Semioses and Social Change. The Relevance of Semiosis on the level of Social Structure and the case of the Generative Metaphorization of Educational Communication Franz Kasper Krönig 1. Society and Semiosis Societal events can be described as sign processes in sign processes. More precisely they must be described in sign processes (how else? ), but not necessarily as sign processes. Whereas it can be argued that “every thought is a sign” (Peirce CP 1.538), a commensurably categorical assertion with reference to society seems to be problematic. Even if we understand society as a “communicatively closed system” (Luhmann 1995: 403) in the sense that every societal operation is communicative, this does not necessarily mean that we describe society when we describe communication. One would not be including the structures of society, its type of differentiation, its devices for the increase of communicative probability, its organizations, its structural couplings, to name but a few examples. Hence, the sociologically decisive question “which selection of signs will be successful in the communicative process” (Luhmann 1995: 160), is not being posed within a semiotic frame but within a theory of symbolically generalized communication media. 1 This media theory seems to provide exactly what sign theories lack in this regard: “Sign systems are far from possessing either an infrastructural or hypostructural virtus” (Rossi-Landi 1992: 248). It is quite clear that a sign theory which has not developed observation instruments for these structural processes in sign systems is unable to find them. However, the theory - or rather: the theories - of symbolically generalized communication media can show how social systems influence or even regulate the probability of communication by means of asymmetrical coding. When, for instance, the coding of legal communication works with the code »legal/ illegal«, the clear preference character of the code side »legal« is a substantial factor in the operativity of the law system. The preference value of this media code not only increases the motivation to participate in legal communication but also carries a reference to the function of this system in every single communication. All cases of change, modulation, or transformation of the preference value would at once be cases of societal change. As soon as the preference orientation of a certain social system (economy, law system, politics, religion, art, education, science) changes, this system focuses on a changed societal problem, delivers a changed performance for the other systems in its environment, increases the probability of the connectivity of other communicative offers and produces different self descriptions. K O D I K A S / C O D E Ars Semeiotica Volume 33 (2010) No. 1 - 2 Gunter Narr Verlag Tübingen Franz Kasper Krönig 4 Briefly, changes of the media code of social systems can be regarded as changes of the social structure. If we observed, for instance, that the educational system changes its coding »teachable/ not teachable« into the coding »better/ worse«, this would not only be an expression of an altered function of the educational system - and this means social change - but rather social change itself. How about the relation of signs to social change? Could one argue that sign processes are more than mere expressions - in our terminology: media - of societal processes? 2. Signs as media Though this paper questions the role of semiosis in social change, there can be no doubt about the fact that social processes are sign processes 2 (though - as already said - in a frame of nonsemiotic structures). What does this imply? Is this assertion similar to the assertion that musical processes are sound processes? Precisely this seems to be the case. The implication is: Always when there are musical events, there is sound 3 ; always when there are societal operations, there are signs brought into play. The relation is, though one of necessity, merely that of a medium/ form difference 4 . Thus, signs are - exactly as the air - a huge pool - Max Bense speaks of “Repertoire” (Bense 1969: 17), of manifold recombinatorable elements which facilitate uncountable formations (here tonal sequences, there meaningful sign chains). This is only possible when these media are transparent, if it is allowed to apply a visual metaphor here. If the air produced sound itself, if accordingly signs arranged themselves meaningfully or in a meaning-disrupting way, they could not be used as media, or at least not satisfactorily. 3. What it would mean to understand signs as a medium The example of music makes clear to which position semioticians move themselves when they conceptualize signs as a medium. Then, on the one hand, semiotics is the most universal approach and always appropriate when talking about both general societal and specific communicative (e.g. musical) issues. On the other hand, this competence only concerns - phenomenologically speaking - a very base constitutional layer on which the object in question is not yet determined essentially and on which it is for us “in seiner spezifischen Bedeutung verschlossen” (Cassirer 1942: 65). Of course, it makes sense to research which forms of music are possible under the condition that we have to presuppose the air as the underlying medium. The medium of air shows properties which do not remain without certain forms of influence in the sense of restrictions for musical form selections. Both the minimum level of loudness and the maximum spatial distance for musical performances are defined by the properties of the medium. It must be taken into consideration that the sounds of different instruments develop in a specific way in the spatial medium of the air so that this medium has a certain influence on orchestration. Moreover, the highly variable parameters of the air like air pressure, humidity, and temperature pose problems for the architecture of musical instruments which are thus not only bound to musical demands but also physical requirements of the medium. Even if there are many more examples like these that depict which types of influence and restrictions the medium of the air has for musical form selections, none of these examples can be interpreted as an instance of determination of the form by the medium. It Semioses and Social Change 5 does not even seem to be possible to argue that certain musical forms are more likely than others for reasons grounded in the medium. The rich stylistic variation music history shows cannot be traced back to a variation in the primary medium of air. 5 What does this mean with respect to semiotic processes? Does the medium of signs restrict the possibility of semiotic behavior? Wilhelm von Humboldt argued that language is nothing but the “Komplement des Denkens” (v. Humboldt 1973: 8) in the strict sense as it was shown later by Benjamin Lee Whorf with regard to grammar, which is “not merely a reproducing instrument for voicing ideas but rather is itself the shaper of ideas” (Whorf 1956: 212). The same holds not only for language but rather for any type of symbolic form, understood as “jede Energie des Geistes (…), durch welche ein geistiger Bedeutungsgehalt an ein konkretes sinnliches Zeichen geknüpft und diesem Zeichen innerlich zugeeignet wird” (Cassirer 1982: 132) as Ernst Cassirer has shown - one is almost tempted to say ‘later’ in an idea-historical sense. In concrete terms this means that Cassirer sees at least in “Mythos, Sprache, Technik, Recht (…), Kunst, Religion, Wissenschaft (…), Historie und (…) ‘Wirtschaft’” (Krois 1988: 19) a necessary oneness of material expression and the expressed. Even if one traces the necessity of materialization of mental (and social) performances as deeply as Cassirer, this never leads to the conclusion that societal processes are in any way determined, steered, instructed, and not even influenced by sign processes/ symbolic forms. The necessary inseparability and coincidence of - in old terms - form and content, thought and language or meaning and materialization which Cassirer argues for with reference to v. Humboldt, excludes any kind of causal relation. The material aspect of signs can in no way determine or influence its own application nor can it cause a change in the latter through the first, since the necessary distance is lacking. When something changes on the semiotic level, one may think of the grammar or the lexicon of a certain language, it makes no sense to describe the societal effects of these changes. Changes in the field of signs are at once societal changes, but not their cause. Thus, there can be no talk of causality in the relation between semiotic and societal processes (Systems theory speaks, as we will see, of semantics and social structure). Rather, it seems to be plausible to understand this relation as a medium/ form-difference. In doing so it is not implied that media have no relevance for the form selections which they facilitate. Of course, the quantity of semiotic elements and their combinatorability (the repertoire in the sense of Max Bense) must be seen as an uncircumventable restriction for that what can be thought and said. But still, the analogy to music holds. The medium of signs is to such an extent rich that one cannot state any limitation of semiotic behavior by the elements of the medium and its combinatorics. If we understand signs in this sense as a medium, we can see no influence of semioses on societal processes. To observe society in general or social change in particular semiotically would accordingly only mean to research the semiotic shape of societal processes without being able to claim any (e.g. circular) causal correlations between these two levels. The semiotic realization of societal processes would be completely arbitrary and sociologically irrelevant. 4. Semiotics, semantics and social structure How would we have to conceptualize semioses in order that they had a scientifically observable influence on social change? They would have to limit that which can be said not only generally and statically but also in a both structural (as opposed to incidentally) and dynamic Franz Kasper Krönig 6 way. One would have to show that semioses make some things which can be said more probable and other things less probable. Moreover, this would have to be something changeable or historic in order to play a role in matters of social change. So, is semiotics only occupied with the observation of the semantic level of processes which are in fact determined on a structural level? There is a broad discussion on the distinction between the levels of social structure and semantics in sociological systems theory (cf. Stäheli 1998; cf. Krönig 2007). As a result of this discourse nobody would easily argue anymore that there is a so-called operative level of communication where the societally relevant and thus structural processes take place opposed to a so-called semantic level which basically only observes and describes the societal operations without being itself of equal structural relevance. However, the solution cannot be to simply brush aside this distinction. It cannot be ignored that not all communications play the same role in the reproduction and the change of society. At least three levels can be distinguished. a) Observations Anything which is uttered, written, or marked in any way can be called observation, since “beobachten heißt einfach (…) Unterscheiden und Bezeichnen” (Luhmann 1998: 69). Thus, whenever there is communication, there is observation. It is clear that the greatest part of all communication is societally completely irrelevant. What A and B tell each other about the weather, but also the majority of scientific publications which do not influence the scientific discourse in any way, are observations which have no effect on social change. More precisely: A certain scientific communication has no societal relevance when it does not influence the probability of other communication at all. b) Semantics The term semantics as it is used here has no connection to the “semantical dimension of semiosis” (Morris 1938: 21). Rather, this concept stands in opposition - though a problematic one - to social structure since it merely observes what is produced on that operational level of society (cf. Luhmann 1998: 556). Among the multitude of observations there are comparatively very few which are taken up again, repeated and which have so to speak a semantic career. In contrast to incidental observations one can speak of semantic structures in order to differentiate the temporality of the elements. Structures are formed by mere repetition of element couplings. Accordingly the highly standardized conversation about the weather is strictly speaking an example of semantics in this sense. The coupling of elements like ‘weather’, ‘could’, ‘be’ and ‘better’ or ‘rain’ and ‘again’ is comparably strict and must be regarded as a strong and popular semantics. However, it makes sense to follow the sociological systems theory and to reserve the concept of semantics for couplings of observations (thus structures) which are observing communication and not only perception (like weather phenomena), since this difference definitely makes a difference. Societally relevant semantics are first and foremost the self-descriptions of the function systems. Especially the change in such self-descriptions, for instance of education, economy, politics, art, and science provides important indications of societal change on the structural and operative levels. When, for example, the educational system begins to observe itself using semantics of performance, competition, output-orientation, and efficiency and reduces the humanistic semantics which observes education as integrative, holistic, and free, social structure has changed. If one does not connect both these levels, social change cannot be seen at all. Semioses and Social Change 7 c) Social structure As already mentioned, the concept of social structure in the systems-theoretical context does explicitly not refer to a static dimension of society as opposed to its dynamic social change. Rather, it defines the difference between societal operations which either reproduce or disturb and change the societal order and those operations which only observe and comment on these structures or changes. Though the latter type of observation can very well develop structures, these structures are called the semantics of society in order to mark the difference between the ‘real’ operational change on the level of social structure and the mere observation of these changes. Stäheli (1998) argued that it does not make much sense trying to differentiate these two levels in such an easy way - especially not in the temporal order of first operations on the level of social structure and then observations on the semantic level. However, it makes sense to describe society as something that has a certain structure. Everybody will admit that we do not live in a society which is primarily structured by stratification or segmentation of its parts. Politics is a social system that is neither defined by being somewhere else in space nor by its superiority or hegemony over other systems. Rather, politics, religion, art, health-care, education, the law system and science relate to each other by their functional differentiation. Each system fulfils a certain function which cannot be taken over by any other system. Functional differentiation can be called the specific social structure of modern society. This structure is, however, not fixed but it is constantly being communicatively developed and confirmed - at least as long as it exists. The reason that the different systems follow their own functional orientation and that they manage to establish their functional role in society lies in the contingent fact that the single communications adhere to and thus stabilize this functional differentiation. Communications which contribute to the consolidation of the differentiation type - and thus the social structure - by focusing on societal problems 6 can be called societal operations in the strict sense. This excludes everyday communication but not even the most insignificant scientific communications, since they are at least confirming the difference between science and, for instance, politics or religion. 5. From semioses to social change? Whatever different sociological theories mean by the term social change, it is necessarily understood as something that concerns the social structure of society, or as already said above: its operational level. This could be, for instance, when communications do not comply with the differentiation type completely. These communications would not be clearly assigned to a certain function system, but would still belong to this system. Let us imagine there was a scientific publication which bluntly appealed to our belief or our political responsibility. Such a contribution would become excluded from scientific communication directly, so that one cannot say that this publication contributes to social change in any way, by, for instance, exemplifying a different and maybe more open understanding of science. How can it be possible for a communication to both belong to a system and to irritate this system? How can it be possible that communications affect the social structure anyway? If we think of the often stated and criticized economization of society, 7 for instance, - what can be the role of semiosis in this global process? First, it must be shown that sign processes can produce a non-arbitrariness of the subsequent sign production, and how they do this must also be demonstrated. It must further be shown how sign processes manage to increase their own probability, to develop structures, and to stabilize themselves. These are features which are Franz Kasper Krönig 8 generally attributed to systems. However, in the following it shall be demonstrated how semioses can gain momentum and influence the social structure without being autopoietic systems and eo ipso having the already mentioned “infrastructural or hypostructural virtus” (Rossi-Landi 1992: 248). 6. The generative metaphor as a sign process relevant on the level of social structure In the light of the foregoing discussion, relevance on the level of social structure means that a sign process anti-symmetrizes the probability of what can be said at all and what is communicatively successful. Furthermore, it must be part of a system without complying with its operativity completely. The generative metaphor does both. The generative metaphor, as described by Schön (1993), is a cognitive instrument that is not bound to language, but a general capacity of thought. Schön describes this kind of quasicognitive structure as the generative process of seeing-as. This means that the generative metaphor is not something observed, but something observing. When these metaphors see something as something else, this takes place in a reciprocal semantization process, where two elements alter continuously by seeing themselves in the light of the other - synchronous continuously altering - element. In cybernetic terms this is a non-linear process. If A is seen as B, and if the semantic effect of this seeing-as on A falls back on B, a feed-back loop has started: The making of generative metaphor involves a developmental process. It has a life cycle. In the earlier stages of the life cycle, one notices or feels that A and B are similar, without being able to say similar with respect to what. Later on, one may come to be able to describe relations of elements present in a re-structured perception of both A and B which account for the preanalytic detection of similarity between A and B, that is, one can formulate an analogy between A and B. Later still, one may construct a general model for which a re-described A and a redescribed B can be identified as instances. To read the later model back onto the beginning of the process would be to engage in a kind of historical revisionism (Schön 1993: 142 f.). In contrast to this metaphor concept, a classical substitution metaphor (Kurz 1982) like ‘man is a wolf’ is just a linear predicate transfer. The substitution of predicates like ‘cruel’ or ‘dangerous’ by ‘wolf’ is completely revertible and has no effect on the concept of ‘man’ or ‘wolf’ at all. Understanding the phrase ‘man is a wolf’ as a generative metaphor would mean that the seeing-as of the man as a wolf changes the concept of man as well as the concept of wolf in a circular feed-back process. This would of course be very uncommon in this example. It is important to see that these metaphors not only fulfil the common falsity criterion, which can - with some problems - serve as a diagnostic criterion to identify metaphors (Beckmann 2001: 55). In the substitution view of metaphor the occurrence of a contradictory statement points to the question whether it is just a wrong or a metaphorical statement. Generative metaphors are, however, not just literally false statements which can be understood when the metaphorical term is substituted by the literal term. These metaphors manage a de-paradoxization of the paradox they are by dynamization of the contradictory identification. When both concepts are re-interpreted by the seeing-as of each other as the other one in a circular process, the false identification, that would be paradox, is de-paradoxized by time. Semioses and Social Change 9 In a schematic outline of the generative metaphor one can speak of a circular reinterpretation of two elements A and B , by which the paradox of the equation “A is B” = is being dynamized and invisibilized: A = B As we have seen above, the specificity of the generative metaphor is that it comes to a feedback of the seeing-as of A as B which reinterprets B on A. As an example of this feed-back in a generative metaphor we can take Schopenhauer’s sentence ‘architecture is frozen music’. The notion of architecture (A) changes, when it is seen as ‘frozen music’ (A B ), as well as the notion of music (B), which then becomes something like ‘architecture in motion’ (B A ): A B = B A This process can be called circular, since the now by B reinterpreted A becomes the starting point for the next ‘round’ of the seeing-as which accordingly does not come upon a B but on an already reinterpreted B A . The result of this semantization process becomes its new starting point. A B, now seen-as B A , which has thus internalized some aspects of B A in order that this equation does not appear as a clear mistake blocking the process becomes A BBA and accordingly B A becomes B AABBA. A BBA = B AABBA The next step which can hardly be verbalized would look like this: A BBABAABBA = B AABBAABBABAABBA It is clear that the part the reinterpretation process adds to the interpreted (this is the original starting point A) becomes more and more prominent. That means that the original distinction between A and B diminishes though it never completely disappears. 7. Economization of education driven by a generative metaphor Speaking of metaphors makes us think of linguistic phenomena. However, the generative metaphor is a device of de-paradoxization of communication which can be applied not only by human perception, understanding and languaging but also by semioses on the level of function systems of society. At least this is the thesis (cf. Krönig 2007) which does not seem to depend on an orthodox and elaborate theoretical systems frame, but only on the common and unopposed 8 understanding that communication in modern society is differentiated functionally. First, we have to differentiate educational from economic communication. We follow one of Niklas Luhmann’s proposals that educational communication is coded by the digital opposition ‘better learning/ worse learning’ 9 . Only when communication focuses on this Franz Kasper Krönig 10 guiding distinction can we talk of educational communication. The economic communication, which focuses on the difference of efficiency/ inefficiency 10 , is of course not connectible to this digital code. In practice this means that a communicative contribution which brings calculations concerning the efficiency of pedagogic work into play is observed as a nonpedagogical communication by the pedagogic system. Of course, teachers are not always communicating pedagogically. They can very well make economical or political observations on the pedagogic system. The point is that when and as long as pedagogic communication, for instance in a class room, is running, all other observation modes are excluded and regarded as disturbance. Alternatively the system, that means the specific communication mode, breaks down and allows space either for a differently coded communication or non-functional communication (every day communication). However, it is impossible to observe in two binary coded observation modes which exclude each other. On the other hand, there seems to be a great pressure on the pedagogic system to integrate economic preferences in its very own observation. How does the pedagogic system deal with the economic demand for effective learning and teaching? It seems that the educational system has found a powerful device for such a connection to the unconnectable: The generative metaphor of quality manages to deparadoxize the paradox ‘better learning is efficient’. »better learning« = »efficient« quality This can explain the development in the educational system which is often described as economization with terms like output orientation 11 , functionalization 12 , competition 13 , result orientation 14 and the inflationary boom 15 of all kinds of tests and competitions of school and educational performance. 16 It has become common to demand as high an educational output as possible in as short a period of education as possible. Every system is confronted with different and necessarily contradicting demands of other systems. Science depends on the education of scientists, religion depends on confessional education, politics demands responsible citizens - or at least active voters - and the economy prefers qualified human capital for little money. This has always been the case and will never change in a functional differentiated society. Similarly it has always been the case that the educational system could easily decline these demands in reference to its very own preference value which cannot be other than incompatible with all other systems’ preferences. The beginning of this development can be dated to the early 1990s 17 , since when the educational system seems to have followed the demand of the economic system in trying to increase its performance in economic categories. Quality is the key concept 18 which invisibilizes the paradoxical attempt to identify the educational preference with the economic preference. It is interesting that the locus for this development which is extensively complained about in the educational system, is education itself. The metaphorical seeing-as of the educational as the economic is nothing but an autonomous attempt of the educational system to connect itself to the economic system. When education modulates its own preference in this way, it does of course not operate in an economic way at all, but merely with the illusion of doing so. Though the self-economization of education must be seen as a paradox attempt which does not reach its goal, this development is far from being without consequences. The conversion of economic noise into educational information distances education from its own Semioses and Social Change 11 preference which is linked to the system’s function, so that this definitely endangers the system’s performance. That pedagogic communication manages to react to economization in a way that not only endangers its very own functional performance but also amplifies the economization of society, can be - as we have seen - traced back to the functioning of a special type of semiosis. As one of the important features of this sign process we have displayed its power to invisibilize the paradox equation of educational and economic preference orientations. When it is true that semioses are not only expressions/ media of societal operations, but rather driving forces for social change, there is an urgent need for a semiotics which fulfils the role of a critical observer and analyser of societal processes. References Beckmann, Susanne (2001): Die Grammatik der Metapher. Eine gebrauchstheoretische Untersuchung des metaphorischen Sprechens. Tübingen: Niemeyer. Bense, Max (1969): Einführung in die informationstheoretische Ästhetik. Grundlegung und Anwendung in der Texttheorie. Reinbek bei Hamburg: Rowohlt. Böttcher, Wolfgang (2001): Ist ein Dialog zwischen Pädagogik und Ökonomie möglich? Eine Diskussion am Beispiel von Anreizsystemen. Zeitschrift für Pädagogik, Heft 4, 2001, 893-911. Cassirer, Ernst (1942): Zur Logik der Kulturwissenschaften. Fünf Studien. Göteborgs Högskolas Å rsskrift XLVIII. Cassirer, Ernst (1982): Der Begriff der symbolischen Form im Aufbau der Geisteswissenschaften. In: Hans-Ludwig Ollig (Ed.): Neukantianismus. Texte. Stuttgart: Reclam, 127-163. Gruschka, Andreas (2001): Bildung. Unvermeidbar und überholt, ohnmächtig und rettend. In: Zeitschrift für Pädagogik, 47. Jg., Nr. 5, 621-639. Habermas, Jürgen (1985): Theorie des kommunikativen Handelns. Vol. 2. Zur Kritik der funktionalistischen Vernunft. Frankfurt a.M.: Suhrkamp. Heider, Fritz (1959). Thing and Medium. In: Psychological Issues, Vol. I, No. 3: On Perception, Event Structure, and Psychological Environment. Selected papers by Fritz Heider. New York: International Universities Press. Helmke, Andreas / Hosenfeld, Ingmar (2005): Leistungsmessung in der Grundschule. In: Klexer. Magazin für die Grundschule, Nr. 14, Januar 2005: 3-5. Humboldt, Wilhelm von (1973): Schriften zur Sprache. Stuttgart: Reclam. Krause, Detlef (2001): Luhmann-Lexikon. Eine Einführung in das Gesamtwerk von Niklas Luhmann, 3. Auflage, Stuttgart: Lucius&Lucius. Krönig, Franz Kasper (2007): Ökonomisierung der Gesellschaft. Systemtheoretische Perspektiven. Bielefeld: Transcript. Krois, John Michael (1988): Problematik, Eigenart und Aktualität der Cassirerschen Philosophie der symbolischen Formen. In: Hans-Jürg Braun; Helmut Holzhey; Ernst Wolfgang Orth (eds.): Ernst Cassirers Philosophie der symbolischen Formen. Frankfurt am Main: Suhrkamp, 15-44. Kurz, Gerhard (1982): Metapher, Allegorie, Symbol, Göttingen: Vandenhoeck & Ruprecht. Luhmann Niklas (1991): Einführende Bemerkungen zu einer Theorie symbolisch generalisierter Kommunikationsmedien. In: Soziologische Aufklärung 2. Aufsätze zur Theorie der Gesellschaft, ed. Niklas Luhmann. Opladen: VS Verlag, 170-192. Luhmann, Niklas (1995): Social Systems. Translated by John Bednarz, Jr. with Dirk Baecker. Stanford University Press: Stanford (CA). Luhmann, Niklas (1997): Erziehung als Formung des Lebenslaufs. In: Bildung und Weiterbildung im Erziehungssystems. Lebenslauf und Humanontogenese als Medium und Form, ed. Dieter Lenzen / Niklas Luhmann. Frankfurt a.M: Suhrkamp, 11-29. Luhmann, Niklas (2001): Das Medium der Kunst. In: Aufsätze und Reden, ed. Oliver Jahraus. Stuttgart: Reclam, 198-217. Luhmann, Niklas (2002): Die Religion der Gesellschaft. Frankfurt a.M.: Suhrkamp. Morris, Charles (1938): Foundations of the Theory of Signs. Volume 1, Number 2. Chicago: The University of Chicago Press. Franz Kasper Krönig 12 Mühlenkamp, Holger (2002): Zum grundlegenden Verständnis einer Ökonomisierung des öffentlichen Sektors. Arbeitsbericht Nr. 1/ 2002, Institut für Haushalts- und Konsumökonomik, Universität Hohenheim. Parsons, Talcott (1980): Zur Theorie der sozialen Interaktionsmedien. Opladen: Westdeutscher Verlag. Peirce, Charles Sanders (1958). Collected Papers of Charles Sanders Peirce, ed. Charles Hartshorne; Paul Weiss; Arthur W. Burke. Cambridge (MA): Harvard University Press. Reinhart, Gunther / Lindemann, Udo / Heinzl, Joachim (1996): Qualitätsmanagement. Ein Kurs für Studium und Praxis, Berlin: Springer. Rossi-Landi, Ferruccio (1992): Between Signs and Non-Signs. Amsterdam; Philadelphia: John Benjamins Publishing Company. Schön, Donald A. (1993): Generative metaphor. A perspective on problem-setting in social policy. In: Metaphor and Thought, second edition, ed. Andrew Ortony. Cambridge: Cambridge University Press, 137-163. Schwarz, Paul (2002): GEW-Zeitung, Rheinland-Pfalz, 9/ 02. Stäheli, Urs (1998): Die Nachträglichkeit der Semantik. Zum Verhältnis von Sozialstruktur und Semantik. In: Soziale Systeme 4, H. 2: 315-339. Terhart, Ewald (2000): Qualität und Qualitätssicherung im Schulsystem. Hintergründe, Konzepte, Probleme. In: Zeitschrift für Pädagogik, 46. Jg., Nr. 6: 809-829. Whorf, Benjamin Lee (1956): Language, Thought and Reality. Selected writings of Benjamin Lee Whorf. Ed. John B. Carroll. Cambridge (MA): MIT Press; New York: John Wiley&Sons. Notes 1 Habermas (1985: 522); Parsons (1980); Luhmann (1991). 2 Doubt could only be communicated in sign processes (performative self-contradiction). 3 One could argue that experiments of music without sound production (John Cage) are nevertheless the attempt at making silence a possible form choice in the medium of sound. 4 We refer to the medium/ form-difference theory as proposed by Niklas Luhmann (2001) as a further development of the work of Fritz Heider (1959) in the field of a thing/ medium-relativity. 5 Not before the 20 th century could one say that the evolution of music is being triggered by the evolution of secondary media (electric and electronic instruments, recording, mixing, sampling and sequencing facilities). 6 Since function is defined as the unity of the difference between problem and (einer “Mehrheit von funktional äquivalenten” (Luhmann 2002: 116)) solutions, the pursuance of a societal problem is the same as a functional orientation. 7 Cf. Krönig 2007. 8 Of course, there are great debates on the relation of functional differentiation to other differentiation types (center/ periphery; stratification; segmentation). 9 This form for the code can be found in Krause (2001: 43). 10 Luhmann speaks of payment/ non-payment, but he points out that payments “always require a counter movement, transferring goods, services, or other monetary variables” (Luhmann 1995: 462). So our proposition of efficient/ inefficient seems to respect this important fact better. Though payments are the operations of the economic system, this does not mean that the coding must be payment/ non-payment as well. Besides that, efficiency makes more sense as a preference value than payment, since payment as such must be regarded as neutral. Only in relation to the ‘what for’ of payments, can payments be preferred or rejected to non payments. 11 Cf. Böttcher, 2001: 893. 12 Cf. Luhmann, 1997: 27. 13 Cf. Gruschka, 2001: 626. 14 Cf. Helmke/ Hosenfeld, 2005: 4. 15 Cf. Schwarz, 2002. 16 PISA, MARKUS, TIMMS, VERA, IGLU, DESI. 17 Cf. Mühlenkamp, 2002: 1; Reinhart/ Lindemann/ Heinzl, 1996: 1. 18 Cf. Terhart, 2000: 809.