organizational communication final essay

kkkkpcy

W12Tayloretal._1996_Thecommunicationalbasisoforganization.pdf

Home >Social Science homework help >organizational communication final essay

Communication Theory

S i x : One

James R. Taylor Fran ois Cooren

Daniel Robichaud Nico c e Giroux

February 1996

Pages: pp. 1-39

The Communicational Basis of Organization: Between the Conversation and the Text

Recent work in the sociology of scientific knowledge leads us to believe that organization exists in the interpretive processes of its members. A theory of communication is presented which views the communicational process as a double translation, from text to conversation and conversation into text. Using this theory as a basis, a conception of complex and extended forms of organization is developed which demonstrates that the text-conversation cycle underlies a complex process of networking out of which the identity of the organization emerges, and social structuring occurs, resulting in the division of labor and coordination. Implications of the theory are briefly considered for research.

When Gilbert and Mulkay (1984) were conducting their research into the unfolding of a scientific controversy, recorded from the recollections of the researchers involved, they uncovered a methodological limitation in the analysis of this kind of interview data: In the words of the authors, “The difficulty with taking any collection o f . . . statements produced by participants as literally descriptive of social action is the potential variability of participants’ statements about any given action” (p. 6). I t is not just that different witnesses give inconsistent accounts about events they have jointly seen or lived, because this would still mean that since some might be more accurate than others we could expect, by careful sifting of the evidence, to eventually arrive at a fairly faithful historical reconstruction. The authors’ point is more radical than this. In their study, a single respondents’s account of the same event would fluctuate, not haphazardly, but depending on the interview context or other dis- course situation. As they said: “Similarities between different statements are just as likely to be the consequence of some similarity in the context of linguistic production as of similarity in the actions described by those statements” (p. 7).

Social action, they went on to argue, is not “directly observable” (p. 8); direct observation does not “free the observer from reliance on the potentially variable discourse of participants.” The reason is that “social

Communication Theory

activities are the repositories of multiple meanings” (p. 9, italics added). The meaning of action is not constant, nor even an initially objective happening o n which subsequent interpretations may be overlaid; the meaning of an action is inherently “variable and context-dependent” (p. 9). There is thus no zero point in social action: The onion has no core. “It seems best, then, to conceive of the meaning of a social action, not as a unitary characteristic of acts that can be observed as they occur, but as a diverse potentiality of acts which can be realized in different ways through participants’ production of different interpretations in different social contexts” (p. 9, italics added). “Actors,” they found in their analy- sis, “continually reinterpret given actions as their biography unfolds” (p. 9). “The social world is not composed of a series of discrete, one- dimensional actions which can be more or less accurately represented (P. 1 0 ) .

We could, of course, read these remarks as merely an observation on the limits of interviewing as a technique, or even as an epistemological statement about discourse-based communication research, but if we did so, we would be missing a much more fundamental point they are mak- ing, that has to do with the ontology of organization-what organization actually is. Seen in this light, Gilberg and Mulkay are challenging one of the most conventional of assumptions in the organizational communica- tion field, something normally so taken for granted by researchers that it goes unquestioned, namely the reality of organization as existing inde- pendently of communication - a constant presupposed ground to ‘the communicational figure (Smith, 1993). Theirs was also an organiza- tional study. (of scientific laboratories) but they were forced to the con- clusion that there was no way they could produce a definitive, scientifi- cally defensible “objective” account of the organization in question. The organization was as variable as the respondents.

It was not the sequential occurrence of the reported events as such that was problematic-on that score, there was more or less general agreement - but rather their organizational meaning. What made the events “organizational” was not an inherent property of the events per se; instead, it was in their interpretation that they become comprehensibly organizational, as we understand the term, that is, endowed with roles, actors, purposes, missions, plans, relationships, and so on. Such inter- pretations were not constant, not even for a single respondent, but varied according to the communicational context of their production. The ac- tivities per se were, as Gilbert and Mulkay said, “repositories of multiple meanings,” a “potentiality” leading to “participants’ production of differ- ent interpretations in different social contexts.” And it was not just a question of “activities”; the identity of the organization also depended not just on who the respondent was but also on when, and in what context of discourse, it was being described. The organization was itserf a “potentiality” a “repository of multiple meanings”

In other words, we err in thinking of communication as a transparent

The Communicational Basis of Organization

window on organization; the properties that we recognize as organ- izational are in the communicational lens, not in the object they are fo- cused on.

Consider a simple event, let us say, a strike of teaching assistants in a large university. The organizational significance of such an event will obviously vary depending on whom you talk to: president, provost, dean, chair, professor, union representative, ordinary students, parents, townsfolk, newspaper reporters, and so on. But it will also vary depend- ing on circumstance: The Provost’s version will change depending on whether he or she is ‘dealing with a student delegation, the board of governors, or the press. As these versions vary in each of these contexts what is being said is not necessarily insincere, or duplicitous; contexts already dictate how accounts are to be rendered. A week later, the ac- tions taken at the start of the strike will take on a quite new meaning: What was initially seen to be prompt action will now look like a “cover- up”; initial indecision may turn out to be no longer waffling but “flexi- bility.”‘

So how does the researcher who wants to use this event as a way of describing university organization proceed? If we credit Gilbert and Mul- kay it would be futile to try to get to the bottom of the controversy, peel away the personal biases, and finally come up with a description of the “real” organization manifested through the events. Instead, we would have to recognize that the organization was being constructed through the interpretations of its members, not as isolated images but as parts of an ongoing negotiation of perspectives on the basis of which a more or less agreed-upon account might finally be produced. Even if something like a consensus were now to be found, it would still be a tentative common ground, hypothetical, always subject to future revision, given a new context of communication.

Developing a Theory of Communication If we take this line of argument seriously it poses two clear challenges. The first of these is to explicate a theory of human communication that makes evident why the observable events of communication are inherently context dependent and variable, from the point of view of their meaning as social action. The second is to show how an organiza- tion, nevertheless, is possible, both micro and macro.

The most common way to conceptualize communication in organiza- tional communication research has been as message exchange.’ The idea of “messaging” as the core of communication originates in information theory (Shannon, 1948). This theory allows the possibility of variable interpretations on the part of sender and receiver- noise, or equivoca- tion- but it treats such variability as a soluble problem, or something to be “corrected.” In such a theory we take for granted the theoretical possibility of perfect (“noiseless”) communication. It is here that we part company with message theory. However, by rejecting the hypothesis

Communication Theory

of nonequivocal communication we are, by this selfsame commitment, obliged to look beyond a messaging metaphor of communication.

The first objective of this paper is to present, in broad outline, an alternative theory of communication, one that is intended to be sensitive to just that dimension of the organizational experience that Gilbert and Mulkay made salient: its continuing construction in the interpretive ac- tivities of its members, situated in networks of communication. We d o not doubt the existence of properties of organization such as hierarchy, purposive performance, control and coordination; it is the locus of these properties that we question. We believe the organization is not in the activities as such, but in their interpretation. And the working through of an interpretation is a social process (in fact the social process we call “communication”) by means of which members both come to an understanding (however tentative) of what the events mean, organiza- tionally, while they simultaneously reconfirm their own positions in the network through the role they play in the interpretive process. We there- fore need to show how the organization emerges in the communication in two ways, both as a more or less shared understanding, and as a dynamic playing out of relationships where identities and roles get nego- tiated.

The process we will describe is, in contrast with messaging theory, not only closed, but also circular. The activities of members that concur in an interpretation are themselves subject to interpretation: As people collectively produce an interpretation they leave their actions open to interpretation. It is this dialectic that forms the basis of our theory. The interpretive activities constitute, in our language, a conversation, while the subject matter and goal of its interpretations are text: The conversa- tion begins as a scaffolding of text. However, because the activities of interpretation are themselves interpretable, the conversation is in turn the subject matter of text (indeed an envelope of texts). Each modality enfolds the other. It is in this dialectic of conversation and text that the organizing occurs. The conversation is mediated by text, the text by conversation. There is what we call a double translation: from the text to the conversation and the conversation to the text.

Our theory of communication must be capable of explaining the emer- gence and sustainability of large, complex organizations. The concept “orgunizution” is rich in ambiguities. A family is an organization. So is a small, autonomous work group. Interpersonal communication has strik- ing organizational properties. Football is an organized sport. Our atten- tion in this paper, however, will be centered elsewhere, on the kind of organization that is more usually the object of organizational communi- cation studies-large, extended enterprises that are a typical feature of the late twentieth-century industrial and governmental landscape in an era of globalization. The challenge that this presents is to show how a communication theory that originates in the explanation of conversa- tional dynamics can be amplified to take account of a quite different

The Communicational Basis of Organization

order of communication universe, that of large, geographically distrib- uted cooperating communities. We need to show both how interpersonal interaction becomes organized and, from within the same conceptual framework, how large-scale organization is thereby born and sustained. We share with Schegloff (1992) the sense that social structure (such as that which is supposed to characterize large, complex organizations) must always be grounded in the everyday talk of members, if it is to be assumed as meaningful at all: It always comes back to conversation. O n the other hand, we recognize the reality of the larger universe for people and believe that this too must be explained.

What makes such a venture problematic is again the thesis of action as inherently variable and context dependent. Conventional organizational theories of the kind generally taught in management schools make n o such assumption. Confronted with the question, “What does it mean to be organized?” they typically take communication as a given and get on to other thing^.^ Management theorist Mintzberg (1979), for example, noted that there are two requirements for the structuring of an organi- zation, division of labor and coordination. From a managerial per- spective, organizing involves getting people - employees - to do certain things, in conformity with some other people’s - management’s- objec- tives. Mintzberg listed five different ways this could be accomplished: through mutual adjustment (leaving the doers to work it out for them- selves- an exceptional case, according to him), direct supervision (most applicable for smallish groups performing well-defined tasks), and, for more diffuse contexts of work, the standardization of work processes (programming), of outputs (objective performance standards or quality control) and of skills (training). The first two involved interpersonal interaction; standardization, whatever form it takes, is more impersonal in character.

We are not management theorists. What he took as given, even cate- gorical (his introductory chapter is titled “The essence of Structure”), is for us what is most problematic in organizational communication re- search, indeed the organizational stakes: getting people to do and think what you want them to d o and think. This, to us, is eminently communi- cational, and is therefore to be explained, not assumed. Mintzberg’s category system strikes us as too prescriptive and cut-and-dried: He sees social action as objective. Yet he succeeded in outlining, in important respects, what must be explained by communication theorists if we are to be taken seriously as organizational researchers. We are not saying there is no organization, just that it is not an objective “thing.” That it must be born and recreated in the equivocal interpretations that are intrinsic to communication does not make it less real-just not “real” in the material sense. The challenge, therefore, is to take the outcomes he described and to show how they might be explained differently.

The first issue is to show how communication events, in supporting social action, come to be variable and context dependent. To this end,

Communication Theory

we propose a theory of communication that counterposes what we un- derstand to be its two central functions, namely to represent and to act.

Representation is accomplished through the production and compre- hension of text; action is mediated by text, but only when the text has been submitted to an interpretation. The text is not the action, but merely its token. The.action unfolds interactively in the context of a conversation. Communication thus has two modalities: text and conver- sation. In the ongoing translation from one modality into another, con- text is crucial (because context supports, and is necessary to, the inter- pretation of the text, not as representation but as action). In such a field of interpretation, variability is inevitable.

The second issue is to show how this theory can be extended to account not just for locally situated conversations but also for intercon- versational dynamics-what Weick (1979, 1985) has referred to as the “loosely coupled” side of organizational communication. This extension has more than one step. T o begin with, we need to consider how Mintz- berg’s “division of labor” and “coordination” can be explained, commu- nicationally, in a way that applies both to locally grounded conversa- tional interaction and to extended networks of conversation. T o accomplish this unification of perspectives, we introduce a concept called modulizution. In an action-based theory of communication, both perfor- mance and belief are grounded in communicational exchanges: What is to be done or thought (division of labor), and how (coordination), is already integral to ordinary conversation.

To apply the concept of modalization to account for more than inter- personal exchange (the subject matter of both speech act and narrative theory, our two initial points of reference), however, an additional phe- nomenon must be explained: how the organization, as a collective com- prising multiple members and interests, comes to speak with a single voice and can thus effectuate division of labor and coordination. How, in other words, does the organization become an actor? (We call this creating an organization.)

From these two preliminary clarifications, we then consider a process called distunciation by means of which the scaling up from locally situ- ated conversation to more complex networks of exchange is accom- plished. The mechanism at work, we postulate, is a sequence of succes- sive conversation-text translations, each of which operates to produce a greater degree of separation from the specific to the general, and thus provides for the emergence of a single organizational identity (and hence the accomplishment of division of labor and coordination for large groups of people, who may be widely distributed, geographically and over time). In this theory, although organization remains locally grounded in conversation, mediated by production and interpretation of text, it has the properties we expect of a complex organization. The real organization, we claim, is still conversationally mediated; the perception of a larger organization is part of the interpretive processes that such

The Cornrnunicational Basis of Organization

conversations depend on. What we are accustomed t o calling orgunizu- tion is at once actual (a network of conversations) and symbolic (a per- ceived actor, embodying the community as a whole). The symbolic di- mension is text mediated.

Such larger social systems (as they have sometimes been thought of) develop material tokens of their existence (buildings, logos, etc.), but they have in and of themselves no objective existence; they are a creation of the communicative translation of text (now thinking of buildings and such as instances of text) into action and hence into social meaning. Organization is thus seen to be the environment of communication (in supplying its interpretive ground, making possible the translation of text into action) but also seen as a product of communication. The process is cyclical.

The paper concludes with observations on the conduct of research. We are presenting what amounts to a discourse theory of organiza-

tion. A particular contribution of the present work is to show conceptual links between two broad classes of discourse theory that are not usually so considered: speech act and narrative theory. Those links will emerge in the presentation of our basic theory of communication.

The First Translation: From Text Into Conversation The making of a conversation involves a translation of text into conver- sation (Figure 1).

Questions are thus posed: Wh?t is text? And: What is conversation? Text In a conversation, the text is what is said. It is the content of the conver- sation: ideas couched in language (syntactics/semantics) and realized through speech or an equivalent channel (pragmatics).

For a text to be generated, there must be a physical support system by means of which the text can be inscribed (spoken or written), thus taking on material form and becoming accessible to others. In the face-to-face speech situation this “support system” or medium is usually the human body itself: voice (through the operation of the vocal cords), intonation and phrasing, gesture and body attitude. There are alternative support systems, even in conversation- paper, for example, or a blackboard or overheads. But although a form of materialization is indispensable to the

Conversation F Figure 1. The transla- tion of text into conver- sation.

Communication Theory

realization of text, it is not the physical medium on which text is in- scribed that makes it text. Text is mediated first by language; it is as discourse that text is apprehended, whatever the material mode of real- ization ~ h o s e n . ~ However, although text is structured in language, it only becomes “text” at the moment of its being produced. Before that it was just an idea-at best, a virtual text.’

One tempting way to think of a conversation, then, would be to see it as a string of texts, collaboratively produced by people who have posi- tioned themselves to interactively generate and collect a sequence of texts that, by common agreement, are supposed to cohere together, reason- ably well, by a quasi-logical principle of ordering where each succeeding text is supposed to link to the ones that preceded it-a kind of text- string. From this perspective, texts are the matter of conversation-the basic material out of which it is composed.6

There is, however, another perspective on the relationship between text and conversation. Although text may be the matter out of which conversation is fashioned, conversation is not, in essence, text at all: It is action (or more accurately, interaction, or transaction). The action could not occur in the absence of text production, to be sure, but the concept of action is not a constituent in how we define text (which is, as we have seen, ideationally based in language). Conversational translation requires text production but transcends it. A conversation, as we see it, is constructed out of speech-mediated acts. We understand such speech acts to be the building blocks of the actual conversation.’

For text to be transmuted into action, therefore, a translation must occur. Actualizing Text: The Translation of Text Into Action The main ideas defining speech as action were laid out by Austin (1962) in a series of lectures delivered at Harvard University in 1955. Austin’s central insight was that, when we communicate, there are simultaneously two things going on. On the one hand, our talk can be thought of as utterance, formulated in the speech code we call “language.” AusM dubbed such an utterance a locution. A locution describes a state of affairs, either virtual or real; this is why it is “semantic.” As a description of reality, it may be true or false. Text, when spoken is composed of locutions.

On the other hand, Austin argued, speech is not merely description, but something quite different, action. This dimension of utterance, that Austin termed illocutzonary, has not received much attention in the orga- nizational communication literature but it behooves us to pay closer heed, because it is in the illocutionary manifestations of speech that the organizing gets done.

Imagine the following exchange:

A. I’ll need that report for Monday morning, Jones. B. I’ll do my best, sir.

The Communicational Basis of Organization

A. I’ll be frank with you Jones. It’s not a question of “best.” I have to have

B. Yes, sir. that report. Do I make myself perfectly clear?

Jones could, if he chose to, analyze this exchange as a string of locu- tions. But it is obvious that, if he were to d o so, he would have missed the organizational point. This is an instance of one person, A, leaning on, pushing, bullying, driving someone else, B. These verbs -leaning on, pushing, bullying, driving- don’t describe speech as such; they describe action. It is clear to us that, if we were in Jones’s shoes, there would be no doubt in our mind what the action amounted to: an order, with an intimation of veiled threat. They may be just words, but they have a sting. It is this sting that Austin means to capture by his notion of an illocutionary force. The “force” is in the leaning, the pushing, the telling what to d o and when to d o it.

Furthermore, consider how much more that you, the reader, already know about this exchange, without anyone having told you in so many words: It is a boss speaking to a subordinate (and there is probably a fair status gap separating them); the exchange is probably occurring in A’s office; it is almost certainly a Friday, during working hours; A is very likely to be older than B, and so forth. Note that none of this framing knowledge was explicit: To make sense of the exchange as action, you automatically began to supply a context (and the context was organiza- tional). That is because, like Jones, you automatically understand utter- ances illocutionally.

The difference between speech as locution and as illocution - and how text is translated into conversation-depends on how we, as analysts, characterize, o r unitize, communication. In a conventional information theory of communication, for example, we could conceive of someone constructing a sentence, making sure that the sentence was syntactically well formed (and expressed one or more propositions, or kernels of meaning, that are semantically interpretable). The sentence would then be spoken: “I’ll need that report for Monday morning.” O n the hearer end, the process would be reversed: from heard, to decoded syntactially, to interpreted semantically. If the process was successfully accomplished, a new equivalence would be established: the state of knowledge of the hearer, with respect to a described world, would now be similar to that of the speaker (always making due allowance for sender-receiver deic- tics): He’ll need that report for Monday morning. It is understandable to Jones, but not as an order. (After all, perhaps someone else will do it.) Failing to grasp that it is an order, however, is almost certainly going to get Jones into a peck of trouble! He would be missing the whole organizational significance (division of labor, coordination). It is a locu- tion to be sure, but it is also an illocution.

To describe an utterance as an illocution, the bare-bones representa- tion of the act of communication we have just outlined will not do. For

Communication Theory

one thing, because it is acontextual, it leaves too much out. To begin with, it leaves out the situational variables, or what Austin called “cer- tain persons” in “certain circumstances”: bosses in offices, for example. T o successfully conduct a marriage ceremony it matters who is being married, who is doing the marrying, who is witnessing the event, and where the whole thing is taking place. In a locutionary view we can leave the context of the “I do” or “I want that report” out of account, because the meaning is assumed to be a function of the spoken sentence- detach- able from its context of production. From an illocutionary point of view, however, part of the meaning comes from who is doing the speaking, and where: “You’re out!” is not the same act when it is pronounced by the drunk sitting beside you in the stands as when it is by the umpire behind the plate (nor is a marriage portrayed on stage legally binding). “I’ll need that report for Monday morning” means something different when spoken by the office boy. Meaning, in this sense, is not detachable from its situational context. A unit of conversation, as opposed to a unit of text, must include markers of situation.

Second, a locutionary perspective abstracts from the presumed state of mind of the person speaking, what Austin called her or his “thoughts, feelings, and intentions.” In ordinary conversational circumstances we take it as normal that the intention behind a statement counts as part of its meaning. What turns a locution into an illocution, and translates text into conversation, is precisely this assumption of intention both on the part of the speaker and the hearer, even though any given locution may be the carrier of different intentions, depending on context, and the attribution of intention made by the hearer is not necessarily identical to that of the speaker. “The door is open,” in speech, may be a statement of fact, or a question, or a request, depending on the intention the speaker intends to express or that the hearer reads into it. An illocutionary view of communication incorporates the variable of intentionality into its rep- resentation even as it recognizes that to actually succeed in communicat- ing intention is problematic. Communicating intention calls into play interpretation; that is why the boss said, “Do I make myself perfectly clear?‘ It was the intention that was at issue: what he wanted Jones to be clear about and could not be sure he had successfully accomplished in his initial utterance.8

Communicating intention thus implies, necessarily, interaction. A lo- cutionary view, however, leaves out of account the interpersonal dynam- ics of speaking: It presumes that the production of a meaningful speech unit can be accomplished by a single individual, in isolation, if necessary. The message, once produced, is assumed to be “on its own,” as it were. It is against this view that Austin (like others such as Bateson, 1972; Garfinkel, 1967; Goffman,; Grice, 1974/ 1975) pits himself. Speech, he pointed out, is not neutral; it imposes. Sit down beside someone in an airplane and strike up a conversation: If they don’t reply-even if they would prefer not to-they risk seeming rude. Ask a question and you

The Communicational Basis of Organization

expect an answer. Ask somebody to d o something and if they don’t do it you wonder why not. Make a promise and you expect to be taken seriously. You always expect a response when you act; that is the point of action (as opposed to mere speech). The unit of conversation is thus minimally a pair (baptized by Sacks, Schegloff & Jefferson, 1978 an “adjacency pair”): action-reaction.’ The Jones-boss sequence is not just four statements; it forms a unit.

This imposing on the attention of someone is why Austin uses the expression illocutionary force. It is what we accomplish in speaking, and not by speaking, even though making sure the intention has been grasped depends on what Austin called the uptake of the hearer: for example, “Yes sir!” Quite independently of what one actually achieves in acting through speech (that Austin calls a perlocutionary effect), such as getting the other passenger sitting beside you to talk ( o r Jones to really do the report), there is a prior condition that must be made good, namely getting the other person to recognize what it is you are trying to achieve.”

Finally, although this is not part of Austin’s proposal, the locutionary view leaves out of account the ongoing nature of action. Utterances are not just isolated sentences or even exchanges; conversation, being interaction, has a history. The unfolding of action follows a logic that is distinct from that of language. The accomplishment of an action is inher- ently sequential: initial stimulus to reaction to action to result. In this unfolding, one person’s action is another’s stimulus (Hulett, 1966). Inter- action is not just a random sequence of disconnected verbalizations; there must be sequential as well as synchronic coherence.

Figure 2 suggests one possible visualization of an illocutionary theory of speaking (where the quotation marks refer to Austin’s terms in his original text). This figure illustrates the relationship of locutionary to illocutionary meaning: What is designated as the “uttering of certain words,” and which thus conforms to the definition of a locution, is seen to be a fully contained component that is embedded within the overall

SPEE HACT F I

THE SETTING

“Certainpersons” r - - - - circumstances“ 1 ”Certain I E

I “Thoughts or “Uttering of “Conventional feelings“ certain words“

Prior I states, events

Figure 2 A schematic representation of a speech act

1 1

Communication Theory

structure o f a speech act. The text (the locution) occurs in a situation (its illocutionary embedding), and this situation has in turn, minimally, four components: ( a ) the antecedent conditions, or setting, prevailing at the time of the act (“certain persons,” “certain circumstances”), ( b ) where we are in the interaction (prior states and events), (c) the presumed intention (“thoughts or feelings”) of the speaker, and ( d ) a convention as to the procedure which will govern the interactive dynamics of the exchange. The (e) consequential, or illocutionary, effect (how the locution is under- stand as action) is determined by the uptake of the hearer. In addition, because we are in the domain of action, the performance of a speech act can also be seen in the context of its subsequent ( f ) behavioral or attitudi- nal (perlocutionary) effects (did Jones actually have the report ready for Monday morning, for example?).

But what are these additional elements that distinguish a speech act from a mere locution? We would suggest that the missing components in a locutionary (or any other information-processing) explanation of communication, which are successfully incorporated in a illocutionary/ perlocutionary view, are precisely what we usually mean by organiza- tion. An organization is composed of certain persons (managers, sales reps, clerks, accountants) in certain circumstances (offices, plant floors, display rooms, boardrooms). An organization, as people see it, is a set of transactional relationships, mediated by interaction: people making requests of others, promising things, passing judgment on others’ perfor- mances, promoting and demoting, hiring and firing, entering into con- tractual arrangements. Finally, an organization is about having effects that may be quite concrete, such as agents involved in manufacturing and transporting and acquiring, and writing reports: things that had t o be organized t o happen, through communication. Organizational com- munication is not just information; it is action in context. The context is the organizational frame.

No wonder Smith (1993), in her review of the organizational litera- ture, discovered that we have mostly thought of communication as being contained within organization; as Figure 2 illustrates, if we characterized communication as centered on locutions and their transmission (and all the left-out parts as merely ancillary), it certainly was! The ancillary part was where the organization was.” Organization: First Reflections Earlier we said that organization was not to be found in the event as such but rather in the event’s interpretation. We can now clarify this assertion. The actual event of conversational communication is the speaking and hearing of a locution; this is what is made material. If we claim it to be also an action, exerting a force on the recipient, it is because of intervening interpretations, and such interpretations bring into play assumptions of identity, circumstance, activity, motive and procedure. Such assumptions, however, are not, in a general way, part of what is explicitly stated (the boss didn’t have to remind Jones he was

The Communicational Basis of Organization

boss); if the text exchange is thought of as the figure of the communica- tion, the rest of what goes into interpreting an utterance as action consti- tutes a taken-for-granted ground. It is the integration of figure and ground that turns text into conversation.

The additional components shown in Figure 2 are what make up this figural ground, for each of the participants-what they take to be the organization in which the encounter is taking place. We could think of this as a list of organizational variables. There is no reason, however, that each of the interactants instantiates the variables in the same way, and many reasons to believe that, in general, they d o not.I2 Why should we imagine that Jones and the boss would be working out of an identical interpretive framework? On the contrary, given the difference in status and experience, they almost certainly are not. But if organization is inscribed in communication as its presuppositional background, and such unspoken frames of apprehension that enter into interpretation are not identical for all the participants, then organization is, as Gilbert and Mulkay (1984) put it, a “repository of multiple meanings” and a “potentiality.” Yet both Jones and the boss would agree without diffi- culty that they were part of the same organization. Part of what this paper is about is demystifying this seeming paradox.

We need to flag another point here because, again, it is part of the communication-organization relation that this paper explores. The weight of organizational variables is not always uniform in different contexts of interaction. In particular, the importance of intention seems to vary inversely with that of convention. Where conventional procedure is highly formalized, as in ritual, the role of individual intention is pro- portionately diminished; because well-established rules permit little vari- ation, not much interpretation is possible, and intention is by the same token rendered more or less irrelevant. Where, alternatively, convention is relatively little binding, the space for expression and reading of inten- tion is enhanced. The boundary between intention and convention is not, however, fixed; it is also sensitive to the situational context. When the home-base umpire is calling balls and strikes, personal intention, according to the rule book, is supposed to play little part in the judg- ment, but occasionally it becomes evident to sports announcers that an umpire is “squeezing” the strike zone on a certain pitcher, and there are reasons advanced by the commentator. It is a rookie or a pitcher who tried to “show up” the umpire, and so on. Even in highly ceremonial circumstances, such as diplomacy, observers become skillful at detecting give-away signs of intentionality (a phenomenon Goffman, 1959, called an “information game”).

We are not the first to observe this trade-off. Bach and Harnish (1 979) go so far as to dichotomize acts as being either communicative (involving attributions of intentionality) or conventional (noncommunciational, in their system, because institutional and therefore conventional). We find this mutual exclusion unacceptable because we consider interpretation to

Communication Theory

invariably take account of institutional factors as well as attributions of intentionality and, moreover, intentionality never to be totally absent, just muted, even in very conventional situation^.'^ Nevertheless, we agree with their emphasis on “institutional.” A question we must address is how organization becomes institutionalized (hence how the role of individual intentionality is reduced in importance and institutionalized organizational control is accordingly augmented).

Actually, the two orders of question have a common thread. Accord- ing to Boden (1994), “ . . . microscopic events as turns at talk embed and enact organizations and institutions and, in so doing, ground global issues in local conditions of concrete action” (pp. 9-10). We agree with her that organization is always, and only, realized in locally situated conversation but this still leaves unanswered how the “global” institution (or organization) is constituted, and how it comes to be realized in con- versation. T o begin to understand that we need to introduce a second translation.

The Second Translation: The Textualization of the Conversation The second translation transforms conversation into text (Figure 3).

We noted in the previous section that the text as a locution is an indispensable basis of conversation but that to be understood as action a translation was involved that turned the locution into an illocution. What made the difference between the explicit linguistic content of the text and the implied actional force of the same text, now thought of as illocution, was background information that, although it was unstated, included identities of the speaker and hearer, physical and other inciden- tal circumstances of time and place, the object of the conversational exchange, conventions of exchange, and the probable intentions of the speaker (we have observed that some of these may be correlated). To arrive at the illocutionary meaning of the text, assumptions about the situation are at work, in the absence of which the text would be incom- prehensible, as far as its actional meaning was concerned. I t is the bring- ing to bear of such assumptions in order to extract the actional sense of the text that constitutes interpretation. It is here that we understand Gilbert and Mulkay’s point: T o an observer, the communication event is merely text; that is all that is recordable, objectively. But the text is not the action. T o understand the actional meaning of the event, the ob-

Figure 3. The transla- tion of conversation into text

Conversation

F‘) 14

The Communicational Basis of Organization

server, like the participants, would have to interpret. This is what we meant when we said there is n o “zero point” in social action.

The person who spoke must have been cognizant of the situation, and indeed have counted on it to produce not just a text but also, through the vehicle of the text, an action. The addressee of the speech act has a task that is the inverse of that of the performer: She or he must do the recognizing of the action that piggy-hacks o n the text (through what Austin called “uptake”).l4 A conversational intervention per se does not supply the key to its own interpretation; it is not like an assemble-it- yourself toy with an instruction sheet. Anyone who has ever tried, on the basis of Austin’s theory, to identify the speech acts in a recorded conversation knows how difficult that task is.” There is no simple one- to-one correspondence between the locutionary form of the speech utter- ance and its illocutionary import. Suppose you want to get someone to d o something, but without giving them a direct order: Think of all the clever ways to convey your intention without being blatant about it, and you have illustrated the problem of the target of your act-and of the analyst!

The only way that the hearer can make sense of the flow of interaction in which he or she is involved is by bracketing (Schutz, 1970), punctuat- ing (Watzlawick, Beavin, & Jackson, 1967), framing (Goffman, 1974, 1981) it so that it is seen to have a beginning, a development and an end. This is where the action is now translated back into a text - interpreted. The bracketing principle or frame in question has a narrative basis: The conversation must be transformed into a story t o be understood. l 6

Figure 4 outlines what we take to be the structure of a simple story. The model is inspired by Mandler and Johnson (1977) and Johnson and Mandler (1980), although it differs from their version in detail. The narrative framing of conversation to produce a comprehensible textual representation of it (and hence make it meaningful) assumes a number of

EPISODIC ~ T R U C T U R E

ermnae

BEG INN IN G DEVELOPMENT END

I T H E 1 iNJ ENTION , , T H E E r E N T 1 muse I

REAC ION GOALFORMATION AC ION RES LT

Figure 4. A frame for translating conversa- tion into text (from Tay- lor, 1993)

Communication Theory

things: ( a ) that there are always CIRCUMSTANCES, (b) that communica- tive interaction occurs in REACTION to an event or a state that requires dealing with ( a BEGINNING), (c) that the protagonist who initiates the event is motivated by an INTENTION (including attitudes, beliefs), and may have formed an objective or GOAL, (d) that the RESULT of action is mediated by a partner’s reaction (the link we show in Figure 4 as [AC- TION-cause-RESULT]) and that these together constitute the main EVENT underpinning the DEVELOPMENT of the EPISODE, and (e) that there is a recognizable outcome-a new state of affairs (the END).

A simple story, for example, might be the following:

We’ve got into computerization lately in our office: It’s changed the way we work (CIRCUMSTANCES). I (PROTAGONIST) have begun to have terrible migraine headaches (BEGINNING OF S T O R Y ) . I decided I couldn’t go on working like that (REACTION, GOAL FORMATION), so I spoke ( A C T I O N ) to my boss (ANTAGONIST), and she has finally agreed to let me split my time between the office and the front desk (RESULT). Since then, my headaches seem to have cleared up (END OF STORY ) .

This is not the conversation any more, but a summary of it, recalled in retrospect. It is a way of making sense of interactions that may have occurred over time but were not as coherent as they are now made to appear. The original conversational sequence no doubt lent itself to a multitude of interpretations, of which this is merely one possibility. The point is that the conversation had to be interpreted to have meaning for the participants-one interpretation per person, of course; there is, we repeat, no zero point in social action.

Note an implication of our model: I t is not that the hearer of an utterance is an empty vessel and that interpretation is limited to the decoding of the locutionary sense of the utterance. On the contrary, to make sense of the utterance as action, it has to be fitted into a larger (narrative) structure to become understandable. This implies that people start with interpretive frames so that the interpretation is already colored by expectations, the sedimentation of previous experience (Ellis, 1992; Werth, 1 9 8 6 , 1 9 9 3 ) .

Figure 4, which is a narrative frame, is really Figure 2 in a new guise. The translation of a text into the acts of a conversation has an inverse, the translation of the acts of the conversation into a textual representa- tion of them. On reflection, this seems obvious. How else could under- standing be achieved? The cycle from text production, by means of which intentions are expressed, into a framework of action (conversa- tion) and back to interpretation, now thought of as virtual text produc- tion, has to be made complete for there to be any understanding of the action at all.” In information-processing theories of communication, we assume complementary encoding-decoding activities; in an actional theory, we assume complementary actions of performance (text into

The Communicational Basis of Organization

action) and attributions of intention - interpretation (action into text). Another way to think of this is as follows: The speaker translates a

perception of the situation into an utterance; for him or her the back- ground is complete with identities, circumstances, a perception of appro- priate action, and a sense of the interactive relationship linking those in conversation. But none of this is explicitly present in what is said: The text comes out as statements. The hearer now has t o make sense of the utterance as action, by narrativizing it. The background perception of situation is no longer exactly the same.

Note that a recursion is implied. The hearer’s interpretation itself becomes a text only when it is produced in another conversation (or the same at a different time). What was unspoken situational (illocutionary) background in the first conversation may now become explicit locution- ary content in the second (cf. the computerization story). In this way one conversation - a context of action - becomes the matter of a second (now merely a description of action): The onion has begun to form.’*

In conversations where the people know each other well, and interac- tion is frequent, background organizational assumptions tend to be con- vergent and the common ground of discourse increasingly homogeneous (Ellis, 1992; Werth, 1993). The narrativization of one conversation by a hearer and its subsequent embedding in a second by a speaker has a double effect: objectification of the background assumptions present in the first conversation (through translation into a narrative that expresses a point of view) and heterogenization of the background assumptions by the addition of a second-level interpretation (that of the hearer in the second conversation).

This “separation” process is one to which we will return; here we have merely introduced a first of several such transpositions.

From Communication Theory to Organization Theory We have now constructed a theoretical platform, using as our planks communicational principles that emphasize the role of interpretation; it remains to be seen how the floor we have built aids in supporting a reasonably rich theory of organization, in the large as well as the small. The Organization Stakes What we need t o explain is what we described earlier, in alluding to Mintzberg (1979), as the stakes of management: how to divide up labor (get certain people to reliably perform certain actions) and how to coor- dinate what has been divided (how to link up the conversations the division of labor created). These are what Lawrence and Lorsch (1967) called differentiation and integration. In this section, we are going to be concerned with how to get many people, involved in many conversations to do (modalization of performance) and think (modalization of the situation) in a way that produces (a) a differentiated structure of work

Communication Theory

and ( b ) an integrated pattern of activity and belief. To accomplish our goal, we proceed in two steps. First, using narrative and speech act theory as our guide, we probe deeper into the dynamics of ordinary conversation to show how interaction usually produces division of labor and coordination. These structuring principles, we find, are not just macrosocial phenomena; they are endemic to communication considered as action, at the micro- as well as the macrosocial level. The second step in our argument is to point out the limitations to this structuring princi- ple when it is extended to wider networks and to identify means by which it could be adapted to this wider context.

Coordination has two objects: the linking of performance and the linking of belief. We consider these two dimensions in turn. The key concept to be developed is that of modalization. The concept modalizu- tion is borrowed from Greimas (Searle, 1970, 1987). In the original French, Greimas identified two kinds of modalization, modalisation du faire and modalisation de l ’ h - e . As the terms indicate, there is a modali- zation of doing, or action ( f a i r e ) , and a modalization of being, or state of affairs ( h e ) . We begin with the first of these: doing, acting, perfor- mance. Modalization of Performance: The Interpersonal Dimension Greimas’s idea comes from an elementary distinction in grammar be- tween ordinary and modal verbs. Modal verbs express possible states and actions-what could or should or might have been or been done, not what actually is, is being, was, or has been. Greimas notes that an actual performance (or an actual state of affairs) exists as a potentiality before it exists in fact (recall, in passing, Gilbert & Mulkay’s 1984, allusion to potentiality). Greimas’s way of conceptualizing this is to say that under- lying any performance there must be competence (the desire, obligation, knowledge, or skill to carry out the act). It is the competence that can turn the potentiality into an actuality. Such competence, organization- ally, is triggered by another person’s communicative act. It is this act that instantiates the state of readiness to perform. This is Mintzberg’s (1979) “coordination,” now restated in a different language.

If this seems somewhat abstract, we can make it more comprehensible by reference to the example of Jones and the boss. The action in question is preparing a report for Monday morning, but at the time of speaking there is in fact no such report: It only exists virtually. Both understand this. For it to become real, Jones must undertake a performance. The exchange initiated by the boss has as its object the transforming of Jones into a “competent” agent who will carry out the performance. The com- petence in question is presumably not the knowledge or skill necessary to prepare the report (since one imagines this is why he was selected to do the job in the first place), but the willingness or intention to do so. This is what has to be triggered. When successfully accomplished, we have a division of labor (Jones’s task), coordination (Monday morning deliv-

The Communicational Basis of Organization

ery), but we also have hierarchical confirmation (boss modalizes, Jones performs).

This kind of exchange, preparatory t o a performance, is termed by Greimas a manipulation and leads to what he terms the qualification of an agent.’’ I t is, as far as we can make out from the reading of Mintzberg ( 1979) and other management theorists, what a communication theorist would describe as the basic communication act underlying an adminis- trative hierarchy, leading, in concrete terms, to the usual distinction between workers and managers: Getting people to do what you want them to do first means getting them to recognize what you want them to d o and then getting them to accept the mandate you want implemented.

The example we have just cited is an instance of direct, overt manipu- lation. There is a term in speech act theory for such an exchange, where one person requests or commands or orders (or even suggests) that the other d o something: It is called a directive (1976, 1979). The comple- ment, promising to carry out the order, is called a commissive. Jones has agreed.20 Modalization of Performance: The Transconversational Dimension Note that this exchange fits Mintzberg’s (1979) category of direct super- vision. If you were reading our analysis from a managerial point of view, you would probably think of direct supervision as having the advantage of explicitness. However, as a managerial strategy, it also suffers from serious disadvantages: It is, for example, costly in time and effort be- cause, being verbally transmitted in face-to-face situations, it only works with individuals or smallish groups. I t is also highly reactive since people never really enjoy being told what to do and are frequently recalcitrant, either overtly or by subterfuge. Jones is unlikely to forget soon the humil- iation of being shoved around. O n the other hand, if the boss chose to mitigate the request by turning it into a polite hint, Jones is left with an out: He didn’t “understand.” That is presumably why the boss was so explicit.

The good news for managers is that there is a much more efficient, subtle way t o instate the virtuality of the action (and it is clear from authors such as Mintzberg that they understand this). I t follows from the logic of translation illustrated in Figure 2. As we described the transla- tion of a text into action, we postulated an interpretation that took account of the situation in which the text production was embedded. Let us represent this dependence of the interpretation on situation as well as text as follows (where Z stands for interpretation, T for text, and S for situation):

Text, we have already argued, underspecifies interpretation because it becomes action only in a situational context (Barwise & Perry, 1983).

Communication Theory

The situation, however, is not what is manifestly expressed, but what is latently presupposed- taken for granted, out of focus for the purposes of the exchange (Werth, 1986). If managers could fix the parameters of the situation for both interactants to the exchange, they would then have gone at least part way toward determining both how the speaker generated the text (in that we assume speakers choose texts that make sense, actionally, in a given context) and how the hearer interpreted it (since, as we have seen, the intention of an act is interpreted by taking account of the situation). By successfully fixing the situational parame- ters of the exchange they would thus have modalized the action occurring therein, that is, made people “competent” in the way they wanted. They would in this way have established an organizational frame that could be applied beyond the limits of a single conversational exchange.

From the theoretical perspective this seems like a much more effective management strategy than direct supervision because it is both nonreac- tive ( n o explicit order has t o be given) and can be generalized to multiple conversational situations, so that the same (or at least very similar) ex- change comes to be carried out everywhere. This must be what Mintz- berg (1 979) meant by “standardization” (programming, quality control, training): it involves fixing the identity of the actors, their circumstances, the object of the exchange, their intentions, and the conventional proce- dure governing the interaction of the participants- everything we de- scribed earlier as the “setting” of action (Figure 2). The question this raises, obviously, is How d o you “fix the situational parameters”? We will return t o a consideration of these questions. Before we do so, how- ever, we need t o say a word concerning the modalization of state. The Modalization of the State of Affairs, or Situation When a speech act is being performed, the speaker produces a text that is the carrier of an intended action; the reaction of the other is presumed but not guaranteed. Individual performance, in other words, underspeci- fies the outcome of the transaction, as it must, when we go from the perspective of a single actor to that of a couple.

The opposite occurs where the reverse translation is concerned: A story, by definition, is the expression of a single protagonist’s point of view, and the implicit complementary story of the partner to the ex- change, the other person’s perspective, is reduced to an embedded func- tion-a “reaction” (Taylor, 1993). We really d o not know the boss’s side of the exchange in the computerization story cited above, for example. All we know is that her manifest reaction to the request to split time between two functions was positive. There is a residual uncertainty as to what she really thought (might she have been furious?). She might have had a different story to tell.

Every conversation, in effect, potentially generates a minimum of two individual texts and is therefore equivocal (Figure 5 ) , as Gilbert and

The Communicational Basis of Organization

Conversation Conversation

Mulkay (1984) claimed: An act is a potentiality, depending on the inter- pretation it is given.

It is this equivocality of interpretation that Greimas (1 987) intends to capture under the heading of modalization of the situation. A modal construction, in this usage of the term, is concerned with what seems, or appears, to be the situation, as opposed to the real state of affairs. Because in social action there is no objective criterion of social reality, deciding what is real, as opposed to mere appearances, becomes an organizational stake. Moreover, because communicative exchanges tend to be what Greimas calls polemical, which is to say characterized not just by different but by opposed perspectives,2’ communicative interaction regularly generates considerable ambiguity (Donnellon, Gray, & Bou- gon, 1986; Eisenberg, 1984)-not exactly what Mintzberg (1979) meant by “coordination”!

One way t o overcome such a state of uncertainty is to make a state- ment affirming a reality and to have the statement accepted, thereby pragmatically clearing up the ambiguity. Making this kind of statement constitutes a class of speech act that has been termed (Searle, 1976, 1979) an assertive. As an act, such a statement is both a locution (asserts the truth of a state of affairs holding in the world) and an illocution (imposes on the hearer a certain interpretation to which she or he is expected to attend). If the affirmation is accepted, communicationally, then a reality has now been created, socially speaking, whether or not that reality is factually true. But since, for social action, the factual basis does not - cannot - exist, the communicationally negotiated equivalent is the best that can be hoped for.”

But why should one side’s interpretation (management’s, let us say) take precedence over the other’s (the employee’s)? Why would the perlo- cutionary effect of any assertion be confirmatory when it means that one reading of the situation takes precedence over another? Making one-on-one assertions (modalizing the situation), we would suspect, is just as inefficient and reactive as one-on-one supervising. An equivalent strategic problem is presented to those who would be managers. They now have to make an organization out of whole cloth by again parame- terizing the communicational situation so as to make the determination

Figure 5. The equivo- cality of the organiza- tional text.

Communication Theory

of reality relatively unproblematical. The way to d o this is, again, to create an organization. Until this is achieved, management is not an actor, and cannot modalize either performance or belief. How does it become an actor? Creating an Organization The key to creating an organization in the large is in the marshaling of illocutionary force. It is the force of an act that produces its illocutionary, and then its perlocutionary, effect, and hence makes directed coordi- nated action possible. So where does such force come

In the transition from text to conversation the speaker of the text is translated into the actor of the act. But the act thus performed, although it is motivated, and even defined, by an intention, is not the intention itself: Because it has had to be mediated through text the latter becomes merely a token of the intention. So the actor is now properly thought of as an agent representing, as the person responsible for the performance of the speech act, a principal’s intention. Often, the agent and the princi- pal are the same person, in which case we are in the domain of individual action. Here, making a distinction between actor and agent may seem artificial. In organizations, however, it is frequently, if not normally, the case that the speaker (or actor) is an agent for someone else; it is not one’s own intention that must be translated but that of the absent princi- pal. Such agency- acting o n another’s behalf-now becomes a compo- nent in the semantic characterization of organizational roles and sta- tuses: An organizational officer is qualified, by definition, to represent certain voices in certain situations (recall Austin’s 1962, “certain persons in certain c i r c u m ~ t a n c e ~ ” ) . ~ ~ The qualification of a person who can speak with authority is, from a different perspective, a constraint o n others’ performances. It seems obvious that Jones would have told someone at the same level of authority as himself ( o r lower in position) to “stuff it” if he had been addressed by them in the same terms as the boss chose. What gave the boss’s speech act the additional punch was that he spoke as an authentic agent of the organization. The punch is what we mean by illocutionary force.2s

There is a bit of prestidigitation involved here. I t should be obvious that an organization, being a collective not an individual identity, can never speak in its own name; only people-individuals-can perform acts of speech (Callon & Latour, 1982). So the organization always speaks through an agent; it has no other possible voice. But how d o we know the agent is correctly representing the client o r principal, that is to say, the whole community of people that goes to make up the organiza- t i o n ? The answer is that we d o not know. People come to speak in the name of the organization because they are accorded the credibility t o d o so-and for n o other reason. I t is a clear case of self-fulfilling prophesy.

In principle, every member of a community is a potential agent of the collectivity and could, on occasion, speak for it. In practice, such speak- ing rights are unequally distributed. That unequal distribution comes

The Communicational Basis of Organization

about, however, strictly through communication. The process is one that Greimas (1987) calls sanctioning: When an agent’s performance is accepted as legitimate in conversation by enough others, then such a performance has been sanctioned, and its identity with the voice of the organizational community thus claimed is legitimated.26 The force of an act, as opposed to its direction, thus expresses the extent to which a given agent’s voice is taken by the community to be legitimately expressive of the collective intention. (It is at this moment that the collectivity has become structured, such structure corresponding to the degree of legiti- macy to speak for the organization accorded to different actors and the constraint thus operating on all others.)

We have now marshaled all the tools we need to explain organization in the large.

The Extended Organizing Process: From Conversation to Text and Back-The Long Way To comprehend how communication comes to develop organizational properties that transcend those of the strictly local conversation we need to look more closely at the dynamic interplay of conversation and text, as it works itself out through the complex modulation of text in a net- work of successive conversational situations. For this purpose, we return to a theme we barely adumbrated earlier, namely the embedding of one conversation in others, through successive phases of textualization - a complex conversation-text cycle. Distanciation The concept of distanciation is borrowed from Ricoeur (1981, 1986). Discourse (expression of ideas formulated in language) differs, he says, according to whether we think of it as event or as work. When he speaks of discourse as event, it is obvious he has in mind what we characterized as the text that forms the basic matter of conversation: temporally con- strained (in the present), referring back to the speaker by a complex set of indicators (such as pronouns) and thus linked to the person who speaks, always about something (a situation or definite “world”), ad- dressed to another person (an interlocutor), part of an exchange-indeed its object. The transition to “discourse as work” occurs when, “by enter- ing the process of understanding, [discourse] surpasses itself as event and becomes meaning” (Ricoeur, 1981, p. 134). As a “work” all that was circumstantial about the discourse- the time, the speaker, the “world,” the interlocutor, the property of exchange-is stripped away: no longer the fleeting event, but the “meaning which endures.” “The very first distanciation is thus the distanciation of the saying in the said” (p. 134). The resulting work need no longer coincide with what the author meant; it is autonomous: Textual and psychological meanings have, as he puts it, “different destinies” (p. 139) (See Figure 6 ) .

The second level of distanciation is a consequence of the dialectic of

Communication Theory

. Figure 6 Discourse as event and discourse as a work (Ricoeur. 1986)

Discourse as event

1. Realized temporally, i n the present

2. Speaker intention and meaning overlap

3. Refers to a specific situation

4. Addressed to someone

Discourse as a work

1. Fixated. materially, in writing

2. Dissociation of intention and meaning

3. Refers to a general “world”

4. Accessible to everyone

speaking and writing. By writing the discourse down-“fixing ” it- discourse is objectified (and hence “sheltered from destruction”). It be- comes part of a practice, the object of a technology, “a material to be worked upon and formed.” As an object, it is decontextualized, dis- tanced from its origins, made accessible to an “unlimited set of readings.” Each reading constitutes a recontextualization and, following the same logic that is illustrated in Figure 2, the meaning thus refound in the new context owes something not only to the text but also to the (different) situation that now frames its reading.

It is the communicational implications of distanciation, in a context of networking, that we now turn to explore. T o do so we develop Ri- coeur’s idea in a slightly new guise that we have termed degrees ofseparu- t i ~ n . ~ ’ Degrees of Separation’’ The first degree of separation we have already dealt with; it occurs when the spoken text is translated into action by becoming the carrier of an intention. The text is not the intention, as we have said, it is its token. This basic rupture, dividing the objectivity of the spoken text from the subjectivity of the intention it translates, is what we mean by an initial ~ e p a r a t i o n . ~ ~

The second degree of separation occurs at the point where the events of the conversation are turned into a narrative representation of them, making it possible to understand the meaning of the exchange. The initial intention is now at two removes-no longer just spoken by an agent but also interpreted by a hearer or a reader. The interpretation becomes a text when it is introduced into a new conversation, where the first conversation has become a locutionary point of reference within the context of a speech act occurring in a situation distinct from the initial one (either bringing into play new people or, if the same people are involved, referring back reflexively to an earlier point in the conversa- tion). This we have already dealt with as well (the computerization story).

It is here that a collective identity begins to take shape. Through the textualization of the conversation, and its subsequent rehearsal in a new context, speakers are reconstituted as actors in a narrative, and such actors are not restricted to individuals: Groups appear, collective actors.

The third degree of separation is facilitated by the transcription of the text onto a permanent or semi-permanent medium permitting storage and hence temporal and physical removal from the original communica-

The Communicational Basis of Organization

tion event. This is what Ricoeur thinks of as objectification. At the second degree of separation, we presumed a co-present speaker and a hearer; now we have entered the domain of the author, where co- presence of author and reader is not required. The situational specificity of earlier separations is accordingly reduced. The locutionary dimension of communication is disconnected from the illocutionary. The distancing from the initial circumstances of conversation now becomes greater. The events are described (the minutes of a meeting, for example) but in a written, not a spoken code; the original vernacular has largely disap- peared.

The fourth degree of separation is the result of the development of specialized media-specific languages leading to the kind of professional- ism associated with the mastery of a medium that has a technological basis (Innis, 1951). As Ricoeur (1981) points out, the existence of spe- cialized media encourages the growth of “a practice,” the object of a techne; no longer just a conversation but a culture o r discipline of con- versations. Newspapers have their codes, where the reporting owes, fre- quently, more to the conventions of journalism than it does to the event being described. Fields such as the sociology of organization strongly encourage generalization, in the form of covering laws, and tend t o discourage representational fidelity to a specific situation. Science, gener- ally, does the same thing. So does management theory. By now, the intentions of the participants have become literally nonrecoverable (Hodge & Kress, 1993): Only the abstract pattern of a professional language remains.

The fifth degree of separation occurs when the text that represented the initial conversation situation, through being transformed into a de- sign, is made the basis for a materialized, physical frame of conversation: an office, a research laboratory, a classroom, a boardroom, an interview room, a sales counter, and so on. Manuals of procedure appear. Con- temporary research in the field of groupware, or computer-supported cooperative work, illustrates the process very well: the literal construc- tion of a computer analogue of human interaction. At this point, conver- sation has mutated from the entirely local and specific to the entirely global and generic (Taylor & Van Every, 1993).

The sixth, and final, degree of separation is publication, dissemina- tion, diffusion, broadcasting. Now we have the basis for standardiza- tion - and for organization in its contemporary postmodern, globalized guise.

Underlying the progressive separation of the text from the conversa- tion is another phenomenon: the successive embedding of conversations, each involving a translation. Conversational Embedding A text may be simultaneously in a conversation (as an illocution) and about a conversation (as a locution). The conversation it is about may, on occasion, be the same as it is in (a phenomenon termed by Watzlawick et al., 1967, metucomrnunicutzon). Mostly it is not. The textual matter

Communication Theory

o f one conversation is normally a representation of another (Taylor & Giroux, 1993). This is the phenomenon of conversational embedding- layered discourse, or what Werth calls the “text-world” of a discourse, within which there are not just worlds but subworlds, and subworlds within subworlds (Taylor & Cooren, 1994; Werth, 1993). The conver- sation may be close to each other (tightly coupled, Weick, 1985, would say), as for example with a couple of colleagues getting together over lunch after the monthly departmental meeting to compare impressions and perhaps come to a common interpretation of episodes that took place in the meeting. Here local detail is retained, if interpreted. But the principle of degrees of separation says that as distance is increased be- tween the conversation and its textual interpretation, specificity is aban- doned in favor of generality. The objectifying character of text ends by producing a reified representation of what is n o longer a situated set of conversations but what has instead become an organizational template so abstract that it can be taken t o represent not just some but all the conversations it refers

The organization has now become an abstract conceptual entity (the kind of object management texts portray), a product of the distanciation accomplished by the textualization of conversation. But how does it come to be accepted (since there is no a priori reason to assume it is everybody’s conception)? The stage is set, but the main actor is still missing; the organization is still virtual. Genesis The fundamental challenge facing every human community is this: How d o the many voices of a (loosely defined) collectivity come to be the one voice of the (now clearly boundaried) social unit. Until it finds a voice, the collectivity is merely a congeries of interests and preoccupations; it is not yet an organization. The key term here is voice. I t becomes an organization at the moment that an agent is authorized t o speak in its name. Then, and only then, is it clearly differentiated from other actors. Only by the ability t o participate in interaction as a recognized voice is identity achieved. Until organization has an agent recognized as acting for it, it remains a diffuse, unrealized potential.

Perhaps in traditional societies the developing of a collective voice to express a commonly felt organizational unity is less problematic. Royalty is born into the role of the voice. I t is neither a function that others can aspire to nor can the chosen easily eschew the responsibility. In our modern world, however, the issue is less clear: People are not always “chosen” through inheritance. Becoming the agent of the organization is something to be achieved, not ascribed. Networking3‘ To be recognized as the voice of the organization, and its legitimate agent, the aspirant must produce a text, and have it authenticated, by a sufficient number of people ( o r at least the right people) to have one’s own right to speak become consensually validated.32 This is not just a

The Communicational Basis of Organization

matter of text production; it is a question of enlisting enough allies on one’s side to become recognized as a legitimate author or spokesperson for the text (Latour, 1987)-while simultaneously having the text recog- nized as the right text. The process is reflexive: The text legitimates its author, the author his or her text. A text that passes the networking test is gradually transformed from an artifact of the network into afact- just part of reality in that organizational

In large collectivities, this usually involves networking (the third and fourth degrees of separation), the network in question being a network of conversation^.^^ The conversation in which the organizational text, and its spokesperson, seeks to become legitimated (termed meta- conversation by Giroux & Taylor, 1993) is one composed of spokesper- sons for their own- other -conversations. In such a metaconversation, because more than one conversation is involved, it is not just one organi- zation that is being reconstituted, but several; there is always multiple embedding.

This is management in the large: politics. Legitimacy has as its target gaining acceptance of a general, or distanciated, text but the process has to be grounded back in local realities through networking. The recruit- ment process works both ways: The aspirant to power aims to enroll adherents, but others in the network are concerned to maintain and enhance their own position by associating themselves with someone in the a~cendancy.~’

The problem with politicking in networks is that it is part of manage- ment’s world-and an unstable part at that. T o reach the ultimate orga- nizational goal, the fifth and sixth degrees of separation are necessary. It is in this way that organization comes to be, to use Ricoeur’s term,“fix- ated” (1981,1986). Fixation: The Parameterization of Organizational Conversations The final stage of separation is the construction of physical frames for interaction, and such constants as programming (the publishing of texts that specify identities, objects, goals and rules of procedure) and quality control and training (the learning and application of conventional proce- dures). Only in this way can management extend its power to coordinate beyond what can be accomplished through costly face-to-face interac- tion. Such frames come to be part of the taken-for-granted situational context of ordinary conversational interaction that we described earlier (Figure 2). Because they are taken for granted by the organizational collectivity they come to be, by the logic of voices, the concrete proof of the legitimacy of management itself. The “direction of fit” is bidirec- tional. Direction of Fit Searle (1989) observed that texts (i.e., a tissue of statements) stand in two kinds of relation to the world they represent, or direction of fit. When the text is a description of an actual situation, the direction of fit

Communication Theory

is word-to-world, that is, the text is assumed to be conformed to that which it represents. When, on the other hand, the world is intended to conform itself to the text, the direction of fit is reversed and it becomes world-to-word. This latter case is related to modalization. The process of text legitimation, and legitimation of its spokespersons, that we have been describing, has as its intended effect the resolution of both direc- tions of fit. The world in question is the conversation. The word, or text, is intended t o both describe the situations of the conversations and to prescribe the performances that go t o compose them. An act of speech that is simultaneously word-to-world and world-to-word is called a dec- laration (Searle, 1989). Organization, as we understand it, is fundamen- tally declarative: It exists because we, collectively, will it to exist. By saying it exists, it exists.36 Translation is merely the medium by means of which this ongoing dynamic is made possible.

The organizational text is, thus, a (standardized) description of per- sons, circumstances, objects, intentions and conventional procedures, now no longer situated circumstantially with respect to a particular con- text, but categorized, generalized. It is also a prescription that determines identities, contexts, objects of value, and correct procedures. If we were now t o pull back t o the long view, we would perceive the effect of textual distanciation from the conversation as a decontextualization-recontextu- alization cycle, such as that described by Ricoeur (1986), complex in its pattern because, for each degree of separation, multiple processes of translation intervene. I t seems that, from a discourse perspective, an organization is a variant on what has been called the hermeneutical circle with this specification that a network of successive interpretations is involved. On the one hand, the perception of reality on which manage- ment decisions are predicated has had to pass through several conversa- tional filters t o be finally validated in a managerial conversation that is many times removed from the ordinary conversations that it will regu- late. O n the other hand, as our earlier discussion emphasized, a text, no matter how meticulously legitimated, is not yet an action until it has been translated into a conversation, and that, as Boden (1994) points out, occurs locally. The organizational text produced by the metaconver- sation of management must be reinterpreted in the local conversational crucible-with, let it be said, results that are not always totally predict- able. Ruth Smith’s Question In her analysis of the literature on organizational communication, Smith ( 1 993) found that the presumed relationship of communication to organ- ization could be classified according to three root metaphors: Communi- cation is something that occurs in organization (and thus organization is seen as a container); communication either produces organization o r organization produces communication (or both); and communication and organization are equivalent (variant expressions for the same real- ity). We undertook the present paper in the spirit of an exploration of

The Communicational Basis of Organization

her question. Our conclusion does not exactly fit any of her alternatives. We believe that communication does occur in organization. If communi- cation is the interaction that makes the conversation, then the organiza- tion is its setting, pointing t o its persons, it circumstances, its objects, and its procedures, but never defining them. That organization, how- ever, is a product of communication, arrived at by a complex process of textualization and conversational embedding, as a result of which the characteristics of organizational structure, although still interpretable, are more or less fixed, identities (including that of the organization) more or less established and the objects of organization more o r less reified through their textualization. Through the materialization of interaction that is enforced by its dependence on text as medium, it constructs its own more or less permanent frames that are the constraining structures within which subsequent communication is channeled.

So what is an organization? How you answer that question depends on the worldview you choose (Taylor, 1 9 9 5 ) . On the one hand, you could think of it simply as a network of conversation: autonomous, an eternal becoming without anything so concrete as a meaning, an unfold- ing happening in which we are swept up, unpredictable in its outcomes. O n the other hand, you could see it as a text-mediated structure of identities, situations, purposes, finalities-heteronomously saturated with meaning, having an existence outside the mere here and now, some- thing we can relate to rationally. Which is the right picture? The answer depends on worldview: One image is simply the inverse of the other, the product of a translation. What makes organization studies intriguing is that translation is an information-losing process. There is a dynamic. We are, it seems, destined to be forever caught between the conversation and the text.

If, as Smith speculates, organization and communication are equiva- lent, it is here that the equivalence is to be found.

implications for Research Recently one of North America’s longest standing organizations - major league baseball - has undergone a prolonged period of internal crisis. It is easy t o find in the extraordinarily messy story instances of the con- structed character of human organization and the central role of dis- course therein. An episode will serve t o illustrate. Montreal Expo out- fielder Larry Walker, who underwent off-season surgery, had been working out a t the club facility in West Palm Beach, Florida, on a sched- ule of rehabilitation. On Saturday, February 1 8 , 1 9 9 5 , a Montreal paper ( T h e Gazette) reported as follows:

Earlier during his drill, Walker had been informed- misinformed- by an Expos minor league coach that he could no longer work out at the stadium. “He . . . said [Expos general manager] Kevin Malone told him to tell us we couldn’t work

Communication Theory

out any more,” Walker said. “That’s typical for them (Expos). They never tell you anything to your face. They don’t have the (courage). They tell you through the media. Or they send a messenger.” Expos trainer Ron McClain, experienced in similar situations, attempted to soothe the simmering all-star outfielder. . . . “It’s a misunderstanding,” McClain assured. ‘‘I’ll speak t o Kevin. Everything will be straightened out.” Walker wasn’t easily pacified. “He (Malone) could have said it in anger” Walker rationalized. “Maybe he doesn’t know the rules. But why don’t they talk to me? I’m here in the clubhouse every day. I take treatments every day. . . . ” As the threesome-Walker, McClain and a reporter-reached the stadium, McClain headed into the Expos’ executive trailer and Walker ex- cused himself.“I’m going to call my agent right away,” Walker said. “And I’m going to call New York (player’s association headquarters). I want to know what’s happening.” McClain emerged moments later after discussing the situa- tion with Expos vice-president Bill Stoneman and confirmed there had been a misunderstanding. Several former Expos now with the Rockies had been work- ing out with Expos. The message the coach was to deliver was that those players were no longer welcome. . . . “SO long as Larry has not signed with another club,” Stoneman said, “at which time I assume that club would want him to work with them, he can continue his rehab with us.” But there is still one problem so far as Walker is concerned. He can use Expos facilities for anything associated with shoulder rehab. This okays throwing and lifting drills and treat- ment from trainers. But no hitting. “I think hitting is part of baseball, isn’t it?” Walker said.

Read this again a n d a s k yourself the following question: Where is t h e organization? If it is t h e E x p o s , is Larry in o r o u t of it? Is his agent a member of t h e organization? If n o t , why n o t ? (It is t h e agent who inter- venes in t h e negotiation of a contract, after all). Is t h e reporter p a r t of t h e organization? Again, if n o t , why n o t ? (He w a s as much involved as t h e trainer in communicating between M a l o n e a n d Walker, after all, although the mediation w a s different, a n d anyway without the ever- present reporter baseball would be something quite different, certainly n o t a billion-dollar business.)

To us, these seem like futile questions, deriving f r o m the unproductive imperative to make h u m a n organization, mediated by h u m a n communi- cation, i n t o a n analogue of physical organizations such as machines, organisms, a n d t h e like. T h e kind of event described in The Gazette is a site, n o t just for multiple interpretations ( t h a t is obvious, o n even the most superficial reading of the story), b u t for t h e playing o u t of intersect- ing organizations, each of which is a n enfolding within itself of the others, so t h a t every organization is constructed o u t of other organiza- tions which, i n t u r n , enfold it. Trying to fix the frontier between them is a n unrewarding exercise a t best.37

Earlier w e referred to t h e tautological basis of t h e organizational con- cept. For organization, tautology is fundamental. W h e n , during the baseball lockout, g r o u p s of minor leaguers elected t o leave spring train-

The Communicational Basis of Organization

ing camp, and others stayed, it was because tautology broke down. The intensive discussions that the strike brought into play (involving, at one point, the president of the United States himself) were not so much reflections of existing collective identities as the attempt t o arrive at a new tautology, and thus to regain stability.

But wait, you may reply, baseball is not typical! This is a very special case of organization! Perhaps, but there is another possibility, namely that every organization is a special case. If that is so (and we believe it to be), then the proper business of an organizational communication researcher is not the study of organization, using communication as data in the search for nonexistent laws, but the study of communication. We should direct our attention, as Latour (1987) puts it, “upstream” to the problematic origins of organization in discourse, not downstream to the by-now reified structures that pass for the object of study in management schools. We need to get behind the categories such as Mintzberg’s to the processes on which they rest. We need to study, not the rock, but the teeming life on which it sits when we overturn it: to focus our attention not on organized baseball but o n organizing baseball, not on the estab- lished structures of enterprise but their convulsions - mergers and acqui- sitions (Dumas, 1995), cooperatives that don’t cooperate (Giroux & Fenocchi, 1994-1995), family successions (Jolicoeur, 1994; Katambwe, in progress), reversals of strategy (Giroux & Cooren, 1995), what lies behind the happy family image of a high-tech firm (Calcagno, 1994), what happens to a production firm when its leader pulls the plug (Key- ton, 1994), the uncomfortable dynamics of the implementation of new technologies (Khalil, 1995; Taylor & Van Every, 1993). Let the others worry about the abstract principles of management; we want to know what management practice is really like. And it is our belief that if we learn to study communication with sufficient insight, we will end up finding organization, because, in John Dewey’s words: “Society not only continues to exist b y transmission, b y communication, but it may fairly be said to exist in transmission, in communication” (Dewey, 1916/ 1944).

When the farmer was dying he told his three heirs there was a pot of gold buried in the orchard. They dug and dug. They never found the gold (there was none) but because of their industrious cultivation of the earth around the roots of the trees the latter thrived and the heirs became rich from the abundant harvest. Finally they realized they had found their pot of gold. It is time, we think, for us also to become more diligent in the cultivation of our own orchard.

James R . Taylor is Professor of Communication ot the Department of Communication, Faculty of Arts and Sciences, Universitk de Montrkal, Postal Box 61 28, Station Down- town, Montrkal (Qukbec), H3C3]7.

Author

Communication Theory

FranSois Cooren IS a recent Ph.D. graduate o f t h e same department, now doing post- doctoral w o r k a t the Unioersiti d e Louvain, in Belgium. Nicole Giroux is Assistant Profes- sor, and Daniel Robrchaud is a Ph.D. candidate at Universiti d e Montrial. Inquiries f o r copies o f t h e article should be addressed to Professor Giroux a t the above address.

‘ For a fascinating illustration of h o w interpretations can evolve in a situation of crisis, see Bechler (1994). T h e author traces the path of a race relations incident in a small university and shows how interpretations enter the process in a fundamental way.

Stohl and Redding (1987), in their review chapter in the Handbook ofOrganizationa1 Communication, noted that while it is obvious that communication, thought of as messag- ing, is the basis of organization (“It is an axiom that organizations cannot exist without communication”) (p. 451), there is in fact not much literature o n messaging as such: “Communication specialists have generally chosen to study correlates or perceptions of message activity rather than actual messages” ( p . 453). It is easy to understand why the literature is so little beholden to a messaging image of communication. T h e message view limits u s to looking a t information exchange, as if this was the only-or even the main- thing going on in communication. Through such a lens there is n o emergent organization; messaging theory predicates the prior existence of organization (otherwise where did the senders, receivers, channels, and codes come f r o m ? ) and thus effectively blocks our dis- covering its genesis in discourse. None of the hierarchial complexity, with its statuses and roles, can be captured in such a theory; it is insensitive to context. T h e meaning of any given event, Gilbert and Mulkay (1984) claim, will vary depending o n the specific commu- nication context within which the event figures. This is precisely what the message meta- phor fails to take into account. ’ When Morgan (1986) compiled his list of the diverse images by means of which the concept of organization has been rendered manageable (the organization is a “machine,” the organization is a n “organism,” the organization is a “brain,” the organization is a “culture,” the organization is a “political system,” etc.) the chapter that was notably absent in his book was the one called “The Organization is Communication”! T h e dominant imagery (machine, organism, brain, etc.) was drawn from other fields: mechanics, biology, cybernetics. Metaphors of machine, organism, and brain have in common one property, deriving as these similes d o from a preoccupation with material objects: They take the events that go to make up the organization they a r e concerned with to be real, objective, substantial and objectively observable. I t is precisely this hard core of object-ness that Gilbert and Mulkay (1984) are calling into question where human organization is con- cerned.

Cf. Halliday and Hasan (1985): “The important thing about the nature of a text is that, although when we write it d o w n it looks as though i t is made of words and sentences, i t is really made of meanings. Of course, the meanings have to be expressed, or coded, in words and structures, just as these in turn have to be expressed over again-recoded, if you like- in sounds or in written symbols. It has to be coded in something in order to be communi- cated; but as a thing in itself, a text is essentially a semantic unit. I t is not something that can be defined as being just another kind of sentence, only bigger” ( p . 10). ’ To again quote Halliday and Hasan (1985): “A text has to be considered from two perspectives at once, both as a product and as a pr T h e text is a product in the sense that it is an output, something that can be nd studied, having a certain construction that can be represented in systematic terms. It is a process in the sense of a continuous process of semantic choice, a movement through the network of meaning potential, with each set of choices constituting the environment for a further set” (pp. 10-1 1).

Cf. Halliday and Hasan (1985): “We would see the text in its ‘process’ aspect as an interactive event, a social exchange of meanings. . . . N o t that dialogue is more important than other kinds of text; but in the last resort, every kind of text in every language is meaningful because it can be related to interaction among speakers, and ultimately to

everyday spontaneous conversation. . . . A text, then, is both a n object in its own and an instance-an instance of social meaning in a particular context of situa-

tion. . . . T e x t is a form of exchange; and the fundamental form of a text is that of dialogue, or interaction between speakers” ( p . 11). ’ In the words of Labov and Fanshel (1977), “Conversation is not a chain of utterances, but rather a matrix of utterances and actions bound together by a web of understandings

Notes

The Communicational Basis of Organization

and reactions. . . . We d o not see conversation as a linguistic form. We have come to understand conversation as a means that people use to deal with one another” (p. 30). Distinguishing between text and conversation thus implies distinguishing between “what is said” and “what is done,” as Labov and Fanshel see it, when people react in a conversation, it is to the latter (the “done”) and not the former (the “ s a i d ) that they are responding. “There are no connections between utterances,” they say (p. 70). Bierwisch (1980) elabo- rates the idea as follows: “Language and communication. . . . are based on different SYS- tems of knowledge . . . Given a domain of language use based on the rules and structure of language, and a domain of communication based on its own principles, rules, and structures belonging to the more inclusive area of social interaction, speech acts must be construed as instances interconnecting the two domains in a particular way. T o put it in simple terms: A speech act makes a linguistic utterance, mainly by virtue of its meaning, the bearer of what would best be called a communicative sense. Notice that a communica- tive sense belongs to the domain of social interaction. . . . Speech act theory is a branch of the theory of communication, viz. that involving linguistic utterances, rather than a part of the theory of language” (p. 3). For Searle (1969) also, “The speech act is the basic unit of communication” (p. 21). * Even when the locution explicitly makes a statement of intention, this is no guarantee of success. The hearer is obliged to interpret. An assertion of the truth of something when we suspect the speaker does not believe what she or he is saying is not a statement of fact, but lying, or irony, or sarcasm, or just babble. A “promise” apparently made without sincerity becomes not a promise but jocularity, or dishonesty.

There is no space in this paper to consider the contribution of conversation analysis (CA) to the understanding of such response patterns. Since the publication of Sacks, Scheg- loff and Jefferson’s paper in 1978 there has emerged a full field which has in certain respects superseded speech act theory. For interpretations of the organizational experience in this framework see, in particular, Drew and Heritage (1992) and Boden (1994). l o Cf. Searle (1969): “Human communication has some extraordinary properties, not shared by most other kinds of human behavior. One of the most extraordinary is this: If I am trying to tell someone something, then. . . . as soon as he recognizes that I am trying to tell him something and exactly what it is 1 am trying to tell him, I have succeeded in telling it to him. Furthermore, unless he recognizes that 1 am trying to tell him something and what I am trying to tell him, I d o not succeed in telling it to him. In the case of illocutionary acts we succeed in doing what we are trying to do by getting our audience to recognize what we are trying to do. But the ‘effect’ on the hearer is not a belief or response, it consists simply in the hearer understanding the utterance of the speaker” (p. 47). ‘ I We believe that a theory of communication which ignores context is not in fact really communicational. For us all communication is inherently organizational. To talk about communication in or within organization is thus not misguided because of what it affirms (that seems to us self-evidently true) but because of the negative implication it encourages, namely that there could ever be communication outside organization. This is the inherent problem of a messaging theory of communication. ’’ We use the terms variables and instantiation with some trepidation in that it may seem that they could be given an exact specification, in the form of constants. This is the approach taken in artificial intelligence. Our view is exactly the opposite: that such vari- ables are necessarily implicit, nonspecifiable, part of a general awareness- ineffable. Try- ing to nail down the unstated situational assumptions people bring to conversation pro- duces, not clarity, but an infinite regression, since new unstated presuppositions to frame the “clarifying” discourse are called into play. l 3 A more nuanced distinction is that developed by Cooren (1995) who discriminates between “rhetorical” and “arhetorical” acts of speech, the former referring to statements that Labov and Fanshel (1977) call “disputable” while the latter are purely conventional interpretations, such as calling a meeting to order.

Note that for something to be “uptaken” it would have had first to be “downput”- by the speaker, in the form of a locution. Another way to conceptualize this (Holt, 1988) is to think of textual material being sequentially consumed and produced in conversation. In the making of a locution, a small portion of accumulated experience and knowledge is used up to produce the locution, and in this sense is consumed. The uptake, by contrast, produces an addition to the accumulated store of knowledge. I s

Austin (1962) in fact dedicates several of his lectures to this issue.

3 3

Communication Theory

Cf. Bruner ( 1 991): “We organize our experience and o u r memory of human happenings mainly in the form of narrative-stories, excuses, myths, reasons for doing and not doing, and so on. Narrative is a conventional form, transmitted culturally and constrained by each individual’s level of mastery and by his conglomerate of prosthetic devices, colleagues, and mentors” ( p . 4). According to Polkinghorne ( 1 9 8 8 ) : “Narrative is the fundamental scheme for linking individual human actions and events into interrelated aspects of an understandable composite” ( p . 1 3 ) .

C f . Ellis (1992): “Interdeterminacy is why individuals must construct their own internal texts, and these texts vary with individual experience” (p. 1). ’ * We have been exploring a possible application of Peirce’s (1955) idea of “phaneroscopy” or “phenomenology” to this process. Peirce argues for three modes of being: possibility, actual fact and “law that will govern facts in the future.” W h a t we have been calling “intention” (to include thoughts and feelings) corresponds to w h a t he calls a “qualitative possibility” (since it is the meaning of what is expressed, but can never itself be actualized otherwise than by an overt act). H e calls thisfirstness. T h e act of speech is what he thinks of as secondness: a “happening” in the “then and there.” Actuality is, as he says, “something brute” ( p . 7 6 ) . Thirdness is what is general, of which the secondness is merely an instance. To interpret Peirce in this way requires, however, a transposition from the individual logic of scientific inquiry (his preoccupation) to that of interpersonal interaction, within which the “logic” is n o longer one of statistical inference, but of narrative. His thought is situated in a context of representation, ours in one of action. In this latter context it becomes evident that one person’s “secondness’ is a stimulus for another: a “firstness.” And how those in interaction come t o share a “thirdness” remains problematical (Robichaud, in progress).

Note that the exchange creates the virtuality of the action ( a t the point where Jones recognizes the thrust of the boss’s utterance), whether or not it ever gets realized. It is the creation of the virtuality of the action that is meant by an illocutionary effect, or force. T h e actualization of the action is w h a t is meant by the perlocutionary effect.

It is arguable that w h a t Greimas’s (1987) terms “modalization” is identical with Austin’s “illocutionary direction and force.” Consider a simple exchange: A. What time is it? B. Three o’clock. T h e giving of the time is the object of the exchange, a performance; the asking for the time is the modalization of the main action. This modalization corresponds to “asking a question” and instantiates the competence of the responder, t o the extent that such competence consists of willingness to carry o u t the action (of course, if he or she wasn’t wearing a watch, the modalization would be unsuccessful). Note again that all communication begins as a n adjacency pair. Such a pair consists of ( a ) a modalization (or instantiation of competence to perform) and ( b ) a performance. This is very evident in examples such a s that just cited, where direction is concerned. It is still true in more apparently ambiguous exchanges, such as making an assertion or expressing oneself, be- cause now the modalized end-state is not a performance but a belief or understanding (or that long-standing favorite of the social psychological laboratory, an “attitude”). T h e difference is between modalization of “doing” and of “being,” to use Greimas’s terms. Communication, in this reading, is invariably manipulative, precisely because of its ac- tional quality: Action is directed to attaining goals and when it takes the form of communi- cation, one person becomes the changer, the other the changee. This is where we part company with theorists such as Habermas (1979) and others whom he has influenced, w h o see strategic as distinct from authentic communication. For a slightly different view of directive and commissives, from a management and artificial intelligence (AI) perspective, see Winograd and Flores ( 1 9 8 7 ) . ’ I T h e “polemical” property of narrative (Greimas, 1970) is a neglected phenomenon in the interpretive literature o n organizational communication, which tends to see conflict as occasional rather than endemic. T h e Greimas view, based in narrative theory, is not so sanguine; it sees adversary relations as an inevitable outcome of the cognitive translation from interaction into understanding, since the semantic frames themselves engender as- sumptions of opposition. Opposition need n o t lead to conflict, especially within the con- straints of organized life ( i t may remain latent, in other words), but it at least supplies a motive. ” Cf. Austin ( 1 9 6 2 ) : “It is essential to realize that ‘true’ and ‘false,’ like ‘free’ and ‘unfree’, d o not stand for anything simple at all; but only for a general dimension of being a right or

2 0

The Communicational Basis of Organization

proper thing to say “as opposed to a wrong thing, in these circumstances, to this audience, for these purposes and with these intentions” (p. 144). 23 The discussion that follows is inspired by two sources: an article by Searle (1989) and a chapter from Durkheim’s (1912/1960) classic study of the origins of religion. We are particularly grateful to Rothenbuhler (1993a, 1993b) for directing our attention to this latter source. 24 We have described this elsewhere as an “accreditive” (Cooren, 1995). The effect of an accreditive speech act is to confer on someone the right to speak, typically by occupying a certain office with which rights and responsib ” How force can be accentuated and mitigated is taken up by Labov and Fanshel (1978). The degree of mitigation of illocutionary force seems to be a reliable indicator of relative status (Linde, 1988). 26 How do we determine what the community is? Is this not tautologous? We will return to this issue later.

Ricoeur (1986) described distanciation phenomenologically, from the perspective of hermeneutics; we are more interested in it as a social process, to produce, through a sequence of mediations, the embedded system of meanings that is one way of thinking about an organization. ’* The ideas expressed here are inspired, broadly, by the writings of Harold lnnis on the gulf dividing the “vernacuiar” from mediated forms of expression. See in particular Innis (1951).

Cf. Derrida ( 1 988): “This structural possibility of being weaned from the referent o r from the signified (hence from communication and its context) seems to me to make every mark, including those which are oral, a grapheme in general; which is to say, as we have seen, the nonpresent remainder [restance] of a differential mark cut off from its putative ‘production’ or origin. And I shall even extend this law to all ‘experience’ in general if it is conceded that there is no experience consisting of pure presence but only of chains of differential marks” (p. 10). ’O Cf. Ricoeur (1981): “What happens to reference when discourse becomes a text? Here we find that writing, and above all the structure of the work, modify reference to the point of rendering it entirely problematic. In oral discourse, the problem is ultimately resolved by the ostensive function of discourse; in other words, reference is determined by the ability to point to a reality common to the interlocutors. If we cannot point to the thing about which we speak, at least we can situate it in relation to the unique spatio-temporal network which is shared by the interlocutors. It is the ‘here’ and ‘now’, determined by the situation of discourse, which provides the ultimate reference of all discourse. With writing, things already begin to change. For there is no longer a situation common to the writer and the reader, and the concrete conditions of the act of pointing no longer exist. . . . The world of the text is therefore not the world of everyday language. In this sense, it constitutes a new sort of distanciation which could be called the distanciation of the real from itself” (pp. 141-142). 31 This section is inspired by the writings of the French school of the sociology of scientific knowledge, located at the Ecole des Mines in Paris. Latour’s (1987) Science in Action is a particularly witty and perspicacious treatment of how networking makes reality (what he calls “closing the black box”). 3 2 Cf. Durkheim (19121 1960): “It is society that speaks by the mouth of those who affirm them [what he calls ‘representations’ and we call ‘texts’] in o u r presence: It is society that we hear when we listen to them and the voice of all has a resonance that no single voice can have” (p. 297; our translation). 3 3 Latour (1987) observed that “a sentence may be more of a fact or more of an artefact depending on how it is inserted into other sentences. By i t s e l f a given sentence is neither a f a c t nor a fiction; it is made so b y others, later o n (italics added). . . . Listeners make sentences less of a fact if they take them back where they came from, to the mouths and hands of whoever made them, or more of a fact if they use them to reach another, more uncertain goal. The difference is as great as going up or down a river”(p. 25). For what we call degrees of separation Latour uses the term modalities. For him, positive modalities are “those sentences that lead a statement away from its conditions of production, making it solid enough to render some other consequences necessary.” Negative modalities “lead a statement in the other direction towards its conditions of production and that explains why

Communication Theory

i t is solid or weak instead of using it to render some other consequences more necessary.” The difference, he points out, is between sentences “devoid of any trace of ownership, construction, time and place” versus statements “uttered by someone situated in time and space.” 34 Our conception of network is thus radically different from the more conventional one; to us an organization is not a network of individuals but rather of conversations mediated by individually negotiated speech acts, involving agency. Is Cf. Schwartzman (1993): “Individuals . . . use meetings to read and/or see their place in particular social systems. We say that an individual is or is not a powerful person, but often we only ‘know’ this based on how we read and interpret events in a meeting. This was certainly the case for participants at Midwest, where there were very few ways outside of meetings for individuals to negotiate and/or determine their status and social ranking, and where their status was frequently in flux. In some cases it was only by astutely ‘reading’ meetings ( e g , who knows about, was/was not attending, calling/canceling, arriving or leaving, and so on) that an individual might learn about his or her place in the status system of the Center” (p. 41).

Cf. Searle (1989): “We ordinary humans d o not have the ability to perform supernatural declarations, but we do have a quasi-magical power nonetheless of bringing about changes in the world through our utterances; and we are given this power by a kind of human agreement. All of these institutions in question are social institutions, and it is only as long as the institution is recognized that it can continue to function to allow for the performance of declarations” (p. 549). ” In our analysis of an assistant dean’s statement (Taylor & Cooren, 1994), we remarked that the discourse was composed of an overlay of superimposed fields of discourse. The most immediate was the actual situation in which the statement occurs. Here, language was phrased in the present tense, with frequent use of pronouns: “We are fully ready to discuss with you. . . . ” Previous exchanges involving the same actors include nominaliz- ations, but there are still deictic references: “I am thinking about a theme that was used frequently in your own reports.” As the fields of discourse become more distant, further separated from the actual conversation, identifying markers of a person, time, and place disappear: the field of communication studies (who is studying whom or what and why is it a field?, we were tempted to ask), graduate studies, research projects, research centers, teaching and course development, and so forth. Each of these is of course a world of conversation in its own right, now reduced to object status. When we had finished the analysis we realized there was no reference in the statement to anything other than embed- ded conversational fields. We also realized that what is embedded is constitutive of the particular perspective on organization: Communication scholars study mass media, but mass media report on academic activities in their constructed world of sense (in French, significantly, the word researcher has two translations: chercheur, for a university re- searcher, and recherchiste, for a mass media researcher). Conversationally, the worlds overlap, without a clearly demarcated frontier; the boundary is in the construction of the discourse, with its us and them. The dean’s organization turned out not to be the depart- ment’s, for example. (For a discussion of strange loops-and tangled hierarchies- see Hofstadter, 1979).

Austin, J. (1962). H o w t o d o things with w o r d s . Oxford, England: Clarendon Press. Bach, K . , & Harnish, R. M. (1979). Linguistic communication and speech acts. Cam-

Barwise, J . , & Perry, J. (1983). Situations and attitudes. Cambridge, MA: MIT Press. Bateson, G. (1972). Steps t o an ecology o f m i n d . New York: Chandler. Bechler, C. (1994, July). Framing and shaping institutional responses t o crisis. Paper

presented at the annual meeting of the International Communication Association, Syd- ney, Australia.

Bierwisch, M. (1980). Semantic structure and illocutionary force. In J. R. Searle, F. Kiefer, & M. Bierwisch (Eds.), Speech act theory and pragmatics. Dordrecht, Holland: D. Reidel Publishing Company.

References bridge, MA: MIT Press.

Boden, D. (1994). The business of talk. Oxford, England: Polity Press. Bruner, J. (1991). The narrative construction of reality. Criticallnquiry, 18, 1-21. Calcagno, C . S. (1994, November). Describing the culture of a high-techfirm: A n assess-

The Communicational Basis of Organization

ment of values. Paper presented at the annual meeting of the Speech Communication Association, Miami, FL.

Callon, M., & Latour, B. ( 1 9 8 1 ) . Unscrewing the big Leviathan: How actors macro- structure reality and how sociologists help them to do so. In K. Knorr-Cetina & A. V. Cicourel (Eds.), Advances in social theory and methodology: Toward an integration of micro- and macro-sociologies. Boston: Routledge & Kegan Paul.

Cooren, F. ( 1 9 9 5 , May). A proposal f o r a transformational grid f o r speech acts: T h e problem of mediation. Paper presented at the 45th annual meeting of the International Communication Association, Albuquerque, NM.

Cooren, F. ( 1 9 9 5 ) Mediation and organization: A proposal f o r a transfornationalgrid f o r speech acts (Working paper). Montreal, Quebec, Canada: Universitt de Montreal, Department of Communication.

Deetz, S. ( 1 9 9 1 ) . Democracy in an age ofcorporate colonization: Developments in commu- nication and the politics of everyday life. Albany: State University of New York Press.

Derrida, J. ( 1 9 8 8 ) . Limited Inc. Evanston, IL: Northwestern University Press. Dewey, J. (1944).Democracy and education. New York: Free Press. (Original work pub-

lished 1916) Donnellon, A., Gray, B., & Bougon, M. G. ( 1 9 8 6 ) . Communication, meaning and orga-

nized action. Administrative Science Quarterly, 31,43-55. Drew, P., & Heritage, J. Talk at work: Interactron in institutional settings. Cambridge,

England: Cambridge University Press. Dumas, D. ( 1 9 9 5 ) . Intkgration planifike et intkgration vkcue. Unpublished master’s thesis.

Universitk de MontrCal, Montreal, Quebec, Canada. Durkheim E . ( 1 9 6 0 ) . Lesformes klkmentaires de la vie religieuse: le systkme totbmique en

Australie. Paris, Quadrige/Presses universitaires de France. (Original work published 1912)

Eisenberg, E. M. ( 1 9 8 4 ) . Ambiguity as strategy in organizational communication. Com- munication Monographs, 51,227-242.

Ellis, D. G . ( 1 9 9 2 ) . Syntactic and pragmatic codes in communication. Communication Theory, 2 ( 1 ) , 1-23.

Garfinkel, H. ( 1 9 6 7 ) . Studies in ethnomethodology. Englewood Cliffs, NJ: Prentice-Hall. Gilbert, G. N., & Mulkay, M. ( 1 9 8 4 ) . Opening Pandora’s box: A sociological analysis of

scientists’ discourse. Cambridge, England: Cambridge University Press. Giroux, N., & Cooren, F. ( 1 9 9 5 ) . La correction du discours strategique. Proceedings of

the Conference of the Association Internationale de Management Stratkgique, Paris. Giroux, N., & Fenocchi, V. (1994-1995). La participation, une realite complexe. Coopdr-

atives et Dkveloppement, 26 ( l ) , 59-81. Giroux, N., & Taylor, J. R. ( 1 9 9 3 ) . Le changement par la conversation strategique.

Proceedings of the Conference of the Association Internationale de Management Stra- tkgique, Lyon, France.

Goffman, E. ( 1 9 5 9 ) . Presentation o f s e l f i n everyday life. New York: Doubleday. Goffman, E. ( 1 9 7 4 ) . Frame analysis. New York: Harper & Row. Goffman, E. ( 1 9 8 1 ) . Forms of talk. Oxford, England: Basil Blackwell. Greimas, A. ( 1 9 7 0 ) . Du sens I . Paris: Seuil. Greimas, A. ( 1 9 8 7 ) . On meaning: Selected writings in semiotic theory. Minneapolis, MN:

University of Minnesota Press. Grice, H. P. ( 1 9 7 5 ) . Logic and conversation. In P. Cole, & J. L. Morgan (Eds.), Syntax

and semantics (Vol. 3 , pp. 86-123). Jordanstown, Northern Ireland: University of Ulster.

Habermas, J. ( 1 9 7 9 ) . Communication and the evolution of society, (T. MacCarthy, Trans.). Boston: Beacon Press.

Halliday, M. A. K., & Hasan, R. ( 1 9 8 9 ) . Language, context, and text: Aspects of language in a social-semiotic perspective. Oxford, England: Oxford University Press.

Hodge, R . , & Kress, G. ( 1 9 9 3 ) . Language as ideology (2nd ed.). London: Routledge. Hofstadter, D. ( 1 9 7 9 ) . Godel, Escher, Bach: A n eternalgolden braid. New York: Vintage. Holt, A. W. ( 1 9 8 8 ) . Diplans: A new language for the study and implementation of coordi-

Hulett, J. E., Jr. ( 1 9 6 6 ) . A symbolic interactionist model of human communication. AV

Innis, H. (1951). The bias of communication. Toronto, Ontario: University of Toronto Press.

nation. ACM Transactions o n Office Automation Systems, 6 ( 2 ) , 109-125.

Communication Review, 14,5-36,203-210.

Communication Theory

Johnson, N . S . , & Mandler, J. M . (1980). A tale of two structures: Underlying and surface forms in stories. Poetics, 9, 51-86.

Jolicoeur, A . (1994). La succession duns les entreprises familiales, un processus de commu- nication. Unpublished master’s thesis, Universite de Montreal, Montreal, Quebec, Canada.

Katambwe, J. M . (in progress). L’entreprise en transition, une analyse du discours manage- rial. Unpublished doctoral dissertation research, Universite de Montreal, Montreal, Quebec, Canada.

Keyton, J. (1994, November). Discord by design. Paper presented to the organizational communication division of the Speech Communication Association at the annual meet- ing of the Speech Communication Association, Miami, FL.

Khalil, M.-J. (1995). Le projet ISI, un point de vue en perspective. Unpublished master’s thesis, Universite de Montreal, Montreal, Quebec, Canada.

Labov, W., & Fanshel, D. (1977). Therapeutic discourse: Psychotherapy as conversation. New York: Academic Press.

Latour, B. (1987). Science in action. Cambridge, MA: Harvard University Press. Lawrence, P. R., & Lorsch, J. W . (1967). Organization and environment: Managing

differentiation and integration. Cambridge, MA: Harvard University Press. Linde, C . (1988). Who’s in charge here? Cooperative work and authority negotiation

in police helicopter missions. Proceedings o f the Conference on Computer-supported Cooperative W o r k . September 26-28, Portland, O R , 52-64.

Mandler, J . M . , & Johnson, N . S. (1977). Remembrance of things parsed: Story structure and recall. Cognitive Psychology, 9 , 111-151.

Minrzberg, H . (1979). The structuring of organizations: A synthesis o f t h e research. Engle- wood Cliffs, NJ: Prentice-Hall.

Morgan, G . (1986). Images of organization. Beverly Hills, CA: Sage. Polkinghorne, D. E. (1988). Narrative knowing and the human sciences. Albany: State

University of New York Press. Ricoeur, P. (1981). Hermeneutics and the human sciences: Essays o n language, action and

interpretation. In J. B. Thompson (Ed. and Trans.), New York: Cambridge University Press.

Ricoeur, P . (1986). From text t o action: Essays in hermeneutics 11. (K. Blarney & J. B . Thompson, Trans.). Evanston, IL: Northwestern University Press.

Robichaud, D. (in progress). La notion de la mkdiation et le problkme de la constitution de l’organisation. Unpublished doctoral dissertation research, Universite d e Montreal, Montreal, Quebec, Canada.

Rothenbuhler, E. W. (1993a). Argument for a Durkheimian theory of the communicative. lournu1 of Communication, 4 3 ( 3 ) , 158-163.

Rothenbuhler, E. W. (1993b). Reading Durkheim’s Elementary forms f o r communication theory. Paper presented at the annual meeting of the Speech Communication Associa- tion,

Sacks, H . , Schegloff, E. A., &Jefferson, G . (1978). A simplest systematics for the organi- zation of turn taking in conversation. In J . Schenkein (Ed.), Studies in the organization of conversational interaction (pp. 7-55). N e w York: Academic Press.

Schegloff, E. A. ( 1 9 9 2 ) . O n talk and its institutional occasions. In P. Drew & J. Heritage ( E d s . ) , Talk at work. (pp. 101-134). Cambridge, England: Cambridge University Press.

Schutz, A . (1970). On phenomenology and social relations. Chicago: University of Chi- cago Press.

Schwartzman, H. B. (1993). Ethnography in organizations. Newbury Park, CA: Sage. Searle, J . (1969). Speech acts: A n essay in the philosophy of language. Cambridge, Eng-

Searle, J . R . (1976). A classification of illocutionary acts. Language in Society, 5 , 1-23. Searle, J . R. ( 1 9 7 9 ) . Expression and meaning: Studres in the theory of speech acts. New

Searle, J. R. (1989). H o w performatives work. Linguistics and Philosophy, 22,535-558. Shannon, C . E . (1948). T h e mathematical theory of communication. Bell System Technical

land: Cambridge University Press.

York: Cambridge University Press.

]ournal, 27(10), 379-423,623-656.

The Communicational Basis of Organization

Smith, R. C. (1993). Images of organizational communication: Root-metaphors of the organization-communication relation. Unpublished. An earlier version of this paper was presented at the 1992 annual meeting of the International Communication Associa- tion, Washington, DC.

Stohl, C., & Redding, W. C. (1987). Messages and message exchange processes. In F. M. Jablin, L. L. Putnam, K. H. Roberts, & L. W. Porter (Eds.), Handbook of organiza- tional communication (pp. 451-502). Newbury Park, CA: Sage.

Taylor, J. R. (1993). Rethinking the theory of organizational communication: H o w to read an organization. Norwood, NJ: Ablex.

Taylor, J. R. (1995). Shifting from a heteronomous to an autonomous worldview of organizational communication: Communication theory on the cusp. Communication Theory, 5(1), 1-35.

Taylor, J. R., & Cooren, F. (1994, July). Reinstating the performatiue/constatiive distinc- tion: An application to the discourse analysis of organization. Paper presented at the Fifth International Conference on Language and Social Psychology, University of Queensland, Brisbane, Australia.

Taylor, J. R., & Giroux, N. (1993, July). The rhetoric of management: Conversation and text. Paper presented at the annual meeting of the European Group Organization Studies, Paris.

Taylor, J. R., & Van Every, E. J. (1993). T h e vulnerable fortress: Bureaucratic organiza- tion and management in the information society. Toronto, Ontario: University of Toronto Press.

Watzlawick, P., Beavin, J., & Jackson, D. (1967). T h e pragmatics of communication. New York: Norton.

Weick, K. (1979). The social psychology of organizing (2nd ed.). Reading, MA: Addison- Wesley.

Weick, K. (1985). Sources of order in underorganized systems: Themes in recent organiza- tional theory. In Y. S. Lincoln (Ed.), Organzzution theory and inquiry: The paradigm revolution (pp. 106-136). Beverly Hills, CA: Sage.

Werth, P. (1986). A functional approach to presupposition: Pulling the plug and holes and filters. In A. Bossuyt (Ed.), Functional explanations in linguistics. Brussels, Belgium: Presses Universitaires de Bruxelles, 239-279.

Werth, P. (1993). Accommodation and the myth of presupposition: The view from dis- course. Lingua, 89, 39-95.

Winograd, T., & Flores, F. (1987). Understanding computers and cognition. Reading, MA: Addison-Wesley.