Interaction versus action in Luhmanns sociology of communication

‘Interaction’ versus ‘action’ in Luhmann’s sociology of communication In: Colin B. Grant (Ed.), Rethinking Interactive Communication: New Interdisciplinary Horizons , Amsterdam: John Benjam ins, forthcoming Loet Leydesdorff Science & Technology Dynamics, University of Amsterdam Amsterdam School of Comm uni cations Research (ASCoR) Kloveniersburgwal 48, 1012 CX Amsterdam, The Netherlands http://www.leydesdorff.net/index.htm ; loet@leydesdorff.net Abstract Both ‘actions’ and ‘interactions’ can be considered as m icro-operations that can be a ggregated from a systemic perspective. Whereas actions operate historically, interactions provide the events retrospectively with meani ng. Luhm ann’s sociology of comm unication sy stems adds t o the approac h of symbol ic interactioni sm the quest ion of w hat global dimensions of comm unication m ean for local interactions. When communication is functionally differen tiated—for example, in terms of media— tensions can be expected t o develo p between local organizat ions and gl obal de velopment s of communication structures. Interfaces enable us t o translate selectively among (provisionally) stabilized representations, for exam pl e, in professi onal practices. ‘Big science’ and ‘hi gh tech’ can be considered as organizational acculturations of an emerging level of s ophistication in global communicat ions. The gl obal dim ension rem ains a hypot hesis, but e ntertainin g this hy pothesis of ‘globalization’ restructur es the local expectations. Introduction The theoretical oppositions between Luhmann (1984) and Habermas (1987) were framed in term s of ‘systems theory’ ve rsus ‘actio n theory’ (cf. Habermas & Luhmann, 1971). ‘Action theory’ may seem less alienating than ‘systems theory’ because ‘actions’ can be inten tional. The analytical distin ction between theories that are based on ‘action’ or ‘interaction’ as micro-opera tions of social systems, however, is more fundamental than the one between action and systems theory. Luhmann’s sociology can be considered as different from other system s-theoretical approaches because it assumes ‘interaction’ as th e basic operation of social systems. Both ‘actions’ and ‘interactions’ can be c onsidered as m icro-operations that can be aggregated from a systemic perspective. Action s can be aggregated, for example, into ‘institutional agency,’ whereas intera ctions may becom e increasingly complex by operating upon one another in a non-linear mode. Action can also be considered as an operation integrating social system s historically (Parsons, 1937; Habermas, 1981), while interactions ma y enable the actors to reproduce differentiation. For example, the constructivist Latour ( 1987) proposed to ‘follow the actors’ in terms of their actions. Actions are then used as a historical explanans . The observation of an interaction, however, a ssumes a perspective from which one can reconstruct the observable events (e.g., actio ns). Like action, interaction occurs in history, but the system of reference for interaction is neces sarily an interhuman construct. Interaction is by definitio n reflexive. The two operations of ‘action’ and ‘interaction’ cannot be reduced to each ot her because of this difference in their epistemological status. From an interactive or networ k perspective, one can attrib ute an action to an acto r, but this attribution can also be reconsider ed. Interaction potentially rewrites the past, for example, from the perspective of a (historical) present. ‘In teraction’ thus provides us with an evolutionary categor y that operates at the network level, whereas actions remain to be attributed to the h istorical development of agency in terms of individuals or groups who carry the evolution of systems of social interactions. While actio ns can be expected to vary, interactions tend to evolve into systems of mutual expectations. The double hermeneutics of sociology The epistemological difference between ‘a ction’ a nd ‘inter ac tion’ was already fundamental to Weber’s Marx-critique. Marx focused on historical action and wished to make predictions on this ba sis. Weber raised the question of the ‘sociological meaning’ of actions. From Weber’s perspective, sociology uses historical instances for u nderstanding the operation of analytical constructs (e.g., ‘idealtypes’). Against Marx, Weber (e.g., 1904 and 1917) maintained that the historical accounts cannot inform us about a system’s logic operating in history. The analyst ‘understands’ th e actions in what can also be called a ‘verstehende Soziologie.’ Understanding raises the question of how pe ople and analysts construct meaning in interactions . A ‘double herm eneutics’ between th e analyst’s and th e participant’s level of action and accounting has since tha t time been a constitu tive problem of sociology (Giddens, 1979). The dimensi on of external observation versus participation can be cross-tab led with the distinction betwee n ‘interaction’ and ‘action’ (Table 1). A participant can also be an observer, but the analytical status of an observation is different fr om that of participation. Participation Observation Action actor report Interaction role discourse Table 1 The generation of a double hermeutics in sociology From the perspective of reflexive interac tion at the network level, ‘action’ by a participant can be considered as a role attributed to or carried by an actor. Expectations with respect to actor s are constructed within the network of communications among the observers. The inte ractive networks operate in term s of non-linear feedback loops on actions. From an action theoretical perspective, 2 however, the network effects are attributed to the intentioned actors in term s of linear cause-effect relations. The unintended consequences of actions remain then unexplained. By considering communication as the unit of analysis—or more precisely as ‘the unit of operation’—of social systems, Luhmann’s sociol ogy shares with symbolic interactionism a focus on inte raction. Sy mbolic interaction ism with its roots in American pragmatism (Blum er, 1969), howev er, has been strongly contrasted to social systems theory (Grathoff, 1978) . Luhmann mainly added to symbolic interactionism the question of what globa l dimensions of communication may mean for local interactions. How and to what extent are the local or ‘first-order’ observations structured by higher-order communications? But in order to ground the next-order level, Luhmann defined the basi c operation of social systems as ‘second- order observations’: how does the network sy stem enable us to m ake distinctions and to provide these distinctions wi th meaning at the network level? Symbolic Interactionism In his authorative study of symbolic interactionism , Herbert Blum er (1969, at p. 8) stated: The importance lies in the fact that social in teraction is a process that forms human conduct instead of being m erely a means or a setting for th e expression and release of huma n conduct. Blumer traced the roots of the interactio n ist approach to George Herbert Mead’s reformulation of the self as the result of a process of social interaction (Mead, 1934, at pp. 26f.). The communicative structure pe rvades action. Society, as Cooley (1902) once argued, exists inside the individual in the form of language and thought. The basic unit of analysis in the interacti onist account was defined as the joint act— the in te rac tio na l ep iso de (Lindesmith, Strauss, & Denzin, 1949, 4 1975, at p . 4 ). Th e interactional episode is part of the larger society. In empirical studies, however, the larger social system was consistently treat ed as a result of interactions in m icro- situations. Blumer (1969, at p. 58), for example, stated: However, in seeing the organization as an organizati on of actions symbolic interactionism takes a different approach. Instead of accounting for the activity of the organization and its parts in terms of organi zational principl es or system principles, it seeks explanatio n in the way the partic ipants define, int erpret, and meet th e situations at their respective points. The linking t ogether of this knowledge of concatenated actions yields a picture of t he organized complex. The resulting ‘picture,’ however, has th e status of an account that can be communicated. This communication i s no longe r necessarily confined to the situation in which it emerged. As noted, the epistemological status of an account is different from an observable action because the observational report is reflexively organized. It contains a knowledge claim th at can be validated by the participants and/or as a contribution to a sociology. 3 The need for a bottom-up approach to the validation does not follo w logically from a focus on ‘interactions,’ but it was implied in the progra mmatic preference of symbolic int era ct ion ism f or th e ana ly sis of ‘mic ro- sit ua tio ns. ’ Knorr-Cetina (1981, a t p. 27 ), for example, argued that the ‘situational appro ach’ is t he road sociology must take for methodological reasons, since: (...) unlike the natural sciences the social sciences cannot hope to get to know the macro-order conceived in terms of em ergent properties: they are methodologically bound to draw upon members’ knowledge and accounts. How can an analyst understand ‘members’ knowle dge and accounts’ other than by situating them in a context that has fi rst to be (re)constructed from these same ‘members’ knowledge and account s’? A reflexive turn is im plied that adds to the analysts’ understanding of the members’ accounts. Whereas the micro-constructivists demand—as a methodological constr aint—that the interpretation be validated locally, the accounts feed back into th e situation from a perspectiv e. This analytical angle makes the observation reflexively availabl e as an observational report in contexts other than the ones in which they originated and were validated. On the one hand, the micro-constructi vists ha ve substantiated their critique of systems approaches, arguing that in order to be us eful for empirical research, a model should be able to account for the specificities of localized action and interaction. The focus on specific episodes has resulted in a richness of substantive understanding which ca nnot easily be brought into a systems perspective. The l atter abstracts from the substance of the accounts by comparing them at the aggregated level. The reports can then be considered as contributions to a discourse. However, the accounts and not the actions reported within them are providing t he variation from this perspective. On the other hand, the situationalist appro ach fails us if we wish to understa nd why interactions are ‘concatenated.’ Some author s in this tradition ha ve tried to specify control as, for example, ‘alignment’ (e.g., Fujimu ra , 1987), but the control mechanisms of the social system (e.g., codification processes) cannot fully be specified from within the situations. The historical report of the sequence only reflects the dynamics that produc ed the sequencing. The systems perspective origin ates from taking a reflexive turn. Ob se rva tio ns , f or example, were defined by Luhmann (1984) fr om his second-order perspective as the operation of first distinguishing and th en indicating the distinction made (“Unterscheiden und Bezeichnen”). The desi gnation provides the di stinctions with meaning. It should be noted that the operati on of ‘observation’ thus defined implies two op er ati ons. By (re)combining the ne twork operation wit h the historical information, the analytical perspective adds to understanding the historical cases. For example, one may also wish to rais e the question why some things did not happen? In addition to the cases that happened to occu r histor ically, one is sometimes—that is, under methodologically specifiable conditions —able to specify expectations about what might have happened. Historical accounts provide the systems theoretician with empirical materials for the formulation of hypotheses. 4 Structuration theor y Some interactions are more likely to o ccur than others; previous interactions ‘constrain and enable’ future interacti ons. ‘S tructures’ seem to operate as constraints both statically, that is, at each moment in time, and dynamicall y, that is, over time. Giddens (1979) proposed in his ‘structuration the ory’ that structure be considered as a virtual operation which ‘constrains and enables’ action ex ante and ‘aggregates’ actions ex post . However, Giddens deliberately re fr ained from a specification of thi s ‘duality of structure’ as a virtual ope ration since—against Marxism and systems theory—the empirical sociologist should, in his opinion, foreground that ‘social reproduction has itself to be explained in terms of the structurally bounded a nd contingently applied knowledgeability of soci al actors’ (Giddens, 1981, at pp. 172 ff.). In order to prevent any specification of struct ure ‘outside time and space’ in e mpirical res ea rch , Giddens (1984) th en re com me nde d as a methodology that structure be described only historically and contextually, that is , substantively operationalized in terms of historical instan ces. In Giddens’s opinion, the mutua l contingencies of structure and action can be studied by ‘br acketing’ the institutio nal dimension when the analysis is at the level of strategic conduct; and vice versa, the latter can be bracketed when one analyzes the former. 1 So, the two perspectives are developed a s different views of the same matter; the tw o pictures together would provide a fuller insight into the mutual contingencies . The definition of structure was thus shifted to ‘a rule of sociological method,’ but Giddens refused to draw the consequence of de fining structure formally, that is, as a network operation. The ‘virtual opera tion’ of structure, however, is analytically different from its substantive instantia tions. Giddens was himself aware of the problem that the core concept of his theory, that is, the ‘dua lity of structure,’ cannot be defined by using ‘bracketing’ (e .g., Gi ddens, 1979, at p. 95). In my opinion, structuration theory contai ns all the elements, but for programmatic reasons, it denies the analyst th e possibility of specifying the operation of ‘structure’ at the level of a social system. Furtherm ore, Giddens warned against making structure the subject of sociological theorizing wh en he strongly for mulated (1984, at p. xxxvii): There can be no doubt about the sophisticati on and importance of the work of some authors currently endeavouring to deve lop Parsons’s wo rk in novel way s, particularly Luhmann and Habermas. But I think it is as necessary to re pudiate the newer versions of Parsonianism as I do the longer established varieties of non-Pars onian structural sociol ogy. Luhmann’s proposal Perhaps even more than Habe rmas, Luhmann has b een deeply influen ced by Parsons’s systems theory. Parsons considered ‘a ction’ as the integrating operator of social systems: the analytical dimensions of an action are instantiated and reorganized in the 1 Giddens (1979, p. 81) compares the concept of bracketing with epoche in the phenomenol ogical tradition. 5 performative dimension. The identific ation of the system with observable ac tion, however, has led to a reification of system s-theoretical approaches. Social s ystems could then be considered as hist orical phenomena (Münch, 1982). Although Parsons (1952) argued strongly th at society should be consi dered as a category sui generis , from his perspective the social remains another dimension of an otherwise naturally given system. The social s ystem was not further analyzed as an interactive and, therefore, cultural construct among human beings. Luhmann (1984) confronted this problem of confusing the historica l level with the analytical by proposing to consider ‘communic ation’ as th e running operator of the social syste m. Interaction can then be considered as a basic operation for producing meaning within social systems. This proposal thoroughly solves the puzzle of combining the explanatory power of systems theory as a theor y about communication and contro l with the richness of the descriptions in studies from interactionist traditions. Th e social system contains instances that were historically realized, but it can be considered as a multi- dimensional space of other options that coul d perhaps be realized in the fut ure. The focus on the dynamics of the network enable s us to integrate the micro- with the macro-approach. Middle-range a pproaches can also be appreciated beca use the analytical definition of the sy stems of ref erence becomes cruc ial to the specification of a research design. Which networks can be considered as re levant for studying a speci fic research question? How can networks be delineated ? Because of the fr eedom to specify expectations on analytical grounds, Luhmann’ s sociology is very di fferent from those of Giddens or Habermas. The latter begi n with historical observations, while Luhmann’s theorizing begins with expect ations t hat are based on ‘horizons of meaning’ (Luhmann, 2002b; cf . Husserl, 1954). This theo ry therefore allows for formalization without losing the relation to the interactive accounts. The interactive accounts provide the variation. From a net work perspective, ‘second-order observations’ refer to a theory about possi ble observations. The observa bles can then be evaluated in relation to the theoret ical expectations. The social system is constructed bottom-u p, but in a network mode. The interactions at the network level add uncertainty to the aggregations of lower-order units. In the formal language of statistics one can formulate that the aggregations contain ‘within group’ variation, but that one e xpects also ‘between group’ variation. A classical example is that of a school expected t o contain more variation than that contained in the sum of the classes with in it. One can expect additional vari ation between the classes, since the classes contain also st ructural variation (Leydesdor ff, 1995). The structural dimensions of the system ma y initially (and partially) be latent for the agents involved, but as the net works develo p by further aggregating, the architecture of a social system can become more apparent (Lazarsfeld & Henry, 1968). A perceptive analyst is able to develop hypot heses about the latent dimensions of a virtually operating structure. The inference by the micro-c onstructivist that one would be unable to specify ‘organ izational principles or s ystems principles’ accounting for the activity of individuals is no lo nger valid from this perspective. 6 The organizational principles can be explaine d historically in term s of how they have been constructed at the network level. Th ey are not given naturally, but constructed historically. However, once constructed the constr uctions may begi n to feed back on agency in a mode very similar to Gi ddens’s ‘duality of structure.’ As agai nst Giddens, however, the focus in Luhmann’s constructivism is not on the construction process, but on what is constructed, that is, the so cial system. Note that our knowledge about this system has the status of a hypothesis. The soci al system should not be reified; it remains operational and under (re)construction. Analytical theorizing about this operation can be i nformed by historical obse rvations, but the systems theoretician takes a refl exive turn. In my opinion, Luhmann’s sociology should primarily be read as a theo ry that informs sociological hypotheses by structuring them into a coherent framewor k. Any knowledge c laim, however, remains itself an operational part of the social sy stem, that is, as another account (Latour, 1988). From this perspec tive, sociological theorizing can be considered as contributing to empirical research by providi ng knowledge claims or hypotheses to be validated. As Luhmann (2002a, at p. 75) formulated it hi mself: The soundness of this reflection, however, arises —and this can still be ascertained by this reflection—from a form of soci al differentia tion that no l onger allows for any bindi ng, authoritative representati on of the world in the world or of soci ety within societ y. Differentiation and integration Accounts by participants to the membership contain an address different from accounts of sociological observers who wish to contribute to the devel opment of their discipline. The social system differentiates in terms of ro les. Neither ‘the syste m’ nor ‘the situation’ (nor ‘everyday language’ or ‘acti on’) necessarily integrates the different (sub)systems. ‘Integration’ is a special case that re quires explanation. In a pluriform society, one expects frictions among discour ses (ba sed on observations and reports from different perspectives). The expected frictions can be observed, for example, in the case of competing paradigms. The structural consequences of previous ac tions and interactions build up over time. People are historically constrained and ena bled by structures that have been constructed at the supra-individual level. These structures are reproduce d (or not) because of their institutionalized social fu nctions. These insights about structure and function, of course, stem directly from Pa rsons. However, when Parsons’s original ‘unit of action’ is replaced wit h ‘interac tion,’ the systems under study are no longer only integrated by the operation. The inte racting systems can both be integr ated locally by action and at the same ti me differentiated in the reproduction. Interaction operates in cycles. The cycles may begin to resonate. Di fferen t levels of nested interactions can be distinguished analytically. The levels can be considered condensations of t hese recursive operations, that is, communica tions a bout communica tions. For example, Luhmann (1975) distinguished between ‘inter action,’ ‘organi zation,’ and ‘society.’ The interacting agents can be expected to remain different, alt hough they are able to 7 exchange using an interface. When the netw orks reproduc e distributions that are base d on differences, the structural characteris tics of these differences can be called ‘differentiations.’ The inte rf aces then also tend to become institutionalized, for example, as organizations. Unlike symbolic interactionism with its pronounced focus on micro-level ‘interactions,’ Luhmann proposed to consid er ‘communication’ as the basic operation of social systems. From this perspecti ve, ‘i nteraction’ can be considered as a specific form of organizing co mmunications, notab ly face-to-face co mmunication in the present. As in Giddens’s structuration theo ry, communications can also be aggregated and structured into contingent organiz a tions and at the macro-level of society. The starting point of this social systems pe rspective is that every action can also be considered as a communication among huma n beings (Luhma nn, 1984, at p. 149). What cannot be communicated, ca nnot be cons idered as part of a social system. It should be observed that this defin ition includes non-verbal communication. Interaction is then the specific form of communication in which the participants are reflexively aware of the contingencies in the communication because of each other’s presence. This ‘double conti ngency’ of the interaction structure s action on both sides as a factor other than the individual lines of action. A structuration of the interactions can also be institutionalized, for example, in marriage. The reflex ive awareness of the double contingencies and asymmetries in the mutual relations t hen induces cognition about the situation for ea ch of the interacting actors. From this reflexive perspective, ac tion can be redefined—as in symbolic interactionism—as attributed by a network of social relations. However, the network perspective stands orthogonally to the actor perspective: agency is no longer considered as the cause and communication as the attribute, but vice vers a, and the system is grounded in communi cation. An actor may take a ction (or not) given one’s position in the network. The c ommunication first provides the events happening with meaning. Meaning can be per ceived by an actor (or not); re ception is more crucial for interaction than taki ng action (Luhmann, 1990). Providing ‘meaning’ to the events is crucia l to all human and in ter-human systems: social systems operate in terms of ge nerating and reproducing meaning. This conclusion can be consid ered as a common heritage shared amo ng Weber ’s sociology, Husserl’s phenomenology, and the American pragmatists. Human beings interact reflexively, that is, in relation to one a nothe r; they evaluate whatever they observe, and although they are able to distinguish bet ween the dimensions of ‘ facts’ and ‘values’, the social science enterprise only takes off whe n the analysts also question what things mean to people. The generation of ‘meaning’ at the social le vel can be consi dered as a consequence of human interaction. Individuals are able to en tertain ‘ meaning’ also discretionarily, but ‘meaning’ is reproduced by communica tion. Using a scheme from cybernetics, Luhmann then inverted the argument a bout the dynamics of meaning from the perspective of systems theory: huma n inte raction can be reorganized by the social system of communications because social meaning is generate d by interacting individuals. As meaning is repeatedly c onstructed bottom-up, the constructed (next- 8 order) system tends to take over control when spe cific configurations can increasingly be stabilized. Social systems and individuals ca n be e xpected to process mea ning differently (Luhmann, 1986). For example, individuals can further de velop as identities that may manage to map meaning one -to-one to their subjectiv ities. The axis for the representation at the social level, however, stands orthogonally to the axis of inter nal processing by an individual. Whereas the individual processe s thoughts and consciousness, the social system enables us t o develop, among other things, discursive knowledge. The generation of a knowledge base When human beings interact, they generate uncertainty at the network level. One is able to handle this uncertainty becaus e one has learned to cope with it by pr oviding meaning to some actions and no t to others. In the sociologi cal literature, this has been discussed under the heading of the double con tingency that provides meaning in social interactions (Parsons and Shills, 1951, at pp. 3-39; Parsons, 19 68; Luhman, 1984, at pp. 148 ff.). Both uncertainty and meaning can be e xpected to circulate am ong human beings. Languages enable us to provide a communication wit h meaning and to distinguish the expected information content of the message at the same ti me. This dual processing can be considered as the evolutionary achie vement that has enabled the social system to develop the complex dynamics of a cu ltural evolution. The system develops in substantive and reflexive layers at the same time, but potentially in a n uncoordinated way. The social system then e merges as a dynamic and flexible coordi nation mechanism among different levels of expectations. The message provides meaning t o the info rmation contained in the message. For example, a word only has meaning in a se ntence. Upon reception, the infor mation can be rewritten as a signal of meaningful info rmation and noise. Th is selective operation is recursive, that is, it can reflexively be applied upon itself. If the operation leaves traces over time, meaning can be invested in specific selec tions. The system can then provisionally be stabilized. In principle, stabilized systems can be further selected for globalization, that is, the historically ac hieved meaning can be compared with a horizon of possible meanings. Knowledge ca n then be developed as a next-order reflection allowing us to distinguish betw een meanings that make a difference and those which can provisionally be discarded as too uncertain. Thus, socially organize d knowledge production further codifies the mea n ing-processing systems at a next-order layer. 2 The stabilization of discursive knowledge in the social subsys tem of scientific communication can be consid ered as a cultural achiev ement of the Scientific 2 Does this m ean that syntax would dri ve semantics ? In a com plex dynamics, the subdynam ics (of syntax and semantics) do not drive each other, but co-prod uce the resulting pheno mena by disturbing and constraining each other. The relative contributio ns of the subdynamics to the manifestations can vary situati onally and over ti me. 9 Revolution of the 17 th century. Individual knowledg e production is then made interactive and in need of validation by communication. Modern sciences can no longer be understood in terms of the knowle dge of si ngle individuals. The study of the development of the sciences in terms of scientific revol utions (Kuhn, 1962) has made us aware of the nature of social syst ems of communication as different from individual consciousness syste ms and thei r sum totals (Leyde sdorff, 2001a). The social system contains surplus value base d on t he interactions among human beings and their aggregates into groups. The social system of interhum an expectations is initially nothing other than a plastic medium in which individuals proces s me aning and uncertainty, for example, by exchanging in these dime nsions. When rep eated over ti me the process can become increasingly structured. The media of social communication can becom e differentiated. A modern soci ety, for example, is highl y structured in terms of carefully constructed balances betwee n different types of communication. The operating structure of the social system is repr oduced at the level of the social system by using our individual contributions as a variety of inputs (e.g., knowledge claims). But the processing is highly struct ured. Thus, human beings are not external to the system, but ‘structurally coupled’ to it i n terms of the dist ribution of their inputs (Maturana & Varela, 1980). The network of communications can be expected to drift into provisional solutions of the puzzle of how to communicate all these inputs in an efficient (albeit perhaps subopti mal) way. The individual contributions provide the variation, while communication structures se lect by reinforcing so me varia tions and not others. The development of cognition as a next-order layer on e ach side—that is, at both the social level and within individuals—pr ovides meaning-proc essing systems with another selective device that can feedback on lower-level selections and underlying variations. However, this mechanism is st ructured i n social systems—Luhmann uses here the word ‘dividuum’—differently from in individuals. While individuals proces s cognition internally, the social system ma nages to construct—under the historical conditions of emerging modernity—discursiv e knowledge as a co ntrol system for (scientific) communications. Functional differentiatio n of the communication What does the social sys tem add when th e inputs are selected for organizing the communications? At this end, the soc iologist can build on metaphors avail able since the founding fathers of the di scipline (e.g., Comte), notabl y, that the social system can be expected to develop evolutionarily in stages. First, ther e was the primitive organization of society based on kinship relations. This can be considered as a segmented system. Next, civilizations were formed based on a hierarchical and stratified structuration of the processing of meani ng. In this stage, the levels of organization provided the main differ entiation. However, the one-to-one correspondence between levels and control fu nctions can be dissolved under historical conditions. 10 When the organization of society could no longer be contained within a single hierarchy (at the end of the Middle Ages), another format was gradually invented in the social system, notably, that of func tional differentiation. This new form was shaped in the 15 th century, for example, when th e House of Burgundy ruled over the Low Countries. The Dukes of Burgundy were neither Emperors (of Germany) nor Kings (of France) and, therefore, they suffe red from a lack of religious legitimation for claiming autonomy. Given the social and power relations of the tim e, monetary unification was invented as a means to bi nd their ‘Em pire of the Middle’ (between France and Germany) together. Philip the Good unified the monetary syst ems of Flanders, Brabant, Holland, and Hainault in 1433. In 1489, the silv er ‘stuiver’ (or ‘sous’) wa s legally standardized as one twentieth of a golden guilder (flori n) in all the Burgundian Netherlands (Groustra, 1995). This monetary union lasted until 1556. Th e coordination eroded because of the inflationary import of silver from the Spanish colonies during the 1540s and the protestant uprisings in the Netherlands in the 1550s. Wh en th e Dut ch revolt gained momentum in th e 1570s and 1580s, the northern provinces also decided that they no longer needed a King ‘by the Grace of God,’ but that they could organize the political system as a republic. The sc iences and the arts , once set free from religious control, could then begin to flourish. The principle of functional differentiation entails that various symbo lically mediated communication systems can operate to solve problems in society in a heterarchical m ode, that is, alongside each other. Over time, these pa rallel systems can develop f u nctionality for one another. Functionality, however, is further develope d along orthogonal di mensions. Thus, one can expect that it will take time to develop from the stage of a breakdown of the horizontally stratified hierarchy into diff erentiation with functions along orthogonal axes as another mode of social orga nization. The different function systems use various codes for providing meaning t o the communication. Whereas th e hierarchical (catholic) syste m had only a single center of control—that was based on a holy text—ec onomic exchange re lations, for example, could now be handled by making payments. The symbolically gene ralized medium of money makes it no longer nece ssar y to communicate by negotiating prices verbally or imposing them by force. The specific ation of a price as an expected market value speeds up the economic transaction processe s by organizing the communication in a specific (that is, functionally codi fied) format. Functional differentiation first had to be inve nted and then also accepted as a solution to coordination proble ms at the level of the social system, for exa mple, by recognizing privacy (e.g., in love relations) vis-à-vi s public relations, market relations for exchange, and political state formation as differen t domains of communication. After the ‘phase transition’ from a hierarchical mode of communication to one in which functionality prevailed, the differentiati on began to feed back on the institutional organization of society, for example, by qu estioning the functionality of the traditional organization. This was then reflected in an emerging disc ourse (during the 18 th century and notably in France) about de sirable forms of social organization. Luhmann has emphasized in a series of studies entitled Gesellschaftsstruktur und Semantik (‘The Structure of Soci ety and Semantics’) that although t he semantic 11 reflection is needed for stabilizing th e functional differentiation, functional differentiation of commu nication should no t be considered as a process within language, but one that precedes language structur ally, that is, at the level of society. The communication bec omes functionally differentiated as a social order; the semantic reflection and codific ation can be e xpected to la g behind. This social process of changing the mode of organizing co mmunications among human beings can be expected to take centuries, and it cannot fully be completed because the complex system builds upon subdynamics that cont ain and reproduce forms of less complex organization as their building blocks. For example, the hierarchical order of th e communication in la nguage with only Latin and then French as the lingua franca , was gradually replaced with a segmented system of ‘natural languages’ which could exist alongside each other as more or less equivalent. A system of nati on states e merged in the 19 th century as a sustainable for m of shaping institutional structures that reinte grated the different functions in specific forms of organization. The prevailing te ndency towards functional di fferentiation, however, is continuously upsetting the historical arrange ments. Functional differentiation allows for hand ling more complexity at the global level since it is based on a next-order reconstruction. The recons truction transforms all ‘natural’ (given) forms by infusing them with knowledge -based inventions . The global system, however, is constrained in terms of the de velopment of retention mechani sms that enable its reproduction. The evolu tionary mecha nism The American and French revol utions can perhaps be considered as t he first deliberate attempts to reorganize a society institutionally so that it would be able to sustain the pluriform multiplicity of functions that characterize a modern society. The functional domains (e.g., markets, sciences) can be considere d as global subsystems of communication, but at lower leve ls specific formats had to be generated in order to optimize the processing of information and meaning locally. While ‘inter action’ occurs also spontaneously bet ween people, organizations have to be constructed. Under the condition of functional differentiati on, three levels can be distinguished at which one can expect that the function sy stems are recombined (c f. Luhmann, 1975): (i) in ‘interaction’ as face-to-face communic a tion; (ii) organization in a social system provides criteria to distinguish those who are within from those exte rnal to a specific domain; and (iii) society can be consider ed as the coordination mechanism am ong functions at the global level. These three levels reconstruct segmentation, hierarchical stratification, and heterarchical differen tiation of meaning processing, respectively. The organization of integration in institutions is thus analytically distinguished from the ongoing processes of functional differ entiation among the globalizing subsystems (such as the economy and the sciences). The interfaces make possible translations among codes that provide different meani ngs to communications. But the interface s have first to be invented a nd developed at specific places. The functional subsystems operate by c oding the communicati ons specifically: for example, the market codes in terms of prices and payments, the sciences code 12 communications in terms of whether they ca n be c onsidered as functional for truth- finding and puzzle-solving, a nd political disc ourses code communication in terms of whether power and legitimation can be organize d. Intimate relations code in term s of love and affection. The in tegra ting levels, however, are not specific in ter ms of what is being coded. They solve the puzzle of how to interface the differences in codings locally. Agents at these different levels of aggregation can be expected to contribute to the differentiation by translating among differently coded meanings. A range of global functions can be ex pected to resonate in inter-human communication. Functional differentiation means that some dimensions ca n be selected and others deselected in speci fi c orders of communication. The integrating mechanisms can be consider ed as functional for organizing the differentiated communications at lower levels. The y serv e the retention of previously achieved levels of sophistication in the communica tion—or they may fail to do so. If they repeatedly fail to do so, an organization can be dissolved and replaced, yet without seriously affecting the dynamics of functiona l differentiation that can be expected to prevail at the global level. This theorizing would remain completely speculative if it were not possible to develop empirical research questions on its basis. The hi storical example of how a monetary standard was developed at the end of the Middle Ages, provided us above with a first example of how one can use this theory as a heuristics for studying evolutionary developments in social processes. But can we also apply these methods more quantitatively and analytically? (Leydesdorff & Oomes, 1999) What does a co mmunication system do when it comm unicates? It selects a syste m’s state for a communication. A so cial communication system can be expecte d to contain a very large number of system states, sin ce the number of possible states increases with the number of the carryi ng age ncies in the exponent. For example, if one throws two dice, one has 6 2 (= 36) possible combinations. N dice would provide us with 6 N possibilities, and similarly a group of ten people with six media for communication would allow for 6 10 , or more than 60 million po ssible combinations. A communication actualizes one or a few of these possibilities. A large number of the actual izations may be volatile. One communication follows upon another without necessarily leaving traces . Selections then rem ain juxtaposed or, in other words, they are not correl ated. However, selections may become correlated (if only by chance) in two respects , notably at the same moment in time and over the time axis. Along the time axis, ‘v ariation’ can be considered as change in relation to stability in the selections. At each moment in tim e, ‘variation’ can be considered as the sum of local disturbances, whereas structu re selects for the function of this input. Structure, however, ha s to be built up histor ically before it can act as a systematic selector. In summ ar y, a stabilized (and therefore observable) system contains two types of selections that operate concurrently: one by the network at each moment in time, and another over the time axis. Note that the medium of communication thus provides us with a first constraint. When written communication is available to a social sys tem, additional m echanisms of transmission become possible other than inte ractions and signaling in the present 13 (Meyrowitz, 1994). Writing, however, has to be historically invented. As long as a communication system is mainly based on direct interaction, the span of communication is limited, and the selected states of the syste ms rem ain mainly juxtaposed. This can be recognized as a segmented order of so cial communications. Writing is highly associated with the intr oduction of a new mode of control of communication, notably, the stabilizati on of a civilization (Innis, 1950). The mechanism of written communication enables cultur es to span tim e periods at the supra-individual level and thus to stab ilize systems of communication. Because the communication can also be saved for c onsiderable periods of time, the new communications can be correlated to older ones and the selection of specific system states can be stabilized. Ti me breaks the symmetry in th e mutual selectio ns of a coevolution. Over time some previous selec tions can be selected for stabilization. In the phase space of possible selections the system then begins to develop along a trajectory. The shape of the historical tr ajectory is contingent upon the selections that the system manages to handle structur ally. For exam ple, a social system in which one is only able to write on clay tabl ets can be expected to develop differently from a social system in which papyrus or parchm ent have been invented. The relations between hierarchical interac tion (command structures) and face-to-face interactions will vary among systems that are differently mediated. Within civilizations based on hierarchies, the top of the hierarchy may be a king or an emperor with divine attributes. Howeve r, the prevalence of comm unication in the command structure can also become reflecte d. The invention of a holy text (e.g., the Bible) that integrates the system at a level more abst ract than the physical presence of an emperor or king changes the co smology. A civilization based on a more abstract set of principles can be considered as a high culture. But the reliance on communication—instead of physical force—as the basis for contro l is self-defeating in the long run because the constructed order needs to be enforced and the communicated order can then be recognized reflexively as historically specific. The invention of new dimensions for the comm unication that can also be codified at the social level can be expected to turn the tables sooner or later (Arthur, 1988 and 1989). When the social system gains an additional degree of freedom, the new dimension allows the communicators to ev ade the dilemm a of the two previously competing orders. At the edges of the sphe res of influence between the Pope and the Emperor, for exam ple, in city-states in No rthern Italy and in the envisaged ‘Empire of the Middle’ in northern Europe one coul d develop trade, art, and sciences. The new communication structures would ev entually challenge the catho lic order spanning a single universe and its corresponding cosmology. The new order of communications can endure different dimensions of communication developing next to each other as different structures. Thus, the system recombines th e advantages of segmented and stratified communication by inventing the mode of functional differentiation. Functional differentiation entails that communications can be distinguished with reference to the function of the communication. This provides new dimensions that were not available in a high culture. For example, when the Netherlands were invaded by the French arm y in 1672, the Prince of Orange needed legitimation for the upcoming negotiations. He sent for 14 Spinoza to join his cortège in order to impress the French generals. That Spinoza had been banned by the jewish and protestant churches for religious reasons was not in the interest of the Prince. In a fun ctionally differentia ted society, the representatives of functions can tolerate moderate conflict b ecause the social system is no longer expected to pr ocess a single solution. An order among the various function systems can be selected and reconstructed in a next round of reflections. If this additional degree of freedom can also be stabilized, this process globalizes the functional differentiation of the system. Some (provisional) stabilizations can be selected for globalization. Globalizatio n , however, does not imply that a global system physically and/or m eta-physically ‘exists.’ The functions refer only to a s upersystem for which the subsystems can analytically be made functional. Initially, the exis tence of this supersystem remained a religious assumption; for example, De scartes’ belief in the Goodness of God ( Veracitas Dei ) which would prevent Him from d eceiving us all the time. Religiou s constructions like Leibniz’ harmonie préétablie would guarantee a cosm ological order in the universe. Since the social system, however , continuously fails ‘to exist’ at the global level in a strong (physical or biological) sense—it remains a system of expectations—the organization of society can be expected to operate with the tensions between functional differentiation and locally organi zed integration of the communication. In the 18 th century, the Constitution was invented as an presumably ‘universal ’ text that would bind all communicating agencies as members of a nation state. Soon, it became clear that each nation would have to develop its own constitution. The constitutions organized institution al systems of checks and b alances that enabled the political economy to further develop on the basis of the level of functional differentiation that was achieved in the first half of the 19 th century. From this perspective, nations can be considered as institutional arrangements that include and exclude on the basis of nati onality. In terms of evolu tion theory, they can be compared with niches. In ecology, niches are functional for the retention because they reorganize the complex environment by stabilizing boundaries. When the system of nation states was completed (by approxim ately 1870), the national systems contained mechanism s for solving the m ajor tensions between the state and civic society so that the function systems could be integrated locally, yet in a competing mode. From 1870 onwards, the social system has developed a new dimension to further improve these ‘nati onal’ solutions. This new dimension can with hindsight be characterized as or ganized knowledge production and control (Whitley, 1984). Technological developments as inter-system dependencies The sciences have developed continuously since the Scientific Revolution of the 17 th century (Price, 1961), but the fully deve loped political economies of the 19 th century provided the sciences with an institutional basis for further development. When the disciplines and the specialt ies then differentiated among themselves and in relation to their social contexts, the idea of a singl e and universal science had gradually to be 15 abandoned. Interfaces with private appropriation by entr epreneurship and public control in science and te chnology policies were increas ingly developed. W ithin science, the proliferation of disciplines and specialties made it possib le to dissolve the idea of a single ‘truth’ to be discover ed by science. One could proceed to a mode of ‘truth-finding’ and empi rical ‘puzzle-solving’ (Sim on, 1969). Thus, the code of scientific communication becam e in ternally differentiated (Gibbons et al ., 1994; Leydesdorff, 2001b). Can the function systems also differentia te and complexify in term s of their interactions? In his 1990 study entitled Die Wissenschaft der Gesellschaft (The Science of Society), Luhmann formulated on p. 340: The differentiation of society changes also the social system in which it oc curs, and this can again be made the subject of scientific theo rizing. However, this is only possible if an accordingly complex sy stems theoretical arrangement can be specified (translation, LL). Is the post-modern order thus eroding the system of functional differentiation (Sevänen, 2001)? When studying the so-c alled ‘techno-scien ces’ as interface organizations with their own dynamics, one leaves the model of functional differentiation behind (Callon, 1998). Algorit hmic models are needed which allow for next-order effects that are neit her intended nor expected. Technological trajectories and regimes (Dosi, 1982), for example, can then be considered as endogenous consequences of non-linear intera ctions at th e interfaces between the sciences (‘supply’) and markets (‘demand’). Luhmann (2000, at p. 396) has discussed the o rganization of interfaces as structural interruptions of the communica tive order at the global leve l. As he formulated it: Society has to develop beyon d functional differenti ation and use a nother pri nciple of systems formation in order to gain the ultrastability and therefore sufficient local capacity to absorb irri tations by provi ding orga nization. What might this ultra-stabilization of an inte raction between functiona lly differentiated sub-syste ms mean? Stabi lity requires a form of integration by organization. Indeed, an important cond ition for the development of modern high- tech sciences seems to be the increa sing integration of political, economic, and scientific orientations in research practices (Gibbons et al. 1994). Professional practices can be considered as organized interaction system s that allow for specific recombinations of integra tion and differentiation in new roles. Integration in the sense of de-differen tiation, however, would be evolutionarily unlikely, since the social system m ight then lose its capacity to handle complexity. Thus, these constructed interaction syst ems are heavily organ ized, but from the perspective of interactions. Alternatively, the constructi ons can be shaped at the level of interactions among organizations. For example, technological de velopments can be considered as the result of inter-systemic resonances which have been stabilized as new functions in the social system during the last centur y. The stabilization of interfaces and the 16 discursive construction of integration can then be considered as instances of an emerging next-order of global communications. This higher-order communication can be expected to contain a new epistèmè (Foucault, 1972 [1969], at p. 191): in addition to the co mmunication of substantive novelty and methodologically warranted codifi cation (‘truth’), high-tech sciences, for example, are translating representations of subsystem s of society into scientific knowledge by modeling them, and vice versa , by legitimating research r esults in ‘trans-epistemic’ cycles of communicati on (Knorr-Cetina, 1982 and 1999). In other words, one is institutionally warranted in changing the code of the communication, for example, because of a flexible division of lab our within th e research community. The emerging patterns of the high-tech scienc es are not expected to replace the older models, but to encompass them and to guide their future developm ent. The next- order regime entrains the trajecto ries on which it builds (Kampmann et al ., 1994). In other words, ‘high tech’ and ‘big science’ can be considered as results of an ‘epistemic drift’ of translations be tween economic innovations and research questions; and vice versa , of the possibility to me rge fundamental and applied research questions in terms of selections of relevant representations (Elzinga, 1985 and 1992). These newly emer ging communication system s contain more than a single codification, and additionally they are able to translate between these codifications internally by using a spiral model of communication. Using computer simulations, for example, developm ents can be analyzed in terms of processes of representation and communica tion within relevant scie ntific-political-economic communities (Ahrweiler, 1995): high-tech sciences develop by communicating in terms of recursive sele ctions on interactively con structed representations. In my opinion, the emergence of ‘big sc ience’ and patterns of international collaborations in science during the second half of the 20th century can be considered as the institutional acculturation of the new epistèmè. The reflexive reorganization of these institution al patterns by using new form s of S&T policies was apparently delayed until the second oil crisis of 1979, when the post-war system entered into a serious crisis at the le vel of the global economy. The gradual development of stable patterns of scientific reproduction in fiel ds like ‘artificial intelligence’, ‘biotechnology’, and ‘advan ced materials’ in the 1980s and 1990s indicates the viability of a new m ode of scientific communication. The globalization of the knowledge base of expectations The local networks of institution s like universities, industries, and govern ments can be considered as carriers of a next order of potentially global communications. These systems can then be expected to go through a phase transition in terms of their need for new communicative competenci es. The translations no longer occur between ‘natural’ languages, but between functional codes of communications that are themselves entrained in a fl ux. This next-oder system emerges within the system as its globalization. The existence of a global system, however, remains a hypothesis. Since this hypothesis is ente rtained and communicat ed, the global level potentially restructures the expectation st ructures in the globalizing systems. By 17 being transformed on this basis, all ‘natura lly giv en’ or ‘historically con structed’ bases of underlying systems tend to beco me increasingly ‘knowledge-based’. Remember that some selections were sele cted for stabilization along the time axis. By globalizing the system entertains th e time axis no longer as a historical symmetry-breaking mechanism, but as anot her dim ension. The local realization can then be evaluated from a global, that is, knowledge-based perspective. The global perspective operates on the pres ent state of the system by enabling us to entertain the idea that what has histo rically been cons tructed ‘ain’t necessarily so.’ In other words, it can always be reconstruc ted on the basis of new insights. A knowledge-based system operates on the basis of the current state of the system as one of its possible representations. Each hist orical representation can be compared to others. The global perspective adds an expectation to the local perspectives. Since the various perspectives compete for the e xplanation of what can be expected to occur, neither the global nor the loca l perspective can claim priority. The perspectives remain analytic ally juxtaposed (as hypothese s!), but they interact. As noted, this means at the level of ‘organ ization’ that the single organization is increasingly networked and that the inter-institutional arrangem ents become more important for the functionality of orga nization than the single perspective. What does this globalization of the knowle dge-base mean for ‘interaction’ ? It seems to me that this can already be observed, fo r example, in the form of the role of e- mail communication as an addition to prev iously existing form s of interaction. We have increasingly become aware that inte raction is mediated and that one can entertain various form s of interaction with different objectives. Furtherm ore, one is increasingly able to anticipate intera ctiv ely the unintended consequences of previous communications. Interactions can thus be expected to become increasingly recognizable as translations among differently coded communications. The programmatic view of symbolic intera ctions that inte r actions can only be concatenated bottom-up in order to inform us about social structure can then no longer be maintained without running in to serious problem s. The methodological restrictions of micro-constr uctivism have practical imp lications. Interactions are situated, and thus next-order levels of nested interactions and communications can be expected to resonate within the obse rvables. The situation is overdeterm ined by expectations based on hypothe tical structures. The systems theoretical prog ram in sociology adds and informs the hypotheses about the feedback loops within the interactions that it studies. References Ahrweiler, P. (1995). Künstliche Intelligenz-Forschung in Deutschland. Die Etablierung eines Hochtechnologie-Fachs . Münster/New York: Waxmann. Arthur, W. B. (1988). Competing technologies. In G. Dosi, C. Freeman, R. Nelson, G. Silverberg, and L. Soete (Eds.), Technical Change and Economic Theory (pp. 590- 607.). London: Pinter. Arthur, W. B. (1989). Competing Technologies, Increasing Returns, and Lock-In by Historical Events. Economic Journal, 99 , 116-131. 18 Blumer, H. (1969). Symbolic Interactionism. Perspective and Method . Englewood Cliffs: Prentice Hall. Braverman, H. (1974). Labor and Monopoly Capit al. The Degradation of Work in the Twentieth Century . New York/London: Monthl y Review Press. Callon, M. (1998). The Laws of the Market. Oxford and Malden, MA: Blackwell. Cohen, H. F. (1994). The Scientific Revolution: A Historiographical Inquiry . Chicago, IL: University of Chicago Press. Cooley, C. H. (1902). Human Nature and the Social Order . New York: Charles Scribner's Sons. Dosi, G. (1982). Technological Paradigms and Technological Trajectories: A Suggested Interpretation of the Determinants a nd Directions of Technical Change. Research Policy, 11 , 147-162. Elzinga, A. (1985). Research, bureaucracy and the dr ift of epistemic criteria. In B. Wittrock & A. Elzinga (Eds.), The University Research System (pp. 191-217). Stockholm: Almqvist & Wiksell. Elzinga, A. (1992). The theory of epistemic drift: a way of relating the social and the epistemic. Revue Roumaine de Philosophie, 36 , 45-61. Foucault, M. (1972 [1969] ). The Archaeology of Knowledge . New York: Pantheon. Foucault, M. (1984). The Politics of Health in th e Eighteenth Century. In P. Rabinow (Ed.), Power/Knowledge: The Foucault Reader . Harmondsworth: Pengu in. Fujimura, J. H. (1987). Constructing 'Do-able ' Problems in Cancer Research: Articulating Alignment. Social Studies of Science, 17 , 257-293. Gibbons, M., C. Limoges, H. Nowotny, S. Schwartzman, P. Scott, P., and M. Tr ow (1994). The new production of knowledge: the dynamics of science and research in contemporary societies . London: Sage. Giddens, A. (1979). Central Problems in Social Theory . London, etc.: Macmillan. Giddens, A. (1981). Agency, institution, and tim e-space analysis. In K. D. Knorr-Cetina & A. V. Cicourel (Eds.), Advances in Social Theory and Methodology. Toward an Integration of Micro- and Macro-Sociologies (pp. 1 61-174). London: Routledg e & Kegan Paul. Giddens, A. (1984). The Constitution of Society . Cambridge: Polity Press. Gouldner, A. W. (1976). The dialectic of ideology and technolo gy . London: Macmillan. Grathoff, R. (Ed.). (1978). The Theory of Social Action. The Correspondence of Alfred Schutz and Talcott Parsons . Bloom ington and London: Indiana University Press. Groustra, I. (1995). Monetaire Unie in de Habsbur g-Bourgondische Nederlanden tussen 1489-1556 . Am sterdam: Faculty of Economics and Econometrics, University of Amsterdam. Habermas, J. (1968). Technik und Wissenschaft als 'Ideologie', Technik und Wissenschaft als 'Ideologie' (pp. 48-103. ). Frankfurt a.M.: Suhrkamp. Habermas, J. (1981). Theorie des kommunikativen Handelns . Frankfurt a.M.: Suhrkamp. Habermas, J. (1987). Excursus on Luhmann, The Philosophical Discourse of Modernity: Twelve Lectures (pp. 368-385.). Cambridge, MA: MIT Press. Habermas, J., & Luhmann, N. (1971). Theorie der Gesellscha ft oder Sozialtechnologie . Frankfurt a.M.: Suhrkamp. Husserl, E. (1954). Die Krisis des europäisch en Menschentums und die Philosophie, In Edmund Husserl, Husserliana , vol. 4, ed. Walter Biemel, 314-348. The Hague: Martinus Nijhoff. Innis, H. A. (1950). Empire and Communications . Oxford: Clarendon Press. Kampmann, C., C. Haxholdt, E. Mosekilde, and J. D. Sterm an (1994). Entrainment in a Disaggregated Long-Wave Model. In L. Leydesdorff & P. van den Besselaar (Eds.), Evolutionary Economics and Chaos Theory: New developments in technology studies (pp. 109-124). London/New York: Pinter. Knorr-Cetina, K. D. (1981). Introduction. The micro-sociological challenge to macro- sociology: towards a reconstruction of social theory and method. In K. D. Knorr- 19 Cetina & A. V. Cicourel (Eds.), Advances in Social Theory and Methodology. Towards an Integration of Micro- and M acro-Sociologies (pp. 1-47). London: Routlege & Kegan Paul. Knorr-Cetina, K. D. (1982). Scientific commun ities or transepistemic arenas of research? A critique of quasi-economic models of science,. Social Studies of Science 12 , 101- 130. Knorr-Cetina, K. D. (1999). Epistemic Cultures: How the Sciences Make Knowledge . Cambridge, MA: Harvard University Press. Kuhn, T. S. (1962). The St ructure of Scientific Revolutions . Chicago: University of Chicago Press. Latour, B. (1987). Science in Action . Milton Keynes: Open University Press. Latour, B. (1988). The Pasteurization of France . Cambridge, MA, and Londo n: Harvard University Press. Lazarsfeld, P. F., & N. W. Henry (1968). Latent structure analysis . New York: Houghton Mifflin. Leydesdorff, L. (1995). The Challenge of Scientometrics: the development, measurement, and self-organization of scientific communications . Leiden: DSWO Press, Leiden University. Leydesdorff, L. (2001a). The Complex Dynami cs of Scientific Communication. In C. B. Grant & D. McLaughlin (Eds.), Language-Meaning-Social Construction: Interdisciplinary Studies (pp. 163-179). Am sterdam/New York: Rodopi. Leydesdorff, L. (2001b). A Sociological Theory of Communication: The Self-Organization of the Knowledge-Based Society . Parkland, FL: Universal Publishers; at . Leydesdorff, L., & P. van den Besselaar (1998). Technological Development and Factor Substitution in a Non-linear Model. Journal of Social and Evolutionary Systems, 21 , 173-192. Leydesdorff, L., & N. Oomes (1999). Is the European Monetary System Converging to Integration? Social Science Information, 38 (1), 57-86. Leydesdorff, L., & P. Wouters (1999). Between Texts and Contexts: Advances in Theories of Citation? Scientometrics, 44 (2), 169-1 82. Lindesmith, A. R., A. L. Strauss, & N. K. Denzin (1949, 4 1975). Social Psychology . Hinsdale: Dryden. Luhmann, N. (1975). Interaktion, Organisation, Gesellschaft, Soziologische Aufklaerung (Vol. 2, pp. 9-20). Opladen: Westdeutscher Verlag. Luhmann, N. (1982). Liebe als Passion. Frankfurt a.M: Suhrkam p. Luhmann, N. (1984). Soziale Systeme. Grundriß einer allgemeinen Theorie . Frankfurt a. M: Suhrkamp. Luhmann, N. (1986). The autopoiesis of social systems,. In F. Geyer & J. v. d. Zouwen (Eds.) Sociocybernetic Paradoxes (pp. 172-192). London: Sage. Luhmann, N. (1990). Die Wissenschaft der Gesellschaft . Frankfurt a.M.: Suhrkam p. Luhmann, N. (2000). Organization und Entscheidung . Opladen: Westdeutscher Verlag. Luhmann, N. (2002a). The Modernity of Science. In W. Rasch (Ed.), Theories of Distinction: Redescribing the descriptions of modernity (pp. 61-75). Stanford, C A: Stanford University Press. Luhmann, N. (2002b). The Modern Sciences and Phenomenology. In W. Rasch (Ed.), Theories of Distinction: Redescribing the descriptions of modernity (pp. 33-60). Stanford, CA: Stanford University Press. Marx, K. (1857). Grundrisse der Kritik der politischen Oekonomie . Moscow: Marx-Engels- Lenin Institute (1939). Maturana, H. R. (1978). Biology of language: the epistemology of reality. In G. A. M. a. E. Lenneberg (Ed.), , Psychology and Biology of Langu age and Thought. Essays in Honor of Eric Lenneberg (pp. 27-63.). New York: Academic Press. 20 21 Maturana, H. R., & F. Varela (1980). Autopoiesis and Cognition: The Realization of the Living . Boston: Reidel. Mead, G. H. (1934). The Point of View of Social Behaviourism. In C. H. Morris (Ed.), Mind, Self, & Society from the Standpoint of a Social Behaviourist. Works of G. H. Mead (Vol. 1, pp. 1-41). Chicago and London: University of Chicago Press. Merton, R. K. (1938). Science, Technology and Society in Seventeenth-Century England . Bruges, Belgium: Saint Catherine Press. Merton, R. K. (1942). Science and Technology in a Democratic Order. Journal of Legal and Political Sociology, 1 , 115-126. Meyrowitz, J. (1994). Medium Theory. In D. Crowley & D. Mitchell (Eds.), Communication Theory Today (pp. 50-77). Stanford, CA: Stanford University Press. Münch, R. (1982). Theorie des Handelns (Understanding Modernity. London/Ne w York: Routledge, 1988 ed.). Frankfurt a.M.: Suhrkam p. Nelson, R. R. (1994). Economic Growth via the Coevo lution of Technology and Institutions. In L. Leydesdorff & P. v. d. Besselaar (Eds.), Evolutionary Economic and Chaos Theory: New directions in technology studies (pp. 21-32). London/ Ne w York: Pinter. Noble, D. (1977). America by Design . New York: Knopf. Parsons, T. S. (1937). The Structure of Social Action . Glencoe, IL: The Free Press. Parsons, T. S. (1952). Superego and the Theory of Social Systems, Social Structure and Personality . London/New York: Routledge. Parsons, T. (1968). Interaction: I. Social Interaction, The International Encyclopedia of the Social Sciences, Vol 7, pp. 429-41. New York: McGraw-Hill. Parsons, T., & Shils, E. A. (Eds.) (1951). Toward a General Theory of Action . New York: Harper and Row. Price, D. de Solla (1961). Science Since Babylon . New Haven: Yale University Press. Sevänen, E. (2001). Art as an Autopoietic Sub-S ystem of Modern Society. Theory, Culture & Society, 18 (1), 75-103. Simon, H. A. (1969). The Sciences of the Artificial . Cam bridge, MA/London: MIT Press. Weber, M. (1904). Die Objektivität sozialwissenschaftlicher und sozialpolitischer Erkenntnis, Gesammelte Aufsätze zur Wissenschaftslehre (pp. 146-214). Tübingen: Mohr, 3 1968. Weber, M. (1917). Der Sinn der 'Wertfreih eit' der soziologischen und ök onomischen Wissenschaften (pp. 489-540). Tübingen: Mohr, 3 1968. Whitley, R. D. (1984). The Intellectual and Social O rganization of the Sciences . Oxford: Oxford University Press.