AN INQUIRY INTO THE ONTOLOGICAL FOUNDATIONS OF ORGANIZATION
Emeritus Professor of Communication
Université de Montréal
Abstract. Globalization, it is generally thought, is transforming organization. This paper offers a theory of organization that lends itself to the study of change. It calls attention to two neglected phenomena: the origin of hierarchy in communication and the role of communication in constituting the identity of the organization as an actor. It argues that organization begins as a co-orientation toward some common object, characterized by relationships of principal to agent. A coorientation system called A-B-X is proposed. A-B-X systems are found to have two boundaries, and it is the existence of communicational boundary conditions that explain the emergence of modern industrial corporatism in the 19th and 20th centuries--boundary conditions that now seem to be eroding. The paper then proposes a reflexive and structurational model of the process by means of which organizations are constituted as actors. The paper concludes with some reflections on the sources of stability and instability of organizations in a global environment.
You could easily convince yourself these days that we were in the midst of a radical transformation of organizational structure and process.
Two March 1999 columns in the New York Times signed by Thomas Friedman recount the story of Lyle Bowlin, who set up a company that operates out of a spare bedroom of his house in Cedar Falls, Iowa. According to Friedman, this little enterprise could soon rival the Internet giants in bookselling, Amazon.com and BarnesandNoble.com. For a minimal investment Bowlin got "www.positively-you.com" up and running on the Net and has been doing a banner business, undercutting his more established competitors by offering an appreciable discount on the going price for a book. Friedmans first column came out on a Thursday, and between 11 P.M. that night, when the column first appeared on the Web, and 2 P.M. Friday, Bowlins page had more than 142,000 hits. By the time Friedmans second column appeared, a week or so later, Bowlin had personally replied to 1,400 E-mails, from all over the world. Friedmans take on this story? "If you think globalization is overrated, youre wrong. You aint seen nothin yet." As Lyle Bowlin can tell you, the minute you start to do business on the Web, you now have to think globally. You have to think about your customers as global, your competitors as global, your readers as global, your suppliers as global and your partners as global."
David Nollers (1997) doctoral dissertation is based on an analysis of what he calls a "virtual organization," a small company specializing in film and video production on consignment to which he gives the pseudonymous name "BrightLight." The whole company is run out of one womans pocket. There are practically no full-time employees, and even those who could be called "permanent" have an ambiguous status, half employee, half friend. The boundaries of the organization fluctuate, depending on what projects are currently underway. Teams form and then, when the project is finished, dissolve. Lindsay, the heart of the operation, is part of an amorphous network of people who specialize in film and video production, in one capacity or another. They all know each other on a first name basis; contractual arrangements are based on trust. Her organization (if that is the appropriate term) has no clear boundaries, nor is it easily identifiable by its physical surroundings. There is a studio, but it is only one component of a larger infrastructure of production facilities that also migrate from one organization to another, depending on the contract. It all works very well, but it doesnt much resemble what we used to think of as a film company.
Pamela Shockley-Zalabak (1999) reports on a team of highly specialized consultants, operating on an international scale, whose internal conversation occurs entirely via electronic media. They never, for the most part, see each other face-to-face, even though their work depends on collaboration. Their performance has been outstanding, but as the years stretch out, they exhibit increasing symptoms of distress, because, it turns out, of the lack of direct interpersonal exchange. They are finally obliged to admit that inter-group communication mediated entirely by technology is too stressful to live with indefinitely. But although they now meet from time to time, they, like many others these days, remain a virtual organization.
Others such as Engeström, Y., Engeström, R. and Vähäaho (1999) see a trend toward what they call "knotworking" in even the highly institutionalized world of health care: loosely joined teams of professionals who form temporary organizations, depending on circumstance. As they describe their concept of "knot" it refers to "rapidly pulsating, distributed and partially improvised orchestration of collaborative performance between otherwise loosely connected actors and activity systems" (p. 2). They argue that knotworking is a "significantly new form of organizing" (p. 2).
Each of these examples makes in its different way a similar point: that the constitutive processes of organization that are now operative in a fully networked society are diverging from the established patterns of those which typified the industrial period stretching from the mid-nineteenth century until very recently. On the other hand, large organization has hardly vanished. The current situation sometimes feels paradoxical: Earlier (Taylor & Van Every, 1993) we reported that enterprise, impelled by globalization, is simultaneously getting bigger, as a result of wave after wave of acquisitions and mergers, and smaller, as a result of a rapidly expanding rate of small and medium size enterprise formation, in every domain one can think of. In both cases, the large and the small, there seems to be a kind of knotworking logic at work: company coalitions that form and dissolve almost kaleidoscopically, depending on the exploitation of opportunity, either to produce mammoth global enterprises or to generate the kind of virtual organizational forms I have just been describing.
It is not my intention in this paper to estimate the dimensions of this transformation nor to address its economic and social implications. Although the booksellers have their shelves full of books devoted to predicting the future of industrial organization I doubt whether many of them will turn out to have been extraordinarily prescient. Instead, I want to take up a different, and more fundamental, question: What might explain both the stability of large organization, during some historical periods, and its susceptibility to rapid change, when the conjuncture of circumstances favors the latter. I want to probe a bit more deeply into those factors that explain the persistence of organization and yet its vulnerability to fragmentation and loss of identity. Why is it that organization forms change but organization persists? Shockley-Zalabak calls contemporary organization "protean" but Proteus remained the same by constantly altering his appearance: Are organizations protean in just this way?
Organization not only is enabled by communication but emerges in it (Taylor & Van Every, 2000). If communication scholars are not simply to be overtaken by events it becomes crucial for us to understand how organization emerges in communication. Demonstrating this emergence, it seems to me, is the challenge that now confronts us.
The easiest way to conceptualize an organization, of course, is to describe some of the features of the organizational entities that surround us, with which we are most familiar. You might think of this as the empirical approach. The problem with the empirical method, however, is that it encourages misleading conceptualizations in a period of rapid transformation, where the emergence of new forms of organization is the norm, not the exception. We risk ending up describing in detail something that is already an anachronism. If, as seems likely, we are now in such a transitional phase, then what is needed instead of description is a return to first principles, and that is a task of theory: a more conceptual than empirical approach. It is such an approach that I am arguing for in this article. In it I will be delineating what I believe to be the elements of an original theory of organizational communication, one that rethinks the concept of organization by placing at its center concepts of agency, coorientation and identity.
To set the stage for the conceptual exercise let us first briefly examine the changes in organizational pattern over the past century that have brought us to our present state.
How We Came to Have the Organizations We Had
The phenomenon of a very rapid passage from one pattern of organization to another has already been documented for the century before ours. Weber (1922/1968), Innis (1950, 1951), Chandler (1962, 1977), Williamson (1972, 1977), Beniger (1986), Jacques (1996) and others have drawn a portrait of the emergence, beginning in the latter half of the 19th century, of the modal form of organization typical of large bureaucracies and corporations in our own century. Before 1840, North American enterprises were typically locally-based, family-owned companies depending for their foreign commerce on independent agents physically located in distant markets, with which they communicated by mail. The largest bank in the United States, the Second Bank, operating in several states, was run by one man, Mr. Biddle, and his entire head office staff consisted of two assistants. According to Chandler (1977), the concept of middle management was unknown in 1840: Owners did their own managing. Government was small--minuscule, actually, compared to now. In the 1830s President Jackson and 665 other civilians ran the three branches of the federal government in Washington (Beniger, 1986, p. 14). When railroads typically had about fifty miles of track, their management could be handled by a single supervisor. Far and away the most important "enterprise" was the family farm.
This all changed overnight when the modern corporation was born. As Chandler (1977, p. 4) puts it: "Rarely in the history of the world has an institution grown to be so important and so pervasive in so short a period of time." Beginning about 1860, and in spite of one war and two recessions, the next half-century witnessed the progressive, and incredibly rapid, emergence of the modern firm or government, with its typical professional administration, and its built-in economies of scale. What had previously been autonomous operations of resource development, manufacturing and sales merged to form complex entrepreneurial ventures that occupied the national work space, becoming not just the typical model of enterprise, but also the symbolic metaphor that we came to take for granted as "normal"--an entity so present to us in our daily lives that, ontologically speaking, we could hardly doubt its existential reality.
By the 1950s the "second industrial" revolution seemed to have run its course (Boulding, 1953). We had arrived at the era of what Whyte (1956) termed the "organization man." Industrialism had come of age. The family farm was a vanishing species. The organizations that dominated the landscape had often grown from initiatives taken in the time of transition, the latter half of the 19th century, but by now they were so imposing, and so well established, that it seemed that they would last forever. A whole new social class had appeared: the professional managers (Djilas, 1957; Gouldner, 1979). The quintessential property of modern organization was hierarchy: a ladder of embedded offices linking senior management to its workers (Weber, 1922; Chandler, 1977). In Chandlers (1977, p. 1) words: "Modern industrial enterprise is easily defined ... It has two specific characteristics: it contains many distinct operating units and it is managed by a hierarchy of salaried executives."
Yet only a few years later the impression of invulnerability and stability came to look more and more like an illusion. During the decade of the 1970s, fully a third of the largest companies went out of business (Birch, 1987; Taylor & Van Every, 1993). The fortresses of enterprise began to crumble. Fortune 500 companies employed 2.2 million fewer people in 1985 than they had just five years earlier, in 1980 (Birch, 1987). The job losses were compensated for by a remarkable increase in the rate of formation of new enterprises; again citing Birch (1987), 1.4 million new enterprises came into existence in a single year, 1985. It was not that large enterprise has vanished; on the contrary the scale of enterprise has expanded very rapidly through consolidation in the form of mergers and acquisitions. What was changing was the organizing pattern, away from hierarchical and toward horizontal.
By the early 1990s it was official: The prophets of business process reengineering (BPR) were heralding the end of hierarchy and the celebration of a process-oriented kind of management where the middle levels of management would simply be "obliterated" (Hammer & Champy, 1993). What had seemed like the triumphant achievement of industrialism only a few years before was now perceived to be an encumbrance (Zuboff, 1988, 1996; Hammer & Champy, 1993; Hammer & Stanton, 1995). To quote Zuboff (1996, p. 14): "The brilliant inventions of the early twentieth century workplace are now the barriers that inhibit the next great leap forward.... Unlocking the promise of an information economy now depends upon dismantling the very same functionalized managerial hierarchy--with its moral vision, social system, entrenched interests, and vertical focus--that once spelled greatness."
How then are we as communication scholars to come to grips with a phenomenon of this extraordinary novelty? As I suggested earlier, we must go back to first principles. In the next section I will suggest how we might begin to achieve such a shift of perspective.
Going Back to First Principles
In their introduction to the Handbook of Organization Studies (1996, p. 3), Clegg and Hardy remark that the kind of changes I have been illustrating are "here to stay" and "have major implications for our understanding of what organization studies constitute." They add: "Gone is the certainty, if it ever existed, about what organizations are; gone, too, is the certainty about how they should be studied, the place of the researcher, the role of methodology, the nature of theory. Defining organization studies these days is by no means an easy task."
For someone whose discipline is communication an immediate
restriction is placed on the "nature of theory" that
can be developed about organization, not to mention the role of
methodology and the place of the researcher. By definition, I
must conceptualize organization through the lens of communication
theory. The same Handbook contains a chapter (Putnam, Phillips
& Chapman, 1996) which enumerates some of the metaphorical
bases underlying the approaches that have been employed by communication
researchers in looking at organization. Putnam et al, building
on an idea of Smith (1993), identify seven "root metaphors"
for the communication/organization relationship:
Each of these metaphors points up an aspect of the communication/organization
relationship. People do think of themselves as being in
an organization and exchanging information there; communication
patterns do influence perception (a phenomenon that March &
Simon, 1993, called "uncertainty absorption"); an organization
may certainly be thought of as a network; clearly for the network
to exist it must be performed; the role of symbolic constructions
in the reflexive construction of organization that people engage
in is certainly important; it is obviously true that in any organization
some voices are heard more clearly than others and this can only
with great difficulty be disassociated from the exercise of power;
and, as Giddens (1976, 1979, 1984) has pointed out, it is in the
interactive exchanges of everyday life that the systems of organization
are both realized, and its structure reinforced and perpetuated.
An adequate theory of communication should account for each of
these perceptions. Nevertheless, it seems to me that none of the
forms of explanation analyzed in Putnam et al's article adequately
accounts for two additional and important factors in the genesis
and maintenance of organizational existence:
Evaluated in relation to these two criteria, it is remarkable how much of the organizational communication literature focuses instead on individuals and their involvement in interactive relationships, as information exchangers and filters, as interconnected nodes within a network, as inter-actors and sense-makers, as speakers enveloped in discourse.
There is nothing wrong with the individual/interaction emphasis, as far as it goes. What it leaves out, however, is the construction of frameworks that transcend the immediate interactive processes of everyday life to become architectures of sociality capable of spanning multiple conversations of organizational members, over expanses of time and space. In Chandler's (1977, p. 7) words: "The existence of a managerial hierarchy is a defining characteristic of the modern business enterprise." Yet in the literature reviewed by Putnam et al you will look in vain for a discussion of the bases of hierarchy in communication. Where the exercise of power is considered, as in the critical literature and writings on dialogue, it tends to be treated as an aberration: a social ill to be eventually corrected. Similarly, the fact that organizations come to have their own identity receives remarkably little attention. Whether one rejoices in them or not these are two facts about organization and they need to be included in a full explanation of it.
In this paper, I am trying to build on this existing literature
to add elements of explanation that may contribute to a re-conceptualization
of the structuring role of communication in the formation and
perpetuation of organization. I will develop a line of argument
that addresses in turn these two factors: the creation of systems
of hierarchical inter-agency networking (which I conceptualize
somewhat differently from network theory, in seeing such systems
as instances of imbrication), and the production of an
organizational voice distinct from that of any of its members.
My argument incorporates a number of assumptions, most of which
are prefigured in Putnam et al's discussion, particularly in its
consideration of performance, symbol, voice and discourse, but
in addition I develop some new principles that emphasize the organizing
properties of communication. It may be useful at this point to
In the pages that follow I will try to develop these ideas to make them more accessible. I am particularly interested in the forces that perpetuate organization or, alternatively, lead to its dissolution. If, as Senge (1990, p. 17) asserts, the survival expectation of a large enterprise is less than forty years, then we need to understand better what it is that sustains the identity of an organization, and alternatively why so many companies lose theirs. This is especially the case in the current time of rapid change.
The modern organization can be seen as an ecology of multiple communities (Krippendorff, 1998), each with its own situated experience of work, and its own local universe of discourse--each, to a significant degree, self-organizing (Weick, 1985). A communitys fund of knowledge is both (1) collectively distributed throughout its network of interconnected components (I will return later to discuss knowledge as a property of networking), and (2) a gamut of opinions "distributed" in a statistical rather than a topological sense--a collection of individual judgments, rather than a property of networking. At the leadership level, however, the organization takes on a different kind of existence as a corporate actor with a unitary voice that claims to represent the intention, no longer of a population of individuals marked by great variety, but of a single acting entity. Any days newspaper is sufficient to illustrate the feature of unitary acting I am referring to: "the Clinton Administration is pushing etc.," "a Senate committee tonight approved etc.," "North Korea repeatedly refuses to sign etc.," "every Indian government has resented etc.," "drug companies are definitely searching for etc.," "the Association of Flight Attendants has said etc.," "Novell plans to say etc.," "Unicredito Italiano said it would offer etc.," " the New Republic hopes a redesign etc.," and so on. We all understand that, behind these affirmative statements, there may lie differences of opinion involving individuals, but we nevertheless tend to take the expression of corporate intention at face value.
So there are two ways to think about organization: as an ecology of local cultures and as a unitary actor. We need to capture both dimensions in our theory. If the work world is collectively grounded in a distributed network of situated practices and understandings, and leadership requires the utterance of a single organizational intention, then the issue is how the diverse knowledge bases of work are translated into a discursive realization of organizational identity. It is not that the organization is rational at the level of management and irrational at the level of ordinary work but that organization can be transformed into an actor only by the rationalization of its distributed intelligence into a symbolic expression of a collective intention (which is what I mean by "rationalization": an authoring of a text suitable to be voiced, and thus producing a singleness of point of view).
Communication figures in this transformation in several ways. But before we consider them in detail we need to lay some groundwork. Let me begin with the local world of situated task-oriented interaction among the members of some working group. The following is my idealized construction of the basic communication process, organizational or other (Figure 1). My explanation starts here, at the simplest level.
First, consider the vertical axis. All communication is rooted in a time and a place, situated in a material world that has to be dealt with. This I think of as the circumstantial axis of communication. It supposes that both the communicators and the objects they focus on are embodied in some ultimate materiality, characterized by spatial coordinates, even where the apparent theme of communication is symbolic, rather than ostensibly physical. Furthermore, whatever people in conversation are dealing with now has a history--a before--and a future--an after. They are situated in time as well as space. To use Heideggers (1996) classic expression, we are thrown into a world. And that "thrownness" comes from being circumscribed by a horizon of time and space.
All communication is grounded in a conversation, therefore in interaction. To communicate, at least two people must be involved, and that is already the starting point (and indeed the manifestation) of a network. I reject the stale metaphor of communication as two or more individuals exchanging messages. My image of the conversational context is much closer to that of Birdwhistell (1952, 1970), Goffman (1959, 1974, 1981), Hall (1959, 1966, 1976), Scheflen (1965, 1973, Scheflen & Ashcraft, 1976), all of whom emphasize that people in interactive talk are doing much more than "messaging" or "performing speech-acts." They are an interactive unit that is revealed as much by its members physical configurations, their gesturing and their dance-like co-positioning, as by the complex and highly interactive verbalizing they engage in to produce more or less coherent strings of talk. The conversational couple engaged in talk is more than the sum of the two individuals; it already forms a social unit, the embryo of an organization.
Now consider the horizontal axis of Figure 1. Obviously, humans are not the only species capable of maintaining a state of interaction. They are, however, the only ones who possess language, and use it as the principal medium of a sustained encounter. I assume with the approaches to cognition of Marr (1982) and Hutchins (1995) that language furnishes the means by which people successively transform representations of circumstance until some resolution of the situation is attained, at least in the view of the communicators. Although in an organizational conversation, language is typically not the only representational medium in play, it nevertheless has properties that none of the others possesses, and which are vital to understanding the organizing ability of communication. Specifically, language is not only descriptive, but it is also modal (I will return to consider the central role of modality shortly). It thus instantiates or establishes, locally, a coorientation system (a topic to which I turn next), even as it mediates the representation of an environment. Language makes it possible for people to not only represent circumstance, but also to represent themselves as a society, a society capable of expressing itself in action ("represent" in the sense of acting as agents for themselves). They become active subjects by their joint orientation to some object and it is language that accomplishes the transformation from individuals into collective actor (a dimension of language that is sometimes overlooked in the literature on sense-making, which tends to focus on the dynamic of the left-hand side of Figure 1: cognition more than pragmatics).
The horizontal axis of Figure 1 should thus be read from left to right. It describes, on the one hand, a representation of circumstance in the frame knowledge people possess to interpret their world and to produce a situation . On the other hand, it describes their transformation into actors, through the generation of what I call their text, voiced to become not merely an expression of their collective intention, a pre-requisite to acting together, each with a nominated responsibility, but the vehicle for their self-transformation into a new kind of actor--a collective actor.
Communication thus occurs at an intersection of axes of conversation and textualization. It is here, in the "talking out" of the conversation, that circumstance is transformed into a discursive representation of it, using frame knowledge, and that conversational talk is transformed into a textual resolution of some kind, such as a "decision", or a "protest", or a "report." The actors are transformed pragmatically through conversation into an actor: Both their individual and their collective identity must be achieved in communication.
Once the text has been realized in the conversational exchanges of the people involved, it then constitutes a new circumstance to be fed back into the conversational cycle. In this sense the conversation is reflexively represented back to itself (Giddens, 1984). And since circumstance is both affected by and affects agencies outside the conversation (Figure 2), including other conversations, communication has organizational properties explained as intra-conversational as well as extra-conversational. There is, as Weick (1985) calls it, both tight (local, based in face-to-face conversation) and loose (organization-wide, and mediated) coupling.
In this first step towards an explanation of the genesis of organization I have argued that an effect of communication is to produce a collective actor. Now I must describe the mechanism by means of which such collective action, and the identification of actors and their responsibilities, emerges.
A Coorientation System
The essential building block of all organization, whatever its manifestation, is coorientation (Newcomb, 1953). A coorientation unit is minimally composed of three elements: two actors (actors may be corporate or individual) and one object to which they are jointly oriented. Newcomb (1953) refers to this as an A-B-X unit.
Two distinct kinds of coorientation are possible. In one instance the link between A and B occurs as the result of a transaction in which X figures as an object of exchange. This is the classic (horizontal) model of the market: Typically the X is compensated for reciprocally by a Y (paying for something), and it is in this manner that the values of X and Y are established. This kind of transaction constitutes the indispensable boundary condition for the formation of organization, in that it gives enterprise its raison dêtre and identity as a unit of value production. It does not, however, explain organization as such.
To understand the basis of organization itself we have to consider a second kind of coorientation, namely that which results in the formation of a bond of agency. The establishment of agency is achieved when one actor B acts for another actor A with respect to some object X to which they are both oriented (Taylor & Van Every, 2000). This link is vertical, in that a principal-agent relationship is established, one in which the roles of the two actors are complementary, one person acting to realize anothers intention. The object is now the carrying through of the horizontal transaction I described above. The X becomes a performance, or objective, not an "object" in the material sense. (For discussion of the concept of object see Engeström 1990.) The object of agency is the transacting itself, a doing, not the eventual object of such a transacting, that is, the object of value. We must therefore distinguish two kinds of object, depending on the nature of the A-B-X unit. For horizontal transactions the object is one of value, whereas for vertical relationships the object is modal. (I will return in a moment to motivate the choice of the term modal as a way of describing the instantiation of responsibility for a doing .)
to form an organization.
Once an agency relationship has been instantiated, in the context of an intended transaction, the basic condition for the existence of organization is satisfied . Different eras use different terms to capture the idea of agency (Jacques, 1996). In colonial America the expressions might have been planter and factor (Beniger, 1986); in the heyday of early industrialism words such as owner and worker were in fashion; contemporary language tends to be phrased in terms of manager and employee. Whatever the lexical choice, in every case there is a complementary relationship linking a principal to an agent responsible for the carrying through of a transaction, and it is because of this link that we know organization exists. The A-B-X triplet is fundamental. It is how organization is established: structured by vertical, boundaried by horizontal relationships.
A number of economic analysts (Weber, 1922/1968; Coase, 1937; Chandler, 1962; Williamson, 1977; Clemons & Row, 1989) have observed that the emergence of modern industrial giants in the late 19th and early 20th centuries can be described as a redefinition of the boundaries of organization. Functions which had previously been left to the marketplace moved over to become the responsibility of salaried employees. This was a mutation from a horizontal to a vertical basis of organization. Williamson (1977) considers that the preference for hierarchy within the bounds of a single administration can be explained as an effect of transaction costs. Whenever it is cheaper and more efficient to hire someone to perform a service than it is to buy it on the open market then the conditions exist for the formation of large companies and public administrations. Others (Porter, 1980, 1985; Clemons & Row, 1989) observe that the same factor which led to the growth of large organization could equally well contribute to its dissolution. Once it is cheaper to buy a service on the market than to pay for it in salaries, downsizing is encouraged. Every activity consists of a chaining of transactions, modal or value-creating. How they come to be boundaried involves a degree of arbitrariness, dictated, among other factors, by economics.
This is economic thinking. What I want to do in the rest of this paper is to delve into the communicational logic of agency that underlies the A-B-X system. Let us first observe that the concept of an A-B-X unit is neutral with respect to its interpretation. The A-B-X model is generic. It applies equally well to corporate and to individual actors. "Management" and "the workforce" make up an A-B-X system composed of collective actors, but equally so does a manager M interacting with an employee E, an exchange involving individual actors. Second, the X may be some specific object, relative to a single exchange, or it may be a category of object, part of an ongoing activity (the usual case in large organizations). Third, actors may be both human and non-human. The ultimate "agent" in the conduct of work is material: a tool or an instrument or a technology. A human agent operating a tool in the performance of a task is as much an exemplar of A-B-X systemness as any other. Finally, and following from the previous observations, every global structure of agency (management in the large) is composed on closer inspection of multiple local instantiations of agency which are contextualized by the larger system but focussed on local objects and set within the constraints of a local system which is part of the larger organization but also distinct from it: This is what I mean when I say that the logic is fractal.
No organization is consistently A-B-X rational everywhere (Taylor, in press). The distinction between horizontal and vertical transactions may be conceptually clear but it is pragmatically fuzzy: Every organization is characterized by internal horizontal as well as by vertical transactions, and every organization has at best ill-defined boundaries, in that relationships with outside actors also manifest elements of verticality as well as horizontality. If the only reason organizations existed was that they were economically rational there would be no large organizations. The economic model of organization is thus a useful guide but it does not deal well with a crucial dimension of organization, its constitution as a social unit. Here modality is a central concept.
The Role of Modality in Communication
The A-B-X unit has two boundaries. Viewed strictly in its B/X-ness, it is framed by an exchange transaction where the A/B unit functions as a single actor in negotiating a transaction with some other actor, C (client or customer, for example), and by doing so establishing the value of X. The X-centered transaction marks the limit of the organization on the B/X dimension: what is often thought of as the system-environment boundary. The value of X is thus circumstantial but the general issue of the origin of value is not problematical. It depends strictly on the transaction. But also very important to my argument is how the second boundary or A edge of A-B-X-ness is established. Why and how is A also a boundary? Here we are obliged to enter into domains of logic that are not tributary on economics (economics being the science of B/X-ness). We have instead to consider agency as a phenomenon of modality literally constituted by language. Modality is an issue of A/B-ness and it is the domain of communication science.
We must first establish what precisely an A/B or agency link is. An A/B unit forms when the intention of one actor A is translated through the mediation of a second actor B into that of a composite or hybrid actor A/B (Latour, 1994). The notion of intention is one of the central contributions of phenomenology, as interpreted by Husserl (1993; for an excellent discussion see Anton, 1999). The phenomenological idea is simply that we relate to our world by being involved in it, purposefully. Our knowledge of the world is part and parcel of our active participation in it, as constructive actors characterized by a stance or an attitude, directed toward some purpose--a concept that Weick (1979) set out to capture in his notion of enactment. If X-ness is always the orienting pole in any A/B relationship it is because of intention. Knowledge and attitude can never be detached one from the other; attitude presupposes the existence of an X and it is the concentration on an X that channels the development of knowledge. Attitude always implies involvement in some situation. Knowledge follows, the result of involvement. Communicatively speaking, then, the formation of an A/B(-X) unit (i.e., "X" motivates and frames A/B) supposes a transaction in which intention is transferred from A to B. When Bs intention mirrors that of A then agency has been instantiated and a collective actor has emerged. A and B have now become a mini-society focused collectively on a shared object, X.
The production of agency is in fact the principal function of communication (Greimas, 1987). As Dewey (1916/1944, p. 4) observed, "There is more than a verbal tie between the words common, community and communication. Men [sic] live in a community in virtue of the things which they have in common; and communication is the way they come to possess things in common."
Secondly, it is a fact of ordinary language that every sentence we speak has two logical levels, more or less explicitly enunciated. At one level, that of its embedded proposition, human speech refers to a world by making statements about it ("The sun is shining"). At another, the embedding frame of the proposition, it expresses an attitude to that world ("I love it when the sun shines"). That component of sentence structure which expresses attitude is known in linguistics as modality (Halliday, 1985; Palmer, 1986; Bybee & Fleischman, 1995; Taylor & Van Every, 2000).
Modality has two dimensions. In its epistemic dimension, modality expresses an attitude toward the facts of a situation ("I think," "I believe," etc.). In its deontic dimension, modality expresses an attitude toward what should be done about them ("Would you please ," "I wish ," "I am afraid ," etc.). Thus when people engage in conversation they knowingly, or unthinkingly, lay the basis for the transmission of intention, simply by talking to each other, using ordinary language. How such a transfer occurs is entirely situational, part of the interpretive dynamics of people interacting with each other. Nevertheless, the constitution of an A-B-X system (or the failure to do so) is the normal outcome of communication.
An A-B-X system is transactional. Jackendoff (1972), in discussing the semantics of ordinary language, observes that a verb such as "to sell" includes as part of its meaning a complementary assumption of "to buy." The structure of "to sell" is called, in formal linguistics, ditransitive (Taylor & Van Every, 2000). The seller is analyzable as an agent (the logical subject of the sentence), who acts on some object (that which is being sold) in order to transfer it to some recipient/beneficiary (the buyer). This is the basic A-B-X formula, except that the directionality of the exchange is A->B. You can't sell, however, if no one buys. By buying, B becomes the agent, X is still the object, and A now becomes a source. Here the directionality of the exchange is A<-B. The combination of A's and B's acts produces a transaction. When it is accomplished there is an A<->B relationship. All I am assuming is that all communication is transactional: If A succeeds in transferring an intention to B, it must be because B has accepted (perhaps even solicited) the intention, and made it his or her own (what Austin, 1962, called a "perlocutionary effect"). This, as I said, is part of the normal transactional dynamic of conversation.
We tend to associate the concept of intention with some persons subjective state of mind. In large organizations, however, the actors may not be individuals but corporate entities. Corporate intention clearly cannot be lodged in the subjective state of some human actor (corporations are not persons, other than in law) and yet it too can be effectively communicated. How do we explain this?
First, the reason organizations form is to accomplish purposes that individuals cannot hope to attain by themselves. This need for cooperative action is as true for a band of bushmen embarking on a hunting expedition in the Kalahari desert as it is for a team of NASA technicians in Houston, Texas, engaged in interplanetary space research. In both instances, the nature of the X-focused transaction imposes the formation of relations of agency. So it is the common orientation to a shared X that establishes the condition making unity of intention a possibility.
A second explanatory factor is the matter of resources and how they come to be mobilized. Giddens (1984) draws a distinction between what he calls allocative and authoritative resources. Allocative resources are all those material instrumentalities by means of which power is exerted in the physical world. Authoritative resources are all those symbolic instrumentalities by means of which power is exerted in the social world. The impulsion to form an organization comes in the first place from the perceived existence of allocative resources, and leads in the second place to the marshaling of authoritative resources.
Consider the example of the meat packer Swift (Chandler, 1962, 1977). Gustavus Swift was a New England wholesale butcher who moved to Chicago in the mid-1870s because he saw an opportunity to market Western beef in Eastern cities. The critical allocative resource in his enterprise was the invention in 1882 of refrigerated cars that would carry beef from Chicago and points west to Boston, New York, and Philadelphia. By the end of the 1880s, and in spite of opposition from local butchers and the railroads, Swift had packing plants in Omaha and Saint Louis and refrigeration plants in all the major cities of the East. Soon other packers such as Armour followed suit. It was the complexity of the operation, with its multiple parts, that impelled the development of a multiplicity of interlocked agencies--a modern corporation, in effect. It was the existence of an allocative resource that made the emergence of a network of authoritative interrelationships plausible.
The result was the formation of an A-B-X system in which management of Swift was perceived to incarnate the intention of the corporation. The employees became the corporations agents. But now a mystery arises. Was this merely a case of one individual, Gustavus Swift, imposing his intention on thousands of other people? Historically yes, one is tempted to reply, but at another level clearly not, since Gustavus Swift was been dead for many long years and yet Swift still operates as a (successful) company. This continuity is possible only if we accept that at some point in the history of his company Mr. Swift, although he began as an entrepreneur, himself became an agent. It is that agency which he embodied in his person that made it possible for his company to continue after he had departed, since the agency survived him and could be embodied in another person.
But if he too was an agent, as well as all his successors, then who was his (and their) principal? The answer, I believe, is the company itself. A company, however, is a virtual, not a real, actor (Livet, 1994). It exists because it is the spiritual manifestation of the community of actors who are identified with it, employees as well as management. The organization, together with the person or people who direct it, is an instrumentality whose existence is justified by the service it provides to those to whom it is transactively linked. The A boundary of an organization is thus a virtual transaction with its own community. This is how the actions of the organization are legitimated.
Earlier I argued that the B/X border can be explained by an environment-related A/B-C-X transaction (horizontal, market-driven); now I argue as well that the A/B border is explained by a virtual C-A/B-X transaction in which C is the corporation or community which has invested its spokesperson or spokespersons with agency, and thus with the authority that comes from speaking for the organization .
the organization itself.
Derrida (1986), in an analysis of the signing of the American Declaration of Independence, observes a fascinating paradox: The representatives who met to sign the Declaration were "representatives" of a non-existent polity in whose name they nevertheless claimed to speak, and yet it was their signing of the Declaration that constituted the body politic for which they spoke, and thus themselves as representatives of it. This is the magic (Searle, 1989) of declarative speech acts, in that they bring into being, at one and the same time, a word (text) and the world (community) it refers to ("word" and "world" are Searle's terms). The founding of an organization is accomplished in exactly this manner. Swift became an agency that transcended the intention of its founder at the moment that it came to stand for the community of people who composed the organization. In the absence of such a virtual actor-network (Callon & Latour, 1981) Mr. Swift would have remained just one small, rather weak individual, incapable to moving such a vast enterprise to action.
The A-B-X relationship supplies a structural (and indeed moral) basis of organization, one which is fractal in character since whatever the level of analysis, macro or micro, an A-B-X system will be present. But, once established, an organization has other mechanisms that serve to perpetuate it as an organism. Its systems of knowledge and action, built up over repeated exposure to some kind of object, and a developing skill in dealing with it, are continually reinforced by the cycle described in Figure 1. Here the A-B-X concept becomes less abstract as I put some flesh on it.
The Importance of Distributed Intelligence
One of the truly original thinkers in 20th century social science was Gregory Bateson. His continuing preoccupation was the nature of mind (G. Bateson, 1972, 1979). Versed in cybernetic theory, Bateson perceived the profound error in the Cartesian view of mind as individual consciousness, physically lodged in a brain. The mind, he said, "is not limited by the skin" (1972, p. 454). Instead he perceived mind to be "immanent" in the interaction of people with their environment. Batesons daughter, Mary Catherine, uses the example of a man with a scythe to illustrate what he is getting at (M.C. Bateson, 1977 p. 65): "A man with a scythe is constrained by the form of the scythe; indeed his own body motion is informed by the curves of his tool, a concrete proposition about the interlocked movement of man and tool through deep growing fields across the generations; as time passes, his own musculature will become a record of the scythes teaching, first in stiffness, then in emerging grace and skill. We need time to understand this system, to get beyond seeing it as simply instrumental."
She goes on to observe that "the particular scythe, unlike the peasant who wields it, cannot be thought of as a system in which mind is immanent, but mind is immanent in the system man plus scythe" (p. 66). Her point is that "in our involvement with the non-persons we encounter in the environment, we are not alone, we are not subject and these others object, but somehow we are involved in a dialogue. Mind is present, immanent in the situation, and we have the choice between seeing ourselves as parts of a thinking system greater than either, or attributing mind to the other" (p. 67).
Let us consider how this principle works out in practice.
Although Barley (1996) bemoans the lack of contemporary empirical studies of ordinary work ("we know more about yesterdays work than we do about todays") there is in fact quite a bit of good research that is concerned with the quotidian business of dealing with a practical world, as part of a working team. What do we learn from this literature? Basically what emerges is an image of the situatedness of purposive activity and the distributed character of human knowledge--very much like the process I outlined in Figure 1: oriented to circumstances, problem-centered with respect to a situation, and developing both knowledge of the situation and a response to it by drawing on the distributed knowledge and skills of the working team.
Consider, for example, field research reported by Heath and Luff (1992, 1996) where the object of study was the computerized control system of the London Underground (with its network of subway lines, stations, and trains). Their research, like that of many others among the new generation of ethnographic investigators, consists of detailed observation of actual work performance, both by direct observation on site, and after the fact, through analysis of video recordings. What they report is what they call the "situated" character of system use, contrasting strongly with a conventional technical perspective on human-computer interaction which has tended to emphasize individual cognitive skills and competencies. On the contrary they find that "tasks are accomplished in and through interaction" and that "the accomplishment of specialized tasks, and the conventional use of complex technologies to support these activities, are dependent on a realm of tacit interactional competencies that inform the very production and intelligibility of organizational conduct" (Heath & Luff, 1996, p. 127).
So crucial is tacit knowledge, in fact, that it is difficult to train new personnel, who are, as a result, "not explicitly taught, but rather left ... to learn in situ" (p. 119). It is not just that the tasks are individually complex, involving as they do a complicated network of train arrivals and departures, and a mastery of sophisticated equipment. The difficulty (and why many trainees initially fail) "derives from the ways in which tasks are systematically coordinated in real time with the actions and activities of colleagues" (p. 119). "Control Room personnel," they observe, "are continually and unavoidably, implicitly and explicitly, gathering and distributing information to each other concerning the "current" operation of the service" (p. 122). Practice and reasoning, they conclude, form what they call "a complex web," which once it has been mastered, then becomes "taken for granted," remaining, necessarily, "an unexplicated resource, "seen but unnoticed" --the foundation of collaborative work and crisis management (p. 119).
Goodwin and Goodwin (1996) equally emphasize both the highly situated character of work and its interdependencies involving both complex sources of information, technically mediated, and what others are doing. Their study focuses on the baggage handling facility of a mid-size airport. Among other things, "situatedness" means for them that "different work positions place the same physical object, a particular airplane, within different webs of accountability" (Goodwin & Goodwin, 1996, p. 65). It is this "plurality of perspectives," where any single category such as "destination" has separate senses that nevertheless "have deep and overlapping connections with each other" that makes the management of airline operations possible. As they say, "For airport personnel, planes do not stand alone as isolated objects. Instead they are defined by their positions in larger webs of activity" (p. 62). Nobody has a "single all-encompassing view of what is happening in the airline" (p. 68). Instead, "one finds multiple, diverse local perspectives, each constituted through the combination of a specific array of tasks, an ensemble of tools for performing those tasks, and an entrainment of workers bodies that encompasses not only their muscles but also phenomena as minute as acts of perception embodied in momentary glances" (pp. 68-9).
They see in this a radical shift of perspective on how work gets done; as they observe (p. 88), "despite a few notable exceptions, the contemporary social sciences typically conceptualize cognition within a Cartesian framework, as something located inside the individual mind, or in Searles (1990) elegant phrase, in "brains in vats." But these authors are advancing a different view: "All of the data examined in this chapter have displayed the interdependence of cognitive processes, tool use, and social organization." For them, cognitive operations are not located in any single mind but instead "emerge through time as a contingent social process" where "cognitive artifacts" (displays, tools) play an important part.
For Suchman (1996, p. 35), "the physical place of a worksite comprises a complex of equipment and action, of spatial orders produced in and informed by the knowledgeable practices of setting members." To her, "place is constituted by, rather than the container for, culturally, historically, and locally meaningful forms of lived activity." Suchman, as for Lave (1988, p. 159) whom she cites on this point, sees problem solving as "part of an articulatory phenomenon constituted between persons-acting and the settings of activity." Dilemmas, Suchman continues (p. 56), "are not so much solved as they are "dissolved" through structuring resources inventively employed."
Hutchins (1995), in his discussion of situated and distributed knowledge, traces the steps that go into fixing the location of a naval vessel as it enters San Diego harbor. He observes first how specialized sighting equipment is pointed at certain landmarks. >From the marks on the screen, the sailors communicate the data as coordinates to a central chart room where it is inscribed on a naval chart, to establish a "fix." The whole operation is transparent, very little individual cogitation is required, and yet the result is "navigation". What makes it cognition is not that the computational cycle occurred in someones head or was done by a computer, but that the output conformed to the requirements of the task as an effect of representational transformation. Hutchins proposes that we think of navigation, and other kinds of work, as a form of "distributed cognition."
Weick and Roberts (1993) similarly report on the dynamics of landing ("recovering") airplanes on a naval aircraft carrier. Many, many people are involved, the tolerances are terrifyingly small, and much of the crew consists of green twenty-year-olds. Yet somehow the task is accomplished flawlessly, most of the time. Weick and Roberts go on to argue for a kind of "group mind." As they point out, nobody knows all the parts of the skills involved and yet the group is enabled to perform as if it were "intelligent." The intelligence, then, is not located in any individuals brain but is, as the Batesons would argue, "immanent" in the interaction of people, technology and environment.
To summarize, each of these studies illustrates steps in the argument I am developing (see Figure 1): (1) the definition of a situation in normal contexts of work is typically an interactive process of sense-making, drawing on the dispersed stores of individual knowledge and skills in order to generate a collective understanding or "distributed cognition"; (2) the situation is resolved when the group has become an effective system of action, with distributed roles; (3) each work group is characterized internally by negotiated agencies, resulting in hierarchy, that is, the group is locally A-B-X; (4) because of their focus on a common object and their temporal continuity as a defined work group, such groups develop significant self-organizing internal mechanisms, independent of their placement within larger structures; (5) each work group described in this literature is in turn embedded in a larger framework of action, where it figures as a B-X component of an encompassing A-B-X system.
Let us now consider briefly how to conceptualize this latter embedding, within the context of coorientation theory. I see it as an instance of what I call imbrication.
The Role of Imbrication in Stabilizing Organization
Consider once again how a principal-agent A-B-X relationship is formed (Figure 4). On the fractal principle, it could be argued that the B-X link might itself be reinterpreted as an A-B-X triplet, where the B in question is some tool, or instrument, or machinery whose function is to realize the purposes of A, the human actor (Engeström, 1990). To avoid confusion let us term this new non-human agent B'. The B'-X relationship may then in turn be expanded to produce B''-Xs and B'''-Xs, and so on. Every time you drive a car you mobilize a whole embedded network of instrumentalities. If the accomplishment of X lends itself to automation then the original B can be sidelined and the interpersonal relationship of A and B is attenuated, and may even vanish. In many respects, this attenuation is the object of much system design, for example, the replacing of a teller in a bank by an ATM.
The A-B link might also, similarly on the fractal principle, be rewritten as an A-B-X system, where A has enlisted an agent, let us say a set of written rules and regulations, to stand in for the principal. Let us term the rules and regulations A'. Or the A-B's may be multiplied to produce the classic "chain of command" of bureaucratic hierarchy. In both cases the salient relationship is A'-B-X, where A' is a surrogate of A, now human, and again the interpersonal character of the A-B exchange has been attenuated.
I think of this recursive (and fractal) embedding of multiple A-B-X triplets, one within the other, as an instance of tiling, or imbrication. The structure I have in mind is illustrated by the shingles on a roof, by the scales on a fish, and by the leaves on a tree. Architecturally, there is an advantage in joining elements where each depends on another, and supports another (as in a brick wall, for example). Imbrication, in its organizational application, creates structures that are durable, disciplined and impervious to outside influence (try penetrating the imbricated structure of a government bureaucracy, for example: as Kafka realized, it is practically impossible to find where any decision gets taken).
A significant effect of imbrication is to mask the basic power relationships within the organization, and to produce a more stable environment for the carrying on of its business. It could be thought of as infrastructure (B-X type embeddings) or institutionalization (A-B type embeddings) .
What I now turn to is the last of my questions, enunciated earlier: How does an organization come to have an identity as an actor, independent of any of its members? Here I continue to draw on the same principles I have been arguing for throughout, but now employed in a different context.
In 1915, Émile Durkheim, then in the twilight of his productive career, published The elementary forms of the religious life (Durkheim, 1960). It is a curious work, in that it offers an analysis written by a French intellectual (and first occupant of a chair in sociology in his country) of the practices of the aboriginal population of Australia, a society which he had not himself visited but about which he had read. And in spite of the title, it is really more a book about the foundations of society than about religion. In one section of the book (Part II, Chapter 7), however, Durkheim outlines what has been interpreted as a theory of organizational communication (Rothenbuhler, 1993a, 1993b).
According to Durkheim, societies, especially ones as loosely associated as the widely scattered nomadic peoples of pre-European Australia, run the risk of becoming fragmented, and losing their sense of identity as a society. Peoples view of things is molded in a circle of family, friends and fellow workers with whom they are intimately associated in dealing with a common environment. It is here that ones beliefs and attitudes take form. The shaping of beliefs occurs within a social world of continuing interaction. But in a society where people spend most of their time in small communities, and little time in communion with larger groupings, as Durkheim argues was the case in aboriginal Australia, the risk of fragmentation is obvious. So Durkheim turns his attention to the mechanism by means of which a larger unity can be attained, and a society produced which transcends the strictly local experience of scattered micro-societies.
First, he observes, the glue which holds societies together is the necessity of working collaboratively, in order to accomplish many important activities, which are beyond the capacity of any single individual or even a single group. (I would think of this as the economic imperative, that I discussed at the beginning of this article.) The power to achieve certain goals is lodged in collective cooperation, and it is this power that both rewards the individuals by bringing them otherwise inaccessible benefits and simultaneously disciplines them and limits their freedom. The collective has, however, in and of itself no voice to make its discipline felt and respected, until someone speaks in its name. When this person is recognized as embodying the will of the collective in his or her speech, then what he or she says is endowed with what Durkheim calls "power."
How is this legitimization achieved? Durkheims answer is to introduce an intermediary agency that he calls "opinion." Out of the multiple overlapping conversations of the many communities making up a society there emerges an amorphous but powerful agency that corresponds to no single set of beliefs but that in some hard-to-define way is recognizable, in spite of its diverse strains of thought, as standing for the collective wisdom of all. It can be recognized, however, only when it is voiced. Its voicing embodies authority when it is understood to state authentically the opinion of the whole collectivity. When voicing is authentic, power is attributed to the individual who has successfully voiced the general opinion. This embodiment of the collectivity is enabled to enforce discipline over recalcitrant members--a discipline needed to make the society as a whole an effective agency for undertaking great tasks. A corporate actor has been produced, and given legitimacy --what Callon and Latour (1981) call a "macro-actor."
There are two ways we can read Figure 8. Its columns portray organization in two guises. In one, the right-hand column, organization is seen to be a heterogeneous dispersion of locally situated communities of work, each enfolded within its own circle of discourse, occupied with its daily activities, supported by its own rich knowledge system, spoken and unspoken--much like the organizations described by Heath and Luff, Goodwin and Goodwin, Suchman, Hutchins, and Weick and Roberts. In this context, knowledge is both shared and distributed, to the extent that it is analyzable as a property of the conversation, and individual-specific, to the extent that it takes the form of opinion. To the degree that the system is self-organizing its autonomy is indigenous (Krippendorff, 1999): inherent in its ongoing activities.
The left-hand column presents organization as an entity that speaks for the corporate society as a whole, translating the latters collective view ("opinion") into an organizational text, or symbolic representation, voiced by a person, or representative. This process of representation (in that it is both a text since it represents a collective intention and a voice since it enunciates the intention, verbally or in writing) constitutes the organization as a single actor, capable of enforcing its discipline on its members. It is the existence of a text that constitutes the collective entity as having an intention; it is the spokesperson (or persons) who gives it its diffusion, who, in "authoring" it, instantiates its authority. There are thus two phases in the production of an organizational identity: a cognitive symbolization of organizational intention, and a pragmatic utterance of it .
This latter perspective incorporates a view of organization that Sachs (1995) describes as "organizational, explicit" as opposed to that which I began by considering, that is, in her terms, "tacit, activity-oriented."
Read as rows, Figure 8 takes on a different signification. The upper row describes an interaction in which an unequal exchange is accomplished, involving on the one hand a governor or governing body and on the other the many working groups of the organization. This reading is what I refer to as a corporate A-B-X system, and it expresses the power structure of the organization. Leadership is legitimate when it establishes itself as an authentic A, whom the rest of the members of the organization, the B, can accept as their true spokesperson. The exercise of power is thus in the realm of "doing," or pragmatics: It is something to be achieved in practice. The lower row is the domain of the cognitive, or, as I prefer to call it, the "textual," since it here that a conceptual translation from the heterogeneous "opinion" of all the members into a homogeneous "symbolic representation" of it takes place, to become, once voiced, the univocal expression of the point of view of the society, or organization, as a whole. Again, its authenticity may be reasonably contested.
It is this expression of the communitys intention in the voicing of the agent who serves as its spokesperson that I postulated to be one of the boundary transactions necessary to the constitution of an A-B-X system. We thus envisage the arrows of Figure 8 as a set of inter-related displacements, or translations (Taylor, Cooren, Giroux & Robichaud, 1996). If the upper row is the domain of the pragmatic and the lower row that of the cognitive, then the dividing line between them is purely instrumental; the pragmatics of communication are already virtual in the cognitive representation of the text, and vice versa. Knowledge and action are interdependent, the two sides of a single coin. We could think of the text acting through management who figure thus as mere spokespersons, with the same plausibility as we think of management acting through its texts which are thus thought of as mere instrumentality--intertextuality versus intersubjectivity, an issue of worldview, in other words (Taylor, 1993). Management and its texts are conceived of as a fusion: an A/B where A (the spokesperson) is impotent without B (the text), and B is inert without A.
There has often in the past been a tendency to emphasize what
might be called the "photographic" function of language
as a way of recording, symbolically, an image of the world, so
that it can be stored and transmitted. This approach is, however,
an incomplete view of its representational function. To "represent"
means two things:
The use of language in a conversational context regularly, systematically and simultaneously mobilizes both of these senses of "represent": Language is as much a tool of action as it is of understanding.
Asking Some Questions that Might Lead to Productive Research
I have constructed in this article a model of organization that is centered on the concept of agency. It is important, however, to clarify the restricted use that I make of this term, a restriction that has perhaps remained implicit throughout my discussion. I mean agency within the context of communication. It is fair to say that agency is of much broader application than its application within the discursive situations of human communication. In a real sense agency is everywhere. A popular song can be thought of as an agent of the singer, or of the band, or of the impresario, or of the radio station; the singer, band, impresario and radio station as agents of the composer; the composer as an agent of a publishing house, the publishing house as agent of an investor, and so on, indefinitely. In this much broader sense of the term every object and event is the carrier of multiple agencies. Deciding what is agent and what is not becomes an arbitrary decision that the observer makes, depending on how he or she has defined the system.
The agency that I have been concerned with is that which is given reality within the discursive formations of some community of talk. It is such a restriction that I have been endeavoring to highlight in my discussion of the A-B-X system. In this context of application, the agency relations that I take to be the building blocks of organization have become recognizable to a community, and are realized concretely in their conversations and texts, in part because the X is understood to be shared, in part because the authoritative voice of the community, the A, is understood to be its authentic agent. We could, as analysts, take the same organization and find in its activities evidence of many other unrecognized agencies at work, a whole virtual network of them. I do not mean to deny this, but it is not the topic I am addressing.
What stirs my imagination is the phenomenon of emergence of a historical entity that we have learned to call an "organization" that to me resembles the poking up of some volcanic island out of the vast sea of undifferentiated agency that characterizes the sweep of human history. Such "islands" are an effect of discourse-in-communication. They emerged within the historically short period of time from roughly 1860 to 1960. They could now sink back into the enveloping sea as fast as they rose up, if the circumstances of communication--and thus of agency stabilization in discourse--changed.
If the survival of large organization remains problematical, as some of the observers I cited at the beginning of this article believe, then we should inquire into the factors that serve to perpetuate organizational unity. One of those factors, I believe, is the combined effect of local self-organization, in combination with the tendency of large systems to become imbricated. Imbrication, supported by local self-organization, has however an inertial effect, and in fact may become a serious brake on innovation, as authors such as Zuboff (1996) and Hammer and Stanton (1995) have pointed out. The other factor that serves to perpetuate organization is the moral force of the A-B-X relationship: People do give their loyalty to the societies and the organizations of which they are a part, as long as they feel themselves to be a part of them. If, as I began by suggesting, we are moving in the direction of a less imbricated form of organization (knotworking, virtual teams, etc.) then it would seem to be this second factor of moral force that takes on increasing importance. But, empirically speaking, does it? The unifying strength of moral force appears to me to be an open question, given the state of our present knowledge.
If indeed there are two factors that may explain the stabilization of organizational form and identity, one indigenous (to use again Krippendorffs term) to the imbricated systems of interrelated work groups, the other a Durkheimian effect of the objectification and voicing of its collectively distributed intelligence, then it becomes reasonable to ask, as a practical researcher, what is driving change in the contemporary world. One causal factor is clearly technology. A principal preoccupation of current system design is the capturing of agency in software design. Applied to practical domains of work, the result is a very rapid increase in the level of imbrication, as more and more tasks get broken down into their component elements and are computerized. At one level the result is an enhancement of stability, especially given the reliability of much new technology, but it is not immune to risk, as the Y2K phenomenon illustrates. There are limits to imbrication as Hutchins and Klausen (1996) point out in their discussion of the operation of a commercial airliner, and disadvantages to infrastructure (Star & Ruhleder, 1996; Bowker & Star, 1999).
This is half the story. The other half is the effect of the radical reshaping of our systems of telecommunications that technology is effecting. The virtual team that Shockley-Zalabak (1999) described in her research found itself in what might be called a "A-B deprived state" as a consequence of its total dependence on mediated interaction. The mobility of enterprise, too, is bound to have effects on the sense of community that people experience and this reduced sense of community in turn renders the exercise of its authority problematical (Perrucci & Stohl, 1998).
It used to be part of ones accepted vision of the world that once one had become an employee of an organization one was there, in all probability, for the rest of ones working career, until the pension took over. Companies such as General Motors, Ford, AT&T, IBM, Westinghouse and their like offered a promise of permanence and security. We make a good deal these days of Japans assumption of lifetime employment in a firm but it was not so long ago that our society in North America was of the same disposition (and Japan is changing too). Somewhere along the line this commitment was abandoned (Bennett, 1990), partly because companies and governments learned to downsize, partly because worker mobility became part of our culture, particularly among the better educated and more professional workers. It seems harder, somehow, to give ones loyalty to a "holding company" than it used to be to the great corporations. If legitimization in some community is one boundary condition of organizational stability, as I have hypothesized, then that consideration has greatly changed as we enter the work world of the 21st century and the 3rd millennium.
There are here, it seems to me, themes to be addressed in research.
Changes in patterns of communication and the erosion of frontiers of discursively stable and identifiable universes of conversation have two effects that I have been attempting to highlight in this paper: They alter the basis of distributed intelligence and thus what we know, and they affect the ability of a society to construct its own hierarchical basis and to discover its authentic voice. As I look around at todays society I think it could be argued that there is evidence of both these trends. The end of the cold war and the greatest prosperity that any society has ever known have not brought about universal peace, or a growing sense of security, or even more happiness. If anything, violence is becoming more widespread than ever, reaching into the heartland of affluent middle class communities whose members used to think of it as part of the culture of poverty, and a trouble confined to the ghettos.
Globalization is upon us, and we need to understand its consequences. However, in another sense, globalization has been going on without interruption for the some five or so millennia of recorded human history. It may be accelerating but it is far from new. Yet in many respects 20th century communication science has tended to treat organization ahistorically. There is a fascinating parallel here with a sister science of communication, linguistics. Ferdinand de Saussure, the so-called "father" of modern linguistics, is credited with making a distinction between the synchronic and diachronic study of language. The latter, diachronic, dimension is historical, in that it deals with the dynamics of language change over time. The synchronic approach makes the simplifying assumption that language may be studied as a system with constant properties. Modern linguistics has largely gone down the synchronic path, turning away from the philological preoccupations of earlier generations. Much the same could be said of organization science, since the massive tome of Max Weber. Contemporary organization science has, like linguistics, been preoccupied with system. Saussure himself though was not arguing for the abandonment of diachronic studies, but a recognition that a perspective is involved (Thibault, 1997). I see a similar need for balance in our own field.
In this paper I too have outlined a synchronic theory of organization. It may now be time to think diachronically. I believe we can do so without abandoning the insights that a synchronic view provides and I have tried to lay the basis for such an inquiry. What I have argued in this paper is that, if organization is an island of stable hierarchical relationships rising out of the ongoing flow of horizontal transactions, then, when the boundary conditions of organization change, such a transformation will almost inevitably alter the nature of organization. I see this ongoing adaptation of organization as a topic that can be addressed scientifically.
I have obviously no more than scratched the surface of a larger problem. This issue of the Electronic Journal of Communication seems to have been inspired by a similar intuition. Its publication may thus mark an important turning point in the development of our science.
 Part of that frame knowledge is who they are: what Sacks
(1964-72) called "membership categories."
Note, however, that Figure 3 describes the building block of
organization, not organization itself. It applies equally well
to interpersonal and institutional forms of communication. That
which is specific to organization in the large remains to be developed.
My point is that, in the large as well as in the small, the same
communicational mechanism is operative.
Anton, C. (1999). Beyond the constitutive-representational dichotomy: The phenomenological notion of intentionality. Communication Theory, 9(1), 26-57.
Austin, J. R. (1962). How to do things with words. Oxford, UK: Oxford University Press.
Barley, S. (1996). Foreword. In J. E. Orr (Ed.), Talking about machines. Ithaca, NY: Cornell University Press.
Bateson, G. (1972). Steps to an ecology of mind. New York: Ballantine Books.
Bateson, G. (1979). Mind and nature: A necessary unity. New York: Bantam Books.
Bateson, M. C. (1977). Daddy, can a scientist be wise? In J. Brockman (Ed.), About Bateson. New York: E. P. Dutton.
Beniger, J. R. (1986). The control revolution: Technological and economic origins of the information society. Cambridge, MA: Harvard University Press.
Bennett, A. (1990). The death of the organization man. New York: William Morrow & Co.
Birch, D. (1987). Job creation in America:: How our smallest companies put the most people to work. New York: Free Press.
Birdwhistell, R. L. (1952). Introduction to kinesics. Louisville: Foreign Service Institute, Louisville.
Birdwhistell, R. L. (1970). Kinesics in context: Essays on body motion communication. Philadelphia: University of Pennsylvania Press.
Boulding, K. (1953). The organization revolution: A study in the ethics of economic organization. New York: Harper & Row.
Bowker, G. C., & Star, S. L. (1999). Sorting things out: Classification and practice. Champaign, IL: University of Illinois.
Bybee, J., & Fleischman, S. (1995). Modality in grammar and discourse. Amsterdam and Philadelphia: John Benjamins.
Callon, M., & Latour, B. (1981). Unscrewing the big Leviathan: How actors macro-structure reality and how sociologists help them to do so. In A. V. Cicourel & K. Knorr-Cetina (Eds.), Advances in social theory and methodology: Towards an integration of micro- and macro-sociologies (pp. 277-303). Boston: Routledge and Kegan Paul.
Chandler, A. D. (1962). Strategy and structure: Chapters in the history of the American industrial enterprise. Cambridge, MA: The MIT Press.
Chandler, A. D. (1977). The visible hand: The managerial revolution in American business. Cambridge, MA: Harvard University Press (Belknap).
Clegg, S. R., & Hardy, C. (1996). Introduction: Organizations, organization and organizing. In S. R. Clegg, C. Hardy, & W. R. Nord (Eds.), Handbook of organization studies (pp. 1-28). London and Thousand Oaks, CA: Sage.
Clemons, E. K., & Row, M. C. (1989). Information technology and economic reorganization. Wharton School of Business, University of Pennsylvania.
Coase, R. H. (1937). The future of the firm. Economica, 4, 386-405.
Derrida, J. (1986). Declarations of independence. New Political Science, 15, 7-15.
Dewey, J. (1944). Democracy and education. New York: The Free Press. (Original work published 1916)
Djilas, M. (1957). The new class: An analysis of the communist system. New York: Praeger.
Durkheim, É. (1960). Les formes élémentaires de la vie religieuse, translated as The elementary forms of the religious life (J. W. Swain, Trans.). Paris, New York: Quadrige/Presses Universitaires de France.
Engeström, Y. (1990). Learning, working and imagining. Helsinki, Finland: Orienta-Konsultit Oy.
Engeström, Y., Engeström, R., & Vähäaho, T. (1999). When the center doesn't hold: The importance of knotworking. Paper presented at the International Communication Association, Organizational Communication Division, San Francisco.
Giddens, A. (1976). New rules of sociological method: A positive critique of interpretive sociologies. London: Hutchison.
Giddens, A. (1979). Central problems in social theory: Action, structure and contradiction in social analysis. Berkeley and Los Angeles: University of California Press.
Giddens, A. (1984). The constitution of society. Cambridge, UK: Polity Press.
Goffman, E. (1959). The presentation of self in everyday life. New York: Doubleday Anchor.
Goffman, E. (1974). Frame analysis. New York: Harper-Colophon.
Goffman, E. (1981). Forms of talk. Philadelphia: University of Pennsylvania Press.
Goodwin, C., & Goodwin, M. H. (1996). Seeing as situated activity: Formulating planes. In Y. Engeström & D. Middleton (Eds.), Cognition and communication at work, (pp. 61-95). Cambridge, UK: Cambridge University Press.
Gouldner, A. (1979). The future of intellectuals and the rise of the new class. New York: Seabury Press, Continuum.
Greimas, A. (1987). On meaning: Selected writings in semiotic theory. Minneapolis: University of Minnesota Press.
Hall, E. T. (1959). The silent language. New York: Doubleday.
Hall, E. T. (1966). The hidden dimension. New York: Doubleday.
Hall, E. T. (1976). Beyond culture. Garden City, NY: Doubleday.
Halliday, M. A. K. (1985). An introduction to functional grammar. London: Edward Arnold.
Hammer, M., & Champy, J. (1993). Reengineering the corporation: A manifesto for business revolution. New York: Harper Collins.
Hammer, M., & Stanton, S. A. (1995). The reengineering revolution. New York: Harper Collins.
Heath, C., & Luff, P. (1992). Collaboration and control: Crisis management and multimedia technology in London underground line control rooms. Computer Supported Cooperative Work, 1(1-2), 69-94.
Heath, C., & Luff, P. (1996). Convergent activities: Line control and passenger information on the London Underground. In Y. Engeström & D. Middleton (Eds.), Cognition and communication at work (pp. 96-129). Cambridge, UK: Cambridge University Press.
Heidegger, M. (1996). Being and time: A translation of Sein une Zeit (J. Stambaugh, Trans.). Albany, NY: State University of New York Press.
Husserl, E. (1993). Cartesian meditations (D. Cairns, Trans.). Boston: Kluwer.
Hutchins, E. (1995). Cognition in the wild. Cambridge MA: The MIT Press.
Hutchins, E., & Klausen, T. (1996). Distributed cognition in an airline cockpit. In Y. Engeström & D. Middleton (Eds.), Cognition and communication at work, (pp. 15-34). Cambridge, UK: Cambridge University Press.
Innis, H. (1950, 1972). Empire and communications (Rev. M. Q. Innis). Toronto: University of Toronto Press. (Original version published 1950)
Innis, H. (1951). The bias of communication. Toronto: University of Toronto Press.
Jackendoff, R. S. (1972). Semantic interpretation in generative grammar. Cambridge, MA: MIT Press.
Jacques, R. (1996). Manufacturing the employee: Management knowledge from the 19th to 21st centuries. London: Sage.
Krippendorff, K. (1999). Beyond coherence. Management Communication Quarterly, 13(1), 135-145.
Krippendorff, K. (1998). Ecological narratives: Reclaiming the voice of theorized others (Working Paper ). Annenberg School of Communication, University of Pennsylvania, Philadelphia PA.
Latour, B. (1994). On technical mediation--Philosophy, sociology, genealogy. Common Knowledge, 3, 29-64.
Lave, J. (1988). Cognition in practice. Cambridge, UK: Cambridge University Press.
Livet, P. (1994). La communauté virtuelle: Action et communication (The virtual community: action and communication). Paris: L'éclat.
March, J. G., & Simon, H. A. (with the collaboration of H. Guetzkow). (1993). Organizations (2nd ed.). Cambridge, MA: Blackwell.
Marr, D. (1982). Vision: A computational investigation into the human representation and processing of visual information. New York: W.H. Freeman and Company.
Newcomb, T. M. (1953). An approach to the study of communicative acts. Psychological Review, 60, 393-404.
Noller, D. K. (1997). Communication and virtual organization. Unpublished doctoral disseretation, University of Colorado at Boulder, Boulder CO.
Palmer, F. R. (1986). Mood and modality. Cambridge, UK: Cambridge University Press.
Perrucci, R., & Stohl, C. (1998). Reconsidering organizational commitment: The community/organization partnership. Paper presented at the National Communication Association, Chicago, IL.
Porter, M. E. (1980). Competitive strategy. New York: Free Press.
Porter, M. E. (1985). Competitive advantage. New York: Free Press.
Putnam, L. L., Phillips, N., & Chapman, P. (1996). Metaphors of communication and organization. In S. R. Clegg, C. Hardy, & W. R. Nord (Eds.), Handbook of organization studies (pp. 375-408). London and Thousand Oaks, CA: Sage.
Rothenbuhler, E. W. (1993a). Argument for a Durkheimian theory of the communicative. Journal of Communication, 43(3), 158-163.
Rothenbuhler, E. W. (1993b). Reading Durkheim's Elementary Forms for communication theory. Paper presented at the Speech Communication Association, Annual Convention, November 18-21.
Sachs, P. (1995). Transforming work: Collaboration, learning and design. Communications of the ACM, 38(9), 36-44.
Scheflen, A. E. (1965). Stream and structure of communicational behavior. Philadelphia: Eastern Pennsylvania Psychiatric Institute.
Scheflen, A. E. (1973). How behavior means. Garden City NY: Doubleday.
Scheflen, A. E., & Ashcraft, N. (1976). Human territories: How we behave in space-time. Englewood Cliffs, NJ: Prentice-Hall.
Searle, J. R. (1989). How performatives work. Linguistics and Philosophy, 12, 535-558.
Searle, J. R. (1990). Collective intentionality and action. In P. R. Cohen, J. Morgen, & M. E. Rollsik (Eds.), Intention in communication (pp. 401-415). Cambridge, MA: MIT Press.
Senge, P. M. (1990). The fifth discipline: The art and practice of the learning organization. New York: Doubleday.
Shockley-Zalabak, P. (1999). Protean places: Teams across time and space. Paper presented at the International Communication Association, San Francisco, CA.
Smith, R. C. (1993). Images of organizational communication: Root metaphors of the organization-communication relation. Paper presented at the International Communication Association, Washington, DC.
Star, S. L., & Ruhleder, K. (1996). Steps toward an ecology of infrastructure: Design and access for large information spaces. Information Systems Research, 7(1), 111-134.
Suchman, L. (1996). Constituting shared workspaces. In Y. Engeström & D. Middleton (Eds.), Cognition and communication at work, (pp. 35-60). Cambridge, UK: Cambridge University Press.
Taylor, J. R. (1993). Rethinking the theory of organizational communication: How to read an organization, Norwood, NJ: Ablex.
Taylor, J. R. (In press). The other side of rationality: Socially distributed cognition, Management Communication Quarterly, 13 (2).
Taylor, J. R., & Van Every, E. J. (1993). The vulnerable fortress: Bureaucratic management and organization in the information age. Toronto: University of Toronto Press.
Taylor, J. R., & Van Every, E. J. ( In press). The emergent organization: Communication as its site and surface. Mahwah, NJ: Lawrence Erlbaum.
Taylor, J. R., Cooren, F., Giroux, N. & Robichaud, D. (1996). The communicational basis of organization: Between the conversation and the text. Communication Theory, 6 (1), 1-39.
Taylor, J. R., Groleau, C., Heaton, L., & Van Every, E. (In press). The computerization of work: A communicational perspective. Thousand Oaks: Sage.
Thibault, P. J. (1997). Re-reading Saussure: The dynamics of signs in social life. London and New York: Routledge.
Weber, M. (1968). Economy and society: An outline of interpretive sociology. New York: Bedminster Press. (Original work published 1922)
Weick, K. (1979). The social psychology of organizing. (Rev. ed.). Reading MA: Addison-Wesley.
Weick, K. E. (1985). Sources of order in underorganized systems: Themes in recent organizational theory. In Y. S. Lincoln (Ed.), Organizational theory and inquiry (pp. 106-136). Beverly Hills: Sage.
Weick, K. E., & Roberts, K. H. (1993). Collective mind in organizations: Heedful interrelating on flight decks. Administrative Science Quarterly, 38, 357-381.
Whyte, W. (1956). The organization man. New York: Simon & Shuster.
Williamson, O. E. (1972). Markets and hierarchies: Analysis and anti-trust implications. New York: Free Press.
Williamson, O. E. (1977). The rise of modern business enterprise: Implications for industrial organization. Philadelphia: University of Pennsylvania Press.
Zuboff, S. (1988). In the age of the smart machine: The future of work and power. New York: Basic Books.
Zuboff, S. (1996). The emperor's new information economy. In W. J. Orlikowski, G. Walsham, M. R. Jones, & J. I. DeGross (Eds.), Information technology changes in organizational work, (pp. 13-17). London, England: Chapman & Hall.
Copyright 2000 Communication Institute for Online Scholarship, Inc.
This file may not be publicly distributed or reproduced without written permission of the Communication Institute for Online Scholarship, P.O. Box 57, Rotterdam Jct., NY 12150 USA (phone: 518-887-2443).