Paper prepared for The Discursive Construction of Knowledge Conference. University of Adelaide, Feb. 21st- 25th, 1994.


John Shotter,
Department of Communication,
University of New Hampshire,
Durham, NH 03824-3586,

"People know what they do; they frequently know why they do what they do; but what they don't know is what what they do does" (Foucault, in Dreyfus and Rabinow, 1986, p.187).

At conferences of this kind, we find talk of theory and of theoretical investigations very familiar; we know how to make sense of and to evaluate talk of that kind. However, following Wittgenstein, I think that there is something very wrong in us attempting to grasp our basic nature as linguistic beings through the activity of us each, as individuals, constructing theories and then debating them - the social and historical conditions making such an activity possible remain unacknowledged. Indeed, as Billig, Condor, Edwards, Gane, Middleton, and Radley (1988) argue in their book, Ideological Dilemmas. "even the main forms of 'post- structuralist' theory have, in their attacks on dominant methodological tendencies, only produced new theoretical monologyes, not displaced them with a perspective in which [a] truly dialogic principle and its necessary conditions are a prime concern" (pp.149-150, my emphasis). What we need, I want to claim, is not knowledge in the form of theoretical representations, but of a very different, much more practical kind. My concern today, then, is with the conditions, the relations between us, that might make possible a more dialogical and involved, less monological and distanced, stance toward our construction of knowledge.

Thus, a part of what I want to explore, is talk of a very different kind to theoretical talk, talk to do with a very different kind of knowing: that which 'floats' around in an uncertain way within the everyday conversational background to our more institutional and disciplinary lives, on the boundaries of, or wnes in between, our separate disciplines and orderly discourses. It is a special kind of knowing that -although it has been more properly recognized and identified in the past -has in more recent times been forgotten. I shall call it a knowing of the third kind. For: i) It is not theoretical knowledge (a "knowing-that" in Ryle's 1949, terminology) -for it is knowledge that is only present to us in our everyday social practices; however, ii) it is not simply a technical knowledge of a skill or craft (a "knowing-how") either -for it is a joint kind of knowledge, a knowledge-held-in-common with others, and judged by them in the process of its use. iii) It is its own kind of knowledge, sui generis, that cannot be reduced to either of the other two.

Rather than to do with us relating ourselves to our physical surroundings, it is primarily to do with us - even when all alone - relating ourselves to each other, with us coordinating our actions together as members of a community. Thus, it is a kind of knowledge one has only from within relationships with others, whether the relationship is actual or imagined.

Unlike the other two forms of knowledge, in which we as individuals stand over against our surroundings and experience them as separate from ourselves, in this third kind of knowing we experience ourselves as involved 'in' a 'something', and we are affected by, or responsive to, what goes on within that 'something'. We find that 'it' seems to 'call out' or to 'demand' various activities of us; we are as much a part of it as it is of us. We might call the kind of knowing involved, a "knowing-from-within," for it determines what at anyone moment we anticipate or expect will happen next from within any situation we are in: It is to do with i) not just what will surprise us and what we and others will merely find familiar, ii) but also what we and they will find disgusting, frightening, iii) as well as delightful and want to celebrate, what we all will count as objective and what subjective, what real and what unreal, what ordinary and what extraordinary, and so on. While the other two forms of knowledge can be said to be disciplined and orderly, and sustained by systematic discourses, conversational knowing seems by contrast to be disorderly and undisciplined.

Richard Bernstein (1983) has called it, "practical-moral knowledge," and related it to Aristotle's notion of phronesis; it is, as Bernstein puts it (following Gadamer, 1975), "knowledge not detached from our being but determinative of what we are" (Bernstein, 1992, p.25) - where who we are must, of course, accord with ways of being others judge as being morally acceptable. In being continuous with, and determinative of, who and what we are, rather than 'in our minds', it is more properly called embodied knowledge. Indeed, it is not unrelated, of course, to what we call our common sense: for it is to do with the 'shape' of the (socially sharable) feelings of anticipation and expectation that we have at anyone moment in time in a social situation, the sense of the possible connections between it and other such situations.

It is the nature of this kind of sensuous, embodied knowledge - and the moment by moment task we face of making a good use of it in bridging 'gaps' between our utterances as we connect them to those of others in our conversational activities - that I want to explore here today. However, I cannot emphasize enough its strange nature, the difficulty involved in focusing upon its workings, the elusiveness of its existence. For the other kinds of knowledge familiar to us - theoretical "knowing-that", and technical "knowing-how" - are manifested in people relating themselves to their surroundings as individuals. It can thus be claimed that these forms of knowledge exists 'in their heads'. As a result, these forms can be theorized and studied as objects, as systems. Knowing-from-within cannot exist as an object, system, or framework; it only ever makes its appearance between us, in the process of our talk. Our 'ways' of talking' do not exist as isolable, temporally constant 'things'; they cannot be arranged and rearranged like building blocks in external relations to each other. They can, perhaps, be thought of as like 'implements' through which we can make contact with each other and our circumstances, like blind people make contact with their surroundings through their sticks and through echos they hear. But if they are spoken of in this way, it can only be as relational things, as 'things' that owe their nature at any moment to their internal or intentional relations to the rest of our living activities at that moment -that is, to their relations both to our earlier activities, and, to what we expect our activities to become. "Our talk," says Wittgenstein (1969), "gets its meaning from the rest of our proceedings" (no.225). How might we study this really peculiar kind of knowledge?

Well, as many of you know, a new and growing movement is afoot in the human sciences and humanities, known as social constructionism (Berger and Luckman, 1966; Coulter, 1979; Gergen, 1985, 1991; Harré, 1983, 1986; Shotter, 1975, 1984, 1993a&b). And what I want to explore with you today, is a version of it - influenced very much by Wittgenstein, Vygotsky, Voloshinov, and Bakhtin - a version that focuses upon the operation of this third kind of knowing in the conversational background to our everyday social lives together. In particular, I want to outline i) a site or focus for its investigation; as well as both ii) a basis (yes - a foundation!) in terms of which we might evaluate claims as to its nature, and iii) a method for its study.

The basis, I shall claim, can be found in the different sensibilities associated with the different 'realities' or 'forms of life' we establish and sustain between ourselves in our everyday conversation itself. Thus, not just any 'way' of talking will do. For, it is in our use of words that we arouse (in others and in ourselves) certain feelings of anticipation and expectation, a sense as to the possible nature of our future conduct - how we will relate what we do both to the others around us, and, to the rest of our circumstances. It is this sense that 'shapes' how it is felt appropriate to respond. And what Wittgenstein realized was, that although we cannot say what these feelings of tendency, of expectation and anticipation, 'are', to the extent that they do shape our conduct - what we do and what we say - then they are shown or expressed in the temporal unfolding of our conduct in quite precise ways. In relating ourselves both to our own circumstances, and, to the others around us, we 'show' the 'movement' of our minds (so to speak), in the pitch, pacing, pausing, and intonation of our speech. And if, as Voloshinov (1986) puts it, "meaning only belongs to a word in its position between speakers [at the moment of its utterance)" (p.102), then the tone in which it is uttered is, for instance, a part of the constructing of the relation between speaker and listener: whether the relation demands submission, invites collaboration, requests refutation ('please say I'm wrong), etc. It also expresses our relation to our own position, our confidence, happiness, uncertainty, and so on. And others - although they may not in any way be conscious of the fact - sense the tendencies toward which a speaker's words gesture.

This 'gestural meaning' (to use a term of Merleau-Ponty's (1962)) of one's words is shown in the fact that such and such 'feels' an appropriate thing to say, while so and so evokes feelings of surprise and awkwardness. For instance, saying: "From what he says, that seems to be his intention, but I doubt it, " raises no problems; however, if you say: "From what! say, that seems to be my intention, but I doubt it," is decidedly odd We find such an utterance senseless. We do not know how to respond to it, how to anticipate the behavior of the person who says such a thing, how to coordinate our actions in with theirs. If some one says to us: "I really mean every word I say, but please don't take me seriously." The anticipations raised by the first part of the utterance are dashed by the second. It is its logical grammar, as Wittgenstein calls it, that is all wrong. For, in knowing of the third kind, it is not our relations as already constructed individuals to our already constructed 'worlds' that is at issue, but the character of the constructing that is to 'go on' between us - the kinds of relationships that will be permitted to develop further between us, and those that will be refused. This is how our claims will be judged: not in terms of truth or falsity in relation to already existing facts, but in terms of the forms of life their speaking anticipates or projects, and whether that form of life will be allowed by the other around us.

Thus, as Wittgenstein (1969) says: "Giving grounds... comes to an end; - but the end is not certain propositions striking us immediately as true, i.e., it is not a kind of seeing on our part; it is our acting, which lies at the bottom of the language-game" (no.204). In other words, for us, our historically developing and developed, linguistically structured and sustained forms of life, are foundational - they are not only determinative of who and what we are, and what counts for us as our world or worlds, but also, of everything else that counts for us in some way. Thus, we cannot turn them around into an object of knowledge, something about which we can have a theory and explain - for, it is only from within such forms of linguistically created and sustained life that theorizing and explaining are possible for us as intelligible activities. Just as Vygotsky (1978) says about the developing child - that "the child begins to perceive the world not only through his [or her] eyes but through his [or her] speech" (p.32) - so also for us: we perceive and act through our different ways of talking; they are formative of the different circumstances in which we have our different ways of being.

To turn now to the question of method: If we cannot work in terms of theoretical monologues, in terms of representations, what can we do? Well, Vygotsky's comment above gives us a clue: we can use a certain kind of talk that will work, in Vygotsky's terms, as a "psychological instrument," as a means or tool through which to influence both an other's, and one's own, behavior. In terms of mind-talk, we might say that such talk, in presenting the 'movement' of one's mind in one's speaking, can also work to 'move the mind' of an other. But putting it in more 'embodied' terms, it is perhaps better to say, in Mead's (1934) terms, that in any living dialogue, "we have this interplay going on with the gestures [or the gestural meaning of words] serving their functions, calling out the responses of the others, these responses becoming themselves stimuli for readjustment, until the final social act itself can be carried out" (p.44, my emphases). Indeed, having mentioned Mead, it is worth also quoting another remark of his upon the responsive nature of practical meaning: "The mechanism of meaning is thus present in the social act before the emergence or consciousness or awareness of meaning occurs. The act or adjustive response of the second organism gives to the gesture of the first the meaning which it has" (pp.77-78). In other words, it is in the actual, practical interplay of voices in an everyday concrete circumstance - not in the play of signifiers within an abstract system - that practical meanings are made.

They are made - and this is the connection between these comments of Mead's and the question of method - by the way in which an utterance of a second person is not only 'linked' or 'connected' as a response to an utterance of a first, but also 'linked' or 'connected' in some way with the circumstances of its utterance. And, to repeat my Wittgensteinian formulation, although we cannot say what these links and connections are, to the extent that they shape our conduct, they are shown in its temporal unfolding, its 'movement', in quite precise ways. Thus it is the gestural meaning of our words-in-their-speaking that work to construct our forms of life: they work practically, both to shape our conduct, and, to draw our attention to connections between aspects of our circumstances that would otherwise go unnoticed. It is this that provides us with a method. For the self-same ways of talking that serve to draw our attention ordinarily to the connections between our utterances and their circumstances, can be used extraordinarily, to draw our attention to how we do in fact make such connections.

Thus what is at stake here, is not so much the grasp by isolated individuals, of an inner 'mental picture' of a state of affairs, but the actual, practical connections between aspects of our own communicative activities - influences that are present and at work in 'shaping' what we say in a particular circumstance. Lacking the appropriate sensibility to notice them, we feel that they require explanation in terms of hidden factors, already existing either within us somewhere, or, within our circumstances - hence, our tendency to seek theories. But, suggests Wittgenstein (1953): "The aspects of things that are most important for us are hidden because of their simplicity and familiarity. (One is unable to notice something -because it is always before one's eyes.) The real foundations of [a person's] enquiry do not strike [them] at all. Unless that fact has at some time struck [them] ..." (no.129). In other words, the meaning of a spoken word -everything important to its practical-moral meaning for us: its demeaning, insulting nature; its demand to be answered; its tangled, puzzling meaning - is not to be found, either in a mental representation in a person's head, nor within a system of signifiers or conventions of meaning, but in its particular use in a particular context.

In other words, we do not need theories for there is nothing to be explained, because - there is nothing hidden or concealed. "How do sentences do it?" Wittgenstein (1953) asks rhetorically. And he answers: "Don't you know? For nothing is hidden." (no.435). For, in his view, words in their speaking are just different 'means' or 'devices' for use in the making of meanings, with different words (like different tools) making available a range of different possible uses. Where everything of importance to us consists in their actual, unique use, in a circumstance. And it is there, in the particular circumstances of their use, that their meaning is to be found. It is to these circumstances that we must attend.

This helps us, I think, identify the crucial site upon which such investigations should focus. They must focus upon that "interactive moment" when a gap or a space is opened up by a change in speaking subjects, when one embodied person stops speaking and another responds to them - the place where what elsewhere I have called "joint action" occurs (Shotter, 1984, 1993a&b). For no matter how mechanical, rule-governed, or systematic the speech of each may be while speaking, there are no rules or principles for bridging that gap: a uniquely creative response is required. Where the response produced will be influenced by the factors at work at that moment, in that space, and the speaker's response to them will be manifested in the 'shape', the 'style' and the 'tone' of their utterance. A listener, perhaps, sensing it as unjustifiably angry, wimpishly feeble, or cunningly misleading, rather than simply fitted appropriately to its circumstances, replies to it accordingly. As Volosinov (1986) puts it, "in point of fact, word is a two-sided act. It is determined equally by whose word it is and for whom it is meant... Each and every word expresses the 'one' in relation to the 'other'... A word is territory shared both by addresser and addressee, by speaker and his [or her] interlocutor" (p.86).

The focus upon living, embodied, dialogical utterances is crucial: utterances are relational-things; they exist only in the interactive space between speakers and listeners; they do not exist 'in the minds' of either; neither do they exist in disembodied discourses. It is in the dialogical interplay of voices that meanings are developed and negotiated, finalized for a moment, and then, perhaps, as a new voice enters the dialogue, answerable for a new position in reality not before noticed, again destabilized.

Here, then, we have a new focus for human agency and the site of its operations: in the voices at work in an interactive moment. In the past, we have attempted to locate human agency either in 'the individual subject' (Kant, for instance), or in recent times, in a 'discourse'. For example, as Foucault (1986) claims: "The author function is [a] characteristic of the mode of existence, circulation, and functioning of certain discourses within a society" (p.108) - as if it is only in already existing discourses of a certain kind, that authorship is possible. Both these ways of talking seem to me to be equally 'blind' (or 'deaf?) to the fact that one's talk is never a matter of innocent description: as if now, instead attending to 'subjects', we should now attend to 'discourses'; both ways of talking are the product of quite peculiar kinds of disciplinary discourses, with their own cultural, ethico-political, and historical dimensions. Indeed, they are both exclusionary, professional ways of talking that have been carved out (not without violence, let it be said) of the ordinary, everyday conversational background to our lives. They both work, in fact, to render what I have called 'the interactive moment' rationallv-invisible. They claim to be justified in this, for they are revealing something 'hidden' that is worth revealing. However, instead of placing the real social and historical processes at the center of our attention, they seduce us - like a good piece of science fiction writing - into talking to each other (as a professional elite) about supposed theoretical events occurring within an abstract framework.

As a result, exiled from those moments where practical meanings are made, we professionals are faced with two possibilities: either i) we adopt systematic or orderly forms of talk, whose words work in terms of already decided meanings, that require the institution of a disciplinary form of life in recompense for the power they confer upon their inmates; or ii) we talk in terms of words whose meanings are forever deferred, that will unendingly circulate within our professional boundaries, but will never pass beyond them into the practical-moral interplay of everyday voices in which history is made(1).

Here, then, is precisely what I think is at stake in Wittgenstein's method of continually drawing our attention to what actually occurs in the circumstances of our talk, but lack the sensibility to notice. For in attempting to make up that lack by theorizing, we open the door to all kinds of claims by elite groups who want to claim, ahead of time, that certain of our words must already have a well-defined meaning; that a monologue can be instituted to replace a living dialogue. Concerned with the enormity of what is happening here - the promotion of deafness to the interplay of everyday voices, and the devaluation of its interactive moments - he 'shouts' at us: "Don't think, but look!" (no.66). Make use of your ways of talking in whatever ways you can, to look through your current ways of talking, to notice what it is in these situations that we are connecting with what, what it is that we are doing to ourselves. For, as he puts it, "a main source of our failure to understand is that we do not command a clear view of our use of our words..." (no.122). But this is a handicap that can be overcome.

For certain of our ways of talking - metaphors, poetic, moving forms of talk, what Wittgenstein calls "perspicuous representations" - can, in our living response to them, produce within us "just that understanding that consists in 'seeing connections'..." (no.122). They can change our sensibilities - help us notice the 'coin' or 'currency', the 'means' of our transactions - thus to focus upon what what-we-do does (Foucault), and to avoid not just misleading ourselves, but entrapping and imprisoning ourselves in inhuman forms of life of our own devising.

In shifting to a concern with the nature of conversational realities then, this has been my concern: to move away from those forms of talk that divert our attention away from what is important to us, that exclude us (or a great number of us) from those moments in which we as ordinary people can participate in the constructing of our own realities - so that we do not have forms of life constructed by elite- others imposed upon us. As professional academics, we really must move away from what can only go on within persons, to what goes on within relationships - even if it means giving up the theories we can each get inside our own heads. Arguing about theories, to repeat, disables us as an 'us'; it excludes those others about whom 'we' argue. For, to the extent that different ways of talking work toward constructing different forms of life, if we deny them a voice in the interplay of voices, in the tradition of argumentation and discussion constituting our culture, then, no matter how charitable we as an elite may have been towards 'them', they will feel themselves to be living in a reality to which they do not 'belong', which is not 'theirs' but 'ours'.

Let me end, then, with the following comments: Common to all the versions of social constructionism, is the central assumption that - instead of i) the study of the inner dynamics of the individual psyche, or, ii) the discovery of the supposed already determined characteristics of the external world - it is iii) the continuous, contingent flow of communicative interaction between human beings, as they cope with each other in different, everyday circumstances, that becomes the central focus of our concern.

What is special about the kind of social constructionism I have outlined, is that it is concerned with the special set of problems raised by the attempt to investigate and articulate the nature of these disorderly, conversational activities, and the influences at work in that interactive moment when the people involved - their joint action - make the connections linking them both to each other, and, to their circumstances. That is, they make connections that are their own connections, rather than merely using those already made by others.

Elsewhere, I have called this version of constructionism, a rhetorical-responsive version (to contrast with those that I see as still concerned with representational-referential ways of talking). I call it this for a number of reasons: i) First, I want to claim that all our linguistic abilities - even our ability as individuals, for instance, to speak representationally, and to depict or describe states of affairs in our surroundings - arises out of us first and primarily speaking in ways that are responsive to the others around us, bodily responsive - the gestural character of our voices matters. ii) But I also want to emphasize that we cannot just talk as we please, that social constructionism does not necessarily give rise to a just 'anything goes' relativism. For we have continually to justify our conduct to the others around us, and this one reason for calling it also a rhetorical version -our ways of talking - and the forms of life they project - are always contested by others. iii) But another reason, is to emphasize that our ways of talking can also 'move' people to action, or change their perceptions: like collective poets, between us we can open up new spaces, new possibilities for being human. And this is of the utmost importance too - but only if the necessary conditions are in place for a truly dialogic 'space' to exist, for everyone to participate in the interplay of voices.

Above, I have gestured toward what I think are a number of the properties of that 'space', the conditions making it possible. They are as follows:

These are just some of the conditions necessary, it seems to me, if we all, both those of us here in this auditorium, and those others our in the world at large, are going to be able to participate in the discursive construction of knowledge.

In setting them out, rather than attending to language considered in terms of previously existing patterns or systems, formed from 'alreadv spoken words', I have focused upon the formative uses to which 'words in their speaking' can be put. My concern has been with the nature of the relationships and relational situations thus created between those in communicative contact with each other in their speakings. Such a focus attends precisely to the political influences at work in deciding the form of connections and contacts made, the possibilities and tendencies they open up, and those they close down. Within systems of already spoken words (in what one might call already-decided-forms-of-talk), those tense moments of uncertainty and instability, when the constructing of those possibilities is decided, is ignored. We study only what has been done, the implications in the system of possibilities already decided - and we can do that as isolated individuals. But we were excluded from the originary interplay of voices that decided the system; if we are not content to live out, or 'work out', its possibilities, we find ourselves powerless to do other than complain. Thus, if we want actively to enter into the constructing of our own forms of life, then we must both: i) locate those sites, those moments when, in the interplay of voices, our voice can count; and, ii) increase our grasp of what what-we-do does (with apologies to Foucault, 1982, p.187). And my concern here today, has been with the ways of talking, the practical means appropriate to a more dialogical way for us, still as professional academics, of conducting our affairs that is not so exclusive of all the others around us.


Bauman, Z. (1987) Legislators and Interpreters. Oxfors: Polity Press.
Berger, P. and Luckman, T. (1966) The Social Construction of Reality. New York: Doubleday and Co.
Bemstein, R.J. (1992) The New Constellation. Cambridge, MA.: MIT Press.
Billig, M., Condor, S., Edwards, D., Gane, M., Middleton, D. and Radley, R. (1988) Ideological Dilemmas. London: Sage Publications.
Coulter, J. (1979) The Social Construction of Mind. London and Basingstoke: Macmillan.
Derrida, J. (1981) Positions. Trans. and annotated by Alan Bass. Chicago: University of Chicago Press. Dreyfus, H.L. and Rabinow, P. (1982) Michel Foucault: Bevond Structuralism and Hermeneutics. Sussex: Harvester Press.
Foucault, M. (1986) What is an author? In P. Rabinow (Ed.) The Foucault Reader. Harmondsworth: Penguin Books.
Gadamer, H-G (1975) Truth and Method. London: Sheed and Ward.
Gergen, KJ, (1985) The social constructionist movement in modem psychology. American Psvchologist,.40, 266-275.
Gergen, KJ. (1991) The Saturated Self. New York: Basic Books.
Harré, R. (1983) Personal Being: a Theorv for Individual Psychology. Oxford: Blackwell .
Harré, R. (1986) The social construction of selves. K. Yardley and T. Honess (Eds.) Self and Identity. Chichester: John Wiley.
Mead, G.H. (1934) Mind. Self and Society. Chicago: University of Chicago Press.
Ryle, G. (1949) The Concept of Mind. London: Methuen.
Shotter, J. (1975) Images of Man in Psychological Research. London: Methuen.
Shotter, J. (1984) Social Accountability and Selfhood. Oxford: Blackwell.
Shotter, J. (1993a) Cultural Politics of Evervdav Life: Social Constructionism. Rhetoric. and Knowing of the Third Kind. Milton Keynes and Toronto: Open University Press and Toronto University Press.
Shotter, J. (1993b) Conversational Realities: Constructing Life through Langage. London: Sage.
Voloshinov, V.N. (1986) Marxism and the Philosophy of Language. Trans. by L. Matejka and I.R. Titunik. Cambridge, Mass.: Harvard University Press.
Wittgenstein, L. (1953) Philosophical Investigations. Oxford: Blackwell.
Wittgenstein, L. (1969) On Certainty. Oxford: Blackwell.

1. Indeed, in this connection, it is worth remarking that Derrida's (1981) way of talking also, is a product of the self-same theoretical-individualism or monologism (that invents analytic concepts to represent otherwise hidden entities or processes), that he seeks in his 'deconstructions', to undo. For instance, in his talk of the mysterious movement of différance he says: "Since there is no presence before and outside semiological différance one can extend to the system of signs in general what Saussure says about language... [But] there is a circle here. for if one rigorously distinguishes language and speech, code and message, schema and usage, etc, and if one wishes to do justice to the two postulates thus enunciated, one does not know where to begin, nor how something can begin in general, be it language or speech. Therefore, one has to admit, before any dissociation of language and speech, code and message, etc a systematic production of differences, the production of a system of differences - a différance - within whose effects one eventually, by abstraction and according to determined motivations, will be able to demarcate a linguistics of language and a linguistics of speech, etc." (p.28). Clearly, whatever else Derrida is doing, he is talking about a something - "différance" - that he suggests has all kinds of properties (that eventually one will be able to demarcate), and is at work within a system of signs. But to talk in this way, of mysterious processes going on behind the backs of those involved in them, is to reinstate the very metaphysical hierarchy that privileges already existing codes, systems, and forms over social life, dialogue, and the creation of new meanings, that he claims to be displacing. As I remarked in the main text, Derrida also leaves us (unstably) fixated, either within a disciplinary, imaginary textual-world, or, unclear as to what it is that we are doing in our talk.