Interview with Gunther Kress

This issue of Designs for Learning features an interview with professor Gunther Kress. Our intention here is to give some further insights regarding interests and influences that form a background to his theoretical work on social semiotics, multimodality and learning theory.

in transformational grammar.It assumed that you would generate a syntactic structure -speech and writing weren´t distinguished -and a kind of semantics was grafted on to that.And that seemed to me implausible.Halliday's model by contrast started immediately from the assumption that there is a system of meaning-choices from which you select.The choices were seen (and laid out) as a complex interrelated meaning-potential, offering choices about representing social relation between me and you; about 'goings on' in the world; and the means which would make the whole thing a text.That started with meaning and went from there to form; to me it seemed an absolutely plausible way of thinking about language and meaning.It didn´t offer an articulated sense of speakers located in social structures; but it did have a clear notion of "I´m in the social world, I want to do certain things, here is a resource for what I want to do".And so I became a Hallidayan.It seemed plausible to me, a strong sense of "yes, this was the way to think about it." I did a post-graduate degree with Michael Halliday; and enrolled to do a PhD with him, on 'theme'.Unfortunately, he left London; I changed jobs, going from the university of Kent to the university of East Anglia in Norwich.That's where I met up with Bob Hodge, who arrived a year after I did.We taught a regular joint seminar in literature and linguistics, within a kind of overarching theoretical frame of Marx and Freud and Whorf and Hallidayand to a lesser extent Chomsky through a realist notion of transformation.Our conception of Marxism was a version of social and economic base producing superstructural categories such as law, literature; and if you translated that to language as a superstructural category, you would see that the shape of the language was related to the social and economic base.So English, for instance, has particular kinds of possessive forms; these might be an effect of an orientation to 'possession' peculiar to English society in some way.That teaching led to what became Language as Ideology (Hodge & Kress, 1979/1993).We were quite clear that we wanted our theorizing to have a social effect and so the notion of critical linguistics emerged.That was about making linguistics socially responsive; to produce a linguistics with social effects, through revealing the structures of power in language use.That was the project of critical linguistics; which had its part in the development of the project of critical discourse analysis: to try and change things by revealing how power worked in representation.
Bob Hodge went to Australia in 1976; but just before he went we said "well we've done this thing on language; but really, meaning rests in many more things than language" and we said "we must do something more.We'll look at all these other ways in which meaning is made."Well, it was difficult to work between England and Australia; but in 1978 I too returned to Australia.He was in Perth and I moved to Adelaide and now the distance wasn't 12000 miles This issue of Designs for Learning features an interview with professor Gunther Kress.Our intention here is to give some further insights regarding interests and influences that form a background to his theoretical work on social semiotics, multimodality and learning theory.
Gunther Kress is professor of semiotics at the Institute of Education, University of London.He has been prominent in the development of social semiotics and multimodality since the 1970's and has written extensively within this field.Among his latter publications are Literacy in the New Media Age (2003), Multimodal Discourse (with Theo van Leeuwen, 2001) and Multimodal teaching and learning: the rhetorics of the science classroom (with Carey Jewitt et al., 2001).In 2009 his new book Multimodality.A social semiotic approach to communication will be published.
FL: Since many of our readers are acquainted with, and interested in, your work within social semiotics and multimodality, it would be interesting to hear a little about your background -where you started theoretically and academically and what led you further towards the development of these theories.It would also be interesting to hear a little about the relationship between your ideas and other things that circulated at that time or that you had immediately around you.

GK:
Well, my first degree was in English literature and as a part of that I did a course on language.In Australia that was a four year degree, where, in the fourth year, you would specialize.I specialized in language which, among other things, meant getting introduced to transformational grammar.So my teacher and I read Syntactic Structures (Chomsky, 1957), Chomsky's first book, line by line literally, sentence by sentence.I read papers which were coming out -mimeographs!! -from MIT.And, feeling not particularly happy with what I thought was a lack of rigour in literary studies and thinking that the study of linguistics would supply the rigour needed in literary theory, I decided to do linguistics.That coincided with a move, after finishing my degree, from Australia to (Germany for one year and) England, to a job as a Research Fellow in Applied Linguistics, at the University of Kent.And after a short while I started a postgraduate degree, part-time, with Michael Halliday, commuting up to London twice a week.
I had been quite unhappy about the way syntax and meaning was separated DOI:http://dx.doi.org/10.16993/dfl.14

Interview with Gunther Kress
By fredrik lindstrand but only 3000.Then I moved to Sydney.We kept on meeting and working intermittently and we thought "what should this thing be called?"We played around with possible titles but in the end it was Halliday's phrase of (language as) 'a social semiotic' which seemed best for our purposes (Halliday, 1978).We felt we wanted to explore, to describe all that was part of a social semiotic.And from Halliday's conception -that there is somebody who chooses from the meaning potential -we took the (there relatively implicit) notion of the significant action of the agent who makes choices from the meaning potential: the agency which insists that you make meaning from existing resources.We took on the idea that meaning is made -expressed, for instance, in another of his book-titles Learning how to mean (Halliday, 1975).
In East Anglia I taught transformational grammar for the seven years I was there.In the late sixties and early seventies there was a big debate whether transformations were 'meaning preserving' or not.Chomsky had said that deep structure has all the meanings of the utterance.That then is transformed, though these transformations have no effect on meaning.So, in a sense, he said the passive sentence has the same meaning as the active sentence.That seemed implausible to both of us.If you do something then something changes; there is an effect.We took the concept of transformations but as operations which had effects on meaning.And so we married Halliday with Chomsky; which wasn't at all permitted at that time.But in Language as Ideology we did that; for instance, we introduced the notion of 'nominalisation' from Chomsky's work -the process that changes a simple sentence into a nominal; and we attempted to explore what meaning consequences that had.We did the same for many transformations -passivization, agent-deletion; relative clause formation; etc.And in each case we were interested in the ideological effects: whose power was at work, for whose benefit.
When we wrote Social Semiotics (Hodge & Kress, 1988) we took the notion of agency, power and representation developed in the theory of Language as Ideology as the agency of anyone who makes any kind of sign.But this was more than just a choosing from existing resources; it was actively making signs.The person who chooses to use the transformations of passivization and agent-deletion to turn 'The police shot the demonstrators' into 'the demonstrators were shot by the police' and then into 'the demonstrators were shot' has made a sign.I had written something in 1977 on the non-arbitrariness of signs, but it now became the idea that signs are made and motivated; so agency was in the making of signs.The sign and the meanings that a sign-maker makes are an expression of their disposition, habitus, identity -of their interest.
We applied that understanding to lots of things -sculptures, photographs, children's drawings, pages from books, newspapers and so on.It was a social semiotics.Unlike existing semiotics which says signs are used -a notion taken over from Saussure -we said signs are made and signs, therefore, are always newly made.And that's more or less where we left it.We also developed ideas of how all this hangs together as text; we developed the term logonomic system which links genres with discourses, provides rules for 'reading' texts, and so on.In a sense that left a still unfinished project, namely to ask "how do we take this further?How is sculpture like a spoken utterance?" In the meantime I continued teaching linguistics, including Hallidayan linguistics.I was particularly interested in his (implicit) notion of the materiality of resources and their social shaping -maybe because I had become interested in children's speech (as all linguists had to be, then) and then the transition for them, to writing (which had not really been a focus for linguistics).It struck me that Halliday's work on speech showed that we really needed to distinguish between writing and speaking; it was his insistence on the materiality of the voice, the physiology of breathing, which showed the relation of material, social working and culture.In breathing out, you exhale, say, five litres of air.You can measure the time this lasts, and establish a kind of average.The expulsion of air, rather than the taking in of air, is the useful bit for speech, in most languages.And with that useful bit of air you do the talking.The exhaled breath becomes a linguistic unit, a unit marked by intonation; and it becomes a semiotic unit, a unit that carries information: an information unit.Intonation can be used to distribute information in different ways.You can see how the natural phenomenon, the expelling of breath, becomes the semiotic entity of information unit, through the use of pitch variation, also a natural phenomenon shaped by social use, as intonation.
So the notion of the materiality of the resource and its social shaping into semiotic use in culture is there -somewhat implicitly -in Halliday's work on speech.That kind of stayed with me and it pointed me towards the need to distinguish between different modes.Because writing, materially, hasn't got breath, it hasn't got intonation and all the things that rely on sound in speech.So if we want to mean broadly the same or similar things in speech and writing, the meaning has to be made by different resources, in writing, word order, for instance.So that was for me the start of thinking about the notion of materiality and mode, though I did not use those terms then.In Learning to Write (Kress, 1982(Kress, /1994)), I thought about the real difficulties that children have in learning to write as analogous to the difficulties of someone learning another language; now I would express that difficulty in terms of the different affordances of the modes of speech and writing: some things similar and many things quite different.
Because I had looked at lots of things that children had written at early stages of learning to write and noticed -difficult not to notice -that they are always accompanied by drawings; and then you can say, because you're a 'linguist', "drawings don't interest me".You wouldn't say that as a father.As a budding semiotician you might say "well there is something here that I can't account for; let me think how I could".And so that became a nagging issue for me.In Social Semiotics we didn't talk about modes, so that questions about the semiotic regularities of these other means of making meaning were not in the forefront of our thinking.
Social Semiotics, the book, was more or less written, by 1986.In 1986, Theo van Leeuwen and I decided that we would set aside one morning a week to start thinking about images.He was living just a few hundred yards from where we lived, in an inner city suburb in Sydney.At seven in the morning he would come to the house and we'd drive off to a beach on the harbour, we had a swim and on the way back we'd pick up some patisserie and have breakfast in our small inner city garden, with my partner, before she went off to work; and we then turned to our work.I can remember the moment we started, sitting in the garden, saying "what do we need?" and "let's look at some women's magazines".I went into the house and got a copy of Australian Women's Weekly.Literally that.It was a method Bob Hodge and I had developed in our joint teaching: take any book on the course -it was a course on 17th century literature -open it and go for it; a make or break method: either the theory works or it doesn't.We had a class of about 20 people and there we all were.We'd take a paragraph and we'd read it out -this is before photo-copying -and say "what can we say about this from an ideological perspective?" Theo and I did that with our materials.So a copy of Australian Woman's Weekly was our beginning.We sort of worked with that for a bit and then some weeks later I went upstairs and got a children's book; a Ladybird book: what can we say about that?And so we worked our way into images.Eventually that became Reading Images (Kress & van Leeuwen, 1990) book, and these early examples are in that book.We wanted to show that images have regularities, but we strongly felt that we could not simply use a theory and descriptions developed to describe a one mode for something that makes use of a quite different material; and transfer categories developed for one material and the work done with that to a totally different material and the work done with that.That was an unusual approach.Michael O'Toole, who had also come with a Hallidayan frame took the former route in his The language of displayed art (O'Toole, 1992).He had taken the Hallidayan theory and the descriptive terms developed for language and used it for the description of image or architecture or sculpture.I think it is important to realize that most people with a Hallidayan ancestry -Kay O'Halloran for instance, now working in Singapore, use Halliday's linguistic theory and its terms, whereas Theo and I, and Bob and I before, used the semiotic aspects of Halliday's theory.That's a bit of the background.

64
FL: To stick with the background for another while and use the term motivation, can you see anything from your personal life that influenced the direction you headed in theoretically?GK: Well, in Learning to Write, I used the theory of Language as Ideology, which added, as I said, some central Chomskian terms into a basically Hallidayan framework.The Chomskian theme I took came out of his theory of language development -you know, his notion that we have an innate disposition to language and an innate structure which is developed by each individual in relation to the mess of language.So the child comes with its innate structures, encounters the mess of ordinary social life and of language and attempts to make regularities And so at each stage when you look at a child's speech, the child has developed structures which are regular -a kind of a grammar, an account of regularities of how she or he uses language; these change as the child continually realises that this regularity doesn't account for these other bits.For me what was crucial in this is that it says, here, actually, is somebody who attempts to theorize about the complexities of the world.And, really, this is what I took from Chomsky: not the innateness bit, but saying "okay, we cannot ignore the efforts of somebody who is a practicing theoretician, constantly revising their accounts, the universe made regular by the grammar that this person developed."So that was one thing.And on the other hand we have the Hallidayan chooser, the person who is agentively making meaning.And you put these together and then you have, say, me looking at my own children and thinking "Are these little people, making their meanings, struggling somehow to become fully competent, or are they, both in a Chomskian and Hallidayan sense -even though very differently seen -doing what people always do, finding regularities, making their meanings, but always with limited resources.In neither the Chomskian nor the Hallidayan account is there a sense of deficiency; in each, makers of meaning are seen as competent in the use of the resources they have -whether as adult or child.And there was the Hallidayan theme, which is, these are the resources and with these resources we shape the world of representation.And that's always with me.People, whether 3 or 30 years old, are intelligent makers of representations through which they represent how they see the world, whether in the way they make sentences or in the way they shape meanings in other modes… yeah?FL: Yeah.GK: I think an ordinary sense of respect demands that we treat what children do as serious attempts, in the way we treat ought to treat all our attempts as se-rious, with whatever resources we have.We make our sense of the world.So it's about dignity assigned to that making of sense.Human dignity, for me.As indeed is Noam Chomsky's political project about social dignity as equality.And then, with that, looking at what children do and therefore never dismissing it but making it my job to attempt to understand what was the regularity and what was the vision of the world or the image of the world.And that's for me the motivation.And that was in Language and Control (Fowler, Hodge, Kress & Trew, 1979) and extended to the working out of power in all kinds of interactions.To see how one could kind of equalize power difference from that position.So that's my motivation.

FL:
In recent years you have worked with several projects focused on school and educational settings of various kinds.How would you formulate this in relation to those kinds of settings and situations?GK: I would formulate it in relation to an understanding of learning.Because I think it translates fairly directly.In thinking of the problems of school, in many places, again, you can either think that children have become stupid or you can say that children actually are as always fiercely interested and of good will to learn.And that they attempt to make sense of what is presented.What sense they make of what is presented to them, in school or out, then becomes a matter of the resources they have, their interest and the attention that they pay to the world presented to them.The sense they make in school of things the school presents isn't necessarily the sense of the holder of power.So then the question is, do you want to say that they have been successful in conforming to authority or should we attempt to understand their sense of what is presented.Is our interest in conformity to power or in understanding their interests, their principles of attention and engagement?These lead to different notions of learning, namely, how have they reshaped or transformed the materials the school presented to them, questions about their transforming of these materials; the need to understand the principles they have applied in their sense-making and learning.Then, on the basis of our understanding of their principles and their interests, how they understand, maybe then take another step and ask how then can we present the things which we think are valuable about our culture?Because I do think that we ought to transmit those things to the next generation but in ways that allows them to make their sense of them, ways that connect with their lives.I don't think we should expect children to invent every kind of wheel on every occasion.But the means of getting there are not by saying "I am telling you, this is what you must be interested in" or "you have not conformed to my authority; you need to be excluded", but by saying "I must attempt to understand the principles you bring to this 66 issue.That understanding can take me a step nearer to bring you towards seeing value in the kinds of things that I would like you to understand."And so, it's around theories of learning that the question of power and communication as teaching and learning appears.Of course this is policed by notions and practices of assessment; and there I think that other kinds of criteria still apply.

FL:
If we look at theories of learning then, what would you say are the main differences between a social semiotic theory of learning and other socially oriented theoretical positions around learning that have become "mainstream" today?GK: For me the question is "How do you get evidence of learning?"And the usual practice is that you measure the distance between what authority says you should learn and what you seem to be able to show as your learning and that is taken as evidence of learning.Or, you ask somebody "what have you learned?" Or, and that is the approach I take, you look at what someone has done and take that as a sign of learning.Put in Peircian terms, you look at the process of semiosis in which the sign made by someone for themselves in their engagement with a specific bit of the world leads to their interpretation, to their interpretant.That interpretant is then the taking off point for the next sign, which is, in my terms a sign of learning.
Social semiotics is close to Peirce in the sense that you process a form from the sign you receive -an interpretant -based on your position in the world, your resources, your interest at this moment.And then the new sign made on the basis of the interpretant kind of gives you an understanding of who this person is and why she or he made the sign like that.So, the notion of signs of learning is an attempt to say 'this sign now, that I can see here, is actually evidence of the process of engagement and transformation of the learner and then sign-maker, you know.An approach of "look at what people do" as evidence of a change in their resources, the result of learning, seems better to me, even if not sufficient, especially in a context where signs are regarded as made on the basis of the sign-maker's interest.

FL:
So what would be needed in order to make it sufficient?How can we complete the picture?GK: Well, I think ethnographers or conversation analyst are not social semioticians.Conversation analysts document what happens in conversation.There isn't a notion of agency.And I think ethnographers, of course every ethnographer does different kinds of things, but in a sense they all ask "Let's see what's going on".In social semiotics there is a sense of, well here is a kind of statement, what was the interest of the sign-maker to make the sign as it is.We have a motivated conjunction of form and meaning, to form a new meaning, and it's that focus on agency and on interest which is the difference between say conversation analysis which asks "what are the mechanisms here?"or Ethnography which says "what's going on?"So that's where I would see the difference and a possible complementarity: the question "what is going on" is, needed, I think, to complement a social semiotic approach, by giving further insight into interest, motivation and the shape of the environments in which signs are made.That comes back, really, to an ethical position.To say that by understanding the conditions in which people make meaning we assert the dignity of the person who does semiotic work.
FL: That seems to be an important part -something central that runs through the perspective itself.It seems like a very humane way of looking at what people do and so on.
GK: I don't know whether it's humane but its absolutely important to me.I think of the dignity of work, and when I walk past somebody who sweeps the street, when I go to work in the morning, I have an entire empathy with that person.I see the dignity of what that person is doing and that speaks to me.What is this person is being paid, what am I being paid?I see the discrepancy, I see a discrepancy in recognition of dignity.I see a discrepancy of social evaluation and in finding means for recognizing and changing that, there lies my political project.

FL:
The political project seems to be very closely connected to the theoretical project, in a way.Perhaps that could be seen as another way of talking about motivation in relation to social semiotics as a theory?GK: For me? FL: Yes.

GK:
Well as you know, social semiotics has many variations already and it would be wrong to assume that everybody who calls herself or himself shares that motivation.Very few people that I have spoken with foreground or share that view, and why should they?FL: Share your view?68 GK: Yeah.Well, it's a political stance.But I think it would be quite right to ask about everybody's motivation.And important.

FL:
The issue of dignity is also related to the notion of interest, as you mentioned before.Can you give a hint of how to define the social semiotic notion of interest in relation to how the term is used in everyday language?GK: Well, when you look at a sign, you see that not everything about the phenomenon that is being represented, seemingly, is represented.Something acts to select, to say "this is what I want to represent, this seems criterial for me at the moment."So I think you need some sort of label which names the process or the principle behind that process of selection, which organises the selections we make at a particular moment, kind of to say "this is what this phenomena is for me at this moment."There are different kinds of principles active and I think when you look at any representation what you see there is always a partial representation, selections have been made.And so you can ask about what the principles of selection were.And then you can ask what motivates the selection.And because I want to have a social theory, I say that a person is socially formed in their history, acts in a socially specific moment, uses socially made resources and all of this kind of comes into the sign-making.It is focused by the prompts to which a person responds; and that gives a particular sort of framing and so, because I don't want to go to psychoanalytic terms, I have chosen interest.And so it is not more, really, than that.Needing a name for the principles which organise the process of the selection about the phenomena I wish to represent, call it whatever.I felt that interest links strongly to agency.

FL:
Sure, but it's obviously motivated.
GK: Yes, because I do think it is interest.It is my position in the world now.You provided me with a prompt.I have chosen to respond and sort of attend to your prompt.I make selections from the prompt.I can't respond to all of your prompt.So I make these selections.What is it that causes my selections if not some or many things in our social history.But not all of our social history.Aspects of our social history condensed at this instant, in the environment of the time, by the social relations active at that moment, you know.Attempting to find a word, I am happy with interest.The question is, is there a plausibility in what lies behind?And if then somebody says "that's implausible, and its implausible for these reasons", then I need to reconsider.

FL:
To briefly touch upon some other central notions within the multimodal and social semiotic framework, I sometimes find it difficult to distinguish bet-ween materials, modes and media in relation to "new" and/or transformed fields of semiotic work.How do I approach this?GK: I think you need to think in terms of 'who is the community?','what is the community?' 'which are the questions?'.'What power is active?' 'When or how does a dialect become a language, or not?'So it's that kind of questions.It's a social question, and a question of power, I think.And I think that Theo and I were right, probably, when we said that the older notion of saying "this is and this isn't 'language' or 'music', this is a mode, no that is not a mode" is much better seen in terms of what a community chooses to regard as a mode, because it is developed to do certain kinds of things in that community which are sufficient for the purposes that it's using the resource for.And, you know, the difference between medium and mode is difficult, for similar reasons; because sound is medium and it yet it gets made into different kinds of modes.And stone is also medium and can be made into mode.So I think it's about shifting frames rather than asking about the older kinds of strict boundaries -that this community does this with these things and, you know that this is medium and this is not.Something may be being used as medium one moment and appear as mode the next -say, font, for instance.Typography might regard typeface as mode.Can you make meaning with type-face?And can you make all kinds of meanings -interpersonal, ideational and textual?I don't know, though I think you can.It is not impossible to work it out.I'm sure that if multimodality or this social semiotic take on it continues, some of these things will become clearer or be made clearer.Of course, the moment you do that you would also move into the area of law-making, and then the social changes in any case, and maybe the social then will move back to much more rigid boundaries and frames.
FL: To finish off, you have said that your theories are in continual development.Have there been any major changes to social semiotics since your work with Bob Hodge?Since then you developed the theory of multimodality together with Theo van Leeuwen and some notions, like logonomic system for example, seem to have disappeared.

GK:
All the things we've talked about are about social semiotics and not all of those are in that book.And you're right, logonomic system is something I haven't taken up much.Largely because I have not continued with description or analysis of text a lot.Because my interests moved more to an understanding of mode rather than a continued interest in text or the social aspects of text.If I were to return to text and its constitution and uses, I am sure I would actually want to have a much clearer and stronger development of logonomic 70 rules, because you do need to account for how genre and discourse for instance come together and in what ways they interact with modal choices and interact mutually and what kind of recognition devices there are in an utterance -you know, the things we pointed to in that book, that allow you to say "this is a joke" or "this is not a joke".So my interests have moved away from that aspect.And in fact, one of the things I am worrying about in the book I'm doing at the moment is that it says next to nothing about text.Because representation always happen as text.You don't see a mode come by itself.And I think, well, Bob's interests have moved in different ways, but I don't think he would find anything in this that he wouldn't recognize or be totally worried about.