DeXUS Summer Institute of Discourse Studies
Aalborg, Denmark (August 2003)

Traversing Discursive Worlds:

Spaces, places, pacing and timing in discursive activity

Jay Lemke

University of Michigan

Shifting Paradigms for Discourse Studies

A theory of discourse is not just about texts. It is about the ways in which we make meaning mediated by semiotic resources such as language and multimedia. In a broader sense, it also includes the ways in which action itself conveys meaning through the unfolding of sequences of cohesive activity. We can even formulate an actional semiotic system insofar as actions themselves have systematic meaning relationships. I want to propose that the emerging paradigm of discourse studies takes as its object of inquiry not texts as such, nor even discursive formations, but discursive activity across settings.

A view of discursive activity

As we live through a day of our lives we find ourselves moving across the category boundaries that define different activities, different settings, different institutional contexts. We may do so slowly and sequentially, without overlap, but it is almost impossible to avoid participating in more than one on-going activity on different timescales, doing several short-term activities in the course of also carrying on one or more longer-term agendas. More and more often, in fact, I believe that we prefer to juggle multiple activities and to find our freedom and our opportunities to be creative moment-to-moment in cycling among these activities on shorter and shorter timescales.

We may remain in one physical place, but our attention may cycle among the conversations of the people around us, work going on in the room, the weather outside the windows, the email on our laptop, a webpage in another window, a CHAT conversation, and the person on our cellphone. This may be an extreme example, but it makes my point. On a slightly longer timescale of activity, we move from one end of the room to the other, from a group setting up a projector to a group examining some books, we may move out into the hallways, continuing a conversation but drawing some example or metaphor from our new surroundings, and we may interrupt our dialogue for a while if we go off to the W.C., and resume it again without a break a few minutes later.

On a still longer timescale, we move from room to room, from seminar room to lecture theatre to hallway to staircase to toilets to cafeteria to outdoor benches to beer pub to automobile to hotel lobby … and in each place there are stereotypical activities and there are affordances for creative variations and novel activities. Moreover, we remember the traversals we have made during the day, we carry with us notes we have made, and copies of documents we have picked up, and laptops with images in them and new emails read earlier in the day, and the meanings we make in each new setting depend in part on the cumulation of meanings we have been making during that day, that week, that year, mediated by both the varied settings and the readable material traces we carry in us and with us, and keep accessible in offices, homes, and many other physical and virtual places.

I am trying to sketch a discursive life, a life lived in real particular places, undertaken in time and through multiple activities, within and across institutional contexts and other settings, in which we are always making meaning with discursive resources, always living on multiple timescales, always juggling and intersecting and concatenating the many agendas we are enacting in any time-span, and integrating these into some sort of coherent wholes through the material-semiotic affordances of artifacts and settings.

Space, place, time, chronotope and text-in-the-making

A long time ago, Mikhail Bakhtin, reading novels and their precursors from many periods of literary history, noticed that each different kind of novel seems to describe the human condition in part through the special way that it treats time and place. He called this aspect of the literary genres their chronotopes: their ways of expressing temporality, action, and movement in and across space and place as a means of showing how people felt and made sense of their lives. The word “chronotope” combines Greek roots for time and place, and also echoes the rhetorical term “topoi”, meaning the familiar idioms, tales, and commonplaces by which orators might show affinity with their audiences.

Social science has long since adapted Bakhtin’s literary theories, including his theories about discourse (dialogism, heteroglossia, ventriloquation, speech genres, etc.), as models for real-life social communication. Bakhtin himself always tried to situate his study of the discourse of literary texts within the larger issues of how people felt and acted as portrayed in those texts. He recognized, through the insights of various authors he studied, that how we feel and how we make meaning and what we choose to do depends not just on general principles or our inner dialogues and our exchanges with others. It also depends on the settings we find ourselves in, the pressures of time, the possibilities of being interrupted and resuming, on pacing and repetition, and on the material affordances of our surroundings as well as their symbolic meanings for us and for others. He saw moreover that the action of stories, like that of life, is not constructed one scene at a time, but flows from scene to scene, making a kind of sense in-the-whole beyond the sense it makes in each scene. The chronotope represents one way to characterize the flow of action through time and space in ways characteristic of a particular genre of novel.

More recently, Henri Lefebvre, while recognizing the fundamental importance of time to how we make meaning in activity, also raised important issues about the construction of social space. He noted the important ways in which the organization of space in society sediments and embodies both our theories of space and place and the history of the social forces which have shaped social spaces. Social space was for him a key linking concept between the larger-scale forces of production and reproduction of the social order and the local activities of people living in social spaces and real, historical places.

The theory of discourse has begun to expand beyond its early focus on verbal texts and immediate situational contexts. We have begun to realize that dialogue creates saliences in settings as much as it is influenced by them, that the larger context of discursive meaning-production is multimodal activity, and often multiple activities nested on multiple timescales, within institutional contexts and across them. From the viewpoint of meaning, which is central to the unity of discourse, we cannot easily separate which meanings are made through language, which through gesture, which by movement and action. When our discursive activity includes drawing or presenting images, donning garments, adopting vocal styles, playing music … then we know that as much meaning is made in the relations among these resources as through each analytically viewed as separate.

The importance of computer-based multimedia technologies has made us especially aware of the complexity of multimodal meaning production and interpretation. New technological opportunities to link communicatively across time and space at every moment of our everyday lives is making us more and more aware that discursive media are intimately integrated into the course of our daily actions. There was a time when we might have thought of a book as a pure, isolated text, a separable bit of discursive reality to be studied on its own. We can always still do that, and for some methodological purposes it remains useful to do so. But increasingly we see the book we carry around as playing a role in our activities, as integrated -- so far as the meanings we make with it are concerned – with short-term agendas such as reading a quotation from it to make a point and long-term ones such as presenting our identity as a literate or intellectual person. Exactly what meaning we make with the text in the book at a particular time and place depends on many things: why we are reading it, what we are doing that involves reading it, who we are with at the moment, what the pressures of time may be, where we are located and where we are moving from and to.

This is not to argue for what we already know, that context influences interpretation, but to argue for a different unit of analysis in studies of discourse: not the text as given, completed, but the text-in-the-making, always still contingent on what may happen next, always evolving moment to moment as we make sense of and with the book, the words in our ear, the webpage before us, as part of several levels of larger encompassing activities carried out across times and places.

Espen Aarseth most recently has argued that there are forms of textuality which are dynamic in different ways than any narrative because they do not present a story which has been written down for us to read, but rather a potential story that is being made jointly by our efforts and those of a text-machine. He calls such literature “ergodic”, meaning made-by-work, where the work is that of the reader, or user, or player, combined with the response of a designed system that is capable, with the combined effort of the user, to produce a readable text. He has in mind here the genres of computer games, in which player choices produce a potentially unique sequence of events, texts, and visual images each time the game-program is used. Many hypertext systems are similarly ergodic, and there are quite a large number of imaginable genres of ergodic literature. (There also exist ergodic prose and non-computer-based ergodic text-making systems.)

So also, once again, in life. The world is not a text-already-written in the mind of God, merely awaiting the next tick of the clock to unfold the next event in our lives. Or if it is, God has cleverly made it appear to be instead a text-in-the-making, the result of our choices and actions, in many respects as unpredictable as our own decisions about what to do next, where to go next, whether to bless or curse. Equally emergent is the dynamic “meaning-text”, the interpreted meaning we make with any object-text (in a book, on a screen, or from a voice) which we use or analyze discursively: it is the contingently emerging product of our activity. Our activity, but also the activities of those around us, and of natural and artifactual forces of many kinds.

Discourse Studies: Method and Object

So I offer here a new paradigm for the study of discourse and textuality. The objects of our study and our units of analysis should not be texts as such. Texts in isolation can only be the traces of some prior discursive activity, useful for making some inferences about how discursive activity proceeds, but not the beast itself. Discourse studies should be the study of discursive activity, the meaningful and meaning-making activity which produces text when it leaves such a trace or record. That trace or record may be a written text, or it may be a change in the arrangement of furniture in the room. In its more complex forms, it is some multimodal combination of words, images, sounds, and their animation in time and space. But the beast itself is discursive activity as it happens, and while we may observe, and even participate in this activity, as with any activity, it is ephemeral and so difficult to study. We study activity by recording it, always imperfectly, and we need particularly to distinguish our record of an activity of producing discursive meanings from the trace of those discursive meanings that was itself produced in the activity we are recording. Both are texts, but they are texts which were produced by two different discursive activities, one of which, the one we are seeking to record and study, was a component, cast as an object, of our second-order meaning-making, our recording activity.

The recording we make is of course one kind of evidence for the analysis of the activity we recorded, and another kind of evidence for our activity of recording. Reflexive research method requires that we consider it in both ways, that we examine both our ways of recording and analyzing activity as well as the activity in focus.

But clearly any such record is incomplete and unsatisfactory in many ways. We do not know how to record how people feel as they act, or what meanings they may be making that have no outward visible or audible sign. And so we ought really to re-play the events for the participants and ask them to narrate, in retrospect, what they were feeling and thinking as they acted. We can also ask them to narrate these meanings and feelings as they act, but in some circumstances we know that this will change the nature of their actions, and that may or may not matter for our purposes.

We can also record ourselves, narrating as we act or not, and review and comment on what we think we were doing and feeling as we did it. And we can get our associates to do so. Always remembering that the narration itself, participation in being recorded, and the post-narration event are different discursive activities from the simple activity we originally set out to study.

What is to be gained by such studies? For one thing, they help us situate meaning in place and time, and contextualize it within activity. When we design documents or systems for people and communities, we will want to take into account where they are, where they are coming from and going to, what they are using the document/system for, what else they are also using their interaction with the document/system for, what timings and pacings, interruptions and resumptions, haste and delay may mean for the kinds of meanings they will make with and from the document/system.

The notion of discursive activity I have sketched here cannot be the whole object of discourse studies. We do not simply want to know how people make discursive meaning in and across places and times. We also want to know what the functions of these activities and meanings are in the larger social system. We want to identify patterns that repeat from instance to instance, and see how they co-vary with other important social categories. We want to move from discursive activity to chronotopes and genres, to heteroglossia, social production and reproduction, cultural maintenance and change.

But to do this we must start from an adequate basis, from examination of discourse-as- phenomenon in enough of its wholeness that we will find its connections to the rest of the social world, rather than seeing it as isolated and unique, or only being able to trace its connections so far and no further. We must hone our skills at studying discursive phenomena defined in sufficient scope that it becomes possible to see how they cohere as meaningful wholes in the real world of space, time, and action.

Both Bakhtin and Aarseth gained insight into these questions from the study of literature, or at least of fictional and created worlds simpler than those of real life, in which discursive activity may show most or all of its native features, but in more limited ways, more amenable to our initial skills of recording and analysis.

I am attracted to the idea of studying how people mean and feel in artificial or virtual worlds, such as those in simulations and computer-video games. These worlds have spaces and places, movement and temporality, nested activities and agendas on different timescales, and multimodal creation of emergent, contingent discursive meaning, which when recorded forms a rich text. Players and users in such worlds can give us accounts, as we ourselves can give them, in real-time and/or retrospectively, of meanings and feelings.

Even from the beginning of such a project, however, we ought to make at least some efforts to place the discursive activities we record and study in the larger context of how the players/users come and go between participation in the virtual world and activity in other places and spaces of their lives. It is characteristic of much current research on human activity that it tries to isolate its phenomena and protect them from interruptions and distractions, but in the model of discursive activity I am proposing here, it is just such “extraneous” matters that we need to understand as part of the real context of meaning-making activity. In real life people are interrupted and resume, people may have time-limits and deadlines, we may play the same game on different computers in different sites, or carry a virtual simulation world around with us on our laptop and use it differently here and there, now and then. We do not understand human activity or discourse phenomena if we snatch them out of lived space and time.

I know that such a research paradigm is not an easy one. But research is not supposed to be easy. It is only supposed to be possible, and worth doing.

Some Kindred Fields of Study

The study of human social activity in several fields deals with similar issues. Ethnography, especially urban and educational ethnography, confronts the problem of following people along the trajectories of their daily lives across very different settings and institutional boundaries. Cultural psychology, especially the tradition descending from Vygotsky and Leontiev’s “activity theory”,  tries to understand how people make meaning in and across activities and settings, subject to the norms and conventions of a community. Closely related are the fields of CSCW and CSCL, computer-supported collaborative work and learning, which look closely at how material-and-semiotic tools and interfaces are integrated into the wider spaces and activities of work and learning. Studies of technology which use ethnographic methods in the workplace, and “usability” research which tries to understand how people make sense of and with computer programs, interfaces, and virtual worlds also address the issues I have highlighted here.

None of these fields of research however pays central attention to discourse, or more generally to what discourse studies can say about the kinds of multimodal meanings available in a setting or being created in dialogue, on screen, or in activity. Researchers tend to rely on their intuitions as members of a culture or community, or their technical knowledge of procedures and affordances of the technology in use, to gauge what meanings are possible and actual in observed or recorded activity. Usually in these fields you do not see detailed discourse analysis or semiotic analysis of multimedia. You do not find sophisticated theories about the relations among representational, interactional, attitudinal, and organizational aspects of various units of meaning, much less detailed accounts of the various means for realizing these aspects of meaning.

The result is that while we know a great deal about what people do in activity, we know very little about the role of specifically discursive or semiotic action in activity as a whole. And while we know a great deal about the formal grammars of linguistic and visual semiotics, and about the meaningful organization of typical text-types and genres of discourse, we know much less about how these are created, interpreted, and deployed in the course of real human activity in lived space and time.

A focus on the study of discursive activity as such should certainly give us a deeper understanding of activity for knowing the role in it of discursive tools and resources, and a deeper understanding of the functional features of discursive systems for observing their integration into ongoing activity. We will understand text better as a unit of meaning when we understand how it is produced and interpreted in real-time in response to the demands and opportunities of situated activity.

History and culture in discursive activity

I hope that no one takes this programmatic proposal as a call for a purely micro-social analysis of discursive activity. I do not believe that such an enterprise is either possible or potentially productive. When we study activity at the normal timescale of human action, from second to second and minute to minute, for durations of up to several hours, we ought to be paying attention to the contextualization of this activity by agendas-in-progress on still longer timescales, from days to lifetimes. What we do and how we act is not determined solely by the immediate setting and the immediately preceding and following moments and minutes, especially not when our activity is mediated discursively. We invoke discourse patterns and discourse contents from other times and places as we make sense of the here-and-now. We imagine and limit our range of actions in accordance with expectations developed over long periods regarding what is socially acceptable and culturally meaningful in various settings. We continue to enact social relationships of long-standing, to perform identities long in the making, to move toward longterm goals, to revise longterm goals. Moreover, we carry with us, in our bodies and in our baggage, the cumulative signs of our other activities, earlier in the day, earlier in the week, earlier in our lives. At any moment, time can skip forward or backward for us, and events and meanings of long past take on immediate significance in present activity.

When we come to consider our embedding in multiple timescales of activity, we also have to consider the material-cultural systems that mediate and provide continuity across such timescales: our wardrobe, the streetscape, systems of transport and communication, the persistence or continuity of institutions such as schools and places of business. We have to consider the embodied habitus that shapes even our spontaneous actions and preferences as a function of our life trajectories, and does so most often differently for people whose lives have been lived in different social positions (by class, by gender and sexuality, by ethnic culture, racial categorization, bodily endowments, etc.). We have to consider, particularly as students of discourse, the discursive norms and persistent typical formations (genres, language ideologies, fashions of speech, rhetorical commonplaces, common opinions, specialized registers, etc.) which persist over long time periods in a culture, and which we have only become able to recognize and mobilize because of our repeated encounters with them over timescales long compared to that of any immediate activity.

I believe that all discursive activity, all activity, is part of larger networks of activity that extend across places and times, and that every activity bears in itself the marks of its connections and its embeddedness in society, history, and culture. In fact I believe that increasingly all activity meaningfully participates in networks that overlap across different institutions, societies, and cultures, and is meaningfully situated at the intersection of many histories. Not every contextualization of activity is evident in every moment, but if we are interested in the coherence of activities and discourses across their moments, then it will not be long before we find ourselves in need of these larger-scale perspectives.

Of traversals and politics

Where are the boundaries of an activity? How do we know when one has ended and another begun? Are activities in fact segmentary at all? I am not sure that it matters. Activity for me is fundamentally a “mass-“ not a “count-“ noun, the name of a process not of a thing. Activity is always taking place, and probably discursive activity is always taking place. Retrospectively we may be able to define criteria to say that some particular kind of activity started or ceased at some particular moment, or in some particular act. This may be important if we have well-defined genres of activity, but in my own experience of analyzing such matters, the boundaries are fuzzy at best and arbitrary at worst. Activity is dynamic. It is also making itself different. There may be cycles, there may be near-repetitions; there are certainly useful patterns to be discerned. We are constantly moving in our on-going activity from one setting to another, from one institutional context to another, and even from one social system or culture to another.

What is interesting to me about the flow of activity are the traversals we make across social boundaries. Not the discontinuities, but the continuities. Not the sense in which we stop doing one thing in one place and start doing something different somewhere else, but the sense in which there is some carry-over, some cumulation, some consequence of earlier activity for later activity. One way of saying this is that we create “traversals” across the socially significant boundaries of our lifeworlds. Another is that we are always engaged in activities on multiple timescales and that some aspects of the longer-timescale activities are always relevant to connections among the shorter timescale ones that they encompass. Perhaps the most important is that the continuities or relationships on many timescales across activities, settings, and institutional contexts are meaningful to us.

The meanings we make along our traversals across activities and contexts provide the possibility of a new kind of text: one that is always in-the-making and never finished or completed, except in retrospect and arbitrarily. It is also a kind of text that catenates elements of different genres or institutionally typical text-forms and discourses, but makes its meanings across these. Not by hybridizing them, but by parasitizing them as elements along the path of the traversal. These are thus post-generic texts, not hybrids but on-going recontextualizations of generic elements. We can best understand the organization of meaning in such traversal-texts by studying, as I suggest here, the discursive activities in which they are produced.

Traversal-texts may indeed instance some of the same kinds of meaning-organizing principles that are found in hypertexts, that is in the traversal-texts created by choosing a pathway through a hypertext environment. For example, as I have argued elsewhere, the logical-semantic relationship between successive webpages or lexia units in a hypertext, is often of the kinds described by Halliday as expansion and projection. Univariate chains may be constructed in this way, but it is relatively difficult to construct multivariate structures because designers of systems of fixed hypertext links cannot normally anticipate user moves beyond a very short range. Thus multivariate structures tend to occur on shorter textscales, and univariate or covariate chains are the principal basis of longer textscale cohesion and coherence.

As we channel-surf television offerings, or web-surf, or just engage in the construction of traversal meaning in the course of normal everyday life, we find that some catenations, some ways of making meaningful relationships among some selections of events or activities, across some sequences of contexts, are more satisfying to us than others. We may be sensing new principles of coherence, or we may be inventing new proto-genres. To the extent that such genres become repeated, typical, and culturally sedimented proto-genres that originate in traversal-meaning can become normal genres (and resources for future traversals). But even if they do not become institutionalized in this way, meaning patterns that form the basis of traversal coherence might be regarded as “semi-genres”, as attractors in the space of possible traversals, which pull us, as it were, it the direction of making meaning sequences of certain kinds, even if we do not usually ever get to complete any fixed and repeatable genre pattern. Normal genres exhibit closure, or at least it is always possible to invoke some closure for them. They are completable sequential patterns. Traversals are not normally completable; they are indefinitely extensible, even if we do find ways to demarcate some boundaries in the ongoing sequence of activity, if only retrospectively and somewhat arbitrarily.

I believe that the increasing importance of traversal meaning and the resultant traversal-texts and their incipient semi-genre patterns is symptomatic of a more fundamental change, or at least development, in postmodern society. Modernist society is characterized, above all, by its complex, interdependent, and relatively well-defined social institutions and their many text-types. Modernist society is a grand pyramid scheme in which even the most basic human survival functions depend on vastly elaborated networks of institutions, activities, and technologies. For this system to continue to operate, large portions of the time of most people must be taken up by ‘serving the machine”, i.e. by performing role-based tasks within specific institutional formations, often by making or making use of specialized text-types. Discursive activity is heavily dominated by the needs of institutional functioning.

Where in this modernist scheme is there room for human creativity and initiative? For people to have a sense that they have made something unique and valuable? For any of us to feel that our lives are other than institutional functions that could as well be performed by someone else? Even leisure and personal relationships are increasingly for many people becoming routinized. Less and less of our lives, as consumers, escapes the same system of institutions that dominates our lives as producers. We have to find our private spaces in the cracks between institutions and their demands, prescriptions, and standardizations.

As those spaces become less and less available to us, and afford us less and less opportunity, we have begun to turn to another possibility: to make our meanings across rather than between institutional spaces. Instead of seeking out the cracks, the not-yet-colonized spaces of everyday life, we can seek to make traversal meanings by appropriating elements from multiple institutional spaces and catenating them together along a unique traversal that affords us a very large space of potential new meanings to be made. This is a form of “tactics” in deCerteau’s terms, as against the “strategies” of the dominant institutions. But it is more than a resistance around the margins, for it is a successor strategy in its own right, a way past the impasse of modernist institutions, the first truly postmodern way of being.

The result is not a post-institutional society, but a trans-institutional one, in which the institutions remain (and are perhaps we may hope reformed), but in which the creative action, the significant meanings, of people’s lives finally come to depend less and less on the internal affordances of the institutions as such, and now more and more on cross-institutional possibilities. We see this on the long timescale of people’s careers, where few young people today expect to remain within the same industry much less the same company for a lifetime or a career. Many also do not see the point of having only a single career at any one time, but plan to participate in multiple institutional systems simultaneously. We see it on the timescale of a day or a week, where people seek a good mix of activities and settings, sometimes by advance plan, but just as often by spontaneously making each next choice on the basis of the emerging continuity or cumulation of what they have done so far, sensing what would contribute to a satisfying traversal of the day or the week. On the much shorter timescales of an hour or two, we increasingly juggle multiple foci of attention, talking on the cellphone while driving or shopping or attending a meeting. Reading email or doing CHAT while in class or at a boring institutional event. And these institutionally secondary or peripheral activities may be for us the primary ones in terms of meaningfulness and how we feel about them as discursive activities.

People long to escape from the hegemony of institutions over their moment-to-moment lives, and there is a new struggle emerging between institutional policing of attention and people’s desire to escape surveillance and control. Institutions with less policing power attempt to use more and more seductive techniques to keep our attention captive, with arresting visual imagery, erotic appeals, emotive sounds, etc. But it always remains possible to decompose these and appropriate them piecemeal as part of our traversals. Moreover, we are capable of multiple simultaneous, or at least rapidly cycled foci of attention, and we can respond to the demands of institutions for productivity, and even to their offerings for our attention, without every wholly giving up the freedom to add them to our attentional mix, while still retaining creative control over the mix as a whole. What we make discursively from that mix is the traversal and its emergent mode of meaning.

For these reasons, I believe that the study of discursive activity is particularly important to help us understand how traversal meanings get made and the important role they may play in postmodern society. For those who believe that the future lies in better institutions, whether by reform or revolutionary replacement of the existing institutions, a focus on traversal meaning may seem like an escape from responsibility for the institutional future. But if you come to agree with me that institutions as such are not the human future, but only one part of the infrastructure for a post-institutional mode of human sociality, then there are some very exciting discoveries that may lie ahead of us.

And whether you are engaged by these historical and cultural speculations about the role of traversals or not, I hope you will at least see that the notion of ongoing emergent meaning in discursive activity is a fundamental one for the study of both discourse and activity.

JLL.