose here is to push beyond such programmatic statements by injecting more explicitness into the idea. The prototypica l description of an adaptive classifying system, and the direct inspiration for the generalization pursued here, is found in Hay ek’s (1952) explanation of how our brains are able to create the array of sensory qualities by which we perceive events. 19 In this remarkable and prescient work, 20 the brain is characterized as a network of interconnected neurons that structurally changes as a result of the p atterns of activity (in the form of electrical impulses transmitted between connected neurons) that are in- duce d in the ne twork by incipient stimuli. The central idea is that this mutable st ructure functions as a ‘ma p’ of the previously experi enced environment in the sense that it instantiates a classification of the stimuli that have impinged on the system from that environment. The map is built from experience, and is modified by strengthening of neur onal connections when new experie nce confir ms old and by the forma tion and detachment of connections wh en new experience produces activation patterns different from those previously experienced. The system is, in this very particula r way, s elf-organizing. The mutability of the map is crucial to its ability to classify. The network paths followed by impulses from stimuli that tend to occur together tend to become connected; conversely, connections rarely invoked tend to decay. This allows for an establishment of similarity and difference between stim- uli, and there is large scope for the building up of subtle gradations of similarity and difference because the stimuli can induce activity in multiple branching and converging neural paths so that there are very many pos- sibilities for the development (or decay) of connections at places where concurrent activations pass sufficiently closely to each other. Further, since both subject and object of classification are patterns of impulses, classifi- cations from one area of the network can be further classified in terms of the activations those (classified) follow-on patterns induce in subsequent neu- ronal groups. This resulting classification is, then, multiple in several senses – any particular stimulus can be a member of multiple classes, an assignment THOMAS J. MCQUADE58 of a particular stimulus to a class may change depending on the presence of concurrent stimuli, and classes (being represented in terms of impulses) can themselves be further classified at subsequent levels. Note that the ability to classify is an emergent property of the system; it is not the property of any neuron or small group of neurons, or of any particular interaction between specific neurons. The tendency to form connections between paths activated by concur- rently experienced stimuli promotes the emergence, from a given stimulus, of an induced pattern of impulses in the network characteristic of that stimulus and of other potential stimuli which have in fact accompanied it in the past. This pattern of impulses generated in the map by the current stimuli can be described, therefore, as a ‘model’ of the current environment because it is characteristic not only of the experienced stimuli but also of the usual implications of these stimuli. The model is, in other words, anticipa- tory and embodies the system’s expectations of likely subsequent stimuli. And, since connections exist to motor neurons at many levels and these connections, like all connections in the map, have been developed as a result of experience (phylogenetic or ontogenetic), the model can result in the selection of motor activity consistent with those expectations. The basic processes of classification described by Hayek as operating in the brain, including particularly the formation of a mutable map of the brain’s environment as experienced in the past and the ability of that map to support an anticip atory model of current experience, have, we claim, their counterpar ts in adaptive social s ystems, implemented differently, of course, but very similar in principle. Social systems are brain-like in a limited but important respect – specifically, the interactions between their components implement a classifying process on stimuli impinging on the system, and this process can induce real changes in component behavior and interaction that, in turn, enge nder adaptive re actio ns of the syst em as a whole to changes in its environment. In short, they are adaptive clas- sify ing systems. 21 4. MARKET AND SCIENCE AS ADAPTIVE CLASSIFYING SYSTEMS In order to convincingly identify a system of social interaction, be it market or science, as an adaptive classifying system, it is necessary first to clearly define what constitutes the system in contradistinction to its environment, then to identify the structure of the system’s map, then to describe how a Science and Market as Adaptive Classifying Systems 59 model of the current environment can be generated in that map, then to characterize the type of classification that the system is performing of the features of the environment to which it is sensitive, and finally to show how the activation of that model can change the map in a way that refines the system’s classification. Delineation of the boundaries of social systems is not a trivial matter, and this is because there are several important differences between physical sys- tems such as brains and the social systems we seek to include with brains in the category of adaptive classifying systems: 1. The most profound difference is that much of the structure in social systems is abstract. The global institutions and conventions which con- dition all participants’ interactions are essential elements of the structure, as are the more local personal habits and routines. These institutions and habits are not physical entities; nonetheless, they can be regarded as having causal efficacy. 22 Their abstract character may make them difficul t to identify and, at the very least, introduces a significant element of the- ory-dependency to any such identification. The maintenance of the in- stitutions may be dependent on other social arrangements – for example, the contract and monetary institutions of the market system are depend- ent on the legal and monetary systems, and effects from these supporting systems can thereby be transmitted to the market system. At least for the exercise of boundary definition, however, these institutions can be taken as given. 2. The active components of any social system are people, and people can (and inevitably do) participate in multiple social systems – an academi c scientist, for example, in addition to publishing and criticizing, may buy groceries, participate in the educational function of the university, vote in an election, and defend against an harassment suit, all on the same day. Again, the identification of which actions belong within which system is a theory-dependent one. 3. The people in social systems are not only the social analogs of the neurons of the brain in that the transactions between them are the impulses of the system’s model, but they also function as the system’s sensory receptors and motor effectors. This means that there is no built-in localization of sensory inputs or motor reactions; it also means that, to the extent that people are changed by their experience in one system, the change can stimulate an input to another system. For example, an economist whose current scientific activity convinces him that, in the big picture, ‘exporting jobs’ is a healthy development may change his political behavior. THOMAS J. MCQUADE60 [...]... inputs from merchants participating in that system 30 For a fuller exposition, see McQuade and Butos (2005) 31 These institutions appear to have taken their current form as recently as the late 1600s, with the advent of the Royal Society and its journal, the Philosophical Transactions – see Merton (19 73, pp 191–2 03, 460–496), Hull (1988, pp 32 3 32 4), and McQuade and Butos (20 03) 32 Characterized by Kuhn... submission and refereeing process in Riyanto and Yetkiner (2002) 35 A possible example here is the belated recognition of the phenomenon of continental drift 36 This table is adapted from McQuade and Butos (2005) 37 For a full discussion of the tension between reductionism and holism in the study of complex systems such as the brain, see Hofstadter (1979), particularly ‘Ant Fugue’ (pp 31 0 33 6) This excerpt,... extended discussion on the differences (and similarities) between markets and science, see Butos and Boettke (2002) Also, Maki (1999) argues (on ¨ generally different grounds to those discussed here) that free market economics does not provide a useful basis for understanding science 10 See Hands (2001, esp p 38 3) For example, as noted by Hands, some sociological and anthropological literature includes... Perspectives on Science, 7, 486–509 Mandeville, B (1724) The fable of the bees London: Pelican Books (1970) Maslow, A H (1966) The psychology of science: A reconnaissance New York: Harper and Row McQuade, T J & Butos, W N (20 03) Order-dependent knowledge and the economics of science, Review of Austrian Economics, 16(2 /3) , 133 –152 McQuade, T J., & Butos, W N (2005) The sensory order and other adaptive classifying... dimension’, and the ‘chaotic and uncertain world of the 21st century’ But there is serious work in the sociology of organizations, Sandelands and Stabelin (1987) in particular, that independently anticipates in part the general approach of this paper and adds credence to our expectation that this approach might have quite wide applicability ACKNOWLEDGMENT Gratitude has been earned by my colleague and frequent... science, and by Kitcher (19 93, pp 87–89) as ‘consensus practice’ 33 This emergence of a reputational assessment in science is the feature of the system that makes the seeking of reputation a significant motivating factor for the individual scientist For other arguments for the cogency and usefulness of modeling scientists as self-interested reputation-seekers, see Hull (1988, pp 281, 30 5 31 0) and McQuade and. .. Aimar (Eds), F.A Hayek as a political economist (pp 1 13 133 ) London: Routledge Cheung, S N S (1975) Roofs or stars: Stated intents and actual effects of a rents ordinance Economic Inquiry, 13, 1–21 Churchland, P M (1992) A neurocomputational perspective: The nature of mind and the structure of science Cambridge: MIT Press Cole, J R., & Cole, S (19 73) Social stratification in science Chicago: University... American Economic Review, 35 (4) 519– 530 Hayek, F A (1952) The sensory order Chicago: University of Chicago Press (1976) Hayek, F A (1967) Studies in philosophy, politics and economics Chicago: University of Chicago Press Hayek, F A (19 73) Law, legislation and liberty, v I: Rules and order Chicago: University of Chicago Press Hayek, F A (1978) New studies London: Routledge and Kegan Paul Hebb, D O (1949)... Lawson, T (1997) Economics and reality London: Routledge Leonard, T C (2002) Reflection on rules in science: An invisible-hand perspective Journal of Economic Methodology, 9(2), 141–168 Machlup, F (1 936 ) Why bother with methodology, Methodology of Economics and Other Social Sciences New York: Academic Press (1978) Maki, U (1999) Science as a free market: A reflexivity test in an economics of economics ¨ Perspectives... the array of market goods and their market prices.27 An individual may intend to develop and sell a particular good, but no individual plans the overall configuration of marketable goods and services, related to each other (as an emergent result of market activity) as inputs and outputs and as complements and substitutes of varying degrees An individual may deliberately set a particular price, but no . pp. 31 0 33 6), Tullock (1994), and Sun (2002) – and the brains of higher animals do fit the description. And so, we claim, do various systems of human social interaction, including markets and. theoretical and taxonomic corpus of established scientific knowledge 32 and in the generally recognized reputations of individual scientists. 33 An individual may intend to develop and expound on a particular. rather than of the individual, can provide a new and useful way of understanding social systems. Cognition and Economics Advances in Austrian Economics, Volume 9, 51–86 Copyright r 2007 by Elsevier