Dialogic control: Power in media evolution

Semiospheric approach

If we are interested in the evolution of media forms, then Lotman has stressed that ‘well-defined and functionally unambiguous systems’ never exist in isolation. Instead, they acquire their role and meaning when perceived as one segment of the continuum of multifaceted, multi-levelled and variegated semiotic formations – that is, when immersed in semiotic space (Lotman, 1990: 123–4). To interpret the dynamics within this space Lotman proposed his concept of ‘semiosphere’. Originally coined by Lotman in an analogy with Vladmir Vernadsky’s ‘biosphere’ and ‘noosphere’, and Bakhtin’s logosphere (Mandelker, 1994), it refers to an abstract ‘semiotic continuum’ that is inherently heterogeneous and enclosed in itself that functions as a self-referential system (Nöth, 2006: 261) but is also in constant interaction with other similar structures (Kotov, 2002: 42). As Hartley (1999: 221) elaborates, the semiosphere should be understood as ‘the whole environment of sense making, required to make any individual utterance possible’. The existence of this environment is a prerequisite for any single act of communication; it is necessary for the existence and functioning of languages and all forms of communication.

As such, it should be emphasised, the concept of semiosphere is not merely a synonym for culture as has been sometimes suggested (Sebeok, 2000: 532). Rather, it refers to the complex relationship between a culture, its different sub-components and its semiotic environment. Relying on the organicist philosophical strategy, it presumes and analyses isomorphic relations between all the structures and levels of a semiosphere (Alexandrov, 2000: 347; Mandelker, 1994: 390). For instance, if we relate a single website as an inherently heterogeneous but bounded textual entity to the whole ‘web-culture’ as another textual entirety and after that to the whole global ‘sphere’ of human culture, their differences should be understood only as quantitative – one ‘level’ cannot exist without the other and one cannot be interpreted without knowledge of the other. Still, as Nöth (2006: 259) explained, Lotman’s hierarchy of levels makes up a system of relational stratifications in a way that higher levels are always conceived as semiotic spaces with more dimensions in relation to the spaces of the lower levels that they embrace. But the dichotomy of super-systems/sub-systems is not linear or simple. Instead, systems of different levels intertwine and relate to one another in complex ways. Super-systems embrace numerous smaller ones. For instance, a carefully designed cross-media strategy as a textual system could consist of a variety of media types and genres (print media, TV, radio, internet, mobile apps, videogames, etc.). At the same time smaller systems could also be perceived as parts of several bigger ones – a website could be seen as part of an internationally recognised webmedia genre, or again as a component of a carefully orchestrated transmedia narrative, or be seen as an entity of a distinct national media system. Therefore the evolution of various groupings of websites as distinct textual domains is also conditioned by the semi-autonomous operations of all these super-systems. What is more, as such the ‘websites’ as well as most individual media platforms constitute dialogic spaces where the various super-systems meet, partly converge and, therefore, are conditioned to co-evolve.

All in all, Lotman argues (1990: 138), the entire space of a semiosphere is transected by boundaries between different levels, sub-sphericules and texts. These boundaries, as they separate, although they also connect and translate (Lotman, 1990: 136–7), are inseparable from the term ‘individuality’ (Torop, 1999). Individuality is seen as the outcome of the autopoietic process where a cultural system identifies itself and its boundaries in space and/or time. It is the self-defined continuum inside the self-generated boundaries that thereafter become the mechanisms of translation – as identifying oneself presumes the realisation that between one’s own domain and an alien domain difference exists, and that the alien domain (‘Theydom’, as Hartley [1996: 107] explicates it) then needs to be understood and translated. It is the coexistence of the infinite number of such sub-systems of culture, with their asymmetrically different languages, discourses and identities that, despite their difference, forces them into dialogues.

Auto-communication

Let us look in more detail into the phenomenon of semiotic self-generation as articulated by Lotman. For denoting this he proposed the concept of ‘auto-communication’ (Lotman, 1990: 123–4), which describes communication from and to oneself where the self-communicating entity can be both an individual and a larger social structure. As several authors have elaborated (Broms and Gahmberg, 1983; Christensen, 1997; Morsing, 2006; Steedman, 2006), in the modern-day context all kinds of communications (such as strategic plans, corporate reports, marketing communications, press releases) that organised bodies or systemic structures might produce could eventually start working auto-communicatively. Even if the communicative act was originally meant not for internal use but for an outside audience, once the message feeds back to its authoring structure the auto-communicative effect has taken place. As Lotman explains, the difference between ‘I–s/he’ communication and ‘I–I’ communication comes down to the fact that in the latter case information is always transferred in time. As a result the information received is eventually qualitatively changed as the circumstances and contexts of the message have changed by the time of its re-articulation by its author. The message has acquired supplementary codes and has the potential to lead to a restructuring of the actual ‘I’ itself (Lotman, 1990: 22). Hence, as it is argued (Christensen, 1997: 202; Morsing, 2006: 175), auto-communication is not primarily oriented towards sending and receiving messages but towards the production and celebration of meta-texts on the identity and nature of the communicating structure. Broms and Gahmberg (1983) suggest that auto-communication turns into a process of organising through which a communicator evokes and enhances its own values and the repetitive use of the same textual form thus produces the mythologies of the communicating structure. As opposed to dialogic communication, auto-communication generates homogeneity at the expense of heterogeneity (Kotov and Kull, 2006: 196).

Unity/plurality

A primary principle that Lotman shares with several system theoretic approaches is that plurality and unity presume and condition each other. As Niklas Luhmann puts it:

(at least) two complexes with divergent perspectives are required to constitute whatever functions in the systems as a unity (unit or element). In reverse, this means that, for analysis of the system, such a unity cannot be dissolved into the divergent complexes constituting it. (1995: 38)

He suggests that, to achieve confidence about this, one can investigate the repercussions of this mutualistic-dialogical, conversational unity and its ‘language’ on the complexes constituting it – in order to study to what extent and within what boundaries these repercussions allow the individualisation of its elements.

This in turn could be understood as a question designed for semiotic analysis. Luhmann is similar to Lotman for arguing that social systems are in effect nothing else than ‘autopoietically’ functioning ‘communications’. From the perspective of Lotman, an act of Luhmann’s ‘communication’ could be understood as an act of language in use. That is, it is bounded by the limits of time and space and can therefore be defined as a ‘text’. But specific to Lotman’s conceptualisation of ‘texts’ is that, whatever may be their modality or materiality, they are always inherently heterogeneous, they are a structure of two at minimum (Lotman, 2009: 2). What this means is that ‘texts’ of all kinds are organised by multiple codes and consist of several semiotic systems. For instance, a prayer is organised – in addition to the logic of verbal language – also by the symbolic message of a particular religion and its specific organising conventions. In a similar way most texts offer for interpreters various intertextual switchers that bring in cultural codes from their ‘extratextual’ outside, which may expand the spectrum of meanings associated with the text.

Another example is poems, which are governed by a vast amount of cultural codes that are not derived from the verbal language: metrics, rhythm and plot, but especially their rhetorical structure – metaphors, comparisons, metonyms and so on, that we can find in the poem and that make it work poetically. These make up a system that functions entirely differently from natural language. In ‘primary’ natural language conventionality (symbolicity in Peirce’s sense) dominates, but in rhetorical figures it is the motivated similarity (iconicity or diagrammatic relations in Peirce’s sense). The poetic innovations of poems come from a tension of novel co-functioning of such different modes and principles (Lotman, 1976). But even natural language is rarely a system representing the world in a direct or simple way as it is always permeated with metaphors and other rhetorical figures (Danesi, 2003; Lakoff and Johnson, 1980; Sebeok, 2001: 58–9) and hence verbal signs are hardly ever ‘primary’ representations of the world (Nöth, 2006: 258) but are literally ‘figures’ that model the world in a variety of ways.

But what if the systems of representation under study are not natural languages but, for instance, the new media forms on our mobile interfaces? In examining this we should first recall the arguments that modern media interfaces, as they are made to ‘remediate’ (Bolter and Grusin, 1999: 45), end up being constellations of modally different topoi from many earlier media (Liestøl, 1999). In other words, these are considered to be increasingly heterogeneous for the representational conventions, modes and media forms they are ‘remediating’. Hence the argument about the inherently heterogeneous nature of all texts becomes more plausible with multimodal media.

Proceeding from that we can suggest, on the one hand, that the theories of Lotman and Luhmann depart from similar premises and, on the other, that Lotman’s semiotics helps to conceive how the ‘communications’ of Luhmann’s system as Lotman’s ‘texts’ guarantee the system’s autopoietic closure as well as its openness simultaneously. It comes from the understanding that, if a system is inherently heterogeneous, it collocates many of the existing languages, it has to be able then to also connect to its environment – where the same or similar language-systems or codes most probably exist. It is the principle that every text is intertextually connected to the rest of the culture. Paradoxically, a text can be perceived, without losing its integrity, not to be identical only with itself but also with a variety of superstructures – with semiotic systems it is a part of. As Luhmann (1990: 13) proposes: ‘communication is an evolutionary potential for building up systems that are able to maintain closure under the condition of openness’. Ultimate isolation is impossible, every social entity (especially those designing communicative forms) has to be in dialogue with others.

Dialogues, auto-communication and emergence of new systems

The suggestion here is that the ability to observe the ‘theydom’, and potentially to understand and talk to ‘Others’, also brings a potential for a dialogue and, subsequently, emergence of new systems. Luhmann explains that ‘every social contact is understood as a system, up to and including society as the inclusion of all possible contacts’ (1995: 15). Therefore when contact between two participants is established, there takes place a dialogue, then there is a possibility for autopoietic closure and an emergent social system.

When we now think about the emergence of various new media out of the dialogic contacts among the variety of societal sub-systems, of institutions and organisations, I first return to the argument that new textual forms repurpose and integrate a variety of representational conventions and rhetorical topoi from the previous or parallel media. Such inherent richness can be suggested to result from the dialogues between the engaged structures. Krippendorff (1995a), for instance, has shown how the vocabulary of modern design discourse stems from several sources – the arts, engineering, ergonomics, advertising, popular culture, software manufacturing and so on. Similarly, any new medium could be understood as a convergent domain that has taken shape in dialogues among a variety of social systems and institutions. Related to this, at the stage when a medium is still young there are not many new and medium-specific norms for the texts as well as for their production practices – since much is simply inherited from ‘parent domains’. However, the new ‘emancipated’ practices and medium-specific conventions are expected to evolve as part of the new system’s auto-communicative process. We can take here as an example Rivett’s suggestion of the challenges to the further evolution of web design in the early 2000s. She proposed that at this point that design as a practice could not survive for long without being sustained by institutions and that, relatedly, its future was dependent on the commodification of the ‘website’.

This is an economic imperative for – if designers are to secure contracts from the commercial sector – they must be seen to provide a recognisable ‘worthwhile’ product. Part of this process of commodification is manifested in the drive (emanating from commercial Web design) to legitimise and professionalise the design/construction of the website. This is demonstrated in the work emerging from the area of website design where there is evidence of the construction and imposition of frameworks for the design of on-line media, within which certain forms are represented as legitimate and others not, a process integral to the professionalisation of this area of design. (Rivett, 2000: 13)

I propose that in most cases the meta-discourse that tries to impose norms for a design from a perspective of a certain social system is also, for the most part, the same discourse that auto-communicatively articulates the identity of the same social system. As Graham and McKenna (2000: 49) propose: ‘the higher the degree of consistency between systematically produced descriptions and individually produced descriptions, the more likely it is that a particular discourse community will maintain an ongoing identity within society’. Krippendorff (1995a) establishes three ways in which design discourse should be instituting its recurrent practices:

Krippendorff also argues that discourse ‘surfaces in textual matter’ which is continuously (re)read, (re)written, (re)produced, (re)searched, (re)articulated, elaborated or rejected. ‘A community continually (re)generates its textual matter and acquires the character of a dynamically connected diversity’ (Krippendorff, 1995a; see also Krippendorff, 2008). I propose that this principle of ‘dynamically connected diversity’ is beneficial for interpreting cultural evolution as it helps us to understand the dynamic processes of social emergence at the borderlines of the existing social structures, discourses and languages. Both the emergent as well as older ‘social systems’ of media production/consumption are inherently heterogeneous in terms of their textual or discursive constellations since they are also in constant dialogue (in multitudes) with outside (since in different ways they are parts of a variety of super-systems), hence the auto-communicative function only reacts to all this diverse dynamic by making suitable connections and rearticulating the functions and identity of the own-domain.

The power of auto-communication

A proposition of this article is that power both emerges from as well as conditions the processes of both auto-communication and dialogic communication. Let us first focus on auto-communication and let us start with Luhmann. He explains that system and environment collaborate constantly, producing every effect, and this relies on the principle that when a system ‘produces’ itself, then it selects some and not all causes that are necessary for specific effects that can be employed under the control of the system.

This difference makes selection possible, and selection makes retention possible. Therefore a complex of ‘productive causes’ can come together as a result of evolution (or subsequently with the help of planning) and, once together, be in a position to assemble appropriate environmental causes. (Luhmann, 1995: 20)

Systems have to select as for them their environment is always more complex than themselves (disorder) and this leads to contingency, which from the perspective of a system can be perceived as a risk to their stability. But what is important is that such selection, every attempt to impose order by a social entity, even if it is done in order to reduce risk, means also an application (or, as Pottage [1998] suggests: the emergence) of power.

This associates with Manovich’s (2001) implicit proposition that in media development there are always two universal stages: a short initial gestation period where it evolves at a rapid pace and develops its main characteristics, and a second stage, where, having acquired its final form, it will thereafter undergo only minor changes during the rest of its existence. Such a way of dividing evolutionary processes of cultural forms analytically into two is reflected in the way Eco (1979: 138) and Lotman (1977) distinguish between ‘grammatically oriented’ and ‘textually oriented’ cultures, and how Kress and van Leeuwen (2001: 113) talk similarly about ‘lexically’ and ‘grammatically’ organised semiotic resources. According to Lotman’s view, textual culture generates texts directly which constitute macro-units from which rules could eventually be inferred. Kress and van Leeuwen add that, in such ‘cultures’ semiotic modes are approached as a paradigm, a loose collection of signs, which functions as a more or less unordered storehouse of resources. In grammatically oriented cultures, in turn, texts are generated by combinations of discrete units and are judged correct or incorrect according to their conformity to the grammatical rules of the particular system.

Grammars … use very broad, abstract classes of items, but provide fairly definite rules for combining them into an infinite number of possible utterances. They are decontextualised and abstract, but also powerful in what can be done with them. Perhaps it is no wonder that grammatically organised modes have tended to be the most powerful modes. (Kress and van Leeuwen, 2001: 113)

The transition in language-systems from one phase to another is understood by Lotman as a system’s movement towards self-description. The fact that Kress and van Leeuwen recognise the increasing power of the grammatically organised modes refers to how auto-communicative functioning is connected to the issue of power. Self-description and normative grammar development are Luhmann’s selections: a system minimising its risk by organising its environment on its own terms, an action legitimised by the power it masters. ‘The highest form and final act of a semiotic system’s structural organization is when it describes itself. This is the stage when grammars are written, customs and laws codified’ (Lotman, 1990: 128). In Lotman’s terms it is a necessary response to the threat of too much diversity within the semiosphere: the system might lose its unity and disintegrate. But this also suggests that one part of the semiosphere – the sub-system that strives to become its dominant centre – is in the process of self-description, of creating its own ‘grammar’, after which it strives to extend these norms (and this way, itself) over the whole semiosphere and, in this way, a partial grammar of one sub-system might become the meta-language of description for culture as such. This makes it a direct application of power that brings along power asymmetry.

There have been several theoreticians who have recognised this kind of phenomenon in new media development. Rivett (2000: 43), for instance, has argued that the increasing circulation of ‘handbooks’, which combine the classification and analysis of websites with the construction of site-design principles, is inextricably linked to particular groups’ attempts to impose their particular vision not only of what the website should be, but of the future of the web itself. The emergence of ‘handbooks’ in Lotman’s terms refers, however, exactly to the stage of ‘grammatical’ orientation of cultures – where the self-descriptions result in detailed codifications of how the new forms of culture should be produced (Eco, 1977: 138; Lotman, 1977). An example of this is also the currently multiplying handbooks on ‘transmedia’ (e.g. Alexander, 2011; Bernardo, 2011; Dowd et al. 2013; Phillips, 2012; Pratten, 2011) – where a set of practitioners closely connected to globally dominant audiovisual industries suggest ‘good’ ways to develop multi-platform narration and to produce transmedia content. In this case we might again have a similar dynamic where the dominant groups within the global industry seek to standardise the specific practices and forms of content of the ‘convergence culture’ (Jenkins, 2006). As the emergence of ‘transmedia’ practices is directly related to the emergence of the networked media it provokes again the question of whether the internet, rather than becoming the radical free-form space predicted by early enthusiasts, is instead being effectively shaped by a variety of dominant cultural forces (Galloway, 2004).

Centre–periphery dynamics

Lotman’s theory opens up still a different perspective for understanding such developments and suggests instead grammatical diversity and ‘unpredictable workings’ for the future of new media. As Andrews (2003: 68) explained, meta-description always gives rise to higher entropy. Once the core of a system starts self-regulating itself and becomes rigidly organised, it starts losing its dynamism. Having exhausted their reserve of indeterminacy, they become inflexible and incapable of further development. But in the ‘periphery’ of a semiosphere, where the agents tend to be in dialogic contact with various ‘others’, the ‘idealised’ norm or a regulative framework of a core will be in contradiction to the semiotic reality lying ‘underneath’, and not a derivation from it. The more to the periphery, the more the power of the core gradually diminishes and the ‘grammars’ of the core become illegitimate. Hence the relationship between semiotic practice and the norms imposed on it becomes ever more strained.

Texts generated in accordance with these norms hang in the air, without any real semiotic context; while organic creations, born of the actual semiotic milieu, come into conflict with the artificial norms. This is the area of semiotic dynamism. This is the field of tension where new languages come into being. (Lotman, 1990: 134)

‘Peripheries’ are the ‘spaces’ in cultural systems that are principally defined by being less regulated (less described) by the dominant ‘cores’ of cultural systems. These are spaces/sub-systems operating relatively autonomously within the auto-communicatively operating super-system and, as such, are more open to also observing other spaces and absorbing semiotic resources from these spaces potentially for their auto-communicative operations.

A relatively recent example of this is the evolution of the mobile web – as I have demonstrated elswhere (Ibrus, 2013a, 2013b) many of the mobile industry’s big infrastructual institutions (telecoms, software developers, search engines) have been interested in designing the mobile web to be as similar as possible to the familiar desktop web – so as to create a cost-effective ecosystem of relatively similarly functioning media platforms. To achieve this goal, these dominant institutions worked on and actively promoted the appropriated web authoring standards that, first, presumed ‘platform agnostic’ content development by content developers and, second, would have put the large infrastructural stakeholders (especially browser vendors) in the position to determine how content is subsequently adapted to various platforms. The only problem was that resistance emerged against this evolutionary trajectory among much smaller content and service providers. They demanded the right to adapt their content themselves to the ‘delivery context’; that is, to maintain control over communication between themselves and their audiences with regard to the form of the message, over what is communicated to whom, when and where. What is more, these ‘grassroots level’ developers, being more immediately aware of the specifics of content development, designed their own appropriated technologies of ‘server-based adaptation’ that were to set the mobile web yet again on a different evolutionary path – one that results in the ‘controlled divergence’ of web-access platforms, especially with regard to forms of content and perceived functionalities of these platforms. I have also identified (Ibrus, 2010, 2013a, 2013b) how, after the dialogical interchange between the two groups, the ‘infrastructure enablers’ started to compromise and ‘accepted’ the potential discontinuities in content forms of the two platforms – a code of practice for the cross-platform web that initially emerged at the periphery of the industry, but gained popularity for being more in touch with the semiotic milieu and the needs of ‘sign-makers’.

Therefore it could be suggested that, generally, all the different avant-garde movements and subcultures that are somewhat independent from the existing power-structures – such as academic cultures (Castells, 2001), open-source movements (Weber, 2004) or simply the communities of creatively engaged end-users who occasionally organise themselves as informal ‘fringe groups’ (Sawhney and Lee, 2005) – ought to be the agents that break the rules, innovate and, in this way, secure the pluralism of grammars and languages of hypermedia and their dynamic development. What is more, sometimes the innovative peripheral system, if its semiotic production gains popularity and its standards are widely accepted, may undermine or even disrupt the existing power-structures in the society/culture. In Lotman’s terms this refers to the ‘maturing’ of the periphery, how peripheral disruptions may become the dominant codes and norms for the whole semiosphere.

This potential is the reason why Schönle (2001, 2003) ascribes to Lotman the status of a potential innovator of western cultural studies. He shows how Lotman, although sharing the post-structuralist premise of the primary role of discourse in founding reality, makes a case that the unavoidable and infinite diversity of a semiotic environment starts eventually mitigating the subject’s dependence on the discourse. ‘Thus subjects act on their impulse to autonomy by playing discourses against each other, recording them in an act of auto-communication that generates novelty in the process’ (Schönle and Shine, 2006: 24). So although people are immersed in systems – discursive or social – agency can still lie in themselves. In Mandelker’s (2006) terms this is enabled by the possibility of personal ‘estrangement’. As she points out, for Lotman the ability to deliberately distance oneself to ‘periphery’, where putting self-reflexivity into dialogue with the Other enables an estranged perspective to be achieved, in turn represents the possibility for an unpredictable, innovative, and free action that enables and empowers the individual.

De-ontologisation of power: dialogic control

This article suggests still that for refinement the centre–periphery power dynamics could benefit from some filtering through Luhmann’s and Foucault’s power theories. From Luhmann’s perspective the construction of self-referential systems results in a need to abandon the idea of unilateral control:

There may be hierarchies, asymmetries, or differences in influence, but no part of the system can control others without itself being subject to control. Under such circumstances it is possible – indeed, in meaning-oriented systems highly probable – that any control must be exercised in anticipation of counter-control. Securing an asymmetrical structure in spite of this (e.g., in power relationships internal to the system) therefore always requires special precautions. (Luhmann, 1995: 36)

What Luhmann here criticises is the so-called classical theory of power, in Foucault’s terms also known as the ‘juridico-political’ power-concept. There are three main assumptions in this image of power (see Foucault, 1990: 94–6). First, assertion of possession: power is conceptualised as a substance that can be possessed or exchanged, which implies an idea of power as a zero-sum game. Second, an assumption of location: power is concentrated in a centre from which it flows (causally and top–down) to the rest of society. Finally, the discourse of sovereignty relies on the contention that power serves purposes of repression: to exercise power is to limit freedom

According to Foucault this model evolved in the feudal era and is how sovereigns of the time preferred to present their power. For Luhmann modern society is primarily differentiated into operationally autonomous sub-systems and is, hence, without an apex or centre. As Borch (2005: 158) notes, this characterisation suggests the need for replacing notions of power that reinstate a conception of a hierarchically differentiated society as the contemporary semantics of power should not reflect a pre-modern social structure.

Foucault’s solution to this dilemma was his concept of governance (Foucault, 2002). Conceptualised as government, power is defined as ‘conduct of conduct’, or ‘action upon action’ and to exercise power is, in the first place, to structure the possible field of action of others, of all the actors in a shared environment. All the systems present in a particular environment face contingency and therefore make their selections so that to reproduce themselves they affect each other by virtue of their own autonomous principle of replication. As Pottage (1998: 22) puts it: each of the actors is dependent on the autonomy of the other. ‘The art of the game is not to dominate the opposing actor, but to anticipate and exploit its interventions, and thus to make one’s own interventions dependent upon an opponent’s restless invention of (counter-)strategies’ (1998: 22). As such power has no headquarters, no core where it sits ex parte, is not owned by any one subject. Instead it is relational, emerging through situated oppositions between autonomous and radically discontinuous processes; it is non-subjective, emergent and contingent.

In this context it should be noted that in Lotman’s writings there are some inclinations towards seemingly ontological takes on power – especially related to the centre–periphery model, where the presumption is that there is always a power asymmetry whereas power is located in the core of the semiosphere, the agents of the core ‘hold it’, while the periphery is relatively powerless. This could be interpreted as an ontological approach, where power is perceived as a substance that could be possessed. At the same time we should bring our attention to Lotman’s (1990: 150) argument that the semiosphere’s elements, its sub-systems, can be at the same time both active and receiving, in one sense centre and in another periphery. This principle offers a way to apply the semiospheric model in understanding the complexities of power dynamics in the modern functionally differentiated society. Namely, if for Luhmann power could be understood simply as ‘communication coded in a certain way’, then Mandoki (2004: 100) defines power as an effect of meaning for a specific subject in a specific situation according to a specific code. She argues that power depends on principles similar to how semiotics understands signs. If for Peirce a sign or representamen is something that stands for something to someone in some respect or capacity then, she argues, we can only speak of power as something that means such to someone in some respect or capacity. It is not an entity or property but a process that mirrors semiosis. ‘Power, however, has to function within a code, a semiosphere or an umwelt, that is, related to a particular context for a particular subject, and has to be actualized every time in order to operate as such’ (Mandoki, 2004: 100). Or, as Krippendorff (1995b) suggests, power is ‘dialogically embodied, emergent in “burdensome languaging” with the Other’. Hence, we can argue, it is in the languages and codes within the semiosphere that the existing power relations are addressed and redefined, enabling in this way mutual adaptation and co-evolution, ‘actions upon actions’. If we now recall that, according to Lotman, every cultural system is inherently infinitely heterogeneous and that every system could be seen as incorporated into different autonomous super-systems, we can posit that every system can participate in a variety of language-systems and thus also in many different power relations that to some extent are independent of each other. A system can in some of these be active and in others be passive, in one respect governing and in another being governed by others, in Foucault’s terms.

Therefore what this discussion suggests is that, when we think about new media artefacts and the (techno)culture around these as an extremely heterogeneous mesh of texts of different levels, modes and materialities, the evolution of this mesh suggests an immensely complex power-dynamic, comprised of multi-levelled mechanisms of control and counter-control between diverse networks of stakeholders of varying goals and backgrounds (Krippendorff, 2008). In some of these relationships the degrees of freedom might be greater for specific actors, while in others they might be fewer. The term ‘dialogic control’ is therefore hereby suggested as a way of understanding and describing such relationships for avoiding the too simplistic use of centre–periphery dynamics. ‘Dialogic control’ refers to co-evolution of auto-communicatively operating intertwined systems that are conditioned to absorb new information from other systems as well as to accommodate the changes effected by these other systems – resulting in an evolutionary process conditioned and controlled by multiple agents, all in dialogic contact.

A recent example of such evolutionary process is the development of the mobile media over the course of the last decade. Initially operator-centric value chains were disrupted by the influx of new players – content and service providers whose bargaining power was on the rise. The associated market horizontalisation started to turn value chains into ‘value nets’, and market competition was seen to be replaced by cooperative models of ‘co-opetition’. This was understood to be interdependent with the parallel process of industry convergence. That is, the industry convergence of institutions with traditionally different functionalities could be suggested to be equivalent to the horizontalisation of their power relations. Although several of the industry fractions tried to reproduce their previous roles and positions (see Ibrus, 2010, 2013a, 2013b), still, since the introduction of 3G air interface standards and appropriated devices there has been a flux in the roles of different stakeholders, in the boundaries separating and defining them, in their social organisation and in their capabilities to mould their environment. There has been a mesh of dialogical relationships, processes of dialogic control between an indefinite number of actors that, in concert, have made up the evolutionary dynamics of the mobile web and its media forms.

Conclusion

The suggestion of this article is that Lotman’s semiotics of culture could contribute usefully to the interdisciplinary approach to ‘media evolution’ that would then be also principally constructivist. I have demonstrated elsewhere (Ibrus, 2010, 2013a) that there could be potential in establishing a dialogue between the semiotics of culture and various evolutionary approaches to social and techno-economic change. This article, however, dealt directly with the question of how may the ‘textual dynamics’ – that is, the changes in the textual and discursive constellations in culture – contribute to the changes in society’s structuration in terms of its social formations, institutional settings and practices of work, especially the practices of production of new forms of media and culture. The special focus was on the phenomenon of power and how it could be perceived both to facilitate as well as to emerge from the dialectics of auto-communication and dialogic communication. Power is emergent in communicative action, but it is also recurring as it results from the recursivity of auto-communicative processes – although then always re-articulated and depending on the changes in the general environment. That is, power is constituted in a system’s response, by its ability to act upon actions by others. And it is therefore the ‘communication upon communication’ dynamics that constitutes inter-system dialogue as a mechanism of control for societal/cultural evolution. The notion of ‘dialogic control’ was therefore suggested in the article.

All in all, the article emphasised the usefulness of Lotman’s notion of dialogue – that leads to change and hence to cultural flux, rather than to social fragmentation. As Schönle and Shine (2006: 24–8) pointed out, Lotman’s theory provides an answer to cultural studies’ age-old dilemma between the hegemonic unity and decentredness of power. For dialogues do not only facilitate homogeneity – convergence of cultural domains – but also the emergence of the new, in effect divergence and heterogeneity. It is a significant paradox of media evolution that the convergence of various media systems and forms does not only mean one-way flows into semantic implosions as Baudrillard (1983) suggests, but also the evolution of new languages, new differentiations and functionalities, new borders, and new institutions or other social formations operating these borders, together with dynamically changing power-asymmetries between these domains.