Social Semiotics

Short Entry

‘Social semiotics’ can refer to two related but distinct entities. ‘Social semiotics’ without capitals is a broad, heterogeneous orientation within semiotics, straddling many other areas of inquiry concerned, in some way, with the social dimensions of meaning in any media of communication, its production, interpretation and circulation, and its implications in social processes, as cause or effect. ‘Social Semiotics’ with capitals is a distinguishable school in linguistics and semiotics which specifically addresses these issues. It is important because it synthesizes these issues, not because it covers those issues in a distinct or authoritative form. Social semiotics makes semiotics more broadly useful, and Social Semiotics assists in this process.

Three texts between them have helped to give currency and substance to the term and field of Social Semiotics. The influential linguist Michael Halliday’s Language as social semiotic, (1978) introduced the term ‘social semiotic’ into linguistics. Bob Hodge and Gunther Kress’s Social semiotics (1988) developed the term in a more fully semiotic framework. >From 1991 the journal Social Semiotics, influenced by both, provided a forum in which over a number of years different academics have given their own inflection to the term.

These three sets of influences have had distinct agendas and trajectories which enrich Social Semiotics, and each opens up specific connections with other fields of inquiry, within semiotics under that name, and beyond. Halliday was primarily a linguist. The ‘social semiotic’ for him contested the usual linguistic separation of language from its social nature, while at the same time beginning to expand its base beyond the verbal. Hodge and Kress were influenced by Halliday, and added a diverse semiotic base, incorporating the sometimes rival traditions of Saussure and Peirce. They also gave a critical inflection to the ‘social’, drawing especially on Marx and the Marxist tradition. The journal derived its name and rationale from ‘Social Semiotics’, but interpreted this generously in practice. This scope has the effect of problematizing any boundary that might have been supposed to exist around the specificity of Social Semiotics. Social Semiotics has become social semiotics, arguably benefiting both.

The broader field of social semiotics is the site of intersection between two more currents, not usually called semiotics but in practice implicitly or explicitly drawing on a social semiotic orientation and tool kit. ‘Discourse’ theory/analysis is widely used in many branches of social research. ‘Cultural studies’ is likewise a popular research tradition. Social Semiotics, which has many affinities with both, can act as a grand node, linking each to the other, reframing them within a wide network of related traditions. At the same time it can serve the salutary effect of bringing these vigorous traditions into the field of semiotics, energising semiotics and making its insights more widely available and appreciated.

Long Entry

1. Halliday’s ‘social semiotic’

Michael Halliday was already arguably Britain’s leading linguistic theorist when he coined the term ‘language as social semiotic’ in 1978. This prestige gave his intervention great impact, while at the same time it has kept the scope of that influence mainly within the study of verbal language. For him ‘the formulation “language as social semiotic” means… interpreting language within a socio-cultural context, in which the culture itself is interpreted in semiotic terms’ (1978:2). Implicit here is a division between ‘language’, understood as verbal language as studied by linguistics, and semiotics as the study of other systems, which interact with verbal language to make up culture.

Halliday here simultaneously illustrates and contests a widespread understanding of linguistics and semiotics as different branches of knowledge, as they often are institutionally, but not conceptually, as in Saussure’s grand scheme, which places linguistics within Semiotics (or Semiology, as he called it). Halliday’s position regarding semiotics is ambiguous. In one interpretation of his project he points to an as-yet undeveloped social semiotics to complete the work of his purely linguistic theory. However, in a more positive interpretation he is opening the way to a more complex relationship between linguistics and semiotics, in which insights into verbal codes, as understood with a more adequate linguistics, will illuminate the study of all other codes. In this sense his linguistic theory, framed to have a more adequate account of social forces and contexts, is already a strand in a Social Semiotics which did not yet exist when he wrote.

In spite of work by some of his followers (e.g., Martin and Rose 2005) the potential of Halliday’s ideas on verbal language has still not been fully realised as part of a general social semiotics, though some writers in Social Semiotics (e.g., Kress and Van Leeuwen) have absorbed Halliday’s ideas so deeply that the full extent of his influence is impossible to determine. The key premises of his linguistic theory, which work equally well as general premises for Social Semiotics, are:

  1. ‘Language is a social fact’ (1978:1) i.e., social relationships constitute language. This is the case with all semiotic codes.

  2. ‘We shall not come to understand the nature of language if we pursue only the kinds of question about language that are formulated by linguists’ (1978:3) That is, autonomous linguistics and semiotics alike are incapable of understanding the nature of their object in disciplinary isolation.

  3. ‘Language is as it is because of the functions it has evolved to serve in people’s lives’ (1978:4). That is, a functional perspective is a key to the inseparable relationship between semiotics and society, structure and function.

  4. There are three functions, or ‘metafunctions’, of language (1978:112): ideational (‘about something’); interpersonal (’doing something’) and textual (‘the speaker’s text-forming potential’). The semiotic interpersonal and textual functions are more obviously social, but are inseparable in semiotic practice from the interpersonal.

  5. Language is constituted as ‘a discrete network of options’ (1978:113). The idea of systems and networks (systems organised as networks) proposed by Halliday before the ‘Network Society’ has applications to all aspects of Social Semiotics that are yet to be fully explored.

2. The roots of Social Semiotics

Social Semiotics (1988) was undoubtedly influenced by Halliday’s ideas. One of the authors, Gunther Kress, had studied with Halliday in the late 1960s. Bob Hodge and Kress had first collaborated in a project primarily concerned with verbal language, called ‘Critical Linguistics’ (Hodge and Kress 1993(1979)). Semiotically this was limited to verbal language, though like Halliday’s work it was already implicitly semiotic. It synthesized a range of schools of linguistics, including warring divisions within mainstream linguistics (Noam Chomsky and his followers, Halliday as a major alternative, Benjamin Lee Whorf’s anthropological linguistics) and socio-linguistics (William Labov, Basil Bernstein, R. Brown and A. Gilman, Dell Hymes, Harvey Sacks, J.L. Austin). It also incorporated theorists of language from outside the discipline of linguistics, such as Herbert Marcuse, Jean Piaget and Sigmund Freud. But crucially this project framed the field with a critical account of society derived from Marx.

Social Semiotics built on Halliday’s five principles, and added distinctive emphases:

  1. Semiotics is the minimal framework for the study of social meanings because there are complex patterns of similarities and differences across different codes, and because social meanings typically flow continually between different codes. Social meanings cannot be tracked only in one code, even in verbal language as the dominant one. The supposed dominance and autonomy of the verbal code is indeed an ideological assumption whose taken-for-granted truth needs to be questioned by social semiotics.

  2. The unitary object of social semiotics is constituted by a series of dialectics: between Halliday’s interpersonal and ideational functions, between ‘text’ (‘a string of messages which is ascribed a semiotic unity’ 1988:263) and ‘discourse’ (‘the social process in which texts are embedded’ 1988:6); and more generally between ‘semiosis’ (‘the processes and effects of the production and reproduction, reception and circulation of meaning in all forms’ 1988:261) and ‘mimesis’ (‘implying some version(s) of reality as a possible referent’ 1988:262).

  3. Power and solidarity are key dimensions of social structures and related meanings, inseparably related in social semiotic practice.

  4. Ideology, a key category in Marxism, is also central in social semiotic analysis, but inflected by social semiotic principles to become the idea of the ideological complex. Instead of the usual assumption that ideology is false consciousness, consistent with itself but misrepresenting reality on behalf of ideologues, the minimal unit of meaning in an ideological complex is its functional set of contradictions, motivated by the need for ideologues to balance issues of power and solidarity for their relations with those they are addressing.

  5. The relationship with reality, treated as a problem for semiotic theory in most forms of semiotics, is seen as constitutive in social semiotic practice. Reality-claims and their contestation are woven into every semiotic act, and determine their social effect. Systems, markers, traces and effects of ‘modality’ (‘the presumed relation of its mimetic content to a world of referents’ 1988:264) are therefore central objects of interest for social semiotic analysis.

  6. Transformations occur everywhere in social semiosis, in texts and systems of classification, as semiosic activity works over different versions of reality for many reasons, all of which have social origins and meanings. The concept of transformations (taken from Chomsky but transformed) is a crucial strategy for analysing the diachronic dimension (time, change) which in its different scales is part of every social semiotic fact, interacting inseparably with relations as they exist within any given time.

3. Social Semiotics and critical discourse analysis

‘Critical Discourse Analysis’ (often abbreviated to CDA) is often treated as distinct from Social Semiotics, and not strictly part of semiotics. Yet there are good reasons both conceptually and genealogically for seeing it as a branch of social semiotics. Both developed at more or less the same time from Critical Linguistics. Norman Fairclough first called his version of Critical Linguistics ‘Critical language studies’ (1989), then ‘Critical Discourse Analysis’ (1995), naming a field that has exploded over two decades, describing a research tool that proved attractive to a wide range of researchers.

‘CDA’ owes much of its appeal to the role it gives to ‘discourse’. This term gained currency through the work of Michel Foucault, perhaps the best known social theorist of his time. However, Foucault was a grand theorist rather than a local analyst, a highly intelligent commentator on a range of issues in the formation of modernity who did not develop or need a method as such, capable of analysing instances of discourse. ‘CDA’ supplied the missing ingredient, a method of analysing linguistic texts to complement Foucault’s theories and concepts.

There are many reasons why CDA ought to have situated itself in a semiotic framework. Power, its major focus of interest, acts through verbal discourse, but not in words alone. The limitation to verbal language ties the analytic hands of CDA. Of the main types of discourse it studies, media discourse, policy discourse, and interactional discourse, only policy is represented mainly in verbal discourse. Increasingly the media are multi-media forms, and interactions have always occurred in multi-semiotic spaces. Ethnography, a form of social semiotics without the name, drew on verbal discourse, collected through interviews, alongside objects and practices viewed through the semiotic tactic of ‘participant observation’. Classic ethnography tended to obscure power-relations, the forte of CDA (and also important for Social Semiotics), but the social semiotic practices of ‘postmodern ethnography’ (Clifford and Marcus 1986) is good social semiotics, aware of relations of power and solidarity in ethnography’s semiosic relationships.

The term ‘critical’ in both ‘Critical Linguistics’ and ‘CDA’ has an ambiguous legacy. On the one hand it creates a hostile relationship of analysts to objects of analysis, since it typically aims to expose the mechanisms of power in the semiotic transaction. In terms of social semiotics this is an advance on classic social science research, which masks this constitutive semiosic relationship under the ideology of ‘objectivity’. Yet semiosis is rarely the pure exercise of power. Relationships of solidarity usually co-exist, including kinds of ‘appreciative’ analysis. It is no accident that ‘critical’ has disappeared from the name ‘Social Semiotics’.

Foucault’s own formulations of discourse seem to take for granted the pre-eminence of verbal language, but they tend to have a capacious social semiotic space around them, which can be seen in his practice also. For instance, one of his most influential ideas has been his highly semiotic analysis of Bentham’s ‘Panopticon’ (1977), an image portraying the design for a gaol in which unseen warders view prisoners from a central tower. The disciplinary practices he analyses, represented in architectural and other spatial arrangements, are all realized through a multiplicity of sign systems, studied as part of social semiotics.

Foucault’s concept of the ‘discursive regime’ (1972) is both a powerful contribution to social semiotics, and in need of a social semiotic framework. A ‘discursive regime’ is an abstract social system which specifies who can speak and what they can speak about, in what circumstances. There are crucial social semiotic questions about who institutes these regimes and how, and what lies outside their scope. The problematic status of ‘reality’ in discourse analysis, which seemingly cannot appeal to ‘reality’ outside of discourse, can be resolved in a Social Semiotic framework, in which there are too many alternative semiotic modes accessing the different objects of discourse for any discursive regime to be fully successful as an instrument of control, and in which the complex mechanisms by which ‘reality’ is specified and controlled (‘modality’ systems) are themselves transparent and available to analysis.

Foucault was treated as a major influence, in effect a social semiotician, in Social Semiotics, as well playing an even more key role in CDA, but with the social semiotic links trimmed back. Given all these over-laps we can ask: does it matter whether Foucault is understood as more a discourse analyst than a social semiotician, or whether CDA, with all its many affinities with Social Semiotics and Critical Linguistics, is declared to be part of social semiotics, or set off against it? Both Social Semiotics and CDA would agree that it does matter, since the social mechanisms and effects of these processes of definition are central objects of analysis in each.

4. Multimodality

Two theorists in particular have worked on the institutional interface between Social Semiotics and CDA. Gunther Kress was a founding theorist of Social Semiotics, a student of Halliday, able to synthesize the different branches of Social Semiotics, with publications using Critical Linguistics that are evidently also CDA. Theo Van Leeuwen was a semiotic Hallidayan who applied Halliday’s ideas to technical aspects of a number of media, music, film and design, and continued to develop a Social Semiotic framework (2005). Together they wrote on ‘Multimodal Discourse’ (2001) in a theory which had a strong social semiotic base, yet used the term ‘discourse’.

Multimodality has two general strengths that have contributed to its popularity. Firstly it has demonstrated the important role played by the semiotic characteristics of communication even in one media, verbal discourse of print media. They showed the systematic role played by layout and design, within a print text and between print and graphic elements. They called these different semiotic channels ‘modes’, so that ‘multimodality’ signals the need for a semiotic analysis, not merely a mono-modal analysis as in discourse analysis (and critical linguistics). Secondly, this view of media texts as always multi-modal applies especially well to the new media, whose multimedia forms, structures and processes severely challenge older mono-modal forms of analysis.

The complex situation of Multimodality as viewed through social semiotic lenses illustrates many contradictory aspects of the case of social semiotics, and semiotics itself. The term ‘multimodal discourse’ declares an affiliation with CDA, though its practitioners typically do not set multimodality squarely within its scope. At the same time it transformationally deletes ‘semiotics’ and ‘social semiotics’ from the title, and downplays it in the description, even though the writers sufficiently declare that this is a development from social semiotics. In their place the key term is ‘multimodality’, which is incomprehensible in everyday discourse.

To begin to explain this we can adapt Foucault’s idea of a ‘discursive regime’, and suppose that there is an abstract entity in current academic discourse which is making ‘semiotics’ and ‘social semiotics’ almost unspeakable. This can be spoken about, and social semiotics taken seriously, in some privileged spaces, such as the present Encyclopedia. Yet (a socio-semiotic claim there is no space to demonstrate here) even this space has shrunken over the past few decades. This is a fact of discursive power, not a judgement on the adequacy of semiotics. In the same time both ‘discourse’ and ‘critical’ have become more speakable, to become the dominant carrier of social semiotics today. Yet sign systems have not simplified down to the single channel, verbal, but the contrary. The entanglement of signs has become ever more complex and pervasive, ever more inseparable from dominant social processes.

To cope with this situation any kind of exclusively verbal analysis is ever less capable of being critical. Yet, for the time being, Social Semiotics withers, and CDA flourishes.

A proposition like this can be asserted within CDA. However, it can only be examined within the framework, Social Semiotics, whose slow demise it deals with. Social semiotics can describe the multi-semiotic terrain in which this process is taking place, the effectivity of the many non-verbal codes, and the cost of their exclusion from the field of study.

In all this it is important to hold out for the importance of the general project of social semiotics, deploying the full range of semiotic insights, whether or not they are officially part of a meta-discipline of semiotics. Social semiotics on this scale is so amorphous and diverse it is almost impossible to capture in any agreed description or definition. But like God, as Voltaire famously said, if it didn’t exist we would have to invent it.


Clifford, J and Marcus, G 1986 Writing culture. Berkeley: University of California Press

Fairclough, N 1989 Language and power. London: Longmans.

Fairclough, N 1995 Critical discourse analysis. London: Longmans.

Foucault, M 1972 ‘The order of discourse’. In Archaeology of knowledge. London: Tavistock.

Foucault, M 1977 Discipline and punish. Harmondsworth: Penguin.

Halliday, M 1978 Language as social semiotic. London: Edward Arnold.

Hodge, B and Kress, G 1988 Social Semiotics Cambridge: Polity

Hodge, B and Kress, G 1993 Language as Ideology, 2 nd ed.. London: Routledge

Kress, G and Van Leeuwen, T 2001 Multimodal discourse. London: Edward Arnold.

Van Leeuwen, T 2005 Introducing social semiotics. London: Routledge.


Bob Hodge